Wnt/Snail Signaling Regulates Cytochrome c Oxidase and ...Wnt/Snail Signaling Regulates Cytochrome c...

Molecular and Cellular Pathobiology

Wnt/Snail Signaling Regulates Cytochrome c Oxidaseand Glucose Metabolism

Su Yeon Lee1, Hyun Min Jeon1, Min Kyung Ju1, Cho Hee Kim1, Gyesoon Yoon3, Song Iy Han4,Hye Gyeong Park2, and Ho Sung Kang1

AbstractWnt signaling plays a critical role in embryonic development, and its deregulation is closely linked to the

occurrence of a number of malignant tumors, including breast and colon cancer. The pathway also inducesSnail-dependent epithelial-to-mesenchymal transition (EMT), which is responsible for tumor invasion andmetastasis. In this study, we show that Wnt suppresses mitochondrial respiration and cytochrome C oxidase(COX) activity by inhibiting the expression of 3 COX subunits, namely, COXVIc, COXVIIa, and COXVIIc.We foundthatWnt induced a glycolytic switch via increased glucose consumption and lactate production, with induction ofpyruvate carboxylase (PC), a key enzyme of anaplerosis. In addition, Wnt-induced mitochondrial repressionand glycolytic switching occurred through the canonical b-catenin/T-cell factor 4/Snail pathway. Short hairpinRNA–mediated knockdown of E-cadherin, a regulator of EMT, repressed mitochondrial respiration and induceda glycolytic switch via Snail activation, indicating that EMT may contribute to Wnt/Snail regulation ofmitochondrial respiration and glucose metabolism. Together, our findings provide a new function for Wnt/Snail signaling in the regulation of mitochondrial respiration (via COX gene expression) and glucose metabolism(via PC gene expression) in tumor growth and progression. Cancer Res; 72(14); 3607–17. �2012 AACR.

IntroductionWnt signaling plays an important role inmultiple embryonic

developmental processes, and its aberrant activation occursin association with the development and progression of manyhuman cancers, including breast and colon cancer (1–3).b-Catenin plays a key role in the canonical Wnt pathway(1, 2). In the absence of Wnt, cytosolic b-catenin is associatedwith a destruction complex containing axis inhibition pro-tein 1/2 (Axin1/2), adenomatous polyposis coli (APC), andglycogen synthase kinase 3b (GSK3b). It is phosphorylated byGSK3b and is rapidly degraded by the ubiquitin-proteasomesystem. When Wnt binds to the Fzd receptor, the b-catenindestruction complex is disassembled, resulting in stabilization

of b-catenin. The stabilized b-catenin is translocated to thenucleus and forms a complex with the transcription factorsT-cell factor (TCF)/lymphoid enhancer factor to induce theexpression of Wnt/b-catenin target genes such as c-Myc andcyclin D.Wnt signaling activates an epithelial-to-mesenchymaltransition (EMT) program that is linked to invasion andmetastasis of tumor cells (4–7). Axin2, one of the Wnt targetgenes, plays an important role in this event. Axin2 acts as anucleocytoplasmic chaperone for GSK3b (6). GSK3b can phos-phorylate the EMT-inducing zinc finger transcription factor,Snail, at consensus motifs (Ser96 or Ser104/Ser107), therebycausing its degradation by the ubiquitin-proteasome system(7, 8). GSK3b localizes to both the cytosolic and nuclearcompartments. If GSK3b is moved to the cytosolic space byAxin2, Snail protein stability in the nucleus is increased andrepresses the expression of E-cadherin and other epithelialmarkers while activating the expression of mesenchymalmarkers, thereby inducing EMT (6).

Mitochondria play a central role in several cellular functionsincluding ATP production, metabolism, and apoptosis. Dysre-gulation of mitochondrial functions has been implicated in anumber of human disorders including cancer development. Intumors, mitochondrial function is downregulated by muta-tions in mitochondrial or nuclear DNA–encoding mitochon-drial proteins (9, 10) and/or by the transcription factors thatregulate mitochondrial proteins encoded by nuclear genes (11,12). Mitochondrial dysfunction favors the shift to glycolyticmetabolism, referred to as "theWarburg effect," a phenomenonwhereby cancer cells rely mainly on aerobic glycolysis togenerate ATP even in the presence of O2 (11–14). Because of

Authors' Affiliations: 1Department of Molecular Biology, College ofNatural Sciences, 2Nanobiotechnology Center, Pusan National University,Pusan; 3Department of Biochemistry and Molecular Biology, Ajou Univer-sity School of Medicine, Suwon; and 4Research Center for Resistant Cells,Chosun University, Gwangju, Korea

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

S.Y. Lee,H.M. Jeon,M.K. Ju, andC.H.Kimcontributed equally to thiswork.

Current address for C.H. Kim: DNA Identification Center, National ForensicService, Seoul 158-707, Korea.

Corresponding Author:Ho Sung Kang, Department of Molecular Biology,College of Natural Sciences, Pusan National University, Pusan 609-735,Korea. Phone: 82-51-510-2275; Fax: 82-51-513-9258; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-12-0006

�2012 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 3607

on October 20, 2020. © 2012 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst May 25, 2012; DOI: 10.1158/0008-5472.CAN-12-0006

http://cancerres.aacrjournals.org/

the low efficiency of glycolysis in generating ATP, tumor cellswith a glycolytic switch increase glucose uptake for ATPproduction, resulting in high lactate production and the acid-ification of the microenvironment, which facilitates tumor cellmigration and invasion. In addition, such alteration in glucosemetabolism confers survival advantages to tumor cells byproviding precursors for the synthesis of nucleotides, lipids,and amino acids, which are required for tumor cell growth (14,15). To support cell proliferation, tumor cells also activate aprocess termed anaplerosis, which replenishes the precursorsthat are extracted from the tricarboxylic acid (TCA) cycle bycataplerosis (16). Two enzymes, pyruvate carboxylase (PC) thatconverts pyruvate to oxaloacetate and glutaminase that con-verts glutamine to glutamate, play a critical role(s) in ana-plerosis (16–21). Thus, a glycolytic switch with higher glucoseuptake correlates withmore aggressive phenotypes and poorerclinical outcomes; however, the mechanism underlying mito-chondrial dysregulation and glycolytic switch remains largelyunknown.

Herein, we show that Wnt/Snail signaling suppresses mito-chondrial respiration and cytochrome c oxidase (COX) activity,with induction of the glycolytic switch and PC expression,providing a new function for Wnt/Snail signaling in tumorgrowth and progression.

Materials and MethodsCell lines

MCF-7, MDA-MB-231, Madin Darby Canine Kidney(MDCK), and L cells were obtained from the American TypeCulture Collection (ATCC; authenticated by short tandemrepeat profiling). The cell lines were passaged 2 timesper week after the receipt and low-passage cultures (passage5–25) were used for the experiments. Wnt3a-secretingL cells and MCF-7 #5 cells were provided by Dr. Min DS(Pusan National University, Pusan, Korea) and by Dr. Yook JI(Yonsei University, Seoul, South Korea), respectively. Thecells were routinely tested negative for mycoplasma usingthe Mycoplasma PCR Detection Kit (iNtRON Biotechnology).Detailed information for cell culture is provided in theSupplementary Methods.

Transfection and short hairpin RNA interferenceThe expression vectors pCR3.1-Snail-Flg, pCR3.1-S96A-

Snail-Flg, pCR3.1-S104, 107A-Snail-Flg, and pcDNA3.1-Wnt1(provided by Dr. Yook JI, Yonsei University) were transfectedinto MCF-7 cells using jetPEI (Polyplus-transfection). pSUPERvectors expressing short hairpin RNA (shRNA) for control,Snail, b-catenin, TCF4, Axin1, Axin2, PC, and E-cadherin weregenerated (Supplementary Table S1) and transfected asdescribed previously (22).

ProteomicsMCF-7 #5 cells were treated with doxycycline and analyzed

by quantitative differential proteome analysis using 2-dimen-sional gel electrophoresis (2-DE) and matrix-assisted laserdesorption/ionization–time-of-flight (MALDI–TOF/MS; Yon-sei Proteome Research Center, Seoul, Korea).

Assays for mitochondrial respiration, COX activity,glucose consumption, lactate production, and ATPproduction

Mitochondrial respiration and COX activity were measuredas described previously (23; for detailed information, seeSupplementary Methods). Glucose, lactate, and intracellularATP levels were determined using a Glucose Oxidation AssayKit (Sigma), a colorimetric- and fluorescence-based LactateAssay Kit (BioVision), and an ATP Bioluminescence Assay Kit(Roche), respectively. The level of ATP produced by aerobicrespiration and glycolysis was determined by measuring lac-tate production and oxygen consumption (24).

Chromatin immunoprecipitation assayChromatin immunoprecipitation (ChIP) assays were con-

ducted using a ChIP Assay Kit (Millipore). Rabbit IgG or rabbitanti-Snail (Santa Cruz Biotechnology, Inc.) were used to immu-noprecipitate DNA-containing complexes. ChIP-enrichedDNAwas analyzed by PCR using primers (Supplementary Table S2)complementary to the promoter regions containing E-box.

ResultsWnt/Snail signaling inhibits mitochondrial respiration

In this study, we examined whether Wnt regulates mito-chondrial respiration in breast cancer MCF-7 cells. Wnt3a-conditioned medium (CM, obtained from Wnt3a-secreting Lcells) induced EMT, themorphologic changes tomesenchymalcells, including loss of intercellular adhesion and formation of aspindle-like cell shape and pseudopodia (Supplementary Fig.S1A), with downregulation of E-cadherin, one of the hallmarksof EMT (Supplementary Table S3). Wnt3a CM markedlydecreased mitochondrial respiration in a time-dependentmanner with inhibition to 77% of control at 7 days (Fig. 1A).Recombinant Wnt3a (rWnt3a) protein also decreased mito-chondrial O2 consumption in MCF-7 cells (Fig. 1B). Similarresults were obtained in MCF-7 cells transfected with Wnt1,another Wnt molecule (Fig. 1C).

As shown by other investigators (6), Wnt3a CM inducednuclear accumulation of Snail proteins inMCF-7 cells (Fig. 1D).The increased nuclear Snail levels were detected at 12 hoursand persisted for up to 7 days. Interestingly, Wnt3a-inducedSnail proteins migrated as a closely spaced doublet (�30 kDa)in SDS-PAGE, possibly due to a posttranslational modificationsuch as phosphorylation. A similar result with 2 to 3 bands wasobtained in Wnt1-transfected cells (Fig. 1E). We investigatedwhether Snail is implicated in Wnt-induced mitochondrialrepression. To this end, we used 2 different Snail shRNAs,which could effectively reduce Snail levels (Fig. 1F; Supple-mentary Table S1). In addition to suppressing Wnt3a/Wnt1-induced EMT (data not shown) and E-cadherin down-regulation (Supplementary Table S3), Snail shRNA preventedWnt3a- (Fig. 1F) and Wnt1- (Fig. 1G) induced repression ofmitochondrial respiration. Wnt is known to induce nuclearSnail levels via the canonical pathway, which includes b-cate-nin, TCF4, and Axin1/2 (6). As shown in Fig. 1, knockdown ofeither b-catenin, TCF4, or Axin1/2 (Fig. 1H) prevented Wnt3a-induced nuclear accumulation of Snail (Fig. 1I), confirming the

Lee et al.

Cancer Res; 72(14) July 15, 2012 Cancer Research3608




canonicalWnt pathway–dependent Snail induction. shRNA forb-catenin, TCF4, Axin1, and Axin2 suppressed Wnt3a-inducedinhibition ofmitochondrial respiration (Fig. 1J), with inhibitionof Wnt3a-induced E-cadherin downregulation (Supplementa-ry Table S3).We examined whether Snail affects mitochondrial function.

To this end, we first used Snail tetracycline-inducibleMCF-7 #5cells (6). Upon Snail induction by doxycycline inMCF-7 #5 cells(Fig. 2A and B), the cells exhibited EMT (Supplementary Fig.S1B), with reduced expression of E-cadherin (SupplementaryTable S3). Increased Snail expression significantly suppressedmitochondrial respirationwith inhibition to 70%of control at 3to 7 days (Fig. 2C). Furthermore, we transiently transfectedMCF-7 cells with pCR3.1-Snail-Flg, and phosphorylation-resistant mutant Snail vectors, pCR3.1-S96A-Snail-Flg andpCR3.1-S104, 107A-Snail-Flg (Fig. 2D). Both wild-type andmutant Snail proteins were detected until 5 days after trans-fection (Fig. 2E). Although wild-type Snail is less stable thanSnail mutant proteins (7), its steady-state levels persisted for 5days because it was overexpressed. Transient transfectionwithSnail expression vectors significantly decreased O2 consump-tion in MCF-7 cells (Fig. 2F). Wild-type Snail could repressmitochondrial respiration to a similar extent as stablemutants.

Similarly, wild-type Snail could induce EMT similar to thatobserved in mutant Snail-overexpressing MCF-7 cells (Supple-mentary Fig. S1C), as shown by other investigators (7). Incontrast, Snail shRNA increased O2 consumption in MCF-7(Fig. 1F andG) andMDA-MB-231 cells (Fig. 2G). Taken together,our results indicate that Wnt represses mitochondrial respira-tion through the canonical b-catenin/TCF4/Snail pathway.

We examined the effects of Wnt signaling on mitochon-drial mass using Green Fluorescent MitoTracker GreenFM. Although Wnt3a and Wnt1 inhibited mitochondrialrespiration, they did not exert any significant inhibitoryeffects on mitochondrial mass (Supplementary Fig. S2),indicating that Wnt may suppress mitochondrial respira-tion without affecting mitochondrial biogenesis therebyresulting in the accumulation of nonfunctional mitochon-dria in the cells.

Wnt/Snail signaling inhibits COX activityChanges in the activity of COX, the terminal enzyme of the

mitochondrial respiratory chain, are closely associated withdysregulation of mitochondrial respiratory activity. Therefore,we examined the effects ofWnt onCOX activity.Wnt3a seemedto inhibit COX activity in MCF-7 cells in a time-dependent

Figure 1. Inhibition ofmitochondrial respiration byWnt/Snail signaling. A–C,mitochondrial respirationwasmeasured inMCF-7 cells treatedwithWnt3aCM (A)or rWnt3a (B) and inMCF-7 cells transfectedwithWnt1 (C). �,P < 0.05; ��,P < 0.01 versus control (untreated andmock). D and E,MCF-7 cells were treatedwithWnt3a CM (D) or transfected with Wnt1 (E), and the nuclear fractions were analyzed by Western blotting for Snail and histone deacetylase (HDAC1).F, MCF-7 cells were transfected with 2 different Snail shRNA constructs (#1 and #2) and analyzed for Snail knockdown by real-time PCR (left). The cells weretreated with Wnt3a CM and analyzed for mitochondrial respiration (right). G, MCF-7 cells were cotransfected with Snail shRNA and Wnt1 andanalyzed for mitochondrial respiration. ��, P < 0.01 versus control (untreated and mock), #, P < 0.05; ##, P < 0.01 versus control shRNA. H–J, MCF-7 cellswere transfected with b-catenin, TCF4, Axin1, and Axin2 shRNA and analyzed by real-time PCR using primers for its respective gene. �, P < 0.05;��,P < 0.01 versus control shRNA (H). The cells were treated withWnt3a CM. The nuclear fractions were analyzed byWestern blotting for Snail and HDAC1 (I).The cells were analyzed for mitochondrial respiration (J). ��, P < 0.01 versus untreated, #, P < 0.05; ##, P < 0.01 versus control shRNA.

Regulation of Mitochondria and Metabolism by Wnt/Snail

www.aacrjournals.org Cancer Res; 72(14) July 15, 2012 3609




manner, with inhibition to 64% of control at 7 days (Fig. 3A).Wnt1 also significantly decreased COX activity in MCF-7 cells(Fig. 3B). Snail shRNA prevented Wnt3a- (Fig. 3C) and Wnt1-(Fig. 3D) induced repression of COX activity. In addition,increased Snail expression significantly reduced COX enzy-matic activity with inhibition to 61% of control at 4 days (Fig.3E). In contrast, Snail shRNA increased COX activity in MCF-7(Fig. 3C and D) and MDA-MB-231 cells (Fig. 3F). These resultsindicate that Wnt/Snail signaling represses mitochondrialrespiration by inhibiting COX activity.

Wnt/Snail signaling represses the expression of COXVIc,COXVIIa, and COXVIIc

Eukaryotic COX is amultimeric protein complex in the innermitochondrial membrane that is composed of 13 differentsubunits encoded by 3 mitochondrial genes and 10 nucleargenes, and its assembly is regulated by the sequential action ofseveral nucleus-encoded assembly factors (SupplementaryTable S3; 25–28). Therefore, we investigated the effects of Wntand Snail on the gene expression of COX subunits and assemblyfactors using real-time PCR to identify the target genes. Wescreened the genes whose levels are altered to less than 70% ofcontrol (Supplementary Table S3). We found that Wnt1,Wnt3a, and Snail commonly decreasedmRNA levels ofCOXVIc,COXVIIa, and COXVIIc, although some other genes, such asCOXVIIb, COX18, and SCO2 are differentially regulated depend-ing on stimuli (Supplementary Table S3). Wnt3a-mediateddecrease in the mRNA levels of COXVIc, COXVIIa, and COXVIIcwas prevented by shRNA for Snail, b-catenin, and TCF4 (Sup-plementary Table S3). Western blot analysis showed thatCOXVIc, COXVIIa, and COXVIIc protein levels, but not those

of other proteins, were reduced in Wnt1-transfected, Wnt3a-treated, or Snail-overexpressing MCF-7 cells (Fig. 4A; Supple-mentary Fig. S3), indicating that Wnt/Snail is involved in theregulation of gene expression of COXVIc, COXVIIa, andCOXVIIc.

We conducted a ChIP assay to examine in vivo binding ofSnail to the promoter of COXVIc, COXVIIa, and COXVIIc.Several Snail-binding sites (E-box)were found in the promotersof these genes (Fig. 4B; Supplementary Table S4). The E-cadherin promoter region containing E-box and the COXVIIaand SCO2 exonic regions (Fig. 4B) were used as a positive andnegative control, respectively. Wnt3a enhanced the binding ofSnail to the E-box sites of the COXVIc, COXVIIa, and COXVIIcpromoters, but not to the E-box of the COXVIIb promoter (Fig.4C). Similar binding was detected in Snail-overexpressing cells(Fig. 4C). Although SCO2 protein levels were not significantlychanged by Wnt3a treatment and Snail overexpression (Fig.4A),Wnt3a induced the binding of Snail to the E-box sites of theSCO2 promoter (Fig. 4C).

In Wnt3a-treated cells, in vivo binding of Snail to thepromoters of COXVIc, COXVIIa, COXVIIc, and E-cadherin wasdetected at 12 hours and persisted for 7 days, showing thesimilar kinetics of Snail induction, the time to steady-statelevels, and E-box DNA binding. These results explain that thedecreasedmitochondrial activity byWnt3a persists 7 days aftertreatment (Supplementary Fig. S4A). In the case of Snailtransfection, E-box DNA binding by Snail mutants wasdetected at 1 to 5 days (Supplementary Fig. S4B); however,E-box DNA binding by wild-type Snail was detected at 1 to 3days, but not at 5 days (even though increased Snail levels wereobserved at this time point; Fig. 2E; Supplementary Fig. S4B).

Figure 2. Repression of mitochondrial respiration by Snail. A–C, MCF-7 #5 cells treated with 0.5 mg/mL doxycycline (DOX) were analyzed by real-timePCR (A) and Western blotting (B) for Snail expression. The cells were analyzed for mitochondrial respiration (C). D–F, MCF-7 cells transfected withwild-type, S96A mutant, or S104, 107A mutant Snail were analyzed by real-time PCR (D) and Western blotting (E) for Snail expression. GAPDH,glyceraldehyde-3-phosphate dehydrogenase. The cells were analyzed for mitochondrial respiration (F). G, mitochondrial respiration was measuredin MDA-MB-231 cells transfected with 2 different Snail shRNA constructs (#1, #2). �, P < 0.05; ��, P < 0.01 versus control (untreated, mock, andcontrol shRNA).

Lee et al.





These results indicate that Snail binding to E-box for 3 days issufficient to inhibit mitochondrial respiration and to induceEMT.

Wnt/Snail signaling induces the glycolytic switchMitochondrial dysfunction favors the emergence of the

glycolytic phenotype (12–14). Thus, we investigated whetherWnt/Snail signaling induces a glycolytic switch. Wnt3a (Fig.5A), Wnt1 (Fig. 5B), and Snail expression either by doxycyclinetreatment inMCF-7 #5 cells (Fig. 5C) or by transfection of S104,107A mutant Snail in MCF-7 cells (Fig. 5D) significantlyincreased glucose consumption and lactate production with-out affecting total ATP production. Wnt and Snail increasedthe ratio of ATP produced by glycolysis versus ATP producedby aerobic respiration (Fig. 5A–D). Snail shRNA caused adecrease in glucose consumption and lactate production inMDA-MB-231 cells with reduced glycolysis-dependent ATPproduction (Fig. 5E). These results indicate that Wnt/Snailsignaling induces a glycolytic switch.

Wnt/Snail signaling induces PCTo identify Snail-responsive targets that are linked to the

glycolytic switch, we conducted quantitative differential pro-teome analysis. Comparative analysis of 2-DE showed that ofthe 1,103 proteins analyzed, 19 were upregulated by more than3-fold and 12 were downregulated by more than 3-fold. One ofthe Snail-inducible proteins was PC (Fig. 6A), a mitochondrialenzyme that replenishes oxaloacetate to the TCA cycle byanaplerosis (16–18); its level was increased 5.5-fold by Snail.Real-time PCR and Western blot analysis show that PC levelsare increased by Snail and Wnt (Fig. 6B–D). Furthermore,Wnt3a-mediated increase in PC levels was prevented byshRNAs for Snail, b-catenin, and TCF4 (Supplementary Fig.S5A), indicating the regulation of PC by the canonical Wnt/b-catenin/TCF4/Snail pathway. To further examine whetherPC gene expression is regulated by Snail, we conducted a ChIPassay. Three E-box sequences were found in the PC promoter(Fig. 6E; Supplementary Table S4). As shown in Fig. 6F, how-ever, the binding of Snail to the E-box sites of the PC promoters

Figure 3. Inhibition of COX activity byWnt/Snail signaling. A and B, COX activity wasmeasured inMCF-7 cells treated withWnt3a CM (A) and transfected withWnt1 (B). C and D, COX activity was measured in MCF-7 cells transfected with 2 different Snail shRNA constructs (#1 and #2) and treated with Wnt3aCM (C) or cotransfected with Snail shRNA and Wnt1 (D). �, P < 0.05; ��, P < 0.01 versus control (untreated and mock), #, P < 0.05; ##, P < 0.01 versuscontrol shRNA. E and F, COX activity wasmeasured inMCF-7 #5 cells treatedwith DOX (E) and inMDA-MB-231 cells transfectedwith 2 different Snail shRNAconstructs (#1and #2; F). �, P < 0.05; ��, P < 0.01 versus control (untreated and control shRNA).






was not detected in Snail-overexpressing or Wnt3a-treatedMCF-7 cells. We examined the role(s) of PC in the growth ofWnt3a-treated MCF-7 cells. PC shRNA seemed to increase thecell doubling time for Wnt3a-treated MCF-7 cells as well as forcontrol MCF-7 cells (Supplementary Table S5), indicating thatPC plays an important role in tumor cell growth. Finally, weexamined whether the PC induction relates to the inhibition ofmitochondrial respiration and COX and the glycolytic switchby the Wnt pathway. PC supplies oxaloacetate to the TCAcycle, thereby maintaining its activity and subsequently mito-chondrial respiration, with an inhibitory effect on lactateproduction (16–18). In line with this, we observed thatshRNA-mediated PC loss (Supplementary Fig. S5B) suppressed

mitochondrial respiration/COX activity and induced a glyco-lytic switch with increased lactate production (SupplementaryFig. S5C)while it did not exert any significant effects onWnt3a-induced regulation of mitochondrial respiration and glucosemetabolism.

EMT is linked to mitochondrial respirationand the glycolytic switch

Because the EMT inducer Snail regulates mitochondrialrespiration and the glycolytic switch, we examined whetherEMT itself is linked to the regulation of mitochondrial respi-ration and the glycolytic switch. E-cadherin shRNA has beenshown to induce EMT (29). We found that in addition to

Figure 4. Inhibition of mitochondrial COXVIc,COXVIIa, andCOXVIIc expression by Wnt/Snail signaling. A, MCF-7 cells were transfected with Wnt1 or treatedwith Wnt3a CM; MCF-7 #5 cells were treated with DOX. The cell lysates were analyzed by Western blotting using the indicated antibodies. B, aschematic model of the human COXVIc, COXVIIa, COXVIIb, COXVIIc, SCO2, and E-cadherin proximal promoter regions. The black boxes representE-boxes. C, MCF-7 and MCF-7 #5 cells were treated with Wnt3a CM and DOX, respectively, and ChIP assays were conducted using rabbit IgG or rabbitanti-Snail and ChIP-enriched DNA was analyzed by PCR using primers for the indicated genes.

Lee et al.





inducing EMT (Supplementary Fig. S6A), E-cadherin shRNArepressed mitochondrial respiration and COX activity andinduced the glycolytic switch (Fig. 7A). We examined whetherSnail is involved in E-cadherin loss–induced mitochondrialrepression and glycolytic switch. E-cadherin shRNA seemedto increase Snail mRNA and protein levels (SupplementaryFig. S6B, Fig. 7B), indicating that E-cadherin, a Snail-repressiblegene, could affect Snail levels, forming a negative-feedbackloop between E-cadherin and Snail levels; the molecular

basis for E-cadherin loss–triggered Snail induction is underinvestigation. We further found that Snail shRNA preventedE-cadherin loss–induced EMT (Supplementary Fig. S6A)and vimentin (a mesenchymal marker) induction (Fig. 7C).Because downregulation of cell adhesionmolecules, other thanE-cadherin, is required for full EMT, we suspect that Snailactivated by E-cadherin shRNA may downregulate othercell adhesion molecules for full EMT to occur. Snail shRNAalso prevented E-cadherin loss–induced mitochondrial

Figure 5. Induction of the glycolytic switch byWnt/Snail signaling. A–E, glucose consumption, lactate production, and total ATP concentrationweremeasuredin MCF-7 cells treated with Wnt3a CM (A) or transfected with Wnt1 (B), in MCF-7 #5 cells treated with DOX (C), in MCF-7 cells transfected with SnailS104, 107A mutant (D), and in MDA-MB-231 cells transfected with Snail shRNA (E). �, P < 0.05; ��, P < 0.01 versus control (untreated, mock, and controlshRNA). The level of ATP produced by aerobic respiration (black bars) and glycolysis (gray bars) was determined.






repression and glycolytic switch (Fig. 7A).We examinedwheth-er E-cadherin shRNA affects the expression of COX subunitsand PC. Real-time PCR (Supplementary Fig. S6B) and West-ern blot analysis (Fig. 7C, Supplementary Fig. S6C) showedthat E-cadherin shRNA reduced the expression of COXVIcand COXVIIa, without affecting the expression of COXVIIcand PC. In addition, Snail shRNA prevented E-cadherin loss–induced downregulation of COXVIc and COXVIIa (Fig. 7C,Supplementary Fig. S6B). The ChIP assay showed thatE-cadherin shRNA enhanced the binding of Snail to the E-box sites of the COXVIc and COXVIIa promoters but not tothose of the COXVIIc promoters (Fig. 7D). Thus, under thisparticular condition, Snail could bind to COXVIc and COX-VIIa promoters but not to the promoter of COXVIIc; a moredetailed mechanism for Snail's differential binding to thepromoters remains to be elucidated. These results show thatEMT influences mitochondrial function and glucose metab-

olism via Snail activation, suggesting that EMT may con-tribute to Wnt/Snail regulation of mitochondria/metabo-lism (Supplementary Fig. S7).

DiscussionWnt/Snail signaling is known to induce EMT, an essential

process for tumor invasion and metastasis (4–7). In this study,we show a new function of theWnt/Snail signaling in suppres-sing mitochondrial respiration and COX activity in MCF-7 andMDA-MB-231 cells (Figs. 1–3). Similar results were observed inMDCK cells (data not shown). COX regulation occurredthrough inhibition of the expression of COXVIc, COXVIIa, andCOXVIIc. In addition, Wnt/Snail signaling induced a glycolyticswitch (Fig. 5). Recently, b-catenin, a downstream mediator ofthe canonicalWnt pathway, has been shown to downregulate 2mitochondrial ATPase subunits (ATP5a1 and ATP5b; 30), and

Figure6. Induction of PCbyWnt/Snail signaling. A,MCF-7 #5cells treatedwithDOXwere analyzedbyquantitative differential proteomics. B–D,PC levelswereanalyzed by real-time PCR and Western blotting in MCF-7 #5 treated with DOX (B) and in MCF-7 cells treated with Wnt3a CM (C) or transfected withWnt1 (D). �, P < 0.05; ��, P < 0.01 versus control (untreated and mock). E, a schematic model of the human PC proximal promoter region. The black boxesrepresent E-boxes. F,MCF-7#5 cellswere treatedwithDOX;MCF-7 cellswere treatedwithWnt3aCM.ChIP assayswere conducted using rabbit IgGor rabbitanti-Snail and ChIP-enriched DNA was analyzed by PCR using primers for the indicated genes.

Lee et al.





to induce the expression of genes involved in glucose/gluta-mine metabolism, including lactate dehydrogenase (LDH; 30),glucose-6-phosphatase, phosphoenolpyruvate carboxykinase(31), glutamine synthetase, ornithine aminotransferase, andglutamate transporter GLT-1 (32). Thus, Wnt signaling seemsto affect mitochondrial respiration and glucose/glutaminemetabolism by b-catenin- and Snail-dependent mechanisms.Wnt/Snail signaling seems to regulate PC, amajor anaplero-

tic enzyme (Fig. 6). PC protein is known to be activated orhighly expressed in human lung tumors (33) and in hepatictumors in rats (34). We observed that PC shRNA suppressedmitochondrial respiration/COX activity and induced the gly-colytic switch with lactate production (Supplementary Fig.S5C). Similar enhanced lactate production was observed incells treated with PC shRNA (35). Although Wnt induces PC, itincreased lactate production (Fig. 5), possibly due to anincrease in LDH levels through b-catenin, which is activatedin response to Wnt (30). Thus, pyruvate seems to be convertedeither to lactate by LDH or to oxaloacetate by PC (Supple-mentary Fig. S7). Although PC could positively regulate theTCA cycle and mitochondrial respiration, mitochondrial func-tion is inhibited by the Wnt/b-catenin/Snail pathway; thus,

only the TCA cycle may be active. The active TCA cycle couldsupply substrates into biosynthetic pathways for tumor cellgrowth (Supplementary Fig. S7). However, because Wnt3a(Supplementary Table S5) and Snail (36) inhibit cell growthwith EMT induction in MCF-7 cells, PC may be implicated inother aspects of cell regulation than cell proliferation. Besidessupplying substrates into biosynthetic pathways for cell pro-liferation (16), PC-mediated anaplerosis is also important forother biologic processes; for example, PC is necessary forglucose-induced insulin secretion by providing oxaloacetateto formmalate in islets and for de novo synthesis of glutamate,an important excitatory neurotransmitter in astrocytes (18). Atpresent, the precise role(s) of PC in Wnt3a-induced cell reg-ulation remains to be characterized.

In most cases, Snail acts as a repressor; however, it also actsas a transcriptional activator (37). Because Snail binds to the E-box in the COXVIc, COXVIIa, and COXVIIc promoter regions(Fig. 4C), these genes are likely to be directly regulated by Snail.In the case of PC, although E-boxes were found in the PC genepromoter, Snail did not bind to the promoter (Fig. 6F). Thus,other Snail-binding elements may be responsible for PC upre-gulation; otherwise, Snail may induce PC expression through

Figure 7. Inhibition of mitochondrial respiration and induction of the glycolytic switch by E-cadherin knockdown. A, MCF-7 cells were cotransfected withE-cadherin shRNA and Snail shRNA and analyzed for mitochondrial respiration, COX activity, glucose consumption, lactate production, and total ATPconcentration. �, P < 0.05; ��, P < 0.01 versus control shRNA; ##, P < 0.01 versus E-cadherin shRNA alone. The level of ATP produced by aerobicrespiration (black bars) and glycolysis (gray bars) was determined. B and C, MCF-7 cells were cotransfected with E-cadherin shRNA and Snail shRNA.The nuclear fractions (B) or whole-cell lysates (C) were analyzed by Western blotting using the indicated antibodies. D, MCF-7 cells were transfectedwith E-cadherin shRNA. ChIP assayswere conducted using rabbit IgG or rabbit anti-Snail andChIP-enriched DNAwas analyzed by PCRusing primers for theindicated genes.






an indirect effect via other transcription factors and/orcoactivators/corepressors.

Tumor cells with dysregulated mitochondria and the gly-colytic switch have been shown to exert apoptotic resistance(38) and exhibit necrosis that could promote tumor progres-sion and aggressiveness upon metabolic stress (39, 40). Wnt(41) and Snail (36, 42, 43) are known to inhibit apoptosis.Furthermore, Snail regulates necrosis in response tometabolicstress (22). Thus, mitochondrial repression by Wnt/Snail maycontribute to its known antiapoptotic and pronecroticactivities.

Snail levels are regulated by cytokines such as TGF-b andtranscription factors such as hypoxia-inducible factor-1, NF-kB, and p53 (44–46). Therefore, these signals may regulatemitochondrial respiration and glucose metabolism via Snailactivation. In fact, TGF-b has been shown to inhibit mitochon-drial respiration (23). In this study, we also showed thatE-cadherin shRNA, acting as an EMT inducer, repressed mito-chondrial respiration and COX activity and induced the gly-colytic switch (Fig. 7A), indicating that EMTmay contribute tothe regulation of mitochondrial function and metabolism byWnt/Snail and possibly other EMT inducers. Interestingly,Snail was found to be implicated in E-cadherin loss–inducedcell regulation (Fig. 7). Recently, Onder and colleagues (2008)showed that E-cadherin loss induces EMT via activiation ofb-catenin and an EMT inducer, Twist (29); thus, E-cadherinloss seems to activate multiple factors, including Snail, b-cate-nin, and Twist, to elicit diverse biologic responses includingEMT and regulation of mitochondria/metabolism.

EMT has been shown to induce generation of cancer stemcells (47–49); Snail is also known to induce acquisition of astem cell–like phenotype in ovarian cancer cells (43). Inaddition, human embryonic stem cells exhibit a high glycolytic

rate similar to cancer cells (50). Thus, Snail-induced mito-chondrial repression and glycolytic switch may be associatedwith alterations in mitochondrial function and metabolism ofembryonic and cancer stem cells.

In summary, we show a new function of theWnt/Snail signalpathway in the regulation of mitochondrial function andglucose metabolism. These findings provide a novel mecha-nistic insight into the mechanism of Wnt/Snail-mediatedtumor progression and aggressiveness and how the Wnt/Snailsignal pathway causes a survival advantage and withstandstherapeutic agents.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: S.Y. Lee, H.M. Jeon, M.K. Ju, C.H. Kim, H.S. KangAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): S.Y. Lee, H.M. Jeon, M.K. Ju, C.H. KimAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): S.Y. Lee, H.M. Jeon, M.K. Ju, C.H. Kim, G. Yoon, S.I.Han, H.G. Park, H.S. KangWriting, review, and/or revision of themanuscript: S.Y. Lee, H.M. Jeon,M.K.Ju, C.H. Kim, H.S. KangAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): H.S. KangStudy supervision: H.S. Kang

Grant SupportThis work was supported by the Bio-Scientific Research Grant funded by the

Pusan National University (PNU, Bio-Scientific Research Grant; PNU-2008-101-103).

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicate thisfact.

Received January 4, 2012; revised April 20, 2012; accepted May 8, 2012;published OnlineFirst May 25, 2012.

References1. Klaus A, Birchmeier W. Wnt signalling and its impact on development

and cancer. Nat Rev Cancer 2008;8:387–98.2. Reya T, Clevers H. Wnt signalling in stem cells and cancer. Nature

2005;434:843–50.3. Li Y, Welm B, Podsypanina K, Huang S, Chamorro M, Zhang X, et al.

Evidence that transgenes encoding components of the Wnt signalingpathwaypreferentially inducemammary cancers fromprogenitor cells.Proc Natl Acad Sci U S A 2003;100:15853–8.

4. Thiery JP, Sleeman JP. Complex networks orchestrate epithelial-mesenchymal transitions. Nat Rev Mol Cell Biol 2006;7:131–42.

5. Huber MA, Kraut N, Beug H. Molecular requirements for epithelial-mesenchymal transition during tumor progression. Curr Opin Cell Biol2005;17:548–58.

6. Yook JI, Li XY, Ota I, Hu C, Kim HS, Kim NH, et al. A Wnt-Axin2-GSK3beta cascade regulates Snail1 activity in breast cancer cells. NatCell Biol 2006;8:1398–406.

7. Yook JI, Li XY, Ota I, Fearon ER, Weiss SJ. Wnt-dependent regulationof the E-cadherin repressor snail. J Biol Chem 2005;280:11740–8.

8. ZhouBP,DengJ, XiaW,Xu J, Li YM,GunduzM, et al. Dual regulation ofSnail by GSK-3beta-mediated phosphorylation in control of epithelial-mesenchymal transition. Nat Cell Biol 2004;6:931–40.

9. Brandon M, Baldi P, Wallace DC. Mitochondrial mutations in cancer.Oncogene 2006;25:4647–62.

10. Gottlieb E, Tomlinson IP.Mitochondrial tumour suppressors: a geneticand biochemical update. Nat Rev Cancer 2005;5:857–66.

11. Gogvadze V, Orrenius S, Zhivotovsky B. Mitochondria in cancer cells:what is so special about them? Trends Cell Biol 2008;18:165–73.

12. DenkoNC. Hypoxia, HIF1 and glucosemetabolism in the solid tumour.Nat Rev Cancer 2008;8:705–13.

13. Warburg O. On the origin of cancer cells. Science 1956;123:309–14.14. Hsu PP, Sabatini DM. Cancer cell metabolism: Warburg and beyond.

Cell 2008;134:703–7.15. Vander Heiden MG, Cantley LC, Thompson CB. Understanding the

Warburg effect: the metabolic requirements of cell proliferation. Sci-ence 2009;324:1029–33.

16. Owen OE, Kalhan SC, Hanson RW. The key role of anaplerosis andcataplerosis for citric acid cycle function. J Biol Chem 2002;277:30409–12.

17. Jitrapakdee S, St Maurice M, Rayment I, Cleland WW, Wallace JC,Attwood PV. Structure, mechanism and regulation of pyruvate car-boxylase. Biochem J 2008;413:369–87.

18. Jitrapakdee S, Vidal-Puig A, Wallace JC. Anaplerotic roles of pyru-vate carboxylase in mammalian tissues. Cell Mol Life Sci 2006;63:843–54.

19. Wang JB, Erickson JW, Fuji R, Ramachandran S, Gao P, Dinavahi R,et al. Targeting mitochondrial glutaminase activity inhibits oncogenictransformation. Cancer Cell 2010;18:207–19.

20. Gao P, Tchernyshyov I, Chang TC, Lee YS, Kita K, Ochi T, et al. c-Mycsuppression of miR-23a/b enhances mitochondrial glutaminaseexpression and glutamine metabolism. Nature 2009;458:762–5.

Lee et al.





21. Cheng T, Sudderth J, Yang C, Mullen AR, Jin ES, Mates JM, et al.Pyruvate carboxylase is required for glutamine-independentgrowth of tumor cells. Proc Natl Acad Sci U S A 2011;108:8674–9.

22. KimCH, JeonHM, LeeSY, JuMK,Moon JY, Park HG, et al. Implicationof snail in metabolic stress-induced necrosis. PLoS One 2011;6:e18000.

23. Yoon YS, Lee JH, Hwang SC, Choi KS, Yoon G. TGF beta1 inducesprolongedmitochondrial ROS generation through decreased complexIV activity with senescent arrest in Mv1Lu cells. Oncogene 2005;24:1895–903.

24. Sariban-Sohraby S, Magrath IT, Balaban RS. Comparison of energymetabolism in human normal and neoplastic (Burkitt's lymphoma)lymphoid cells. Cancer Res 1983;43:4662–4.

25. Herrmann PC, Herrmann EC. Oxygen metabolism and a potential rolefor cytochrome c oxidase in the Warburg effect. J Bioenerg Biomembr2007;39:247–50.

26. Pecina P, Houstkova H, Hansikova H, Zeman J, Houstek J. Geneticdefects of cytochromecoxidaseassembly. Physiol Res2004;53Suppl1:S213–23.

27. Sacconi S, Trevisson E, Pistollato F, Baldoin MC, Rezzonico R,Bourget I, et al. hCOX18 and hCOX19: two human genes involved incytochrome c oxidase assembly. Biochem Biophys Res Commun2005;337:832–9.

28. Xu F, Morin C, Mitchell G, Ackerley C, Robinson BH. The role of theLRPPRC (leucine-rich pentatricopeptide repeat cassette) gene incytochrome oxidase assembly: mutation causes lowered levels ofCOX (cytochrome c oxidase) I and COX III mRNA. Biochem J2004;382:331–6.

29. Onder TT, Gupta PB, Mani SA, Yang J, Lander ES, Weinberg RA. Lossof E-cadherin promotesmetastasis viamultiple downstream transcrip-tional pathways. Cancer Res 2008;68:3645–54.

30. Chafey P, Finzi L, Boisgard R, Cauzac M, Clary G, Broussard C, et al.Proteomic analysis of beta-catenin activation in mouse liver by DIGEanalysis identifies glucose metabolism as a new target of the Wntpathway. Proteomics 2009;9:3889–900.

31. Liu H, Fergusson MM, Wu JJ, Rovira II, Liu J, Gavrilova O, et al. Wntsignaling regulates hepatic metabolism. Sci Signal 2011;4:ra6.

32. Cadoret A, Ovejero C, Terris B, Souil E, Levy L, Lamers WH, et al. Newtargets of beta-catenin signaling in the liver are involved in the gluta-mine metabolism. Oncogene 2002;21:8293–301.

33. Fan TW, Lane AN, Higashi RM, Farag MA, Gao H, Bousamra M, et al.Altered regulation of metabolic pathways in human lung cancer dis-cerned by (13)C stable isotope-resolved metabolomics (SIRM). MolCancer 2009;8:41.

34. Liu KJ, Kleps R, Henderson T, Nyhus L. 13C NMR study of hepaticpyruvate carboxylase activity in tumor rats. Biochem Biophys ResCommun 1991;179:366–71.

35. Hasan NM, Longacre MJ, Stoker SW, Boonsaen T, Jitrapakdee S,Kendrick MA, et al. Impaired anaplerosis and insulin secretion in

insulinoma cells caused by small interfering RNA-mediated suppres-sion of pyruvate carboxylase. J Biol Chem 2008;283:28048–59.

36. Vega S, Morales AV, Ocana OH, Valdes F, Fabregat I, Nieto MA. Snailblocks the cell cycle and confers resistance to cell death. Genes Dev2004;18:1131–43.

37. Du F, Nakamura Y, Tan TL, Lee P, LeeR, YuB, et al. Expression of snailin epidermal keratinocytes promotes cutaneous inflammation andhyperplasia conducive to tumor formation. Cancer Res 2010;70:10080–9.

38. Pelicano H, Xu RH, Du M, Feng L, Sasaki R, Carew JS, et al. Mito-chondrial respiration defects in cancer cells cause activation of Aktsurvival pathway through a redox-mediated mechanism. J Cell Biol2006;175:913–23.

39. Zong WX, Thompson CB. Necrotic death as a cell fate. Genes Dev2006;20:1–15.

40. Vakkila J, Lotze MT. Inflammation and necrosis promote tumourgrowth. Nat Rev Immunol 2004;4:641–8.

41. Chen S,GuttridgeDC, YouZ, Zhang Z, Fribley A,MayoMW, et al.Wnt-1 signaling inhibits apoptosis by activating beta-catenin/T cell factor-mediated transcription. J Cell Biol 2001;152:87–96.

42. EscrivaM, Peiro S, Herranz N, Villagrasa P, Dave N,Montserrat-SentisB, et al. Repression of PTEN phosphatase by Snail1 transcriptionalfactor during gamma radiation-induced apoptosis. Mol Cell Biol2008;28:1528–40.

43. Kurrey NK, Jalgaonkar SP, Joglekar AV, Ghanate AD, Chaskar PD,Doiphode RY, et al. Snail and slug mediate radioresistance andchemoresistance by antagonizing p53-mediated apoptosis andacquiring a stem-like phenotype in ovarian cancer cells. Stem Cells2009;27:2059–68.

44. Luo D, Wang J, Li J, Post M. Mouse snail is a target gene for HIF. MolCancer Res 2011;9:234–45.

45. Wu Y, Deng J, Rychahou PG, Qiu S, Evers BM, Zhou BP. Stabilizationof snail by NF-kappaB is required for inflammation-induced cell migra-tion and invasion. Cancer Cell 2009;15:416–28.

46. KimNH,KimHS, Li XY, Lee I, Choi HS, KangSE, et al. A p53/miRNA-34axis regulates Snail1-dependent cancer cell epithelial-mesenchymaltransition. J Cell Biol 2011;195:417–33.

47. Mani SA, Guo W, Liao MJ, Eaton EN, Ayyanan A, Zhou AY, et al. Theepithelial-mesenchymal transition generates cells with properties ofstem cells. Cell 2008;133:704–15.

48. Polyak K, Weinberg RA. Transitions between epithelial and mesen-chymal states: acquisition of malignant and stem cell traits. Nat RevCancer 2009;9:265–73.

49. Singh A, Settleman J. EMT, cancer stem cells and drug resistance: anemerging axis of evil in the war on cancer. Oncogene 2010;29:4741–51.

50. ZhouW, Choi M, Margineantu D, Margaretha L, Hesson J, CavanaughC, et al. HIF1alpha induced switch from bivalent to exclusively glyco-lytic metabolism during ESC-to-EpiSC/hESC transition. EMBO J2012;31:2103–16.






2012;72:3607-3617. Published OnlineFirst May 25, 2012.Cancer Res Su Yeon Lee, Hyun Min Jeon, Min Kyung Ju, et al. Metabolism

Oxidase and GlucosecWnt/Snail Signaling Regulates Cytochrome

Updated version

10.1158/0008-5472.CAN-12-0006doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2012/05/25/0008-5472.CAN-12-0006.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/72/14/3607.full#ref-list-1

This article cites 50 articles, 19 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/72/14/3607.full#related-urls

This article has been cited by 2 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/72/14/3607To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-12-0006

http://cancerres.aacrjournals.org/content/suppl/2012/05/25/0008-5472.CAN-12-0006.DC1

http://cancerres.aacrjournals.org/content/72/14/3607.full#ref-list-1

http://cancerres.aacrjournals.org/content/72/14/3607.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/72/14/3607


Wnt/Snail Signaling Regulates Cytochrome c Oxidase and ...Wnt/Snail Signaling Regulates Cytochrome c...

Documents

Transcript of Wnt/Snail Signaling Regulates Cytochrome c Oxidase and ...Wnt/Snail Signaling Regulates Cytochrome c...