Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

Systemic Elevation of PTEN Inducesa Tumor-Suppressive Metabolic StateIsabel Garcia-Cao,1 Min Sup Song,1,9 Robin M. Hobbs,1,7,9 Gaelle Laurent,2,9 Carlotta Giorgi,3,9 Vincent C.J. de Boer,2

Dimitrios Anastasiou,4 Keisuke Ito,1 Atsuo T. Sasaki,4 Lucia Rameh,5 Arkaitz Carracedo,1,8 Matthew G. Vander Heiden,6

Lewis C. Cantley,4 Paolo Pinton,3 Marcia C. Haigis,2 and Pier Paolo Pandolfi1,*1Cancer Genetics Program, Beth Israel Deaconess Cancer Center, Departments of Medicine and Pathology,

Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA 02215, USA2Department of Cell Biology, Paul F. Glenn Laboratories for the Biological Mechanisms of Aging, Harvard Medical School,

Boston, MA 02115, USA3Department of Experimental and Diagnostic Medicine, Section of General Pathology, Interdisciplinary Center for the Study of Inflammation

(ICSI) and LTTA center, University of Ferrara, 44100 Ferrara, Italy4Department of Systems Biology, Department of Medicine, Division of Signal Transduction, Beth Israel Deaconess Medical Center,

Harvard Medical School, Boston, MA 02115, USA5Boston Biomedical Research Institute, 64 Grove Street, Watertown, MA 02472, USA6Koch Institute for Integrative Cancer Research at MIT, Cambridge, MA 02139, USA7Present address: Australian Regenerative Medicine Institute and Monash Immunology and Stem Cell Laboratories, Monash University,

Clayton, VIC 3800, Australia8Present address: CIC bioGUNE, Technology Park of Bizkaia, 48160 Derio, Bizkaia, Spain and IKERBASQUE,Basque Foundation for Science, 48011 Bilbao, Bizkaia, Spain9These authors contributed equally to this work

*Correspondence: ppandolf@bidmc.harvard.edu

DOI 10.1016/j.cell.2012.02.030

SUMMARY

Decremental loss of PTEN results in cancer suscep-tibility and tumor progression. PTEN elevation mighttherefore be an attractive option for cancer preven-tion and therapy. We have generated several trans-genic mouse lines with PTEN expression elevatedto varying levels by taking advantage of bacterialartificial chromosome (BAC)-mediated transgenesis.The ‘‘Super-PTEN’’ mutants are viable and showreduced body size due to decreased cell number,with no effect on cell size. Unexpectedly, PTENelevation at the organism level results in healthymetabolism characterized by increased energyexpenditure and reduced body fat accumulation.Cells derived from thesemice show reduced glucoseand glutamine uptake and increased mitochondrialoxidative phosphorylation and are resistant to onco-genic transformation. Mechanistically we find thatPTEN elevation orchestrates this metabolic switchby regulating PI3K-dependent and -independentpathways and negatively impacting two of the mostpronounced metabolic features of tumor cells: gluta-minolysis and the Warburg effect.

INTRODUCTION

PTEN, a tumor suppressor frequently mutated or deleted in

human cancer, is a main negative regulator of the phosphoinosi-

tide 3-kinase (PI3K) signaling pathway by dephosphorylating

the 30 position of phosphatidylinositol-3,4,5-trisphosphate

(PIP3) (Maehama and Dixon, 1998). The PI3K pathway trans-

duces intracellular signals for growth, proliferation, and cell

survival (Leevers et al., 1999). Somatic inactivation of PTEN

occurs in a wide range of tumors, including glioblastoma, mela-

noma, prostate, and endometrial neoplasia (Bonneau and

Longy, 2000; Cantley and Neel, 1999; Simpson and Parsons,

2001). Furthermore, germline mutations of PTEN are the under-

lying genetic cause of three related multiple hamartoma disor-

ders: Cowden disease, characterized by an increased risk of

breast and thyroid cancers; Bannayan-Zonana syndrome; and

Proteus syndrome (Eng, 2003). Homozygous deletion of Pten

in mice results in embryonic lethality, and Pten heterozygous

mutant mice develop dysplasia in a wide spectrum of tissues

and have a high incidence of prostate and colon tumors (Di Cris-

tofano et al., 1998; Podsypanina et al., 1999; Suzuki et al., 1998).

It has been previously reported that PTEN dose is a key determi-

nant in prostate cancer progression (Trotman et al., 2003). Inter-

estingly, a more recent study shows that even a slight reduction

in PTEN levels dictates cancer susceptibility (Alimonti et al.,

2010). These studies highlight the crucial dose-dependent role

of PTEN in cancer progression.

Studies in Drosophila melanogaster reveal a novel role for

PTEN in the control of tissue growth (Gao et al., 2000; Goberdhan

et al., 1999; Huang et al., 1999). The phenotypes of flies carrying

mutations for various components of the PI3K-PKB/Akt pathway

have shown that this pathway positively controls cell number and

cell size (Bohni et al., 1999; Scanga et al., 2000; Verdu et al.,

1999; Weinkove et al., 1999). Consistent with its role as an

antagonist of this pathway, Drosophila PTEN (dPTEN) loss-of-

function mutants display increased cell and organ size, whereas

overexpression of dPTEN yields the opposite phenotype.

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. 1

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

Although consequences of gradual PTEN loss have been

extensively studied, the consequences and potential benefits

of elevating PTEN in the whole organism remain unknown. It

has been reported in Drosophila that ubiquitous overexpression

of PTEN results in lethality during embryonic and larval stages

(Gao et al., 2000; Huang et al., 1999). Importantly, the tumor

suppressor PTEN maintains cellular homeostasis through the

regulation of biological processes both in the cytoplasm and

within the nucleus (Salmena et al., 2008). However, it is currently

unknown whether fluctuations in PTEN dose would impact its

nuclear functions and/or its ability to modulate metabolic cues

at the organismal level. This information is critically needed, as

the elevation of PTEN is in principle an option for tumor preven-

tion and therapy.

Tumor cells have a metabolism that is remarkably different

from that in normal differentiated cells. Transformed cells take

up and metabolize nutrients such as glucose and glutamine at

high levels that support anabolic growth (Tong et al., 2009). In

contrast to normal differentiated cells that rely primarily on mito-

chondrial oxidative phosphorylation to generate energy needed

for cellular processes, most cancer cells instead rely on aerobic

glycolysis, a phenomenon termed ‘‘the Warburg effect’’ (War-

burg, 1956). Themetabolic alterations and adaptations of cancer

cells create a phenotype that is essential for tumor cell growth

and survival, altering flux along key metabolic pathways such

as glycolysis and glutaminolysis. Indeed there is mounting

evidence for the therapeutic potential of targeting cancer meta-

bolic reprogramming (Tennant et al., 2010).

Here we report that PTEN elevation is unexpectedly com-

patible with adult life and triggers a systemic metabolic regrog-

ramming that results in a healthy and tumor-suppressive

anti-Warburg state through the modulation of both PI3K-depen-

dent and -independent pathways.

RESULTS

Generation of Super-PTEN MiceTo elucidate the pathophysiological impact of PTEN elevation,

we attempted to generate transgenic (TG) mice carrying addi-

tional copies of this critical tumor suppressor gene (referred to

as Super-PTEN mice). To maintain the regulation properties of

the endogenous Pten gene, we made use of large genomic

fragments containing the entire Pten locus carried by bacterial

artificial chromosomes (BACs). These large genomic fragments

protect the gene of interest from chromatin positional effects,

preserving in every respect the pattern of expression of the

endogenous gene. A genomic insert containing Pten (Figure 1A)

was isolated from a mouse BAC genomic library. We obtained

different TG lines containing varying numbers of the entire Pten

locus. We next generated mouse embryonic fibroblasts (MEFs)

from these lines to determine the PTEN expression level, which

we found to vary from 1.1- to 3.5-fold above the endogenous

level (Figures 1B and 1C).

PTEN Regulates Mammalian Body Size by ControllingCell Number but Not Cell SizeSurprisingly, elevation of PTEN levels in the mouse was compat-

ible with life but resulted in reduced body weight and size,

2 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

a phenotype that was already evident during embryonic devel-

opment (Figure 1D). Increased PTEN levels were observed in

all adult tissues analyzed as well as during embryogenesis (Fig-

ure 1E) with no changes in its subcellular distribution and expres-

sion pattern (Figures 1F and 1G). All organs examined from TG

mice with high PTEN levels weighed below normal, whereas

the ratio of their weight to whole body weight was indistinguish-

able from that observed in wild-type (WT) individuals (Figure S1A

available online). Serum leptin, GH, and IGF-1 levels were normal

in Super-PTEN mice, therefore indicating that the effect of PTEN

on body size is unlikely to be mediated indirectly through effects

on hormone production (Figure S1B).

Interestingly, the effect on body weight and size (Figures 2A

and 2C) was more severe as PTEN levels increased, revealing

that PTEN controls body size in a dose-dependent manner in

a mammalian organism (Figures 2A and 2B).

A reduction in organ size can be the result of a decrease in cell

size, cell number, or both. Flow-cytometry analysis on dissoci-

ated cells from tissues (TG mice from line 3) revealed normal

cell size (as measured by forward scatter), whereas the total

number of cells was decreased (Figure 2D). Thus, PTEN eleva-

tion results in a reduced body size in mammals due to reduced

cell number, with no effect on cell size.

PTEN Elevation Results in Reduced Cell Proliferationand Decreased c-Myc Levels and Confers CancerResistancePTEN regulates a variety of biological processes to ensure

correct cell homeostasis, and alterations of these functions con-

tribute to cancer initiation and progression (Salmena et al., 2008).

In line with this notion and with our previous results, Super-PTEN

MEFs showed a significantly slower growth rate than their WT

counterparts (Figures 3A and 3B: growth curve and serial 3T3

cultivation). The ability of primary MEFs to form colonies when

seeded at low density is a reliable way to measure their prolifer-

ative potential. Indeed, Super-PTEN MEFs display a lower

plating efficiency than those derived from WT embryos

(Figure 3C).

We next examined the impact of PTEN overexpression on

oncogene-mediated cellular transformation. Employing classical

focus-formation assays with E1A and Ras oncogenes in Super-

PTEN andWTMEFs, we found the number of foci of morpholog-

ically transformed cells in Super-PTEN cells to be decreased

(Figure 3D). To validate this result in vivo, we evaluated the

susceptibility of Super-PTEN mice to develop tumors upon

chemical carcinogenesis (induction of fibrosarcomas by injec-

tion of 3MC, 3-methyl-cholantrene). As shown in Figure 3E, WT

mice started developing tumors at week 14 after 3MC injection,

whereas Super-PTENmice developed tumorswith a significantly

longer latency, starting at week 21. It has been observed that

3MC treatment induces the methylation and corresponding

loss of expression of tumor suppressor genes in rats (Liu et al.,

2010). However, PTEN expression ismaintained in 3MC-induced

tumors from both WT and Super-PTEN mice (Figure 3E; western

blot), suggesting that the tumor resistance phenotype of

Super-PTEN mice is due to an enhanced PTEN-dependent

tumor-suppressive effect rather than a reduced chance for

Pten inactivation by 3MC. These data demonstrate that it is

Figure 1. Generation of Super-PTEN Mice

(A) Map of the genomic insert carried by BAC RP23-215F15 clone (RPCI library, C57BL/6J) used to generate mice with increased gene dosage of PTEN

(‘‘Super-PTEN’’ mice). The genomic insert is 218.50 Kb long and contains the entire Pten locus (red boxes represent PTEN-coding sequence). Gray lines indicate

the BAC vector (pBACe3.6) sequence. AscI was used to linearize the BAC clone. Primers used for detection of the transgene are shown (T7 side: F1,R1; SP6

side: F2,R2).

(B) Quantification of PTEN levels in the different BAC-PTEN TG lines generated. MEFs were obtained from each line, and total protein lysates were probed with

antibodies toward PTEN and b-actin. The graph shows the fold increase in PTEN levels in each TG line relative to WT littermates.

(C) Representative immunoblotting ofMEFs derived from lineswith no expression (line 1), moderate expression (line 2), or high expression of the transgene (line 3).

(D) Representative images of WT and TG embryos harvested at 13.5 days post-coitum (dpc). Note reduced body size in TG embryo compared to WT.

(E)Western blot showing PTEN expression during embryogenesis and in tissues from adult mice (WT and TG from line 3). *In the case of heart, Ponceau S staining

is shown as loading control.

(F) Western blot showing PTEN levels in cytoplasmic (C) and nuclear (N) extracts from WT and TG MEFs (line 3).

(G) PTEN expression by immunohistochemistry in tissues from WT and TG mice (line 3).

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. 3

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

Figure 2. A Super-PTEN State Is Compatible with Life but Results in Decreased Organ Size due to Reduced Cell Number

(A) Growth curves indicate that PTEN overexpression results in reduced body mass. Note that the effect is more severe as PTEN dose increases.

(B) Western blot showing PTEN levels in different tissues (spleen, liver) from different PTEN TG lines (line 1: 1.2-fold; line 2: 2-fold; line 3: 3.5-fold above the

endogenous PTEN level).

(C) Effect of PTEN overexpression on body size (line 3).

(D) Cell number and cell size in organs isolated fromSuper-PTEN andWTmice (line 3, n = 3 per genotype). BM=bonemarrow. FSC= forward scatter. CD4/8 cells:

CD4+/Cd8+, KSL cells: Lin�/c-Kit+/Sca-1+, AT2 cells: Sca1�/CD45�/PECAM�/Autoflhi, B cells: B220mid+/IgMmid+.

Error bars in (A) and (D) denote standard deviation (SD). See also Figure S1.

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

possible to confer resistance to oncogenic transformation both

in vitro and in vivo by increasing PTEN levels.

The reduced organ weight, size, and cellularity observed in

Super-PTEN mice were reminiscent of the phenotype observed

in c-Myc hypomorphic mice (Trumpp et al., 2001). c-Myc plays

a key role in direct reprogramming of somatic cells into induced

4 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

pluripotent stem cells (iPSCs) (Banito and Gil, 2010). iPSC re-

programming is achieved by coexpressing pluripotency factors

and oncogenes, including c-Myc, and can provide a readout of

transformability and oncogenic potential (Banito and Gil, 2010),

whereas it has been recently reported that tumor suppressors

act as a barrier to reprogramming (Banito et al., 2009; Hong

Figure 3. PTEN Overexpression Results in Reduced Growth Rate in Culture, Resistance to Oncogenic Cellular Transformation, and

Decreased c-Myc Levels and Confers Cancer Resistance In Vivo

(A) Growth curves of WT and TG MEFs (n = 3 per genotype).

(B) Serial 3T3 cultivation of primary WT and TG MEFs (n = 3). The figure shows the accumulated population doublings (PDL) at each passage.

(C) Colony-formation efficiency of WT and TG MEFs (n = 4).

(D) Transformation susceptibility of WT and TG MEFs (n = 4). The picture shows the numbers of neoplastic foci formed after transfection with a combination of

oncogenic Ras and E1A.

(E) Susceptibility to 3MC-induced fibrosarcomas. Mice of the indicated genotypes were injected intramuscularly with 3MC in one of the rear legs, and the latency

for the development of fibrosarcomas was scored (line 3, n = 5 per genotype). Super-PTEN mice developed tumors with a significantly longer latency than WT

mice, as determined by Gehan-Breslow-Wilcoxon test (*p < 0.05). Western blot shows PTEN expression levels in tumors derived fromWT and Super-PTENmice.

(F) Soft-agar assay in primary MEFs transformed by E1A+Ras (top) or c-Myc+E1A+Ras (bottom). The graph shows the number of colonies per well (6-well plates,

triplicate samples). A representative picture from the assay is shown.

(G) Immunoblotting of protein lysates from WT and TG MEFs (top) and tissues (bottom, mice from line 3). Total cell extracts were probed with antibodies toward

c-Myc, PTEN and b-actin. WAT = white adipose tissue, BAT = brown adipose tissue.

Error bars in (A), (B), (C), (D), and (F) denote SD. See also Figure S2.

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. 5

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

et al., 2009; Kawamura et al., 2009; Li et al., 2009; Marion et al.,

2009; Utikal et al., 2009). Overexpression of PTEN in cells re-

sulted in a reduced number of iPSC colonies upon expression

of reprogramming factors Oct4, Sox2, and Klf4 (Figure S2A),

whereas, importantly, the negative effect of PTEN expression

in iPSC formation was overcome by c-Myc transduction. To

determine the involvement of c-Myc in the resistance of

Super-PTEN cells to oncogenic transformation, we performed

soft-agar assays to evaluate anchorage-independent growth

in cells infected with E1A-Ras alone or in the presence of

c-Myc (cMyc+E1A-Ras). As shown in Figure 3F (top), Super-

PTEN cells form fewer colonies in soft agar when compared to

WT cells. This resistance to transformation in Super-PTEN cells

is completely rescued by the addition of c-Myc (Figure 3F,

bottom). On the basis of these observations, we hypothesized

that PTEN opposes c-Myc function or suppresses its expres-

sion. Indeed, c-Myc expression was profoundly reduced in

various tissues both in Super-PTEN MEFs as well as in vivo

(Figure 3G).

Super-PTENMice Exhibit Increased Energy ExpenditureWe next examined the effect of a Super-PTEN state on body

fat accumulation and energy metabolism. Given the inhibitory

role of PTEN on signaling pathways critical for growth factors

and glucose uptake (Leevers et al., 1999), we anticipated that

Super-PTEN mice would exhibit reduced metabolic activity.

Surprisingly, Super-PTEN mice exhibited reduced body fat

accumulation compared to WT counterparts, as determined by

EchoMRI (Figure 4A), suggesting that PTEN elevation impacts

nutrient adaptation and utilization. Furthermore, indirect calorim-

etry analysis revealed that Super-PTEN mice presented higher

energy expenditure than WT mice (Figures 4B and S3A),

although locomotor activity was not significantly affected (Fig-

ure S3B). Food intake was similar or even slightly higher in

Super-PTEN mice (Figure S3C), indicating that the difference in

energy expenditure observed between WT and Super-PTEN

mice is not due to changes in food intake and rather due to an

elevated metabolic state.

FAO (fatty acid oxidation) represents a crucial process in

energy metabolism and fat storage (Ruderman and Flier, 2001).

Fatty acids can be either used for lipid synthesis and protein

modification or degraded through mitochondrial b-oxidation,

which produces substrates that maintain ATP generation

through oxidative phosphorylation. To determine a potential

contribution of altered FAO to the metabolic state of Super-

PTENmice and cells, we initially measured the expression levels

of key enzymes in fatty acid metabolism (SCD1, CPT-1a,

PPAR-a, PPAR-d, PDK4, MCAD, Acox1). However, we could

not find significant differences in the expression of these genes,

with the exception of Cpt-1a (Figure S3D), which is slightly

increased in Super-PTEN cells. Interestingly, PI3K signaling

reportedly suppresses Cpt-1a expression (Deberardinis et al.,

2006). We could not find significant differences in the rate of

FAO between WT and Super-PTEN cells, either primary MEFs

(Figure S3E) or hepatocytes (Figure S3F). We next examined

the use of glucose as a precursor for lipid synthesis and found

that Super-PTEN cells show a significant reduction in the contri-

bution of (6-14C)-glucose to lipid synthesis when compared to

6 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

WT cells (Figure S3G). This decreased rate of lipogenesis is

consistent with the reduction in body fat accumulation observed

in Super-PTEN mice. We next evaluated serum lactate levels as

an indicator of glycolytic activity in Super-PTEN mice. As shown

in Figure 4C, we observed a reduction in lactate levels in Super-

PTEN mice compared to WT littermates. Pyruvate generated

from glucose catabolism can be reduced to lactic acid (anaer-

obic glycolysis) or further metabolized by the mitochondria

(oxidative phosphorylation). These data suggested that the

increase in energy expenditure observed in Super-PTEN mice

could reflect increased mitochondrial oxidative phosphorylation

and a concomitant reduction in anaerobic glycolysis.

PTEN Elevation Shifts Cellular Energy Metabolismtoward Mitochondrial RespirationTo determine the contribution of mitochondrial respiration to

cellular bioenergetics, we measured oxygen consumption in

Super-PTEN and WT MEFs. The difference in energy expendi-

ture in vivo was consistent with the increased mitochondrial

oxygen consumption observed in primary Super-PTEN MEFs

(Figure 5A), which was accompanied by increasedmitochondrial

ATP production (Figure 5B) and the generation of reactive

oxygen species (ROS) (Figure S4A). It has previously been

shown that hypoxia-inducible factor 1 (HIF-1) stimulates glyco-

lytic energy production and negatively regulates mitochondrial

biogenesis and O2 consumption (Denko, 2008). Because activa-

tion of the PI3K-Akt pathway leads to increased HIF-1a levels

and activity (Zundel et al., 2000), we compared the levels of

HIF-1a in Super-PTEN versus WT MEFs. However, we found

no substantial differences in HIF-1a protein levels (Figure S4B)

or HIF-1a target genes (Figure S4C) between Super-PTEN and

WT MEFs.

The increased mitochondrial ATP production could not be

ascribed to a more efficient mitochondrial Ca2+ uptake (Griffiths

and Rutter, 2009) as this parameter was if anything slightly lower

in Super-PTEN MEFs (Figure 5C). Moreover, Super-PTEN and

WT MEFs showed similar mitochondrial membrane potential

(DJm) (Figure 5D). Strikingly, however, Super-PTEN MEFs ex-

hibited greater mitochondrial biogenesis. Indeed, total network

mitochondrial volume was higher in Super-PTEN MEFs due to

an increase in mitochondrial number, whereas average mito-

chondrial volume and morphology were unaffected (Figure 5E).

Microarray analysis revealed that Super-PTEN MEFs ex-

hibited a significant PGC-1a gene enrichment signature (Fig-

ure S4D), consistent with the fact that Akt suppresses PGC-1a

function (Li et al., 2007). PGC-1a is a key regulator of energy

metabolism that promotes mitochondrial oxidative phosphoryla-

tion and mitochondrial biogenesis (Puigserver and Spiegelman,

2003; Wu et al., 1999).

To functionally assess differences in mitochondrial activity, we

forced cells to rely on oxidative phosphorylation alone for energy

production by substituting glucose for galactose in the growth

media (Marroquin et al., 2007). In line with an increased mito-

chondrial function, Super-PTEN cells showed less of a reduction

in growth in galactose versus glucose compared to WT cells

(Figure S4E). Thus, through the generation of the Super-PTEN

mouse we have identified a role for PTEN in energy homeostasis

and mitochondrial biogenesis and function.

Figure 4. Super-PTEN Mice Show Increased Energy Expenditure

(A) Percentage of fat and leanmass was determined in young (left panel: WT, n = 7; TG, n = 7) and oldmice (right panel: WT, n = 12; TG, n = 10) by EchoMRI (line 3).

(B) Indirect calorimetry of Super-PTEN (red line) and WTmice (black line). Oxygen consumption (VO2), carbon dioxide release (VCO2), respiratory exchange ratio

(RER; VCO2/VO2), and energy expenditure per kg of body weight were determined in Super-PTEN and WT mice in metabolic chambers (line 3, n = 4 per

genotype).

(C) Serum lactate levels in WT (n = 10) and TG (n = 9) mice (line 3).

Error bars in (A) denote SD; error bars in (B) denote standard error of the mean (SEM). See also Figure S3.

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

The Super-PTEN State Is Accompanied by RepressedPI3K-Akt Signaling and Reduced Glucose Uptakein Spite of High Energy UtilizationWe then aimed to define the biochemical features of Super-

PTEN mice that are responsible for this unexpected metabolic

phenotype. PTEN is the main negative regulator of the PI3K

pathway, a highly oncogenic and metabolic node (Engelman

et al., 2006). Dephosphorylation of PIP3 by PTEN impairs Akt

activation and thereby opposes the PI3K-Akt signaling pathway.

Consistently, cells overexpressing PTEN show reduced levels

of the substrate PIP3 (Figure S5A), reduced PI3K activity (as

measured by phosphorylation of the downstream component

Akt) (Figure S5B), and impaired glucose uptake (Figure S5C).

The analysis of glucose and lactate metabolites in the extracel-

lular media revealed that Super-PTEN cells consume less

glucose and extrude less lactate into the media than WT cells

(Figure 6A). These data demonstrate that a Super-PTEN state

is coherently accompanied by repressed PI3K-Akt signaling

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. 7

Figure 5. PTENOverexpression Results in IncreasedMitochondrial OxygenConsumption,Mitochondrial ATP Production, andMitochondrial

Number

(A) OCRwasmeasured (Seahorse XF24 analyzer) in primaryWT and TGMEFs (n = 3 per genotype) under basal conditions and after addition of oligomycin, FCCP,

and Rotenone.

(B) Mitochondrial ATP production in WT and TGMEFs after agonist stimulation as described in the Experimental Procedures. Where indicated, cells were treated

with 100 mM ATP. Data are expressed as a percentage of the initial value. WT: 162% ± 8%; TG: 251% ± 18%. n = 15 from three independent experiments and

p < 0.05 (mean ± SEM).

(C) Mitochondrial Ca2+ homeostasis measurements after agonist stimulation as described in the Experimental Procedures. Where indicated, cells were treated

with 100 mM ATP. WT: [Ca2+]m peak 155 ± 8 mM. TG: [Ca2+]m peak 131 ± 5 mM. n = 12 from three independent experiments (mean ± SEM).

(D) Analysis of DJm in WT and TG MEFs. Cells where loaded with TMRM as described in the Experimental Procedures. Where indicated cells were treated with

FCCP to collapse completely the DJm. The traces are representative of single-cell responses (WT, n = 28; TG, n = 33).

(E) Analysis of total and single mitochondrial volume and mitochondrial numbers as described in the Experimental Procedures (WT, n = 43; TG, n = 40 from three

independent experiments and p < 0.05). Mitochondrial morphology in WT and TGMEFs was revealed bymitochondrial targeted GFP visualization. Mitochondrial

fragmentation index was calculated as described in the Experimental Procedures (WT, n = 43; TG, n = 40 from three independent experiments). Ordinates for the

graphs of network volume, average mitochondrial volume, and mitochondrial number are voxel/cell, voxel/object, and N� object/cell, respectively.Error bars in (A) denote SD; error bars in (D) and (E) denote SEM. See also Figure S4.

8 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

Figure 6. PTEN Elevation Induces a Tumor-Suppressive Metabolic State by Regulating PI3K-Dependent and -Independent Pathways

(A) Glucose, lactate, and glutamine levels in culture media were measured in WT and TG cells (n = 2) and normalized to cell number.

(B) Effects of pharmacological inhibition of mTORC1 and PI3K-Akt pathway on glycolytic and glutaminolytic enzymes. Cell lysates fromWT and TGMEFs treated

with DMSO (Cont.) or 20 nM rapamycin (Rapa.) for 24 hr or 100 nM wortmannin (Wort.) for 8 hr were subjected to immunoblotting with antibodies against PKM2,

PFKFB3, GLS, p-S6, S6, p-Akt, Akt, PTEN, and b-actin.

(C) Effects of TSC2 depletion-mediated mTORC1 activation on PKM2 protein level. Cell lysates from GL3-shRNA (control shRNA against the luciferase gene)

and Tsc2-shRNA infected WT and TG MEFs were subjected to immunoblotting with antibodies against PKM2, TSC2, PTEN, and b-actin.

(D) Fold change ratio (TG versus WT) in Pten, Pfkfb3, and Gls mRNA levels determined by qRT-PCR of total RNA from WT and TG MEFs (left). Proteasome-

mediated degradation of GLS and PFKFB3 is shown. Cell lysates from WT and TG MEFs treated with 10 mM of the proteasome inhibitor MG132 for 4 hr were

subjected to immunoblotting with antibodies against GLS, PFKFB3, PTEN, and b-actin (right).

(E) Impact of PFKFB3 overexpression in cell growth (48 hr after plating) in primary MEFs transformed by E1A+Ras. Percentage of growth relative to WT

MEFs+E1A+Ras is shown.

(F) PTEN enhances APC/C-Cdh1-mediated ubiquitination of GLS.PTEN-deficient PC-3 cells were cotransfectedwith GLS, His-Ub, CDH1, and PTEN and treated

withMG132 (10 mM) for 4 hr before harvesting. His-Ub-conjugatedGLSwas purified from cell lysates using anNi2+-NTA spin column under denaturing conditions.

(G)Cdh1 silencing recovers the levels of GLS and PFKFB3 in TGMEFs. Cell lysates fromWT and TGMEFs transfected with siRNAs for Renilla luciferase (siCont.)

or Cdh1 were subjected to immunoblotting with antibodies against GLS, PFKFB3, PTEN, Cdh1, and b-actin.

(H) Number of colonies (fold change ratio: TG versus WT MEFs+Ras) formed in soft-agar assay after Tsc2 or Cdh1 knockdown in immortalized TG MEFs

transformed by oncogenic Ras.

Error bars in (A), (D), (E), and (H) denote SD. See also Figure S5.

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. 9

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

and reduced glucose uptake. Although the prediction is that this

would result in low energy utilization, it was surprisingly not the

case in Super-PTEN cells and mice (see above).

PTEN Regulates PKM2 Levels through mTORC1One key characteristic of cancer cells is the way glucose is

utilized, switching from energy-efficient oxidative phosphoryla-

tion to inefficient lactate production (Warburg effect), allowing

for the accumulation of glycolysis intermediates and their redi-

rection to biosynthetic pathways. Our data show that Super-

PTEN cells take up less glucose, and yet they redirect a greater

fraction of glycolytic products into mitochondrial oxidative

phosphorylation. These features could be regarded as an

anti-Warburg status, which would correlate with decreased

proliferation and a cancer-resistant state. One of the central

components in the metabolic switch is the enzyme responsible

for converting phosphoenolpyruvate into pyruvate, pyruvate

kinase (PK). The switch from pyruvate kinase M 1 (PKM1)

expression to PKM2 in cells results in a lower rate of PK activity,

allowing the accumulation of glycolytic precursors for the

production of redox power and biomass (Vander Heiden et al.,

2009, 2010). We therefore studied the status of PKM2 in

Super-PTEN cells. Interestingly, Super-PTEN cells had reduced

levels of PKM2 (Figure 6B) and decreased PK activity (Fig-

ure S5D), suggesting that altered activity of this glycolytic regu-

latory node may contribute to the observed metabolic changes

in Super-PTEN cells.

As the PKM1/M2 isoforms are generated through alternative

splicing of two mutually exclusive exons (Christofk et al., 2008;

Clower et al., 2010; David et al., 2010), we next compared

splicing patterns of the Pkm gene in WT and Super-PTEN

MEFs. However, Pkm2 was the predominant mRNA isoform in

MEFs regardless of genotype (Figure S5E). This result was unan-

ticipated given that PKM2 splicing is under control of c-Myc

(David et al., 2010) and that we find Super-PTEN cells to express

lower levels of c-Myc (Figure 3G). Interestingly, in the iPSC re-

programming experiments, addition of c-Myc rescues PKM2

expression in Super-PTEN cells (Figure S2B). However, splicing

pattern analysis (Figure S2C) suggests that the c-Myc-depen-

dent rescue of PKM2 levels in Super-PTEN cells under iPSC re-

programming conditions is mediated through a mechanism

distinct to a switch in splicing patterns (see below).

In an effort to account for changes to PKM2 expression in

Super-PTEN cells, we investigated the potential involvement of

other PTEN-regulated cellular pathways in the control of PKM2

expression. Rapamycin treatment was found to reduce PKM2

levels in MEFs, suggesting that decreased activity of mTORC1,

a downstream target of the PI3K-Akt pathway (Zoncu et al.,

2011), could account for the reduced levels of PKM2 in Super-

PTEN cells (Figure 6B). To test this hypothesis, we analyzed

PKM2 expression in WT and Super-PTEN MEFs upon knock-

down of Tsc2, a negative regulator of mTORC1 (Tee and Blenis,

2005). Importantly, hyperactivation of the mTORC1 pathway by

Tsc2 knockdown rescued PKM2 levels in Super-PTEN MEFs

and further increased PKM2 levels in WT cells (Figure 6C), con-

firming that reduced mTORC1 activity in Super-PTEN cells

accounts for the decreased expression of PKM2. Interestingly,

Tsc2 knockdown also increased c-Myc expression in Super-

10 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

PTEN cells, although levels still remained substantially lower

than those of WT cells (Figure S5F). These data are consistent

with the described role of mTORC1 in regulating c-Myc transla-

tion (West et al., 1998). However, the inability of Tsc2 knockdown

to completely rescue c-Myc expression in Super-PTEN cells

indicates that additional pathways downstream of PTEN known

to regulate c-Myc levels are co-ordinately active (Gregory et al.,

2003; Sears et al., 2000). Conversely, the ability of c-Myc to

rescue PKM2 expression in Super-PTEN cells during iPSC re-

programming may be related to enhanced mTORC1 signaling

at multiple levels (Jones et al., 1996; Ravitz et al., 2007).

PTEN Controls both Glycolysis and Glutaminolysisthrough Regulation of PFKFB3 and GLS StabilityThe regulation ofmetabolism and growthmust be tightly coupled

to guarantee efficient use of energy and anabolic substrates

throughout the cell cycle. PFKFB3 (6-phosphofructo-2-kinase/

fructose-2,6-biphosphatase isoform 3) potently stimulates

glycolysis by catalyzing the formation of fructose 2,6-bisphos-

phate, an allosteric activator of 6-phosphofructo-1-kinase

(PFK-1), a rate-limiting enzyme and essential control point in

glycolysis. PFKFB3 has been shown to be abundantly expressed

in human tumors (Atsumi et al., 2002) and required for the high

glycolytic rate and anchorage-independent growth of Ras-trans-

formed cells (Telang et al., 2006). As shown in Figure 6B,

PFKFB3 levels are downregulated in Super-PTEN cells.

Cancer cells depend on a high rate of glucose uptake and

metabolism to maintain their viability despite being maintained

in an oxygen-replete environment. Another remarkable meta-

bolic feature of tumor cells is glutamine addiction. Interestingly,

Super-PTENcells showan impaired glutamine uptake (Figure 6A)

and reduced levels of glutaminase (GLS), the first enzyme in the

glutaminolysis pathway (Figure 6B). Therefore, PTEN elevation

negatively impacts both glycolysis and glutaminolysis through

the regulation of PFKFB3 and GLS.

Both GLS and PFKFB3 are substrates targeted for degrada-

tion by the E3 ubiquitin ligase anaphase-promoting complex/

cyclosome-Cdh1 (APC/C-Cdh1) (Colombo et al., 2010; Najafov

and Alessi, 2010), and PTEN has recently been demonstrated

to promote APC/C-Cdh1 activity (Song et al., 2011). This func-

tion of PTEN is dependent on its ability to directly promote

APC/C-Cdh1 complex assembly and independent of its ability

to inhibit the PI3K signaling pathway through dephosphorylation

of PIP3. Consistently, both PFKFB3 and GLS levels were

unaffected by treatment with the PI3K inhibitor wortmannin

(Figure 6B). We found that the reduction in PFKFB3 and GLS

levels was rescued by MG132 treatment, and no significant

changes in mRNA levels were found in cells overexpressing

PTEN compared to WT cells, indicating a proteasome-

dependent downregulation of PFKFB3 and GLS (Figure 6D).

Super-PTEN cells are resistant to transformation and show a

reduced growth rate in response to the oncogenic combination

E1A+Ras. Importantly, the addition of PFKFB3 is able to com-

pletely rescue the deficient growth of transformed Super-PTEN

cells (Figure 6E). An in vivo ubiquitination assay confirmed GLS

as a substrate for APC/C-Cdh1 and increased GLS ubiq-

uitination in the presence of PTEN (Figure 6F). PFKFB3 and

GLS levels were rescued by depletion of Cdh1 (Figure 6G),

Figure 7. PTEN Induces a Tumor-Suppressive Metabolic State by

Regulating PI3K-Dependent and -Independent Pathways

Model by which PTEN elevation negatively impacts two of the most noticeable

metabolic features of tumor cells: glutaminolysis and the Warburg effect.

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

indicating that PTEN negatively regulates PFKFB3 and GLS

through the APC/Cdh1 complex.

Reduced PKM2, PFKFB3, and GLS protein levels characterize

the tumor-suppressive metabolic state of Super-PTEN cells. As

discussed above, PKM2 levels are rescued by Tsc2 knockdown

(Figure 6C), whereas PFKFB3 and GLS levels are normalized by

Cdh1 inactivation (Figure 6G). As previously described (Li et al.,

2008; Zhang et al., 2003), Tsc2 or Cdh1 knockdown resulted in

a marked reduction in growth rate of primary MEFs (Figure S5G),

although this effect was more pronounced with WT than with

Super-PTEN cells in the case of Tsc2 inactivation. To assess

whether Tsc2 or Cdh1 inactivation could rescue the resistance

to oncogenic transformation of Super-PTEN cells, we studied

anchorage-independent growth of immortalized WT and

Super-PTEN MEFs after introduction of oncogenic Ras. Strik-

ingly, Tsc2 inactivation renders Super-PTEN cells as susceptible

to oncogenic transformation as WT cells, whereas Cdh1 inacti-

vation leads to a partial rescue in colony-forming ability in soft

agar (Figure 6H).

On the basis of these findings, we propose a model (Figure 7)

by which PTEN elevation induces a tumor-suppressive meta-

bolic state by regulating PI3K-dependent and -independent

pathways and negatively impacts two of the most noticeable

metabolic features of tumor cells: glutaminolysis and the

Warburg effect.

DISCUSSION

Our study reveals the profound impact of PTEN elevation in

organismal homeostasis and cancer development. Studies in

D.melanogasterhavehighlighted rolesof theAktpathway inposi-

tively controlling cell number and cell size (Bohni et al., 1999; Gao

et al., 2000; Goberdhan et al., 1999; Huang et al., 1999; Scanga

et al., 2000; Verdu et al., 1999; Weinkove et al., 1999). However,

our data show that the reduction in organism and tissue size in

Super-PTEN is due solely to a reduction in cell number, not cell

size. Interestingly, cell growth phenotypes of some Drosophila

mutations are not always paralleled by their mammalian counter-

parts. For instance, reduction of c-Mycexpression inmice results

in an overall decrease in body size due to a reduction in cell

number (hypoplasia) without detectable changes in cell size

(Trumpp et al., 2001), whereas Drosophila dmyc mutants are

smaller as a result of reduced cell size (hypotrophy) (Johnston

et al., 1999). Our Super-PTEN model displays an obvious

reduction of c-Myc expression as well, yet reduced cell number

but not cell size is the outcome at the organismal level. Neverthe-

less, the c-Myc downregulation is a distinctive component of the

Super-PTEN phenotype. Accordingly, c-Myc overexpression

rescues the defective three-factor iPSC reprogramming of

Super-PTEN MEFs. Further, c-Myc controls genes regulating

glucose metabolism and glutaminolysis, including PKM2 and

GLS, which are decreased in Super-PTEN cells (David et al.,

2010; Gao et al., 2009).

Tumor cells exhibit an altered metabolism that allows them

to sustain higher proliferative rates and resist cell death signals

(DeBerardinis et al., 2008; King and Gottlieb, 2009; Tennant

et al., 2009). Microarray analysis revealed that genes of the

glycolysis pathway are overexpressed in the majority of clinically

relevant cancers (Altenberg and Greulich, 2004). Among these

genes is PK, which regulates the rate-limiting final step of glycol-

ysis. Recent work demonstrated that expression of the type II

isoform of the Pkm gene is a critical determinant of the

metabolic phenotype of cancer cells and confers a selective

proliferative advantage to tumor cells in vivo (Christofk et al.,

2008). Our data show that Super-PTEN cells have reduced levels

of PKM2, and that this key regulator of glycolytic flux is under

the control of the mTORC1 pathway. Another key regulator of

glycolysis is PFKFB3, which potently stimulates glycolysis by

catalyzing the formation of fructose 2,6-bisphosphate, the allo-

steric activator of PFK1 (Hue and Rider, 1987), a rate-limiting

enzyme of glycolysis. It has been previously reported that

PFKFB3 is constitutively expressed by neoplastic cells and

serves as an essential downstream metabolic mediator of onco-

genic Ras (Telang et al., 2006). Importantly, we demonstrate

that PTEN elevation triggers PFKFB3 degradation through the

APC/Cdh1 complex. Super-PTEN cells show reduced protein

levels of PFKFB3, with no changes at the mRNA level. Beyond

the PTEN-dependent inhibition of glycolysis, we find that

Super-PTEN cells have reduced levels of GLS, the first enzyme

in glutaminolysis. Once again, this key tumor-suppressive meta-

bolic switch that opposes glutaminolysis is triggered by PTEN

elevation through GLS degradation by the APC/Cdh1 complex.

Interestingly, GLS has been reported to have increased activity

in several tumor types, and it is upregulated in c-Myc-trans-

formed cells (Gao et al., 2009; Wise et al., 2008), highlighting it

as a potential therapeutic target.

In summary, our data unexpectedly identify PTEN as a key

node for the control of obesity and tumorigenesis. Super-PTEN

mice exhibit an unexpected cancer-resistant and unique meta-

bolic state, which is the outcome of the ability of PTEN to regu-

late metabolism at multiple levels both from the cytosol and from

the nucleus. On the one hand, these mice exhibit increased

oxygen consumption and energy expenditure. On the other

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. 11

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

hand, Super-PTEN mice and cells are less prone to transforma-

tion and cancer development. Thus, the features observed in

Super-PTEN mice and cells resemble an ‘‘anti-Warburg state,’’

in which less glucose is taken up but is more efficiently directed

to the mitochondrial tricarboxylic acid cycle. Moreover, PTEN

elevations induce mitochondrial biogenesis and thus increased

mitochondrial ATP production. Thus PTEN couples mitochon-

drial function and dynamics to cellular metabolism through

PI3K-dependent and PI3K-independent mechanisms. It is

tempting to speculate that the unique metabolic state resulting

from PTEN elevation contributes to the increased cancer resis-

tance observed in these mice and cells by opposing cancer-

associated metabolic reprogramming (Figure 7). PTEN elevation

hence represents a potentially attractive therapeutic approach

that could both prevent cancer development and increase

energy expenditure to oppose fat accumulation and obesity.

EXPERIMENTAL PROCEDURES

Please refer to the Extended Experimental Procedures for more extensive

information.

BAC Transgenesis

For transgenesis, a large genomic insert (218.50 Kb) containing the entire

Pten locus and cloned into the BAC vector pBACe3.6 was isolated from

a mouse BAC genomic library (BAC RP23-215F15 clone; RPCI library,

C57BL/6J). After digestion with AscI, linearized BAC DNA was used for micro-

injection into the pronuclei of fertilized oocytes, derived from intercrosses

between (C57BL/6 3 CBA)F1 mice.

shRNAs

WT and Super-PTEN MEFs were infected with lentiviruses expressing shRNA

against mouse Cdh1 (clone ID TRCN0000027712, Open Biosystems) or rat

Tsc2 or corresponding control shRNA against luciferase (Di Nardo et al.,

2009). To prepare lentiviral particles, 3 3 106 293T cells were plated per

10 cm culture dish, and then shRNAs transfected with Lipofectamine 2000

(Invitrogen). For infection, MEFs were plated at a density of 6 3 105 cells per

10 cm culture dish and infected by virus from 293T cells 48 hr after transfec-

tion. After reseeding, MEFs were used for the different assays.

In Vivo Ubiquitination Assay

PTEN-deficient PC-3 cells were transfected with a combination of pCMV6-

XL4-GLS (SC114750: OriGene), HA-PTEN, Myc-CDH1, and His-ubiquitin

(Ub). His-Ub-conjugated GLS was purified from cell lysates using an Ni2+-NTA

spin column (QIAGEN) under denaturing conditions. The extent of ubiq-

uitination was then analyzed by immunoblotting.

Chemical Carcinogenesis

For 3MC-induced carcinogenesis, 4-month-old mice received a single intra-

muscular injection into one of their rear legs of a 100 ml solution containing

3-methyl-cholanthrene (Sigma), dissolved at a final concentration of 10 mg/ml

in sesame oil (Sigma) as previously described (Garcıa-Cao et al., 2002; Wexler

and Rosenberg, 1979).

Measurement of Oxygen Consumption

For oxygen consumption, 2 3 104 cells were plated, and 24 hr later, oxygen

consumption rate (OCR) was measured with the Seahorse XF24 instrument

(Seahorse Bioscience) under basal conditions and after addition of oligomycin

(1 mM), FCCP (2.5 mM), and Rotenone (1 mM). All the chemicals were purchased

from Sigma.

Measurement of Glucose, Lactate, and Glutamine

Glucose, lactate, and glutamine levels in culture media were measured using

the BioProfile FLEX analyzer (NOVA biomedical) and normalized to cell

12 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

number. Fresh media were added to a 6-well plate of cells and analyzed after

24 hr (triplicate samples).

Measurement of ROS Production

ROS production was measured by flow cytometric assessment of DCF

(20,70-dichlorodihydrofluorescein) fluorescence. The details of this assay are

described in the Extended Experimental Procedures.

FAO

FAO was determined by measuring 3H2O produced during cellular oxidation

of [3H] palmitate (Finck et al., 2006; Gerhart-Hines et al., 2007). Please refer

to the Extended Experimental Procedures for details.

Lipid Synthesis Assay

Lipid synthesis was assayed bymeasuring the amount of 14C incorporated into

lipids after 2 hr incubation with [6-14C]glucose (Hatzivassiliou et al., 2005).

Please refer to the Extended Experimental Procedures for details.

Statistical Analysis

In vitro and in vivo data were analyzedwith an unpaired t test (GraphPad Prism,

GraphPad Software Inc.). Values of p < 0.05 were considered statistically

significant (*p < 0.05; **p < 0.01; ***p < 0.001). Correlation analysis was per-

formed with SigmaPlot 5.0 software (SPSS Inc.).

ACCESSION NUMBERS

Microarray data have been deposited in MIAME under accession number

GSE35670.

SUPPLEMENTAL INFORMATION

Supplemental Information includes Extended Experimental Procedures and

five figures and can be found with this article online at doi:10.1016/j.cell.

2012.02.030.

ACKNOWLEDGMENTS

We thankM. Bhasin for help with microarray analysis, C. Clower for assistance

with PKM splicing assays, W.J. Haveman for mouse genotyping, and Kaitlyn

Webster for IHC analysis. I.G.-C. was supported by fellowships from the

Human Frontier Science Program (HFSP) and the Spanish Ministry of Educa-

tion and Science (MEC). This work was supported by NIH grants R01 CA-

82328-09 to P.P.P., P01-CA089021 and R01-GM41890 to L.C.C., and Italian

Association for Cancer Research (AIRC) grant to P.P.P.

Received: April 27, 2011

Revised: November 23, 2011

Accepted: February 7, 2012

Published online: March 6, 2012

REFERENCES

Alimonti, A., Carracedo, A., Clohessy, J.G., Trotman, L.C., Nardella, C., Egia,

A., Salmena, L., Sampieri, K., Haveman, W.J., Brogi, E., et al. (2010).

Subtle variations in Pten dose determine cancer susceptibility. Nat. Genet.

42, 454–458.

Altenberg, B., and Greulich, K.O. (2004). Genes of glycolysis are ubiquitously

overexpressed in 24 cancer classes. Genomics 84, 1014–1020.

Atsumi, T., Chesney, J., Metz, C., Leng, L., Donnelly, S., Makita, Z., Mitchell,

R., and Bucala, R. (2002). High expression of inducible 6-phosphofructo-2-

kinase/fructose-2,6-bisphosphatase (iPFK-2; PFKFB3) in human cancers.

Cancer Res. 62, 5881–5887.

Banito, A., Rashid, S.T., Acosta, J.C., Li, S., Pereira, C.F., Geti, I., Pinho, S.,

Silva, J.C., Azuara, V., Walsh, M., et al. (2009). Senescence impairs successful

reprogramming to pluripotent stem cells. Genes Dev. 23, 2134–2139.

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

Banito, A., and Gil, J. (2010). Induced pluripotent stem cells and senescence:

learning the biology to improve the technology. EMBO Rep. 11, 353–359.

Bohni, R., Riesgo-Escovar, J., Oldham, S., Brogiolo, W., Stocker, H., Andruss,

B.F., Beckingham, K., and Hafen, E. (1999). Autonomous control of cell and

organ size by CHICO, a Drosophila homolog of vertebrate IRS1-4. Cell 97,

865–875.

Bonneau, D., and Longy, M. (2000). Mutations of the human PTEN gene. Hum.

Mutat. 16, 109–122.

Cantley, L.C., and Neel, B.G. (1999). New insights into tumor suppression:

PTEN suppresses tumor formation by restraining the phosphoinositide

3-kinase/AKT pathway. Proc. Natl. Acad. Sci. USA 96, 4240–4245.

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

tumour growth. Nature 452, 230–233.

Clower, C.V., Chatterjee, D., Wang, Z., Cantley, L.C., Vander Heiden, M.G.,

and Krainer, A.R. (2010). The alternative splicing repressors hnRNP A1/A2

and PTB influence pyruvate kinase isoform expression and cell metabolism.

Proc. Natl. Acad. Sci. USA 107, 1894–1899.

Colombo, S.L., Palacios-Callender, M., Frakich, N., De Leon, J., Schmitt, C.A.,

Boorn, L., Davis, N., and Moncada, S. (2010). Anaphase-promoting complex/

cyclosome-Cdh1 coordinates glycolysis and glutaminolysis with transition to S

phase in human T lymphocytes. Proc. Natl. Acad. Sci. USA 107, 18868–18873.

David, C.J., Chen,M., Assanah,M., Canoll, P., andManley, J.L. (2010). HnRNP

proteins controlled by c-Myc deregulate pyruvate kinase mRNA splicing in

cancer. Nature 463, 364–368.

Deberardinis, R.J., Lum, J.J., and Thompson, C.B. (2006). Phosphatidylinosi-

tol 3-kinase-dependent modulation of carnitine palmitoyltransferase 1A

expression regulates lipid metabolism during hematopoietic cell growth.

J. Biol. Chem. 281, 37372–37380.

DeBerardinis, R.J., Lum, J.J., Hatzivassiliou, G., and Thompson, C.B. (2008).

The biology of cancer: metabolic reprogramming fuels cell growth and

proliferation. Cell Metab. 7, 11–20.

Denko, N.C. (2008). Hypoxia, HIF1 and glucose metabolism in the solid

tumour. Nat. Rev. Cancer 8, 705–713.

Di Cristofano, A., Pesce, B., Cordon-Cardo, C., and Pandolfi, P.P. (1998). Pten

is essential for embryonic development and tumour suppression. Nat. Genet.

19, 348–355.

Di Nardo, A., Kramvis, I., Cho, N., Sadowski, A., Meikle, L., Kwiatkowski, D.J.,

and Sahin,M. (2009). Tuberous sclerosis complex activity is required to control

neuronal stress responses in an mTOR-dependent manner. J. Neurosci. 29,

5926–5937.

Eng, C. (2003). PTEN: one gene, many syndromes. Hum. Mutat. 22, 183–198.

Engelman, J.A., Luo, J., and Cantley, L.C. (2006). The evolution of phosphati-

dylinositol 3-kinases as regulators of growth andmetabolism. Nat. Rev. Genet.

7, 606–619.

Finck, B.N., Gropler, M.C., Chen, Z., Leone, T.C., Croce, M.A., Harris, T.E.,

Lawrence, J.C., Jr., and Kelly, D.P. (2006). Lipin 1 is an inducible amplifier of

the hepatic PGC-1alpha/PPARalpha regulatory pathway. Cell Metab. 4,

199–210.

Gao, P., Tchernyshyov, I., Chang, T.C., Lee, Y.S., Kita, K., Ochi, T., Zeller, K.I.,

De Marzo, A.M., Van Eyk, J.E., Mendell, J.T., and Dang, C.V. (2009). c-Myc

suppression of miR-23a/b enhances mitochondrial glutaminase expression

and glutamine metabolism. Nature 458, 762–765.

Gao, X., Neufeld, T.P., and Pan, D. (2000). Drosophila PTEN regulates cell

growth and proliferation through PI3K-dependent and -independent path-

ways. Dev. Biol. 221, 404–418.

Garcıa-Cao, I., Garcıa-Cao, M., Martın-Caballero, J., Criado, L.M., Klatt, P.,

Flores, J.M., Weill, J.C., Blasco, M.A., and Serrano, M. (2002). ‘‘Super p53’’

mice exhibit enhanced DNA damage response, are tumor resistant and age

normally. EMBO J. 21, 6225–6235.

Gerhart-Hines, Z., Rodgers, J.T., Bare, O., Lerin, C., Kim, S.H., Mostoslavsky,

R., Alt, F.W., Wu, Z., and Puigserver, P. (2007). Metabolic control of muscle

mitochondrial function and fatty acid oxidation through SIRT1/PGC-1alpha.

EMBO J. 26, 1913–1923.

Goberdhan, D.C., Paricio, N., Goodman, E.C., Mlodzik, M., and Wilson, C.

(1999). Drosophila tumor suppressor PTEN controls cell size and number by

antagonizing the Chico/PI3-kinase signaling pathway. Genes Dev. 13, 3244–

3258.

Gregory, M.A., Qi, Y., and Hann, S.R. (2003). Phosphorylation by glycogen

synthase kinase-3 controls c-myc proteolysis and subnuclear localization. J.

Biol. Chem. 278, 51606–51612.

Griffiths, E.J., and Rutter, G.A. (2009). Mitochondrial calcium as a key regulator

of mitochondrial ATP production in mammalian cells. Biochim. Biophys. Acta

1787, 1324–1333.

Hatzivassiliou, G., Zhao, F., Bauer, D.E., Andreadis, C., Shaw, A.N., Dhanak,

D., Hingorani, S.R., Tuveson, D.A., and Thompson, C.B. (2005). ATP citrate

lyase inhibition can suppress tumor cell growth. Cancer Cell 8, 311–321.

Hong, H., Takahashi, K., Ichisaka, T., Aoi, T., Kanagawa, O., Nakagawa, M.,

Okita, K., and Yamanaka, S. (2009). Suppression of induced pluripotent

stem cell generation by the p53-p21 pathway. Nature 460, 1132–1135.

Huang, H., Potter, C.J., Tao, W., Li, D.M., Brogiolo, W., Hafen, E., Sun, H., and

Xu, T. (1999). PTEN affects cell size, cell proliferation and apoptosis during

Drosophila eye development. Development 126, 5365–5372.

Hue, L., and Rider, M.H. (1987). Role of fructose 2,6-bisphosphate in the

control of glycolysis in mammalian tissues. Biochem. J. 245, 313–324.

Johnston, L.A., Prober, D.A., Edgar, B.A., Eisenman, R.N., and Gallant, P.

(1999). Drosophila myc regulates cellular growth during development. Cell

98, 779–790.

Jones, R.M., Branda, J., Johnston, K.A., Polymenis, M., Gadd, M., Rustgi, A.,

Callanan, L., and Schmidt, E.V. (1996). An essential E box in the promoter of

the gene encoding the mRNA cap-binding protein (eukaryotic initiation factor

4E) is a target for activation by c-myc. Mol. Cell. Biol. 16, 4754–4764.

Kawamura, T., Suzuki, J., Wang, Y.V., Menendez, S., Morera, L.B., Raya, A.,

Wahl, G.M., and Izpisua Belmonte, J.C. (2009). Linking the p53 tumour

suppressor pathway to somatic cell reprogramming. Nature 460, 1140–1144.

King, A., and Gottlieb, E. (2009). Glucose metabolism and programmed cell

death: an evolutionary and mechanistic perspective. Curr. Opin. Cell Biol.

21, 885–893.

Leevers, S.J., Vanhaesebroeck, B., and Waterfield, M.D. (1999). Signalling

through phosphoinositide 3-kinases: the lipids take centre stage. Curr. Opin.

Cell Biol. 11, 219–225.

Li, H., Collado, M., Villasante, A., Strati, K., Ortega, S., Canamero, M., Blasco,

M.A., and Serrano, M. (2009). The Ink4/Arf locus is a barrier for iPS cell reprog-

ramming. Nature 460, 1136–1139.

Li, M., Shin, Y.H., Hou, L., Huang, X., Wei, Z., Klann, E., and Zhang, P. (2008).

The adaptor protein of the anaphase promoting complex Cdh1 is essential in

maintaining replicative lifespan and in learning and memory. Nat. Cell Biol. 10,

1083–1089.

Li, X., Monks, B., Ge, Q., and Birnbaum, M.J. (2007). Akt/PKB regulates

hepatic metabolism by directly inhibiting PGC-1alpha transcription coactiva-

tor. Nature 447, 1012–1016.

Liu,W.B., Ao, L., Zhou, Z.Y., Cui, Z.H., Zhou, Y.H., Yuan, X.Y., Xiang, Y.L., Cao,

J., and Liu, J.Y. (2010). CpG island hypermethylation of multiple tumor

suppressor genes associated with loss of their protein expression during rat

lung carcinogenesis induced by 3-methylcholanthrene and diethylnitrosamine.

Biochem. Biophys. Res. Commun. 402, 507–514.

Maehama, T., and Dixon, J.E. (1998). The tumor suppressor, PTEN/MMAC1,

dephosphorylates the lipid second messenger, phosphatidylinositol 3,4,5-tri-

sphosphate. J. Biol. Chem. 273, 13375–13378.

Marion, R.M., Strati, K., Li, H., Murga, M., Blanco, R., Ortega, S., Fernandez-

Capetillo, O., Serrano, M., and Blasco, M.A. (2009). A p53-mediated DNA

damage response limits reprogramming to ensure iPS cell genomic integrity.

Nature 460, 1149–1153.

Marroquin, L.D., Hynes, J., Dykens, J.A., Jamieson, J.D., and Will, Y. (2007).

Circumventing the Crabtree effect: replacing media glucose with galactose

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. 13

Please cite this article in press as: Garcia-Cao et al., Systemic Elevation of PTEN Induces a Tumor-Suppressive Metabolic State, Cell (2012),doi:10.1016/j.cell.2012.02.030

increases susceptibility of HepG2 cells to mitochondrial toxicants. Toxicol.

Sci. 97, 539–547.

Najafov, A., and Alessi, D.R. (2010). Uncoupling the Warburg effect from

cancer. Proc. Natl. Acad. Sci. USA 107, 19135–19136.

Podsypanina, K., Ellenson, L.H., Nemes, A., Gu, J., Tamura, M., Yamada,

K.M., Cordon-Cardo, C., Catoretti, G., Fisher, P.E., and Parsons, R. (1999).

Mutation of Pten/Mmac1 in mice causes neoplasia in multiple organ systems.

Proc. Natl. Acad. Sci. USA 96, 1563–1568.

Puigserver, P., and Spiegelman, B.M. (2003). Peroxisome proliferator-

activated receptor-gamma coactivator 1 alpha (PGC-1 alpha): transcriptional

coactivator and metabolic regulator. Endocr. Rev. 24, 78–90.

Ravitz, M.J., Chen, L., Lynch, M., and Schmidt, E.V. (2007). c-myc Repression

of TSC2 contributes to control of translation initiation and Myc-induced trans-

formation. Cancer Res. 67, 11209–11217.

Ruderman, N., and Flier, J.S. (2001). Cell biology. Chewing the fat—ACC and

energy balance. Science 291, 2558–2559.

Salmena, L., Carracedo, A., and Pandolfi, P.P. (2008). Tenets of PTEN tumor

suppression. Cell 133, 403–414.

Scanga, S.E., Ruel, L., Binari, R.C., Snow, B., Stambolic, V., Bouchard, D.,

Peters, M., Calvieri, B., Mak, T.W., Woodgett, J.R., and Manoukian, A.S.

(2000). The conserved PI30K/PTEN/Akt signaling pathway regulates both cell

size and survival in Drosophila. Oncogene 19, 3971–3977.

Sears, R., Nuckolls, F., Haura, E., Taya, Y., Tamai, K., and Nevins, J.R. (2000).

Multiple Ras-dependent phosphorylation pathways regulate Myc protein

stability. Genes Dev. 14, 2501–2514.

Simpson, L., and Parsons, R. (2001). PTEN: life as a tumor suppressor. Exp.

Cell Res. 264, 29–41.

Song, M.S., Carracedo, A., Salmena, L., Song, S.J., Egia, A., Malumbres, M.,

and Pandolfi, P.P. (2011). Nuclear PTEN regulates the APC-CDH1 tumor-

suppressive complex in a phosphatase-independent manner. Cell 144,

187–199.

Suzuki, A., de la Pompa, J.L., Stambolic, V., Elia, A.J., Sasaki, T., del Barco

Barrantes, I., Ho, A., Wakeham, A., Itie, A., Khoo, W., et al. (1998). High cancer

susceptibility and embryonic lethality associated with mutation of the PTEN

tumor suppressor gene in mice. Curr. Biol. 8, 1169–1178.

Tee, A.R., and Blenis, J. (2005). mTOR, translational control and human

disease. Semin. Cell Dev. Biol. 16, 29–37.

Telang, S., Yalcin, A., Clem, A.L., Bucala, R., Lane, A.N., Eaton, J.W., and

Chesney, J. (2006). Ras transformation requires metabolic control by 6-phos-

phofructo-2-kinase. Oncogene 25, 7225–7234.

Tennant, D.A., Duran, R.V., Boulahbel, H., and Gottlieb, E. (2009). Metabolic

transformation in cancer. Carcinogenesis 30, 1269–1280.

Tennant, D.A., Duran, R.V., and Gottlieb, E. (2010). Targeting metabolic

transformation for cancer therapy. Nat. Rev. Cancer 10, 267–277.

Tong, X., Zhao, F., and Thompson, C.B. (2009). The molecular determinants of

de novo nucleotide biosynthesis in cancer cells. Curr. Opin. Genet. Dev. 19,

32–37.

Trotman, L.C., Niki, M., Dotan, Z.A., Koutcher, J.A., Di Cristofano, A., Xiao, A.,

Khoo, A.S., Roy-Burman, P., Greenberg, N.M., Van Dyke, T., et al. (2003). Pten

dose dictates cancer progression in the prostate. PLoS Biol. 1, E59.

14 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

Trumpp, A., Refaeli, Y., Oskarsson, T., Gasser, S., Murphy, M., Martin, G.R.,

and Bishop, J.M. (2001). c-Myc regulates mammalian body size by controlling

cell number but not cell size. Nature 414, 768–773.

Utikal, J., Polo, J.M., Stadtfeld, M., Maherali, N., Kulalert, W., Walsh, R.M.,

Khalil, A., Rheinwald, J.G., and Hochedlinger, K. (2009). Immortalization

eliminates a roadblock during cellular reprogramming into iPS cells. Nature

460, 1145–1148.

Vander Heiden, M.G., Cantley, L.C., and Thompson, C.B. (2009). Under-

standing the Warburg effect: the metabolic requirements of cell proliferation.

Science 324, 1029–1033.

Vander Heiden, M.G., Locasale, J.W., Swanson, K.D., Sharfi, H., Heffron, G.J.,

Amador-Noguez, D., Christofk, H.R., Wagner, G., Rabinowitz, J.D., Asara,

J.M., and Cantley, L.C. (2010). Evidence for an alternative glycolytic pathway

in rapidly proliferating cells. Science 329, 1492–1499.

Verdu, J., Buratovich, M.A., Wilder, E.L., and Birnbaum, M.J. (1999).

Cell-autonomous regulation of cell and organ growth in Drosophila by

Akt/PKB. Nat. Cell Biol. 1, 500–506.

Warburg, O. (1956). On the origin of cancer cells. Science 123, 309–314.

Weinkove, D., Neufeld, T.P., Twardzik, T., Waterfield, M.D., and Leevers, S.J.

(1999). Regulation of imaginal disc cell size, cell number and organ size by

Drosophila class I(A) phosphoinositide 3-kinase and its adaptor. Curr. Biol.

9, 1019–1029.

West, M.J., Stoneley, M., and Willis, A.E. (1998). Translational induction of

the c-myc oncogene via activation of the FRAP/TOR signalling pathway.

Oncogene 17, 769–780.

Wexler, H., and Rosenberg, S.A. (1979). Pulmonary metastases from autoch-

thonous 3-methylcholanthrene-induced murine tumors. J. Natl. Cancer Inst.

63, 1393–1395.

Wise, D.R., DeBerardinis, R.J., Mancuso, A., Sayed, N., Zhang, X.Y., Pfeiffer,

H.K., Nissim, I., Daikhin, E., Yudkoff, M., McMahon, S.B., and Thompson, C.B.

(2008). Myc regulates a transcriptional program that stimulates mitochondrial

glutaminolysis and leads to glutamine addiction. Proc. Natl. Acad. Sci. USA

105, 18782–18787.

Wu, Z., Puigserver, P., Andersson, U., Zhang, C., Adelmant, G., Mootha, V.,

Troy, A., Cinti, S., Lowell, B., Scarpulla, R.C., and Spiegelman, B.M. (1999).

Mechanisms controlling mitochondrial biogenesis and respiration through

the thermogenic coactivator PGC-1. Cell 98, 115–124.

Zhang, H., Cicchetti, G., Onda, H., Koon, H.B., Asrican, K., Bajraszewski, N.,

Vazquez, F., Carpenter, C.L., and Kwiatkowski, D.J. (2003). Loss of Tsc1/

Tsc2 activatesmTOR and disrupts PI3K-Akt signaling through downregulation

of PDGFR. J. Clin. Invest. 112, 1223–1233.

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.

Zundel, W., Schindler, C., Haas-Kogan, D., Koong, A., Kaper, F., Chen, E.,

Gottschalk, A.R., Ryan, H.E., Johnson, R.S., Jefferson, A.B., et al. (2000).

Loss of PTEN facilitates HIF-1-mediated gene expression. Genes Dev. 14,

391–396.

Supplemental Information

EXTENDED EXPERIMENTAL PROCEDURES

Assays with MEFsMEFs were isolated from embryonic day (E) 13.5 embryos and cultured in DMEM medium supplemented with 10% FBS, penicillin/

streptomycin, and 2 mM L-glutamine using standard techniques (Todaro and Green, 1963). For serial 3T3 cultivation, 106 cells were

passaged into 10 cm dishes every 3 days. Growth curves were generated by seeding 0.25 3 105 cells per well in 12-well plates, in

triplicate. Plates were stained with crystal violet at each indicated time point. The dyewas extracted with 10%acetic acid followed by

plate reading at 590 nm. The values were normalized to the absorbance at day 0. For galactose experiments, cells were cultured in

glucose-free DMEM with 4.5 g/l galactose (Sigma) in addition to the above additives. For colony-formation assays, 104 cells were

plated in triplicate in 10 cm dishes. Two weeks later, plates were stained with Giemsa and the number of visible colonies was scored.

For transformation assays, MEFs were plated at a density of 106 cells per 10 cm dish and cotransfected with 10 mg of plasmid

expressing E1A (pCMV-E1A) and 10 mg of plasmid expressing oncogenic H-Ras (pAL8) by a standard calcium-phosphate procedure

(Graham and van der Eb, 1973). Plasmids pCMV-E1A and pAL8-H-Ras were generously provided by Mariano Barbacid

(CNIO, Spain). Foci were scored after 3 weeks by Giemsa staining. For insulin and serum stimulation, MEFs were starved (0.1%

FBS) overnight followed by stimulation with insulin (1.2 mg/ml, Sigma) or 10% FBS for different periods as indicated (5 min,

15min, 30min, 1 hr, and 3 hr after stimulation). For PFKFB3 overexpression, MEFs were infected with retrovirus containing the empty

vector, pLHCX, or PLHCX-PFKFB3 (generously provided by Bin Zheng, Columbia University, USA). For anchorage-independent

growth in soft agar, the bottom layer was obtained by covering 6-well dishes with 3 ml of 0.6% agar in complete medium. The

following day, 105 cells were plated on this bottom layer in triplicate in 2ml of 0.3% agar in complete medium. Colonies were counted

after 3–4 weeks under the microscope. All assays were performed from MEFs derived from line 3.

Generation of Mouse iPSCsReprogramming of primary MEFs was performed as previously described (Li et al., 2009; Marion et al., 2009). In brief, primary MEFs

were seeded in 6-well plates (0.53 105 cells per well). They were infected four times in the next 2 days with a cocktail of the retroviral

constructs pMXs-Klf4, pMXs-Sox2, pMXs-Oct4 (three factors), or with three factors plus pMXs-c-Myc (Addgene). After infection was

completed, medium was replaced by embryonic stem cell (ESC) medium composed of DMEM supplemented with 15% FBS, LIF

(1000 U/ml), L-glutamine, nonessential amino acids, and b-mercaptoethanol. Reprogramming was assessed 2 weeks after infection

by alkaline phosphatase staining (Millipore).

ImmunohistochemistryFor immunohistochemistry, mouse tissues were fixed in 4%paraformaldehyde overnight, subsequently washed twice with PBS, and

transferred into 70% ethanol. Tissues were embedded in paraffin, sectioned, and stained with anti-PTEN antibody (M3627: DAKO).

Immunoblot AnalysisFor immunoblotting, cells were lysed in RIPA buffer (50mMTris [pH 7.6], 150mMNaCl, 1%Nonidet P40, 0.5% sodiumdeoxycholate,

0.1% SDS, and protease inhibitor cocktail; Roche). Western blot analysis was carried out according to standard methods. The

following antibodies were used: anti-PTEN (#9559: Cell Signaling Technology), anti-phospho-serine 473 of Akt (#9271: Cell Signaling

Technology), anti-panAkt (#4691: Cell Signaling Technology), anti-b-actin (A2228: Sigma-Aldrich), anti-c-Myc (1472-1: Epitomics),

anti-PFKFB3 (AP8145b: Abgent), anti-GLS (12855-1-AP: ProteinTech Group), anti-phospho-serine235/236 of S6 (#2211: Cell

Signaling Technology), anti-S6 (#2217: Cell Signaling Technology), anti-Cdh1 (CC43: Calbiochem), anti-Lamin B (ab16048: Abcam),

anti-Hsp90 (H1775: Sigma-Aldrich), anti-TSC2 (sc-893: Santa Cruz Biotechnology), anti-HIF1a (NB100-479 Novus Biologicals).

Anti-PKM2 antibody was kindly provided by L.C. Cantley.

Nuclear/Cytoplasmic FractionationThe cell pellets of WT and TG MEFs were suspended in buffer A (10 mM HEPES [pH 7.9], 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA,

1 mM DTT, 1 mM PMSF, and protease inhibitor cocktail). Cells were swollen on ice for 15 min, after which 10% NP40 was added to

a final concentration of 0.5%. The homogenate was vortexed for 10 s and centrifuged for 5min at 4,000 rpm. The nuclear pellets were

lysed in RIPA buffer containing 1% SDS.

Metabolic StudiesActivity, food consumption, oxygen consumption rate (VO2), and respiratory exchange ratio (RER) weremeasured under a consistent

environmental temperature and light cycle using an indirect calorimetry system (TSE systems). Studies were started after 2 to 4 days

of acclimation to the metabolic chamber using an air flow of 1 l/min. VO2 was measured in individual mice at 15 min intervals during

a 48 hr period and normalized with respect to body weight. RER is the ratio of VCO2 to VO2, which changes depending on the energy

source the animal is using (usually around 1 for carbohydrates and 0.7 for fatty acids). Activity was measured on x, y, and z axes by

using infrared beams to count the beam breaks during a specified measurement period. Feeding is measured by recording the differ-

ence in the scale measurement of the center feeder from one time point to another. Fat and lean tissue masses were determined in

living, nonanesthetized mice by using a magnetic resonance imaging (MRI) machine (Echo Medical Systems, Houston, TX, USA).

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. S1

Serum lactate was measured using a Lactate Reagent Kit (Trinity Biosciences). Serum hormones were assayed via RIA (radioimmu-

noassay). Animal care and experiments were carried out in accordance with both institutional and federal animal care regulations and

were approved by the Harvard Medical Area Standing Committee on Animals.

Measurement of ROS ProductionCells were washed once with warm PBS and incubated with PBS containing 1 mM CM-H2DCFDA (Invitrogen) for 30 min at 37�C.Cells were then washed once with warm PBS, growth medium was replaced and cells were incubated for another 30 min at

37�C. Following trypsinization, cells were centrifuged, resuspended in cold PBS, and kept on ice until analysis by flow cytometry

within 30 min after harvesting.

Mitochondrial Luciferase MeasurementsMEFs grown on 13 mm round glass coverslips at 50% confluence were transfected with mitochondrial luciferase (mtLuc). Measure-

ments were performed 36 hr after transfection. Cell luminescence was measured in a luminometer as previously described (Jouaville

et al., 1999). Cells were constantly perfused with a modified KRB containing 20 mM luciferin.

Mitochondrial Aequorin MeasurementsMEFs grown on 13mm round glass coverslips at 50% confluence were transfected with the mitochondrial targeted chimera (mtAEQ)

as previously described (Pinton et al., 2007). All aequorin measurements were carried out in KRB (Krebs-Ringer modified buffer:

125 mM NaCl, 5 mM KCl, 1 mM Na3PO4, 1 mM MgSO4, 5.5 mM glucose, 20 mM HEPES [pH 7.4], 37�C) supplemented with

1 mM CaCl2. The experiments were terminated by lysing the cells with 100 mM digitonin in a hypotonic Ca2+-rich solution (10 mM

CaCl2 in H2O), thus discharging the remaining aequorin pool. The light signal was collected and calibrated into [Ca2+] values, as

previously described (Pinton et al., 2007).

Measurements of DJm

MEFs were seeded onto 24 mm diameter round glass coverslips and DJm was measured based on the accumulation of TMRM.

MEFs were loaded with 10 nM TMRM for 30 min, then analyzed on a confocal microscope (Nikon Swept field fast confocal system).

The signal was collected as total emission > 570 nm.

Microscopic Analysis of Mitochondrial Volume, Structure, and NumberMEFs were seeded on 24mm glass coverslip, and after 24 hr, the cells were infected with the adenovirus expressing a mitochondrial

targeted GFP. Reporter expression was allowed for 48 hr, then MEFs with labeled mitochondrial network were live imaged with the

Nikon Swept field fast confocal system. GFP were excited with a 488 nm solid state laser under 603magnification and aquired with

theDU885 Andor CCDCamera. To obtain a volumetricmeasurement z axis were acquired in 11 planeswith 0,6 mm for step. Collected

images were analyzes by the ImageJ software. Initially were subtracted of background, then volumetric object count were obtained

by the ‘‘Object count 3D’’ plugins. Morphometric measurements were obtained on the maximum intensity projection analyzed

morphometric plugins. For each object revealed we obtained the ‘‘roundness’’ and ‘‘aspect ratio’’ parameters, useful to quantify

circularity or elongation of an object respectively. We created a ratio of these parameters in order to obtain a ‘‘fragmentation index’’

that could be equally sensitive for the circular or elongated shape of a mitochondrion.

Phosphoinositide AnalysisMEFs were seeded at 1.3 3 106 cells per 10 cm dish. The following day, they were labeled with [32P]-inorganic phosphate (1 mCi/

10 cm dish) in 5% FCS phosphate-free DMEM for 4 hr. The cells were stimulated with 100 ng of PDGF per ml for 5 min. The reaction

was stopped with 1 ml of 1N HCl. Lipids were extracted with 1 ml of CHCl3/MeOH (1:1) and deacylated as described (Serunian et al.,

1991). Phosphatidylinositides were separated by anion-exchange high-performance liquid chromatography (Beckman), detected by

a flow scintillation analyzer (Perkin-Elmer), and quantified with ProFSA software (Perkin-Elmer) as described (Serunian et al., 1991).

The [32P]-PIP3 peak counts were normalized against the total phosphoinositide counts.

qRT-PCRRNA was isolated with the RNeasy Protect kit (QIAGEN) and included a DNase digestion step using the RNase-free DNase kit

(QIAGEN). cDNA was obtained with Transcriptor (Roche). TaqMan probes were obtained from Applied Biosystems Inc. Amplifica-

tions were run in a 7900 Real-Time PCR System (Applied Biosystems Inc.). Each value was adjusted using b-glucuronidase levels

as a reference.

Radioactive RT-PCR AssaySplicing patterns of the Pkm gene in WT and TG cells were analyzed essentially as previously described (Clower et al., 2010). Two

micrograms of total RNA was extracted from MEFs using Trizol reagent (Invitrogen). Contaminating DNA was removed by treatment

with DNase I (Invitrogen). Reverse transcription was carried out using ImPromp-II reverse transcriptase (Promega). Semiquantitative

S2 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

PCR using AmpliTaq polymerase (Applied Biosystems) was performed by including [a-32P]-dCTP in the reactions. The specific

primer sets anneal to exons 8 and 11, and their sequences are as follows:

mPKMF: 50ATGCTGGAGAGCATGATCAAGAAGCCACGC-30;mPKMR: 50CAACATCCATGGCCAAGTT-30.After 22 amplification cycles, the reactions were separated into four aliquots for digestion with NcoI, PstI (New England Biolabs),

both, or neither. The products were analyzed on a 5% native polyacrylamide gel and visualized by autoradiography.

Measurement of Glucose MetabolismCellular glucose metabolism rates were measured by following the conversion of 5-3H-glucose to 3H2O as previously described

(Vander Heiden et al., 2001). The assay was performed with cells attached to tissue culture plates. Briefly, the cells were washed

once in PBS, before incubation in Krebs buffer without glucose for 30 min at 37�C. The Krebs buffer was then replaced with Krebs

buffer containing 10mM glucose spiked with 10 mCi of 5-3H-glucose. After 1 hr, triplicate samples of media were transferred to PCR

tubes containing 0.2N HCl and the amount of 3H2O generated was determined by diffusion.

Measurement of PK ActivityPK activity was measured by a continuous assay coupled to lactate dehydrogenase (LDH). The change in absorbance at 340 nm

owing to oxidation of NADHwasmeasured using a Victor3 1420Multilabel Counter spectrophotometer (PerkinElmer). Kinetic assays

for activity determinations (triplicate samples) contained cell lysate (1–2mg), Tris (pH 7.5) (50 mM), KCl (100mM), MgCl2 (5 mM), ADP

(0.6 mM), PEP (0.5 mM), NADH (180 mM), FBP (10 mM), and LDH (8 units).

Hepatocyte IsolationPrimary hepatocytes were isolated using a two-step perfusion protocol, based on a previous method (Lin et al., 2004). Livers were

perfused first with Hanks balanced salt solution (HBSS, pH 7.4), containing glucose (1.0 g/l), EDTA (0.2 g/l), HCO3 (2.1 g/l), and KCl

(0.4 g/l) for 5 min. Next, livers were perfused for 15 min with a collagenase buffer (pH 7.4, Invitrogen). After perfusion, livers were

dissected, minced, filtered, and hepatocytes purified using Percoll (Sigma) and plated (500,000 cells per well) on collagen-coated

6-well plates (BD Biosciences) in DMEM (4.5 g/l glucose) containing 10% FBS, 2 mM pyruvate, 2% penicillin/streptomycin, 1 mM

dexamethasone, and 100 nM insulin. Two hours after plating, medium was replaced with maintenance medium (DMEM with

0.2% BSA, 2 mM pyruvate, 2% penicillin/streptomycin, 0.1 mM dexamethasone and 1 nM insulin). FAO assays were performed

1 day after isolation.

FAOCells were incubated overnight in culture medium containing 100 mM palmitate (C16:0) and 1 mM carnitine. In the final 2 hr of incu-

bation, cells were pulsed with 1.7 mCi [9,10(n)-3H] palmitic acid (GE Healthcare), and the medium was collected to analyze the

released 3H2O, formed during cellular oxidation of [3H] palmitate (Finck et al., 2006; Gerhart-Hines et al., 2007). In brief, medium

was TCA precipitated, and supernatants were neutralized with NaOH and loaded onto ion-exchange columns packed with DOWEX

1X2-400 resin (Sigma). The radioactive product was eluted with water and quantitated by liquid scintillation counting. Oxidation of

[3H] palmitate was normalized to protein content using Bio-Rad DC protein assay.

Lipid Synthesis AssayFormeasurement of glucose-dependent lipid synthesis (Hatzivassiliou et al., 2005), 105 cells per well were seeded in 12-well plates at

t = �24 hr. At t = �1 hr, the medium was changed to fresh growth medium. At t = 0, the medium was replaced with fresh growth

medium containing 4 mCi/ml [6-14C]glucose (final concentration = 72 mM). Cells were then incubated for 2 hr at 37�C, 5% CO2.

The labeling medium was then removed, cells were washed once with PBS at room temperature, and lipids were extracted twice

with 500 ml hexane:isopropanol (3:2 v/v), extracts were pooled and dried under nitrogen. Dried lipids were resuspended in 100 ml

chloroform, and radioactivity was measured in a scintillation counter.

Microarray AnalysisGene expression was assessed in WT and TG MEFs (n = 3 per genotype) with GeneChip HT MG-430 PM 24-Array Plate (Affymetrix,

#901257). Data were processed using the Affymetrix Gene Chip Operating System.

SUPPLEMENTAL REFERENCES

Graham, F.L., and van der Eb, A.J. (1973). A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52, 456–467.

Jouaville, L.S., Pinton, P., Bastianutto, C., Rutter, G.A., and Rizzuto, R. (1999). Regulation of mitochondrial ATP synthesis by calcium: evidence for a long-term

metabolic priming. Proc. Natl. Acad. Sci. USA 96, 13807–13812.

Lin, J., Wu, P.H., Tarr, P.T., Lindenberg, K.S., St-Pierre, J., Zhang, C.Y., Mootha, V.K., Jager, S., Vianna, C.R., Reznick, R.M., et al. (2004). Defects in adaptive

energy metabolism with CNS-linked hyperactivity in PGC-1alpha null mice. Cell 119, 121–135.

Pinton, P., Rimessi, A., Romagnoli, A., Prandini, A., and Rizzuto, R. (2007). Biosensors for the detection of calcium and pH. Methods Cell Biol. 80, 297–325.

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. S3

Serunian, L.A., Auger, K.R., and Cantley, L.C. (1991). Identification and quantification of polyphosphoinositides produced in response to platelet-derived growth

factor stimulation. Methods Enzymol. 198, 78–87.

Todaro, G.J., and Green, H. (1963). Quantitative studies of the growth of mouse embryo cells in culture and their development into established lines. J. Cell Biol.

17, 299–313.

Vander Heiden, M.G., Plas, D.R., Rathmell, J.C., Fox, C.J., Harris, M.H., and Thompson, C.B. (2001). Growth factors can influence cell growth and survival

through effects on glucose metabolism. Mol. Cell. Biol. 21, 5899–5912.

S4 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

Figure S1. Related to Figure 2(A) Ratio of each organ weight to whole body weight in WT and TG mice (n = 3 per genotype, line 3).

(B) Leptin (WT, n = 14; TG, n = 11), GH (WT, n = 10; TG, n = 7) and IGF-1 (WT, n = 10; TG, n = 9) serum levels determined by RIA (line 3).

Error bars denote SD.

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. S5

Figure S2. Related to Figure 3

(A) iPSC colony formation in WT and TG cells after expression of 3F and 3F+c-Myc. Representative image shows tissue culture wells with alkaline phosphatase-

stained iPSC colonies.

(B) Western blot showing PKM2 expression in WT and TG MEFs at indicated time points during iPSC induction protocol compared to mock-infected controls.

Three factors (3F) refers to infection with Oct4, Sox2, and Klf4, whereas four factors (4F) additionally includes infection with c-Myc.

(C) PKM splicing assay from WT and TG MEF at day 7 after infection with iPSC factors. A 502 bp fragment of the Pkm transcript was amplified by RT-PCR then

subject to restriction digest with the indicated enzymes. Pkm1 and Pkm2 isoforms are differentially cleaved by the enzymes to produce DNA fragments of the

indicated sizes.

S6 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

Figure S3. Related to Figure 4

(A) Energy expenditure measured in WT and TG males (line 3, n = 4 per genotype). The graph represents the average area under the curve (AUC) (dark and light

cycle). Similar results were obtained in an independent experiment.

(B) No changes in home-cage locomotor activity were observed between WT and TG mice (line 3, n = 4 per genotype). Similar results were obtained in an

independent experiment. Locomotor activity was studied by using an automated combined indirect calorimetry system.

(C) Food intake in WT and TG mice (line 3, n = 4 per genotype). The graph shows food intake (g/g body weight) in WT and TG mice during dark and light cycles.

(D) Fold change ratio (TG versus WT) in Scd1, Cpt-1a, Ppar-a, Ppar-d, Pdk4,Mcad, and Acox1mRNA levels determined by qRT-PCR of total RNA from primary

WT and TG MEFs (n = 3).

(E) FAO measured in WT and TG primary MEFs (n = 2). FAO was determined by measuring 3H2O produced by WT and TG primary MEFs incubated with

[9,10(n)-3H] palmitate for 2 hr. Eto indicates the addition of the FAO inhibitor etoxomir.

(F) FAOmeasured in primary hepatocytes derived fromWT and TGmice (line 3, n = 2). FAOwas determined bymeasuring 3H2O produced byWT and TG primary

hepatocytes incubated with [9,10(n)-3H] palmitate for 2 hr.

(G) Contribution of glucose to lipid synthesis in primary WT and TG MEFs. Lipid synthesis was assayed by measuring the amount of 14C incorporated into lipids

after 2 hr incubation with [6-14C]glucose.

Error bars in (A), (C), (D), (E), (F), and (G) denote SD. Error bars in (B) denote SEM.

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. S7

Figure S4. Related to Figure 5

(A) ROS production was measured by flow cytometric assessment of DCF (20,70-dichlorodihydrofluorescein) fluorescence (AFU = arbitrary fluorescence units) in

primary WT and TG MEFS (n = 2 per genotype).

(B) HIF-1a levels in WT and TG MEFs in steady-state culture conditions (Cont.), after treatment with the proteasome inhibitor MG132 or in response to the

hypoxia-mimetic CoCl2.(C) Fold change ratio (TG versusWT) inHk2, Pfkm, Pfkl, Pdk1, and LdhamRNA levels determined by qRT-PCR of total RNA from primaryWT and TGMEFs (n = 3).

(D) PGC-1a gene enrichment signature in TG versus WT MEFs by microarray (n = 3 per genotype); FDR < 4%.

(E) Percentage of growth (galactose versus glucose) in primary WT and TG MEFs (n = 2).

Error bars in (A), (C), and (E) denote SD.

S8 Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc.

Figure S5. Related to Figure 6

(A) PIP3 detection in WT and TG MEFs, serum starved, and stimulated with 100 ng insulin for 5 min. Phosphatidylinositides were separated by anion-exchange

high-performance liquid chromatography.

(B) Phospho-Akt(S473) levels in WT and TG MEFs in basal conditions and after serum (top panel) or insulin (bottom panel) stimulation. MEFs were starved

(0.1% FBS, 16 hr) and then stimulated with 10% FBS or insulin (1.2 mg/ml) for the indicated period of time.

(C) Glucose metabolism rates of primary WT and TG MEFs.

(D) PK activity (in the presence of FBP) in lysates from primary WT and TG MEFs (n = 2 per genotype).

(E) PKM splicing assay from WT and TG MEF under steady-state growth conditions. A 502 bp fragment of the Pkm transcript was amplified by RT-PCR then

subjected to restriction digest with the indicated enzymes. Pkm1 and Pkm2 isoforms are differentially cleaved by the enzymes to produce DNA fragments of the

indicated sizes.

(F) Effects of TSC2 depletion-mediated mTORC1 activation on c-Myc protein level. Cell lysates from GL3-shRNA and Tsc2-shRNA infected WT and TG MEFs

were analyzed by western blot using the indicated antibodies. Quantification of c-Myc levels corrected to the Hsp90 loading control and normalized to GL3-

shRNA WT cells is indicated. Phospho-ribosomal S6 protein is used as readout of mTORC1 activity.

(G) Growth curves of sh-Control (black line), sh-Tsc2 (green line), and sh-Cdh1 (orange line) infected WT (filled circles) and TG (open circles) MEFs. Western blot

shows expression levels of the indicated proteins after infection (day 4).

Error bars in (A), (C), (D), and (G) denote SD.

Cell 149, 1–14, March 30, 2012 ª2012 Elsevier Inc. S9