Discrete mechanisms of mTOR and cell cycle regulationby AMPK agonists independent of AMPKXiaona Liua, Rishi Raj Chhipaa, Shabnam Pooyaa, Matthew Wortmanb, Sara Yachyshina, Lionel M. L. Chowa,Ashish Kumarc, Xuan Zhouc, Ying Sund, Brian Quinnd, Christopher McPhersone, Ronald E. Warnicke, Ady Kendlerf,Shailendra Girig, Jeroen Poelsh, Koenraad Norgai, Benoit Violletj,k,l, Gregory A. Grabowskid, and Biplab Dasguptaa,1

aDepartments of Oncology, cExperimental Hematology and Cancer Biology, and dHuman Genetics, Cincinnati Children’s Hospital Medical Center, Cincinnati,OH 45242; bDepartments of Internal Medicine and fPathology and Laboratory Medicine, University of Cincinnati, Cincinnati, OH 45229; eDepartment ofNeurosurgery, Brain Tumor Center, University of Cincinnati Neuroscience Institute and Mayfield Clinic, Cincinnati, OH 45229; gDepartment of LaboratoryMedicine/Pathology, Mayo Clinic, Rochester, MN 55902; hDepartment of Animal Physiology and Neurobiology, Katholieke Universiteit Leuven, 3000 Leuven,Belgium; iPaediatric Oncology, Antwerp University Hospital, 2000 Antwerp, Belgium; jInstitut National de la Santé et de la Recherche Médicale, Unité 1016,Institut Cochin, 75014 Paris, France; kCentre National de la Recherche Scientifique, Unité Mixte de Recherche 8104, 75014 Paris, France; and lUniversitéParis Descartes, 75270 Paris, France

Edited by Gregg L. Semenza, The Johns Hopkins University School of Medicine, Baltimore, MD, and approved December 4, 2013 (received for reviewJune 13, 2013)

The multifunctional AMPK-activated protein kinase (AMPK) is anevolutionarily conserved energy sensor that plays an importantrole in cell proliferation, growth, and survival. It remains unclearwhether AMPK functions as a tumor suppressor or a contextualoncogene. This is because although on one hand active AMPKinhibits mammalian target of rapamycin (mTOR) and lipogenesis—two crucial arms of cancer growth—AMPK also ensures viability bymetabolic reprogramming in cancer cells. AMPK activation by twoindirect AMPK agonists AICAR and metformin (now in over 50clinical trials on cancer) has been correlated with reduced cancercell proliferation and viability. Surprisingly, we found that com-pared with normal tissue, AMPK is constitutively activated in bothhuman and mouse gliomas. Therefore, we questioned whetherthe antiproliferative actions of AICAR and metformin are AMPKindependent. Both AMPK agonists inhibited proliferation, butthrough unique AMPK-independent mechanisms and both re-duced tumor growth in vivo independent of AMPK. Importantly,A769662, a direct AMPK activator, had no effect on proliferation,uncoupling high AMPK activity from inhibition of proliferation.Metformin directly inhibited mTOR by enhancing PRAS40’s associ-ation with RAPTOR, whereas AICAR blocked the cell cycle throughproteasomal degradation of the G2M phosphatase cdc25c. To-gether, our results suggest that although AICAR and metforminare potent AMPK-independent antiproliferative agents, physiolog-ical AMPK activation in glioma may be a response mechanism tometabolic stress and anticancer agents.

metabolism | glioma

AMP-activated protein kinase (AMPK) is a molecular hub forcellular metabolic control (1–4). It is a heterotrimer of

catalytic α, regulatory β, and γ subunits. The rising AMP:ATPratio during energy stress leads to AMP-dependent phosphory-lation of the catalytic α subunits. This activates AMPK whichthen phosphorylates numerous substrates to restore energy ho-meostasis. It phosphorylates acetyl CoA carboxylase (ACCα) toinhibit fatty acid (FA) synthesis (5) and TSC2 and RAPTOR(6, 7) to inhibit mammalian target of rapamycin (mTOR)C1.Because fatty acid synthesis and mTORC1 activity are essentialfor cell proliferation and growth (8), AMPK activation with twoindirect AMPK agonists AICAR and metformin have been cor-related with suppression of cell proliferation and growth (9–11).AICAR is metabolized to an AMP mimetic, ZMP that acti-

vates AMPK (12). Although AICAR does inhibit proliferation(11–15), it also causes AMPK-independent cellular and meta-bolic effects (12, 16) including inhibition of glucokinase, glyco-gen phosphorylase, and nucleotide biosynthesis (17, 18). WhetherAICAR requires AMPK to suppress proliferation is question-able because although both AICAR and 2-deoxyglucose activated

AMPK, only AICAR inhibited proliferation of trisomic mousefibroblasts (11). Moreover, although AICAR strongly increasesglucose uptake through AMPK activation in muscle cells, it re-duced fluorodeoxyglucose-PET signals and inhibited gliomagrowth in vivo (9), suggesting that reduced PET signals could bedue to its AMPK-independent antiglioma action.The antiproliferative mechanisms of metformin also remain

unclear. It is argued that because metformin inhibits mitochon-drial respiration (19), it induces an energy crisis (metabolicstress), leading to AMPK activation, mTOR inhibition, andsuppression of proliferation (20). However, Dykens et al. (21)showed that net cellular ATP is not affected by metformin.Other suggested mechanisms include disruption of cross-talkbetween GPCRs and insulin receptors (22), inhibition of theErbB2/IGF1 receptor (23), and mTOR inhibition by blockingRAG function (24). In vivo, metformin and the direct AMPKagonist A769662 delayed onset but not progression of lymphomain Pten+/−;LKB1+/− mice (25) (LKB1 is the upstream kinase thatactivates AMPK). Moreover, these experiments were not con-ducted on AMPK-deficient animals, making it unclear whetherthe drug effects were AMPK dependent. Contrary to these re-sults, metformin prevented tumorigenesis without activatingAMPK in lung tumors (26), and in fact, LKB1-deficient lungtumors were actually more responsive to the metformin analogphenformin (27). The latter results suggest that the LKB1–AMPK

Significance

Cancer cells reprogram their metabolism for optimal growthand survival. AMPK-activated protein kinase (AMPK) is a keyenergy sensor that controls many metabolic pathways includingmetabolic reprogramming. However, its role in cancer is poorlyunderstood. Some studies claim that it has a tumor suppressorrole while others show its protumor role. Two AMPK-activatingcompounds (includingmetformin, now in many clinical trials) arewidely used to suppress cancer cell proliferation. We found thatAMPK is abundantly expressed in high-grade gliomas and, incontrast to popular belief, these two AMPK activators suppressedglioma cell proliferation through unique AMPK-independentmechanisms.

Author contributions: B.D. designed research; X.L., R.R.C., S.P., M.W., S.Y., A. Kumar, X.Z.,B.Q., S.G., and J.P. performed research; L.M.L.C., C.M., R.E.W., A. Kendler, B.V., and G.A.G.contributed new reagents/analytic tools; M.W., Y.S., A. Kendler, K.N., and B.D. analyzeddata; and B.D. wrote the paper.

The authors declare no conflict of interest.

This article is a PNAS Direct Submission.1To whom correspondence should be addressed. E-mail: Biplab.DasGupta@cchmc.org.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1311121111/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1311121111 PNAS | Published online January 13, 2014 | E435–E444

BIOCH

EMISTR

YPN

ASPL

US

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

pathway protects cancer cells from antiproliferative agents andmay support tumorigenesis.In line with the above idea, genetic studies showed a procancer

role of AMPK in the in vivo growth of H-RAS–transformedfibroblasts and astrocytic tumors, in pancreatic cancer, and ina subtype of renal cell carcinoma (28–31). Additional geneticstudies also underscore the requirement of AMPK in cancer cellmetabolic programming (32, 33); cell division (34–37); migration(38), protection against stress; and anticancer therapy (39–41).However, in Myc-driven mouse lymphoma, AMPK was shownto function as a tumor suppressor (42), suggesting a context-dependent role of AMPK in cancer.To definitively determine whether AMPK is necessary for the

antiproliferative actions of AICAR and metformin, we conducteda comprehensive pharmacogenetic study in glioma. First, we foundthat gliomas express constitutively active AMPK, and thatAICAR and metformin inhibit proliferation by distinct AMPK-independent and unique mechanisms. Second, A769662, a directAMPK activator (43) showed no antiproliferative effects. Therefore,

many agents that inhibit proliferation with concomitant AMPKactivation may not require AMPK for their action. Instead,AMPK activation could be a response mechanism to counterstress induced by anticancer agents.

ResultsActive AMPK Is Abundantly Expressed in High-Grade Gliomas. Met-abolic stress in solid tumors like gliomas poses a formidablechallenge for tumor cell survival. Because metabolic stress acti-vates AMPK, we examined AMPK’s activation state in gliomas.Immunohistochemistry (IHC) of grade IV human gliomas (calledglioblastoma or GBM) showed abundant active (phosphorylated)AMPK in all GBM tissues (Fig. 1A; IHC of 6 of 12 GBMs shown).Compared with normal astrocytes, human GBM cell lines alsoexpressed significantly higher levels of phosphorylated AMPKand ACC (an AMPK substrate) (Fig. 1B). Furthermore, in agenetically engineered mouse model of high-grade glioma (Fig.1C; ref. 44), we observed high pAMPK and pACC levels withinthe tumor compared with normal cortical tissue (Fig. 1 D and

Fig. 1. Phosphorylated (active) AMPK is abundantly expressed in gliomas and the direct AMPK activator A769662 does not inhibit glioma growth. (A) IHC ofactive AMPK (pAMPK) in six human GBMs (12 tumors were analyzed). (Inset) High magnification of A. (B) Immunoblot shows pAMPK in glioma cell lines andnormal astrocytes. Histology (C), IHC (D), and immunoblot (E) of mouse high-grade gliomas (HGGs). Magnification: A, 20×; A, Inset, 60×; C, 10×; D, 40×. A totalof 12 tumors were analyzed. N, contralateral normal brain; T, tumor tissue. (F) Immunoblot analysis of glioma cells treated with AMPK agonists and (G) theeffect AICAR, metformin, and A769662 on the proliferation of glioma cells. *P ≤ 0.005. Data shown is representative of three to six independent experiments.Error bars represent mean ± SD.

E436 | www.pnas.org/cgi/doi/10.1073/pnas.1311121111 Liu et al.

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

E), akin to another report in a rat model of glioma (45). Thesedata demonstrate that active AMPK is common in GBMs.

Glioma Cell Proliferation Is Suppressed by AICAR and Metformin butNot by the Direct AMPK Activator A769662. To examine whether theextent of AMPK activation correlates with the antiproliferativeeffects of AMPK activators in glioma, we conducted a dose- andtime-dependent immunoblot analysis using AMPK and ACCantibodies and parallel cell proliferation assays. AICAR, met-formin, and A769662 activated AMPK similarly at all doses(AICAR: 1 and 2 mM; metformin: 1, 2.5, 5, and 10 mM; andA769662: 100 and 200 μM; Fig. S1 A and B). Whereas AICARinhibited proliferation similarly at all doses, metformin’s effectwas significant only at 10 mM (Fig. S1C). Based on these re-sults, we chose 1 mM AICAR, 10 mM metformin, and 100 μMA769662 to conduct the time-course analysis. All three agonistsactivated AMPK leading to ACC phosphorylation, with maximalactivation occurring at 48 and 72 h (Fig. 1F). The earliest timepoint AICAR and metformin suppressed proliferation was at48 h which was maintained until 72 h. (Fig. 1G). This reductionin viable cell numbers was not due to increased cell death (mea-sured by Trypan blue exclusion at each time point; Fig. S1D) butlargely due to the cytostatic effects of these agents. Surprisingly,despite robustly activating AMPK (Fig. 1F and Fig. S1B), A769662did not suppress proliferation (Fig. 1G and Fig. S1E). Akt acti-vation can oppose AMPK action; however, A769662 did not

increase Akt phosphorylation (Fig. S1F). Our results suggestthat the indirect AMPK activators AICAR and metformin in-hibit proliferation possibly through pleiotropic actions, whereasthe direct activator A769662 does not suppress cell proliferation.

Inhibition of Lipogenesis Is Not a Likely Mechanism of AICAR andMetformin’s Antiproliferative Action. AMPK inhibits ACCα andHMG-CoA reductase, to suppress FA and cholesterol synthesis,respectively. ACCα and FA synthase (FASN) work in conjunc-tion to form palmitate. To test whether endogenous FASN isrequired for optimal proliferation of glioma cells and whetherAICAR and metformin suppresses proliferation by inhibiting thispathway, we silenced FASN and ACC and examined whether si-lencing this pathway confers cellular resistance to these agents.shRNA-mediated silencing of FASN (Fig. 2A, Inset) did notreduce proliferation in normal growth medium containing 10%serum (Fig. 2A and Fig. S2A). Remarkably, regardless of theserum content, metformin and AICAR’s effects were essentiallysimilar in control and FASN shRNA-expressing cells (Fig. 2Aand Fig. S2 A and B). Silencing of ACC (Fig. 2B, Inset) or ex-pression of a phosphorylation-incapable S79A mutant of ACC1(AMPK target site) also did not suppress proliferation and didnot confer resistance to AICAR or metformin’s action (Fig. 2Band Fig. S2B). Addition of lipogenesis end products (palmitateand mevalonate) also did not reverse AICAR and metformin’santiproliferative effects (Fig. S2 D–F). Furthermore, the FASN

Fig. 2. Inhibition of lipogenesis is not a mechanism of AICAR and metformin’s antiproliferative action. Proliferation of FASN-silenced (A) and ACC-silenced(B) T98G and U87EGFRvIII glioma cells treated with AICAR and metformin. nt, nontarget. Immunoblot with FASN (A, Inset) and ACC (B, Inset) antibodies. nt,nontarget shRNA. *P ≤ 0.005. (C) Proliferation of glioma cells in the presence of lipogenesis inhibitors [C75 (10 μg/mL) and atorvastatin (1 μM)]. Data shown isrepresentative of two to four independent experiments. Error bars in A–C represent mean ± SD.

Liu et al. PNAS | Published online January 13, 2014 | E437

BIOCH

EMISTR

YPN

ASPL

US

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

inhibitor had little effect on glioma cell proliferation (Fig. 2C).Collectively, these results indicate that AICAR and metforminsuppresses proliferation through other mechanisms.

AMPK-Dependent mTOR Inhibition Is Not Required by AICAR andMetformin to Suppress Glioma Cell Proliferation. mTORC1 pro-motes cell proliferation and growth through the phosphorylationof its effectors 4EBP1 and S6. AMPK inhibits mTOR through phos-phorylation of two mTORC1 regulators—TSC2 and RAPTOR.We first examined the importance of mTORC1 in glioma byusing the mTORC1 inhibitor rapamycin. Low nanomolar con-centrations of rapamycin completely suppressed S6 but had littleeffect on 4EBP1 phosphorylation or proliferation (Fig. S3 A andB). This suggests that complete inhibition of 4EBP1 but not S6

phosphorylation may be necessary to halt proliferation. Toexamine whether AMPK agonists suppresses proliferation byAMPK-dependent mTORC1 inhibition, we first examined theireffect on phosphorylation of S6 and 4EBP1 in glioma cells. Asexpected, all three AMPK agonists activated AMPK leading toRAPTOR phosphorylation (Fig. 3A and Fig. S3C). However,despite similar RAPTOR phosphorylation, metformin but notAICAR or A769662 significantly inhibited mTOR (suppressingboth 4EBP1 and S6 phosphorylation; Fig. 3A and Fig. S3C).Inhibition of mTORC1 by AICAR and A769662, if any, wasincomplete (Fig. 3A and Fig. S3C). Dephosphorylated 4EBP1sequesters the translation initiation factor eIF4E (a CAP-bindingprotein) to inhibit CAP-dependent translation. Predictably, highamounts of 4EBP1 were bound to immunoprecipitated eIF4E

Fig. 3. AMPK-dependent mTOR inhibition is not required by AICAR and metformin to suppress glioma cell proliferation. (A) Immunoblots showing theeffects of AICAR, metformin, and A769662 on phosphorylation of AMPK substrates (ACC and RAPTOR) and mTOR effectors (S6 and 4EBP1) in T98G gliomacells. In B, eIF4E was immunoprecipitated with m7GDP-Sepharose and bound 4EBP1 was detected with 4EBP1 antibody. (C, Upper) Proliferation of control (ntshRNA) or AMPKβ1 shRNA cells treated with AMPK agonists. nt, nontarget. *P ≤ 0.005. (C, Lower) Immunoblots show AMPK, ACC, and RAPTOR phos-phorylation by AMPK agonists. Densitometry of pAMPK levels is also shown. Error bars represent mean ± SD. (D) Immunoblots demonstrate the effects ofAMPK agonists on mTOR effectors (S6 and 4EBP1) in control (nt) and AMPKβ1 shRNA T98G glioma cells. (E) The CAP-binding assay was done as in B in controlor AMPKβ1 shRNA T98G cells. Data shown is representative of two to five independent experiments.

E438 | www.pnas.org/cgi/doi/10.1073/pnas.1311121111 Liu et al.

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

in metformin-treated cells but not in A769662-treated cells(Fig. 3B). The high sensitivity of this assay showed that AICARdoes inhibit 4EBP1 phosphorylation but to a significantly lesserextent than metformin. Based on these results, we reason thatbecause both AICAR and rapamycin partially inhibited mTORbut only AICAR suppressed proliferation, AICAR’s suppressiveactions are likely mTOR independent. Metformin on the otherhand, completely suppressed mTORC1 and proliferation, sug-gesting that mTOR inhibition could be necessary for its antipro-liferative actions.To examine whether mTOR inhibition by metformin is de-

pendent on AMPK, we knocked down the AMPKβ1 subunit torepress AMPK function. We targeted this subunit because first,the β-subunits regulate the stability of the catalytic α-subunits ofAMPK and second, glioma cells expressed significantly more β1than β2 (Fig. S3D). Indeed, by silencing β1 we achieved an 80–90% reduction of active AMPK levels (Fig. S3 E–G). AICARand metformin could not sufficiently activate AMPK in β1-silenced cells (Fig. 3C), yet they inhibited proliferation similar to(in fact more than) control cells (Fig. 3C). This clearly indicatesthat AICAR and metformin suppress proliferation regardless ofAMPK. To rule out the possibility that these agents use theremaining AMPK β2 subunit or directly the AMPK α-subunits,we silenced both β1/β2 or α1/α2 subunits in T98G cells (Fig.S3G) and also used mouse fibroblasts derived from β1/β2 doubleknockout (DKO) animals. AMPK agonist-dependent phos-phorylation of ACC and RAPTOR was considerably reduced inβ1/β2 and α1/α2 silenced cells compared with NT cells (Fig. S3H and I). Remarkably, both AICAR and metformin but notA769662 inhibited glioma cell proliferation in the double-silenced cells that was comparable to NT cells (Fig. S3 J and K).Similar results were observed in mouse embryonic fibroblasts(MEFs). As expected, the catalytic α-subunits of AMPK becamecompletely unstable in the β1/β2 DKO MEFs and consequentlyAMPK substrate phosphorylation (ACC and RAPTOR) wascompletely inhibited (Fig. S4A). Even in these cells both AICARand metformin robustly inhibited proliferation. In fact, AMPK-null MEFs were inhibited more than wild-type (WT) MEFs (Fig.S4B). Together, these results confirm that AMPK is not requiredfor the antiproliferative actions of AICAR and metformin. Be-sides, the relatively greater sensitivity of both AMPK-silencedglioma cells and AMPK-null MEFs to AICAR and metforminshows that AMPK probably provides resistance to the detrimentalactions of anticancer agents as shown by others (41).Pharmacological AMPK activation by all three AMPK ago-

nists caused RAPTOR phosphorylation. However, only AICARand metformin inhibited proliferation. In any case, there weredifferential effects of these agonists on the two mTOR effectors(S6 and 4EBP1). Therefore, we examined whether the AMPK–mTOR axis is preserved during physiological AMPK activation.A time kinetics of glucose deprivation (GD) revealed that al-though AMPK was activated (along with phosphorylation ofACC and RAPTOR) at 24 h (and likely much earlier), phos-phorylation of AMPK and its substrates decreased significantlyat 48 and 72 h, suggesting a resetting of the bioenergetic equi-librium after acute metabolic stress (Fig. S3L). Surprisingly,despite these effects, mTOR was completely inhibited at all timepoints following glucose withdrawal, independent of AMPK(note: absence of pS6 and p4EBP1 in both nt and β1/ β2 shRNAlanes during GD). We also consistently noticed reduced levels oftotal 4EBP1 protein in AMPK β-subunit–silenced T98G gliomacells. Together, these results suggest that both AMPK-dependentand -independent mechanisms of mTOR inhibition exist in theseglioma cells.Depending on the context, mTORC1 can be inhibited in-

dependent of AMPK (46). Because we predicted that metforminlikely requires mTOR to suppress glioma proliferation and haveshown that it does so regardless of AMPK, we wanted to verify

that its mTOR inhibition was AMPK independent. Indeed,metformin completely inhibited mTOR similarly in control [non-target (NT)] and AMPK-silenced glioma cells (Fig. 3 D and E).GD experiments in MEFs showed both similarities and dissim-ilarities with glioma cells. Similar to glioma cells, levels ofphosphorylated AMPK, ACC, and RAPTOR increased at 24 hbut were reduced at 48 and 72 h (Fig. S4C). However, mTORinhibition was considerably more coupled to AMPK activation inthese normal fibroblasts. Phosphorylated levels of both S6 and4EBP1 were reduced in WT MEFs at 48 and 72 h of GD (likelydue to AMPK activation at an earlier time point), but the levels(particularly that of 4EBP1) persisted in the AMPKβ1/β2 DKOMEFs. These contrasting results between glioma and normalcells suggest that probably the complex wiring of signaling net-works, including the AMPK–mTOR axis, renders cancer cellsdifferentially responsive to antiproliferative agents comparedwith normal cells.

Induction of Chronic Energy Stress Is Not a Mechanism by WhichAICAR and Metformin Repress Glioma Cell Proliferation. It hasbeen suggested that AICAR and metformin induce energy stressto activate AMPK and inhibit proliferation (10, 11). We there-fore examined whether these reagents cause energy stress. HPLCanalysis showed that surprisingly, neither AICAR nor metformindecreased ATP:AMP ratio acutely (4 h) or chronically (72 h). Infact, we observed an increase in ATP:AMP ratio followingtreatment with these agents (Fig. 4A). Total ATP content mea-sured by luciferase assay was also increased (Fig. S5 A–C), sug-gesting that these agents directly stimulate ATP production orAMPK-dependent response mechanisms exist in these cells tocounter energy stress caused by these agents.To further validate these results, we examined energy-producing

pathways by measuring the extracellular acidification rate (ECAR),a measure of glycolysis, and the O2 consumption rate (OCR) usingthe Seahorse XF-analyzer (Seahorse Bioscience). Metformin butnot AICAR sharply reduced OCR (Fig. 4B and Fig. S5D), butmetformin also increased compensatory ECAR (Fig. 4C and Fig.S5E). These effects of metformin on ECAR and OCR wereAMPK independent. (Fig. 4 D and E). To provide additionalevidence that metformin does not inhibit proliferation by sup-pressing mitochondrial OCR, we created mitochondria-deficientglioma (Rho-0) cells. Our results show that metformin’s growthinhibition does not require functional mitochondria as it in-hibited proliferation of Rho-0 cells similar to mitochondria-proficient cells (Fig. 4F).

AICAR and Metformin Suppresses Glioma Growth in Vivo Independentof AMPK. We next determined whether the AMPK-independentgrowth inhibitory effects of AICAR and metformin are alsotrue in vivo. To test this, we created flank xenografts usingU87EGFRvIII glioma cells expressing NT or AMPKβ1 shRNAand treated mice with AICAR, metformin, or vehicle. BothAICAR and metformin significantly suppressed tumor growth atall time points (P ≤ 0.03). Consistent with our in vitro results,both agents, particularly AICAR, caused a greater growth in-hibition of AMPK-silenced tumors than control tumors (Fig. 5 Aand B). We conclude that AMPK is not required by AICARand metformin to inhibit glioma proliferation in vitro or growthin vivo.

AICAR Blocks Cell Cycle by AMPK-Independent Proteosomal Degra-dation of cdc25c. We next examined the mechanism of AICAR’sgrowth suppression by conducting cell cycle analysis. AICARblocked the cell cycle at G2M consistently in all cell lines (Fig.5C). This effect of AICAR was also AMPK independent (Fig.5D). Additionally, cell proliferation experiments using the thy-midine analog 5-ethynyl-2′-deoxyuridine that incorporates intonewly synthesized DNA showed no significant S-phase arrest by

Liu et al. PNAS | Published online January 13, 2014 | E439

BIOCH

EMISTR

YPN

ASPL

US

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

AMPK activators (Fig. S6 A and B). To examine the mechanismof cell cycle arrest we studied several cell cycle regulators in-cluding Cyclin A, D1, D2, and E; CDK2; p16; p21; p27; p53; andRb phosphorylation. AICAR had no significant effect on any ofthese proteins (Fig. S6C).Dephosphorylation of the kinase cdc2 by the phosphatase

cdc25c is crucial for G2M transition. We observed that AICAR-treated cells had significantly reduced cdc25c levels (Fig. 5E).This effect of AICAR was independent of AMPK (Fig. 5F).Reduced cdc25c levels resulted in increased phosphorylation

and inhibition of cdc2 (Fig. 5 E and F). Cyclin B1 (whichassociates with cdc2) was not affected. To examine how AICARsuppresses cdc25c, we studied ROS because cdc25c is down-regulated by ROS (47) and AICAR was shown to cause ROSproduction (48) However, AICAR did not produce ROS inglioma cells (Fig. S6D). RT-PCR analysis confirmed thatAICAR does not affect cdc25c transcript levels (Fig. 5G).Because cdc25c is also regulated by the proteasome pathway(49), we examined whether AICAR degrades cdc25c through theproteasome. Indeed, the proteasome inhibitor MG132 (Sigma)

Fig. 4. AICAR and metformin do not suppress glioma proliferation by inducing chronic energy crisis. (A) HPLC analysis shows the energy content ofT98G glioma cells treated with AMPK agonists. (B) O2 consumption (OCR) and (C ) glycolysis (ECAR) of T98G glioma cells treated with AICAR ormetformin. OCR (D) and ECAR (E ) were also measured in control (nt) and AMPKβ1 shRNA T98G cells. nt, nontarget. (F ) Proliferation of control andRho-0 T98G cells treated with AICAR or metformin. Inset shows RT-PCR analysis of mitochondrial transcripts [cytochrome oxidase II (CoxII) and NADH de-hydrogenase 4 (ND4)] in control and Rho-0 cells. Data shown is representative of at least three independent experiments. *P ≤ 0.001. Error bars in A–F representmean ± SD.

E440 | www.pnas.org/cgi/doi/10.1073/pnas.1311121111 Liu et al.

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

restored cdc25c levels in AICAR-treated T98G cells and U87cells (Fig. 5H and Fig. S6E) and this effect was independent ofAMPK (Fig. 5H). To validate that cdc25c down-regulation hasa functional consequence on glioma cell proliferation, we si-lenced cdc25c (Fig. 5I, Inset). Three independent clones ofcdc25c shRNA significantly arrested glioma cells at G2M and re-pressed glioma cell proliferation by about 35%. (Fig. 5I). Together,our findings demonstrate that AICAR’s growth-suppressive effectsare independent of AMPK and occur through multiple mech-anisms, important among which is a previously unidentified mech-

anism that involves degradation of a key cell cycle protein bythe proteasome.

Metformin Enhances PRAS40–RAPTOR Association to Inhibit mTORand Suppress Glioma Proliferation, Independent of AMPK. Because4EBP1 phosphorylation by mTOR is crucial for cancer cellproliferation and growth (50) and metformin strongly inhibited4EBP1 phosphorylation in glioma cells, we tested whethermTOR and 4EBP1 are necessary for metformin’s suppressiveaction by knocking down these proteins (Fig. 6 A and B and Fig.S7 A and B). Predictably, mTOR silencing reduced and 4EBP1

Fig. 5. Metformin and AICAR inhibit glioma growth in vivo and AICAR inhibits glioma proliferation by degrading cdc25c independent of AMPK. (A)Metformin and AICAR’s effect on growth of control (NT shRNA) and AMPKβ1 shRNA expressing U87EGFRVIII glioma xenografts in Nu/Nu mice (n = 10 percondition). *P ≤ 0.03 shown for both metformin and AICAR in NT and shRNA tumors. (B) Photomicrographs of representative control and treated gliomas.Histogram of the cell cycle analysis of three glioma cells (C) and control (nt) and AMPKβ1-silenced T98G glioma cells (D), treated with AMPK agonists. *P ≤0.005. Immunoblots show the effects of AICAR on G2M regulators in two glioma cells (E), and in control (nt) and AMPKβ1shRNA T98G glioma cells (F). (G)Effect of AICAR on cdc25c RNA levels. (H) Effects of AICAR alone or in the presence of the proteasomal inhibitor MG132 on protein levels of cdc25c in control(nt) and β1 shRNA T98G glioma cells. (I) Proliferation and cell cycle analysis of control (nt) and cdc25c shRNA expressing T98G glioma cells. (I, Inset) Immu-noblot shows knockdown of cdc25c protein by three independent shRNA. Data are representative of at least three independent experiments. nt, nontarget.*P ≤ 0.005. Error bars in A and I represent mean ± SD.

Liu et al. PNAS | Published online January 13, 2014 | E441

BIOCH

EMISTR

YPN

ASPL

US

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

silencing enhanced cell proliferation (Fig. 6C and Fig. S7C).Remarkably, the effect of metformin was significantly blunted inmTOR-silenced (>90% rescue) and 4EBP1-silenced (>70%rescue) cells (Fig. 6C and Fig. S7C), suggesting that metforminrequires mTOR to suppress proliferation. Unlike metformin,AICAR’s inhibitory effects were maintained in mTOR and4EBP1 knockdown cells (Fig. 6C and Fig. S7C).Next, we explored the mechanism of metformin’s mTOR in-

hibition. In the in vitro kinase assay using recombinant 4EBP1and mTOR, metformin did not inhibit mTOR kinase activity(Fig. S7D). Kalender et al. (24) showed that, similar to aminoacid starvation, metformin inhibits mTOR by blocking RAGGTPases in HEK293T cells. As expected, amino acid starvationinhibited mTOR in control cells that was rescued by expressionof two independent constitutively active RAGB–GTP constructs(Fig. S7E). However, metformin still inhibited mTOR (Fig. S7E)and repressed proliferation (Fig. S7 F and G) similarly in control

and RAGB–GTP-expressing glioma cells, suggesting that met-formin uses other mechanisms to repress glioma proliferation.Growth factors activate mTOR through Rheb–GTP. However,

expression of WT or a constitutively active Rheb–GTP (Q64L)did not prevent metformin’s inhibition on mTOR or proliferation(Fig. S7 H and I). Metformin also did not reduce protein levelsof the mTORC1 complex (mTOR, RAPTOR, PRAS40, andDEPTOR; Fig. S7J). The Akt substrate PRAS40 binds toRAPTOR to negatively regulate mTORC1 (51, 52). Because cellstressors (like 2DG, oligomycin) inhibit mTOR by increasingPRAS40 binding with RAPTOR (52), we questioned whethermetformin inhibits mTOR by enhancing this association. Indeed,both metformin and its analog phenformin significantly en-hanced PRAS40–RAPTOR association (Fig. 6D). To examinewhether this is a crucial mechanism of metformin’s inhibitoryaction in glioma cells, we tested whether PRAS40-silenced cellsresist metformin’s inhibitory action. Knocking down PRAS40significantly rescued glioma cells from metformin’s inhibition(Fig. 6E), whereas a second shRNA that did not silence PRAS40failed to rescue these cells.Our overall findings provide compelling evidence that AMPK

is not required for the antiproliferative actions of two widelyused AMPK agonists. Based on our unique results, we proposea model (Fig. 6F) in which the direct AMPK activator A769662has little effect on mTOR or proliferation; AICAR and met-formin suppress proliferation independent of AMPK: AICAR byproteasomal degradation of cdc25c and metformin by increasingPRAS40-mediated mTOR inhibition.

DiscussionThe function of AMPK in cancer is not fully understood. HighAMPK activity is observed in colorectal, cervical, and braincancers (refs. 30, 45, and 53–56 and this study), whereas lowAMPK activity has been observed in others. Because LKB1, thekinase that activates AMPK is a tumor suppressor, AMPK ac-tivation by agonists is perceived to have therapeutic benefits.This belief remains despite the puzzling observation that incontrast to canonical tumor suppressors, LKB1 is required forcellular transformation (1) and LKB1- or AMPK-deficient can-cer cells can be sensitive to chemotherapeutic agents (27, 31, 57).Ironically, these agents include the very AMPK agonists thatare used to activate AMPK. To unravel this conundrum weexamined whether these AMPK agonists work independent ofAMPK. We found that two indirect AMPK agonists, AICARand metformin (that do not activate AMPK in cell-free assays)(58), inhibit proliferation independent of AMPK and to oursurprise, the direct AMPK activator A769662 has no suchinhibitory effects in normal or glioma cells. We discover thatalthough AICAR blocks the cell cycle by a unique mechanismthat involves proteasomal degradation of the G2M phospha-tase cdc25c, metformin suppresses proliferation by inhibit-ing mTOR through enhanced PRAS40 binding to RAPTOR.We conclude that in line with other observations (12, 15),AICAR and metformin’s growth inhibition in vitro does notrequire AMPK.The growth suppressive mechanisms of the AMPK agonists in

vivo are as enigmatic as the role of AMPK itself in tumorigen-esis. In PTEN+/−;LKB1 hypomorphic mice that develop lym-phoma (25), A769662, metformin, and its analog phenformindelayed the onset but not progression of tumor growth. It is notknown whether these agents sufficiently activated AMPK insidetumors. In fact, metformin prevented lung tumorigenesis withoutactivating AMPK inside tumors, likely due to its systemic effectson insulin signaling (26). It is possible that the growth-inhibitingeffects in our GBM xenografts are a combination of the AMPK-independent mTOR-dependent cellular and systemic effects ofmetformin. Contrasting results about AMPK’s antitumor role inlymphoma (42) and protumor role in brain cancer (30) suggest

Fig. 6. Metformin suppresses proliferation through PRAS40-mediated mTORinhibition. (A and B) Immunoblot using mTOR and 4EBP1 antibodies showingshRNA-mediated knockdown of mTOR and 4EBP1 in T98G glioma cells. Actinwas used as a loading control. (C) Proliferation of control (NT) and mTORshRNA-expressing and 4EBP1 shRNA-expressing T98G cells treated withAICAR and metformin. (D) Immunoprecipitation of RAPTOR followed byimmunoblot analysis showing the effect of metformin and other cell stres-sors on RAPTOR–PRAS40 association. Loading control lysates are shown inthe bottom three panels. (E) Proliferation of metformin-treated glioma cellsexpressing NT or PRAS40 shRNA. (E″) Immunoblot showing PRAS40 knock-down. (F) The model shows active AMPK is highly expressed in glioblastoma.The direct AMPK activator A769662 or the indirect activator AICAR does noteffectively suppress mTOR. Although A769662 has no effect on gliomaproliferation, AICAR suppresses glioma proliferation by degrading acrucial G2M phosphatase cdc25c through the proteasome, independent ofAMPK. The other AMPK agonist metformin represses glioma proliferationthrough mTOR inhibition by increasing PRAS40-RAPTOR interaction, inde-pendent of AMPK. Data are representative of two to four independentexperiments. *P ≤ 0.001. Error bars in C and E represent mean ± SD.

E442 | www.pnas.org/cgi/doi/10.1073/pnas.1311121111 Liu et al.

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

that AMPK likely plays context-dependent diverse roles intumorigenesis and, depending on the model, much of the in-hibitory effects of the AMPK agonists in vivo are independentof AMPK.Similar to our results, AICAR suppressed GBM xenograft

growth (9). In that pharmacological study, AICAR was shown toinhibit lipogenesis in vitro and lipogenesis inhibitors reducedproliferation in vitro. Whether AICAR’s action required AMPKwas not examined. In contrast, we found that in fact, AMPK-silenced GBM tumors are hypersensitive to both AICAR andmetformin. Furthermore, knockdown of lipogenic genes (ACCand FASN) did not interfere with glioma proliferation, akin tothat observed in human breast cancer cells (59). However,AICAR and metformin were still inhibitory on these ACC- orFASN-silenced cells indicating that they clearly work throughother pathways. We speculate that regulation of FA utilization incancer cells (de novo synthesis vs. import) is far more complexthan that in normal lipogenic cells and AMPK activation isallowed at specific stages during the evolution of certain tumorsto endure metabolic stress without significant inhibition of FAsynthesis at the level of the AMPK substrate ACC.Our study shows that mTOR regulation by AMPK activators

in glioma cells (and probably in other cancer cells) is differentfrom that in normal cells. By comparing the effects of rapamycin,AICAR, A769662, and metformin on mTOR, we conclude thatinhibition of 4EBP1 but not S6 phosphorylation by metformin,independent of AMPK, is critical for the suppression of gliomacell proliferation. Unlike metformin, mTOR inhibition by AICARand A769662 was at best partial or marginal. To unravel this dis-crepancy, we examined the extent of TSC2 and RAPTOR phos-phorylation (the two AMPK substrates that inhibit mTOR) bythe AMPK agonists. We found that although phospho-TSC2antibodies yielded multiple nonspecific bands and were thusunreliable, the RAPTOR antibody was specific. All three AMPKagonists phosphorylated RAPTOR. It is unclear at this time whythe signal from RAPTOR phosphorylation did not completelyreach mTOR in glioma cells. Discrete localization of the variousAMPK complexes and mTOR substrates in specific subcellularcompartments, together with signal interference from the onco-genic pathways in cancer cells may explain some of these results.It is worth noting that AMPK silencing did not increase baselinemTOR signaling in glioma cells, suggesting that in certain con-texts basally active AMPK may allow mTOR activation. It is alsoclear from our results that although the AMPK–mTOR axis isstrongly preserved during physiological metabolic stress in nor-mal fibroblasts, it is less so in glioma cells. Future studies willunravel this intriguing uncoupling of the AMPK–mTOR axisin glioma.We discovered two AMPK-independent mechanisms by which

AICAR and metformin represses glioma proliferation. Wefound that AICAR blocks glioma cells at G2M by activating theproteasomal pathway through a yet unidentified mechanism.This leads to down-regulation of the critical G2M phosphatasecdc25c (and possibly other proteins). Indeed, akin to otherstudies (60), knockdown of cdc25c in glioma cells caused G2Marrest and suppressed proliferation. Additional studies, includingthe effects of AICAR on protein ubiquitination, will be requiredto determine how AICAR activates the proteasomal pathway inglioma cells. In contrast to AICAR, metformin clearly requiredmTOR to repress proliferation. mTOR is regulated by multiplemechanisms and the dominant mechanism might vary accordingto cell type and context. We found that metformin treatmentenhanced binding of PRAS40 with RAPTOR, an associationknown to be enriched during stress to inhibit mTOR activity.Although metformin inhibits mitochondrial respiration, we did

not find any chronic energy reduction in metformin-treatedglioma cells. It is possible that metformin causes other types ofcell stresses (e.g., unfolded protein response, DNA damage re-sponse, and endoplasmic reticulum stress) that are transmittedthrough PRAS40 to mTOR inhibition. Finally, we surmise thatAMPK plays context-specific roles in tumorigenesis that dependon tumor stage and energy state, as well-nourished vasculartumors undergo AMPK-dependent metabolic reprogramming(32) and evolve into substrate-limited, hypoxic, and invasivetumors. To delineate these processes and define the role ofAMPK in cancer, genetic studies and use of specific AMPKmodulators will be crucial. Metformin is used in several can-cer clinical trials. Our study warrants careful interpretation ofdata on the anticancer mechanisms of this useful drug.

Experimental ProceduresPatient Samples. GBM samples were obtained under an institutional reviewboard-approved protocol from The University of Cincinnati Brain TumorCenter. Informed consent was obtained from the patients whose sampleswere used in our research. For details, see SI Experimental Procedures.

Mouse Strains. All animal procedures were carried out in accordance withthe Institutional Animal Care and Use Committee-approved protocol ofCincinnati Children’s Hospital Medical Center. AMPKβ1−/− and AMPKβ2−/−

mice have been described (35, 45). Quadruple conditional knockout micewere derived by crossing triple knockout mice (GFAP-CreER; PtenloxP/loxP;Trp53loxP/loxP; Rb1loxP/loxP) (44) with Rb110−/− mice. Filial-1 progeny wereintercrossed to generate the quadruple homozygously targeted strain.Deletion of conditional alleles was induced by i.p. tamoxifen injections.

Cell Culture and Growth Analysis. Cell culture methods are described in SIExperimental Procedures. β1−/−;β12−/− MEFs were generated by crossing β1−/−

and β2−/− mice. For proliferation and viability analysis, a fluorescence-basedmethod (Cell-titer-fluor; Promega) and direct counting using Trypan blue wereused. Drugs were added 24 h postseeding and cell viability was determined atindicated times.

Immunoblot Analysis. Western blot analysis was carried out following stan-dard methods (35). For details, see SI Experimental Procedures.

shRNA and Lentivirus. ThefollowingshRNAcloneswerepurchasedfromtheLenti-shRNA Library Core [Cincinnati Children’s Hospital Medical Center (CCHMC)]:AMPKβ1 (TRCN0000004770), FASN (TRCN0000003127 or TRCN0000003128),mTOR (TRCN0000038674 or TRCN0000038677), ACCα (TRCN0000004766,TRCN0000004767 and TRCN0000004769), PRAS40 (TRCN0000165347),cdc25c (TRCN0000002432, TRCN0000002432 and TRCN0000002434), andpLKO.1-puro scrambled (NT); 4EBP1 shRNA (TRCN0000040203) was a giftfromGeorge Thomas (University of Cincinnati, Cincinnati). For details, seeSI Experimental Procedures.

Metabolic Experiments. ATP and AMP were analyzed by HPLC as describedbefore (45). ATP was additionally measured by using the ApoSENSOR ATPLuminescence Assay Kit (BioVision). ECAR and OCR were analyzed by usingthe Seahorse XF-Analyzer. For detailed methods, see SI ExperimentalProcedures.

Cell Cycle Analysis. Cell cycle analysis was done in a FACScan analyzer (BD). Fullmethod is in SI Experimental Procedures.

Statistical Analysis. Student t test was used to calculate statistical significancewith P < 0.05 representing a statistically significant difference.

ACKNOWLEDGMENTS. We thank Nancy Ratner for manuscript review;George Thomas for the 4EBP1 shRNA plasmid, and P53−/− and TSC2−/− MEFs;Paul Mischel for U87EGFR and U87EGFRvIII cells; Russell Jones for AMPKα1/α2shRNA plasmid; Carol Mercer for the control GST plasmid; Peter Vogt for theRheb plasmids; and Nissim Hay for the ACC1 S79A plasmid. This work wassupported by the CancerFreeKids, the Smith–Brinker Golf Foundation,a CCHMC Trustee Scholar grant, and National Institutes of Health Grant1R01NS075291-01A1 (all to B.D.).

Liu et al. PNAS | Published online January 13, 2014 | E443

BIOCH

EMISTR

YPN

ASPL

US

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1

1. Shaw RJ, et al. (2004) The tumor suppressor LKB1 kinase directly activates AMP-acti-vated kinase and regulates apoptosis in response to energy stress. Proc Natl Acad SciUSA 101(10):3329–3335.

2. Jones RG, Thompson CB (2009) Tumor suppressors and cell metabolism: A recipe forcancer growth. Genes Dev 23(5):537–548.

3. Viollet B, et al. (2010) AMPK inhibition in health and disease. Crit Rev Biochem MolBiol 45(4):276–295.

4. Hardie DG (2011) AMP-activated protein kinase: An energy sensor that regulates allaspects of cell function. Genes Dev 25(18):1895–1908.

5. Foretz M, Carling D, Guichard C, Ferré P, Foufelle F (1998) AMP-activated proteinkinase inhibits the glucose-activated expression of fatty acid synthase gene in rathepatocytes. J Biol Chem 273(24):14767–14771.

6. Inoki K, Zhu T, Guan KL (2003) TSC2 mediates cellular energy response to control cellgrowth and survival. Cell 115(5):577–590.

7. Gwinn DM, et al. (2008) AMPK phosphorylation of raptor mediates a metaboliccheckpoint. Mol Cell 30(2):214–226.

8. Sabatini DM (2006) mTOR and cancer: Insights into a complex relationship. Nat RevCancer 6(9):729–734.

9. Guo D, et al. (2009) The AMPK agonist AICAR inhibits the growth of EGFRvIII-expressing glioblastomas by inhibiting lipogenesis. Proc Natl Acad Sci USA 106(31):12932–12937.

10. Swinnen JV, et al. (2005) Mimicry of a cellular low energy status blocks tumor cellanabolism and suppresses the malignant phenotype. Cancer Res 65(6):2441–2448.

11. Tang YC, Williams BR, Siegel JJ, Amon A (2011) Identification of aneuploidy-selectiveantiproliferation compounds. Cell 144(4):499–512.

12. Guigas B, et al. (2007) AMP-activated protein kinase-independent inhibition of he-patic mitochondrial oxidative phosphorylation by AICA riboside. Biochem J 404(3):499–507.

13. Jones RG, et al. (2005) AMP-activated protein kinase induces a p53-dependent met-abolic checkpoint. Mol Cell 18(3):283–293.

14. Rattan R, Giri S, Singh AK, Singh I (2005) 5-Aminoimidazole-4-carboxamide-1-beta-D-ribofuranoside inhibits cancer cell proliferation in vitro and in vivo via AMP-activatedprotein kinase. J Biol Chem 280(47):39582–39593.

15. Santidrián AF, et al. (2010) AICAR induces apoptosis independently of AMPK and p53through up-regulation of the BH3-only proteins BIM and NOXA in chronic lympho-cytic leukemia cells. Blood 116(16):3023–3032.

16. Guigas B, et al. (2006) 5-Aminoimidazole-4-carboxamide-1-beta-D-ribofuranoside andmetformin inhibit hepatic glucose phosphorylation by an AMP-activated proteinkinase-independent effect on glucokinase translocation. Diabetes 55(4):865–874.

17. Barankiewicz J, Jimenez R, Ronlov G, Magill M, Gruber HE (1990) Alteration of purinemetabolism by AICA-riboside in human B lymphoblasts. Arch Biochem Biophys 282(2):377–385.

18. Shang J, Lehrman MA (2004) Activation of glycogen phosphorylase with 5-amino-imidazole-4-carboxamide riboside (AICAR). Assessment of glycogen as a precursor ofmannosyl residues in glycoconjugates. J Biol Chem 279(13):12076–12080.

19. Guigas B, et al. (2004) Metformin inhibits mitochondrial permeability transition andcell death: A pharmacological in vitro study. Biochem J 382(Pt 3):877–884.

20. Dowling RJ, Zakikhani M, Fantus IG, Pollak M, Sonenberg N (2007) Metformin inhibitsmammalian target of rapamycin-dependent translation initiation in breast cancercells. Cancer Res 67(22):10804–10812.

21. Dykens JA, et al. (2008) Biguanide-induced mitochondrial dysfunction yields increasedlactate production and cytotoxicity of aerobically-poised HepG2 cells and humanhepatocytes in vitro. Toxicol Appl Pharmacol 233(2):203–210.

22. Kisfalvi K, Eibl G, Sinnett-Smith J, Rozengurt E (2009) Metformin disrupts crosstalkbetween G protein-coupled receptor and insulin receptor signaling systems and in-hibits pancreatic cancer growth. Cancer Res 69(16):6539–6545.

23. Liu B, et al. (2011) Potent anti-proliferative effects of metformin on trastuzumab-resistant breast cancer cells via inhibition of erbB2/IGF-1 receptor interactions. CellCycle 10(17):2959–2966.

24. Kalender A, et al. (2010) Metformin, independent of AMPK, inhibits mTORC1 in a ragGTPase-dependent manner. Cell Metab 11(5):390–401.

25. Huang X, et al. (2008) Important role of the LKB1-AMPK pathway in suppressingtumorigenesis in PTEN-deficient mice. Biochem J 412(2):211–221.

26. Memmott RM, et al. (2010) Metformin prevents tobacco carcinogen—induced lungtumorigenesis. Cancer Prev Res (Phila) 3(9):1066–1076.

27. Shackelford DB, et al. (2013) LKB1 inactivation dictates therapeutic response of non-small cell lung cancer to the metabolism drug phenformin. Cancer Cell 23(2):143–158.

28. Laderoute KR, et al. (2006) 5′-AMP-activated protein kinase (AMPK) is induced by low-oxygen and glucose deprivation conditions found in solid-tumor microenvironments.Mol Cell Biol 26(14):5336–5347.

29. Kato K, et al. (2002) Critical roles of AMP-activated protein kinase in constitutivetolerance of cancer cells to nutrient deprivation and tumor formation. Oncogene21(39):6082–6090.

30. Ríos M, et al. (2013) AMPK activation by oncogenesis is required to maintain cancercell proliferation in astrocytic tumors. Cancer Res 73(8):2628–2638.

31. Bardella C, et al. (2012) Cells lacking the fumarase tumor suppressor are protectedfrom apoptosis through a hypoxia-inducible factor-independent, AMPK-dependentmechanism. Mol Cell Biol 32(15):3081–3094.

32. Liu L, et al. (2012) Deregulated MYC expression induces dependence upon AMPK-related kinase 5. Nature 483(7391):608–612.

33. Jeon SM, Chandel NS, Hay N (2012) AMPK regulates NADPH homeostasis to promotetumour cell survival during energy stress. Nature 485(7400):661–665.

34. Banko MR, et al. (2011) Chemical genetic screen for AMPKα2 substrates uncoversa network of proteins involved in mitosis. Mol Cell 44(6):878–892.

35. Dasgupta B, Milbrandt J (2009) AMP-activated protein kinase phosphorylatesretinoblastoma protein to control mammalian brain development. Dev Cell 16(2):256–270.

36. Daub H, et al. (2008) Kinase-selective enrichment enables quantitative phosphopro-teomics of the kinome across the cell cycle. Mol Cell 31(3):438–448.

37. Lee JH, et al. (2007) Energy-dependent regulation of cell structure by AMP-activatedprotein kinase. Nature 447(7147):1017–1020.

38. Nakano A, et al. (2010) AMPK controls the speed of microtubule polymerization anddirectional cell migration through CLIP-170 phosphorylation. Nat Cell Biol 12(6):583–590.

39. Bungard D, et al. (2010) Signaling kinase AMPK activates stress-promoted transcrip-tion via histone H2B phosphorylation. Science 329(5996):1201–1205.

40. Egan DF, et al. (2011) Phosphorylation of ULK1 (hATG1) by AMP-activated proteinkinase connects energy sensing to mitophagy. Science 331(6016):456–461.

41. Kim HS, et al. (2008) Inhibition of AMP-activated protein kinase sensitizes cancercells to cisplatin-induced apoptosis via hyper-induction of p53. J Biol Chem 283(7):3731–3742.

42. Faubert B, et al. (2013) AMPK is a negative regulator of the Warburg effect andsuppresses tumor growth in vivo. Cell Metab 17(1):113–124.

43. Sanders MJ, et al. (2007) Defining the mechanism of activation of AMP-activatedprotein kinase by the small molecule A-769662, a member of the thienopyridonefamily. J Biol Chem 282(45):32539–32548.

44. Chow LM, et al. (2011) Cooperativity within and among Pten, p53, and Rb pathwaysinduces high-grade astrocytoma in adult brain. Cancer Cell 19(3):305–316.

45. Jang T, et al. (2011) Calaoagan JM, Kwon E, Samuelsson S, Recht L, Laderoute KR. 50-AMP-activated protein kinase activity is elevated early during primary brain tumordevelopment in the rat. Int J Cancer 128(9):2230–2239.

46. Dasgupta B, et al. (2012) The AMPK β2 subunit is required for energy homeostasisduring metabolic stress. Mol Cell Biol 32(14):2837–2848.

47. Savitsky PA, Finkel T (2002) Redox regulationofCdc25C. J Biol Chem 277(23):20535–20540.48. Kim WH, et al. (2007) AICAR potentiates ROS production induced by chronic high

glucose: Roles of AMPK in pancreatic beta-cell apoptosis. Cell Signal 19(4):791–805.49. Xiao D, Johnson CS, Trump DL, Singh SV (2004) Proteasome-mediated degradation

of cell division cycle 25C and cyclin-dependent kinase 1 in phenethyl isothiocyanate-induced G2-M-phase cell cycle arrest in PC-3 human prostate cancer cells. Mol CancerTher 3(5):567–575.

50. Hsieh AC, et al. (2010) Genetic dissection of the oncogenic mTOR pathway revealsdruggable addiction to translational control via 4EBP-eIF4E. Cancer Cell 17(3):249–261.

51. Sancak Y, et al. (2007) PRAS40 is an insulin-regulated inhibitor of the mTORC1 proteinkinase. Mol Cell 25(6):903–915.

52. Vander Haar E, Lee SI, Bandhakavi S, Griffin TJ, Kim DH (2007) Insulin signalling tomTOR mediated by the Akt/PKB substrate PRAS40. Nat Cell Biol 9(3):316–323.

53. Baba Y, et al. (2010) Prognostic significance of AMP-activated protein kinase ex-pression and modifying effect of MAPK3/1 in colorectal cancer. Br J Cancer 103(7):1025–1033.

54. Vazquez-Martin A, et al. (2009) Mitotic kinase dynamics of the active form of AMPK(phospho-AMPKalphaThr172) in human cancer cells. Cell Cycle 8(5):788–791.

55. Mizrachy-Schwartz S, Cohen N, Klein S, Kravchenko-Balasha N, Levitzki A (2011) Up-regulation of AMP-activated protein kinase in cancer cell lines is mediated throughc-Src activation. J Biol Chem 286(17):15268–15277.

56. Neurath KM, Keough MP, Mikkelsen T, Claffey KP (2006) AMP-dependent proteinkinase alpha 2 isoform promotes hypoxia-induced VEGF expression in human glio-blastoma. Glia 53(7):733–743.

57. Kuznetsov JN, Leclerc GJ, Leclerc GM, Barredo JC (2011) AMPK and Akt determineapoptotic cell death following perturbations of one-carbon metabolism by regulatingER stress in acute lymphoblastic leukemia. Mol Cancer Ther 10(3):437–447.

58. Göransson O, et al. (2007) Mechanism of action of A-769662, a valuable tool for ac-tivation of AMP-activated protein kinase. J Biol Chem 282(45):32549–32560.

59. Kuhajda FP (2006) Fatty acid synthase and cancer: New application of an old pathway.Cancer Res 66(12):5977–5980.

60. Chou YW, et al. (2013) Androgens upregulate Cdc25C protein by inhibiting its pro-teasomal and lysosomal degradation pathways. PLoS ONE 8(4):e61934.

E444 | www.pnas.org/cgi/doi/10.1073/pnas.1311121111 Liu et al.

Dow

nloa

ded

by g

uest

on

Feb

ruar

y 14

, 202

1