Metformin induces microRNA-34a to downregulate the Sirt1/Pgc-1α/Nrf2 pathway, leading to increased...

Original Contribution

Metformin induces microRNA-34a to downregulate theSirt1/Pgc-1α/Nrf2 pathway, leading to increased susceptibility ofwild-type p53 cancer cells to oxidative stress and therapeutic agents

Minh Truong Do, Hyung Gyun Kim, Jae Ho Choi, Hye Gwang Jeong nQ1

Department of Toxicology, College of Pharmacy, Chungnam National University, Daejeon 305-764, Republic of Korea

a r t i c l e i n f o

Article history:Received 25 March 2014Received in revised form10 June 2014Accepted 14 June 2014

Keywords:MetforminMiR-34aSirt1Nrf2ApoptosisOxidative stressFree radicals

a b s t r a c t

Sirtuin 1 (Sirt1) plays an important role in cellular redox balance and resistance to oxidative stress. Sirt1exhibits oncogenic properties in wild-type p53 cancer cells, whereas it acts as a tumor suppressor inp53-mutated cancer cells. Here, we investigated the effects of metformin on Sirt1 expression in severalcancer cell lines. Using human cancer cell lines that exhibit differential expression of p53, we found thatmetformin reduced Sirt1 protein levels in cancer cells bearing wild-type p53, but did not affect Sirt1protein levels in cancer cell lines harboring mutant forms of p53. Metformin-induced p53 protein levelsin wild-type p53 cancer cells resulted in upregulation of microRNA (miR)-34a. The use of a miR-34ainhibitor confirmed that metformin-induced miR-34a was required for Sirt1 downregulation. Metforminsuppressed peroxisome proliferator-activated receptor γ (PPARγ) coactivator-1α (Pgc-1α) expressionand its downstream target Nrf2 in MCF-7 cells. Genetic tools demonstrated that the reduction of Sirt1and Pgc-1α by metformin caused Nrf2 downregulation via suppression of PPARγ transcriptional activity.Metformin reduced heme oxygenase-1 and superoxide dismutase 2 but upregulated catalase expressionin MCF-7 cells. Metformin-treated MCF-7 cells had no increase in basal levels of reactive oxygen speciesbut were more susceptible to oxidative stress. Furthermore, upregulation of death receptor 5 bymetformin-mediated Sirt1 downregulation enhanced the sensitivity of wild-type p53 cancer cells toTRAIL-induced apoptosis. Our results demonstrated that metformin induces miR-34a to suppress theSirt1/Pgc-1α/Nrf2 pathway and increases susceptibility of wild-type p53 cancer cells to oxidative stressand TRAIL-induced apoptosis.

& 2014 Elsevier Inc. All rights reserved.

Sirtuin 1 (Sirt1)1, a mammalian NADþ-dependent histone deace-tylase, is involved in diverse cellular processes such as metabolism,cellular redox balance, resistance to oxidative stress, aging, oncogen-esis, and cancer development [1,2]. Sirt1 regulates important tran-scription factors such as p53 [3,4], peroxisome proliferator-activatedreceptor γ coactivator 1α (Pgc-1α) [5], forkhead homeobox type O(FOXO) proteins [6], and nuclear erythroid factor 2-related factor 2(Nrf2) [7], which regulates the transcription of pro- and antioxidantenzymes, by which the cellular redox state is affected [1].

Sirt1 plays a critical role in cancer initiation, progression, anddrug resistance by blocking senescence and apoptosis and pro-moting cancer cell growth and angiogenesis [2,8] through inhibi-tion of the tumor suppressor p53, FOXO1, and Ku70-mediatedfunctions [3,4,9,10]. Sirt1 is overexpressed in human breast, colon,non-small-cell lung, and prostate cancer cells [9,11] and a sirtinol-or nicotinamide-specific inhibitor of Sirt1 increased senescence-like growth arrest in human breast, lung, and prostate cancer cells[9,12]. Downregulation of Sirt1 by antisense oligonucleotidesinhibited the growth and viability of human prostate cancer [9],induced apoptosis, and enhanced radiation-induced antiprolifera-tive effects in human lung cancer cells [13]. Moreover, pharmaco-logical inhibition of Sirt1 or Sirt1 knockdown induced apoptosis inleukemia stem cells and suppressed growth in vitro and in vivo[14,15]. Sirt1-knockout mice exhibited p53 hyperacetylation,and Sirt1-deficient cells enhanced radiation-induced thymocyteapoptosis, suggesting that Sirt1 can facilitate tumor growth bysuppressing p53 function [16]. However, several studies haveshown that Sirt1 has tumor-suppressive effects. Activation ofSirt1 by resveratrol inhibited growth of BRCA1-deficient and

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

67686970717273747576777879808182838485868788899091

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/freeradbiomed

Free Radical Biology and Medicine

http://dx.doi.org/10.1016/j.freeradbiomed.2014.06.0100891-5849/& 2014 Elsevier Inc. All rights reserved.

Abbreviations: ChIP, chromatin immunoprecipitation; CHOP, C/EBP homologyprotein; DR5, death receptor 5; HO-1, heme oxygenase-1; IP, immunoprecipitation;miR-34a, microRNA-34a; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazo-lium bromide; Nrf2, nuclear erythroid factor 2-related factor 2; PARP, poly(ADPribose) polymerase; Pgc-1α, peroxisome proliferator-activated receptor γcoactivator-1α; PPARγ, peroxisome proliferator-activated receptor γ; PPRE, PPAR-responsive element; ROS, reactive oxygen species; SOD2, superoxide dismutase 2;Sirt1, sirtuin 1; TNF, tumor necrosis factor; TRAIL, tumor necrosis factor-relatedapoptosis-inducing ligand

n Corresponding author.E-mail address: [email protected] (H. Gwang Jeong).

Please cite this article as: Truong Do, M; et al. Metformin induces microRNA-34a to downregulate the Sirt1/Pgc-1α/Nrf2 pathway,leading to increased.... Free Radic. Biol. Med. (2014), http://dx.doi.org/10.1016/j.freeradbiomed.2014.06.010i

Free Radical Biology and Medicine ∎ (∎∎∎∎) ∎∎∎–∎∎∎

www.sciencedirect.com/science/journal/08915849

www.elsevier.com/locate/freeradbiomed

http://dx.doi.org/10.1016/j.freeradbiomed.2014.06.010



mailto:[email protected]





p53þ /� tumor cells [17] and reduced tumorigenesis in p53þ /�

mice [18]. Paradoxically, a recent study reported that resveratrolinduced apoptosis in wild-type p53 Hodgkin lymphoma cells,which was related to Sirt1 inhibition, p53, and FOXO3a hyperace-tylation [6]. These results suggest that Sirt1 has an oncogeniceffect in cells expressing wild-type p53 but has a tumor-suppressive effect in mutated p53 cells. Therefore, downregulationof Sirt1 expression or inhibition of Sirt1 activity might be aneffective approach to the treatment of cancers harboring wild-type p53.

Nrf2 is a redox-sensitive transcription factor regulating theexpression of a battery of cytoprotective genes, including hemeoxygenase-1 (HO-1) and antioxidative enzymes such as super-oxide dismutase 2 (SOD2) [19]. Constitutive Nrf2 activation inmany tumors enhances cell survival and resistance to anticancerdrugs [20]. Nrf2 expression is positively regulated by Sirt1 [7], andNrf2 was shown to be involved in tumor necrosis factor-relatedapoptosis-inducing ligand (TRAIL) resistance in cancer cells [21].Downregulation of Sirt1 resulted in induction of death receptor 5(DR5) and sensitized cancer cells to TRAIL-induced apoptosis [10].In this study, we investigated the effect of metformin on Sirt1expression in several cancer cell lines expressing different forms ofp53. Our results indicate that metformin upregulates miR-34ato downregulate the Sirt1/Pgc-1α/Nrf2 pathway and increasessusceptibility of wild-type p53 cancer cells to oxidative stress.Metformin induces C/EBP homology protein (CHOP) and DR5expression, enhancing TRAIL-induced apoptosis in wild-type p53cancer cells.

Materials and methods

Materials

Metformin, TRAIL, and hydrogen peroxide (H2O2) were pur-chased from Sigma Chemical Co. (St. Louis, MO, USA). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)was purchased from USB Corp. (Cleveland, OH, USA), and a lactatedehydrogenase (LDH) release detection kit was obtained fromRoche Applied Science (Indianapolis, IN, USA). The plasmid pCMV-β-gal was obtained from Clontech (Palo Alto, CA, USA). Lipofecta-mine 2000 and nitrocellulose membranes were purchased fromInvitrogen (Carlsbad, CA, USA). Oligonucleotide polymerase chainreaction (PCR) primers were custom-synthesized by Bioneer(Seoul, South Korea). A protein assay kit was purchased fromBio-Rad Laboratories (Hercules, CA, USA). Antibodies against Sirt1,Pgc-1, PPARγ, catalase, SOD2, DR5, lamin B1, and β-actin wereobtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA).Anti-HO-1 antibody was obtained from Calbiochem (San Diego,CA, USA). Antibodies against p53, acetyl-p53, and poly(ADP-ribose) polymerase (PARP), as well as the secondary antibodieshorseradish peroxidase (HRP)-linked anti-rabbit and anti-mouseIgG were purchased from Cell Signaling Technologies (Beverly, MA,USA). An antibody against Nrf2 was obtained from Abcam (Cam-bridge, MA, USA). All other chemicals and reagents were ofanalytical grade.

Cell culture and treatment

The human breast cancer cell line MCF-7 and human lungadenocarcinoma cell line A549, which contain wild-type p53;MDA-MB-231 cells, which harbor mutant p53; and the p53-nullhuman ovarian cancer line SKOV3 were obtained from the Amer-ican Type Culture Collection (Rockville, MD, USA). The humancolon cancer cell line HCT 116 containing wild-type p53 andp53-knockout HCT 116 cells (HCT 116 p53�/�) were a kind gift

from Dr. Vogelstein at Johns Hopkins University (Baltimore, MD,USA). All cells were cultured in the appropriate RPMI 1640,Dulbecco’s modified Eagle’s, or McCoy's 5A medium in a humidi-fied 5% CO2 incubator at 37 1C in complete medium supplementedwith 10% heat-inactivated fetal bovine serum (FBS) (Invitrogen) to70–80% confluence. Metformin was dissolved in water. TRAILwas dissolved in phosphate-buffered saline containing 0.1% bovineserum albumin. The working concentrations were added directlyto serum-free culture medium. Control cells were treated withvehicle only.

Measurement of cell viability and cytotoxicity

MCF-7 cells were cultured in medium containing 10% FBS in96-well plates at 37 1C. After incubation for 24 h, the growthmediumwas renewed with serum-free medium and the cells werepretreated with various concentrations of metformin for 24 h,followed by treatment with H2O2 or TRAIL for an additional 24 or48 h at 37 1C, respectively. After treatment, the cells were treatedwith MTT solution (final concentration, 0.5 mg/ml) for 1 h. Thedark blue formazan crystals formed in intact cells were solubilizedwith dimethyl sulfoxide, and absorbance at 570 nm was measuredwith a microplate reader (Varioskan; Thermo Electron, Waltham,MA, USA). Cell supernatants were used in lactate dehydrogenaseassays, with measurement of absorbance at 490 nm using amicroplate reader (Varioskan; Thermo Electron). Percentage cellviability or cytotoxicity was calculated based on absorbancerelative to that of vehicle-treated control cells

Luciferase and β-galactosidase assays

Cells were transfected with 0.5 μg of PPRE-luciferase vectorand/or 0.2 μg of pCMV-β-gal per well using Lipofectamine 2000.The medium was renewed 6 h after transfection. The transfectedcells were treated with metformin (1–5 mM) for 24 h and lysed.The lysed cell preparations were then centrifuged (12,000 rpm,10 min), and the supernatant was assayed for luciferase and β-galactosidase activities. Relative luciferase activity was determinedby normalization to β-galactosidase activity.

Quantitative real-time RT-PCR (qRT-PCR)

After 24 or 48 h of treatment with 1–5 mM metformin, totalRNA was isolated from untreated and metformin-treated cellsusing RNAiso Plus reagent (Takara, Tokyo, Japan) according tothe manufacturer’s protocol. The concentration and purity ofextracted RNA were measured using a Nanodrop instrument. AfterRNA isolation, cDNA was synthesized using a reverse transcriptionkit (Promega). Product formation during PCR was monitoredcontinuously using Sequence Detection System software (version1.7; Applied Biosystems, Foster City, CA, USA) by monitoringincreases in reporter dye (SYBR) signals. The mRNA levels of Pgc-1α, Nrf2, CHOP, and DR5 in metformin-treated cells were com-pared with those in control cells at each time point using thecomparative cycle threshold (Ct) method. The following primerswere used: human Pgc-1α forward, 50-AACAGCAGCAGAGA-CAAATGCACC-30, reverse, 50-TGCAGTTCCAGAGAGTTCCACACT-30;human Nrf2 forward, 50-TACTCCCAGGTTGCCCACA-30, reverse, 50-CATCTACAAACGGGAATGTCTGC-30; CHOP forward, 50-CAACTGCA-GAGATGGCAGCT-30, reverse, 50-CTGATGCTCCCAATTGTTCA-30; DR5forward, 50-GCCCCACAACAAAAGAGGTC-30, reverse, 50-GGAGGT-CATTCCAGTGAGTG-30; 18S rRNA forward, 50-GCTGGAAT-TACCGCGGCT-30, and reverse, 50-CGGCTACCACATCCAAGGAA-30.The quantity of each transcript was calculated as described inthe instrument manual and normalized to the mRNA level of the18S rRNA housekeeping gene.

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

M. Truong Do et al. / Free Radical Biology and Medicine ∎ (∎∎∎∎) ∎∎∎–∎∎∎2





For detection of miR-34a, total RNA was extracted from thecells using the miRNeasy Mini kit (Qiagen) according to themanufacturer’s instructions. Total RNA was reverse transcribedusing the QuantiTect Reverse Transcription kit (Qiagen). Levels ofmiR-34a expression were detected using the miScript SYBR GreenPCR kit (Qiagen) with specific primers for miR-34a (Bioneer).Values for miR-34a expression were normalized using specificprimers to RNU6B (Bioneer) as an endogenous reference RNA.

Preparation of nuclear extracts and Western blotting

Nuclear protein extraction was performed according to ourprevious study [22]. Whole-cell lysates were prepared in lysisbuffer (50 mM Tris–HCl, pH 7.4, 1% NP-40, 0.25% sodium deox-ycholate, 150 mM NaCl, 1 mM EDTA), and the nuclear proteinfraction was prepared for Western blot. Protein concentrationswere determined at 595 nm using the Bio-Rad protein assay kit.The denatured protein samples were separated by SDS–PAGE on a10% polyacrylamide gel and blotted onto nitrocellulose mem-branes. The membranes were blocked with 5% skim milk andthen incubated with the appropriate primary antibodies, followedby an HRP-conjugated secondary antibody. The blots were visua-lized using an enhanced chemiluminescence Western blot kitaccording to the manufacturer’s protocol. To investigate multipleprotein targets under the same treatment conditions, the blot wasstripped and reused. Equal sample loading was confirmed bymeasuring β-actin levels in whole-cell lysates and lamin B1 innuclear fractions. The integrated optical density for the proteinband was calculated by ImageJ software, and the values werenormalized to the housekeeping gene β-actin or lamin B1.

Immunoprecipitation (IP)

For IP, MCF-7 cells were grown to 70% confluence and subse-quently treated with metformin in fresh medium for 24 h. Thecells were harvested and nuclear extracts prepared with a com-mercial kit according to the manufacturer’s instructions. Thenuclear fraction was precleared with 30 μl of protein G plus/protein A agarose for 1 h at 4 1C. IP was performed at 4 1C for3 h by incubation of 1 μg anti-Pgc-1 antibody or normal IgG withcell extracts containing 2 mg protein, followed by addition of 30 μlof protein G plus/protein A agarose for another 2 h. Immunopre-cipitates were collected and washed five times with lysis buffer,resuspended in SDS sample buffer, and boiled for 5 min at 95 1C.PPARγ was detected in bound proteins by Western blotting.

Chromatin immunoprecipitation (ChIP)

A ChIP assay was performed using the EZ ChIP kit (Millipore,Billerica, MA, USA) according to the manufacturer’s protocol.Briefly, cells were cross-linked with a formaldehyde solution,and the chromatin was sheared by sonication to isolate DNAfragments that averaged 300–500 bp. An anti-PPARγ antibodywas added to aliquots of precleared chromatin and incubatedovernight. Input samples were incubated with the negative-control IgG. The immune complexes were captured by incubationwith protein G agarose for 1 h at 4 1C. After the cross-links werereversed, DNA samples from the immunoprecipitates were iso-lated, and RT-PCR was performed using the following specificprimers flanking the PPRE of Nrf2 with 35 cycles of PCR amplifica-tion: Nrf2 forward, 50-CGAGAGCGCTGCCCTTATTT-30, and reverse,50-GGGGGACCTAGAGGAGGTCT-30, and the control GAPDH primers(Millipore).

Small interfering RNA (siRNA) transfection

SiRNA targeting Sirt1 and the siRNA transfection reagent wereobtained from Santa Cruz Biotechnology. DR5 siRNA was obtainedfrom Bioneer. MCF-7 cells grown to 60% confluence were trans-fected with Sirt1 siRNA or a nonspecific control siRNA according tothe manufacturer’s instructions. The cells were harvested 48 hafter transfection.

Sirt1, Pgc-1α, and PPARγ overexpression

The Sirt1 expression vector was a gift from Dr. Kwang Youl Lee(Chonnam National University, South Korea). The Pgc-1α expres-sion vector was purchased from Addgene (Cambridge, MA, USA).The PPARγ expression vector was purchased from OriGene Tech-nologies (Rockville, MD, USA). MCF-7 cells were transfected withSirt1, Pgc-1α, PPARγ, or the pCMV6 or pcDNA4 empty vector usingLipofectamine 2000 in antibiotic-free medium and cultured for48 h before experiments.

Transfection of miR-34a inhibitor and mimic

MiR-34a inhibitor and miR-34a mimic oligonucleotides werechemically synthesized by Bioneer. MCF-7 cells were transfectedwith 200 nM oligonucleotides using Lipofectamine 2000, accord-ing to the manufacturer’s protocol. One day after transfection, thecells were treated with 5 mM metformin or left untreated for anadditional 24 h. The cells were harvested for subsequent experi-ments 48 h posttransfection.

Measurement of intracellular reactive oxygen species (ROS)

The fluorescent probe 20,70-dichlorodihydrofluorescin diacetate(H2DCFDA) was used to measure intracellular ROS generation inMCF-7 cells after treatment with metformin or H2O2. Briefly,confluent MCF-7 cells in 96-well plates were preincubated with20 μM H2DCFDA for 30 min at 37 1C, followed by treating the cellswith metformin (1–5 mM) or H2O2 (100 μM) for 1 h. The fluores-cence intensity was measured at excitation and emission wave-lengths of 485 and 530 nm, respectively, using a fluorescencespectrophotometer. The generation of ROS in MCF-7 cells after24 h of treatment with metformin was also tested.

Statistical analysis

The data are expressed as means 7 SD. One-way analysis ofvariance was used to determine the significance of differencesbetween treatment groups. The Newman–Keuls test was used formultigroup comparisons. A p value of o0.05 was considered toindicate statistical significance.

Results

Metformin suppresses Sirt1 expression in p53 wild-type cancer cells

We elucidated the effects of metformin on Sirt1 expression inseveral cancer cell lines expressing different forms of p53: MCF-7,HCT 116, and A549 cancer cells expressing wild-type p53; MDA-MB-231 cells harboring a mutated p53; p53-knockout HCT 116cells; and null-p53 SKOV3 cancer cells. Cells were treated withmetformin (1–5 mM) for 24 h and Western blots performed todetermine the levels of Sirt1 protein in cell lysates. Interestingly,metformin reduced Sirt1 protein levels in all tested MCF-7, HCT116, and A549 cancer cell lines harboring wild-type p53 (Figs. 1A,C, and E), whereas Sirt1 protein levels in cell lines with altered p53

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

M. Truong Do et al. / Free Radical Biology and Medicine ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3





expression were not affected by metformin treatment (Figs. 1B, D,and F). These results suggest that the effect of metformin on Sirt1expression in cancer cells depends on the status of p53.

p53-dependent induction of miR-34a is required for the reduction ofSirt1 by metformin

Sirt1 regulates apoptosis by deacetylating molecular targetssuch as p53 [3,4], and p53 can regulate Sirt1 expression via apositive-feedback loop through the induction of miR-34a [23,24].We performed qRT-PCR to determine the effects of metformin onmiR-34a expression in wild-type p53 MCF-7 and p53-mutantMDA-MB-231 breast cancer cells. These cells were treated withmetformin (0.5, 1, and 5 mM) for 24 h and harvested for miR-34aanalysis. Metformin significantly induced miR-34a levels in wild-type p53 MCF-7 cells but did not change the miR-34a levels inp53-mutant MDA-MB-231 cells (Fig. 2A). Western blot dataindicated that metformin increased the acetylated form and p53protein levels after 24 h of treatment in wild-type p53 MCF-7 cellsbut did not affect p53 expression in p53-mutant MDA-MB-231cells (Figs. 2B and C). Additionally, metformin induced p53 proteinlevels early in wild-type p53 A549 lung cancer cells, with thestarting point at 2 h and clearly at 14 and 24 h after treatment(Fig. 2D). To confirm the critical role of p53 in upregulation of miR-34a by metformin, p53 wild-type HCT 116 and p53-knockout HCT116 colon cancer cells were used to assess the role of p53 status.Metformin significantly induced miR-34a levels in p53 wild-typeHCT 116 cells but not in p53�/� HCT 116 cancer cells (Fig. 2E).Western blot results confirmed that p53 expression was knockedout successfully with no induction of p53 in p53�/� HCT 116cancer cells, whereas increased p53 levels were observed in p53wild-type HCT 116 cells (Fig. 2F). Furthermore, to elucidate the roleof miR-34a induction in metformin-induced Sirt1 downregulation,

we transfected MCF-7 cells with a miR-34a inhibitor for 24 h andthen treated the cells with 5 mMmetformin for an additional 24 h.As shown in Fig. 2G, the miR-34a inhibitor reversed metformin-mediated downregulation of Sirt1 protein levels. Additionally,a miR-34a mimic robustly reduced Sirt1 protein levels after 24 hof transfection in MCF-7 cells (Fig. 2H). These results demonstratethat p53-dependent-induction of miR-34a is required for thereduction of Sirt1 by metformin.

Downregulation of Sirt1 by metformin inhibits Nrf2 expression andincreases susceptibility of wild-type p53 cancer cells to oxidativestress

Our previous study demonstrated that metformin reduced Nrf2expression in several human cancer cell lines [22]. Because arecent investigation reported that Nrf2 expression was regulatedby Sirt1 in human cancer cells [7], their roles in metformin actionwere investigated in this study. MCF-7 cells were treated withmetformin (1–5 mM) for 24 h, and the cells were harvested forWestern blot analysis. Metformin markedly reduced Nrf2 proteinlevels in a dose-dependent manner (Fig. 3A, top) and reduced Nrf2protein levels after long-term treatment at 48 and 72 h (Fig. 3A,bottom). We performed qRT-PCR to verify that metformin sig-nificantly reduced Nrf2 mRNA levels in MCF-7 cells after 24 h oftreatment (Fig. 3B). To confirm that Sirt1 positively regulates Nrf2expression, MCF-7 cells were transfected with Sirt1 siRNA for 48 h.Knockdown of Sirt1 notably reduced Nrf2 protein levels. Addi-tionally, MCF-7 cells were transfected with a Sirt1 expressionvector, which remarkably induced Nrf2 protein levels (Fig. 3C).These results suggest that downregulation of Nrf2 expression bymetformin was mediated through a reduction of Sirt1 in cancercells.

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 1. Metformin downregulates Sirt1 expression in wild-type p53 cancer cells. (A, C, and E) Effects of metformin on Sirt1 protein expression in MCF-7, HCT 116, and A549cancer cells harboring wild-type p53. (B, D, and F) Effects of metformin on Sirt1 protein expression in MDA-MB-231 cells harboring mutant p53, p53-knockout HCT 116 cells,and null-p53 SKOV3 cells. Sirt1 and β-actin expression was analyzed in cell lysates by Western blotting.






We next investigated the effects of metformin on the expres-sion of cytoprotective HO-1 and antioxidative SOD2 enzymes thatwere regulated by Nrf2 [19]. Metformin reduced HO-1 and SOD2protein levels after 24 h of treatment in MCF-7 cells (Fig. 3D).However, metformin upregulated catalase protein levels under thesame conditions of treatment (Fig. 3D). Because the expression ofSOD2, catalase, and HO-1 plays an important role in controllingthe balance of ROS and cellular protection against oxidative stress,we determined intracellular ROS production in metformin orH2O2-stimulated MCF-7 breast cancer cells. As shown in Fig. 3E,metformin (1–5 mM) did not affect basal ROS, whereas ROS

production was increased significantly fivefold by stimulation withH2O2 (100 μM) compared with the control after 1 h of treatment.Metformin also did not change ROS generation after 24 h oftreatment (data not shown). Interestingly, pretreatment withmetformin for 24 h significantly enhanced H2O2-induced cellcytotoxicity and apoptosis in wild-type p53 MCF-7 breast cancercells (Figs. 3F–I). However, pretreatment with metformin did notenhance H2O2-induced cell cytotoxicity in p53-mutated MDA-MB-231 cells (Figs. 3J and K). These results indicated that metforminincreases the susceptibility of wild-type p53 cancer cells tooxidative stress.

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 2. p53-dependent induction of miR-34a is required for Sirt1 reduction by metformin. (A) Effects of metformin on miR-34a levels in wild-type p53 MCF-7 and p53-mutated MDA-MB-231 breast cancer cells. MiR-34a levels were analyzed by qRT-PCR. All experiments were performed in triplicate. Bars indicate mean 7 SD; np o 0.05 vscontrol. (B and C) Effects of metformin on p53 protein expression in wild-type p53 MCF-7 cells and MDA-MB-231 cells harboring a p53 mutation. p53, acetyl-p53, and β-actin protein levels in cell lysates were assayed by Western blotting. (D) Western blot analysis of p53 and β-actin proteins in cell lysates showed that metformin induced p53protein levels in wild-type p53 A549 lung cancer cells in a time-dependent manner. (E) Effects of metformin on miR-34a levels in wild-type p53 HCT 116 and p53-knockoutHCT 116 colon cancer cells. MiR-34a levels were analyzed by qRT-PCR. All experiments were performed in triplicate. Bars indicate mean 7 SD; np o 0.05 vs control.(F) Effects of metformin on p53 protein expression in wild-type p53 HCT 116 and p53-knockout HCT 116 colon cancer cells. p53, acetyl-p53, and β-actin protein levels in celllysates were assayed by Western blotting. (G) Anti-miR-34a reverses metformin-mediated downregulation of Sirt1. (H) MiR-34a mimic downregulates Sirt1 protein levels inMCF-7 breast cancer cells. Sirt1 and β-actin protein levels in cell lysates were assayed by Western blotting.






Metformin inhibits Nrf2 expression mediated by suppressionof Pgc-1α

Previous studies demonstrated that Sirt1 regulation of Nrf2expression is mediated through a transcriptional coactivator, Pgc-1α [7,25], and Sirt1 controls the transcription of Pgc-1α through aPgc-1α autoregulatory loop [5]. Therefore, we elucidated theeffects of metformin on Pgc-1α expression in MCF-7 breast cancer

cells. Metformin reduced Pgc-1 protein levels in MCF-7 cells after24 h of treatment (Fig. 4A), and 5 mM metformin also decreasedPgc-1 protein levels at 48 or 72 h of treatment (Fig. 4B). Further-more, metformin significantly suppressed Pgc-1α mRNA levelsafter 24 h of treatment (Fig. 4C). Notably, Sirt1 overexpressionincreased Pgc-1 protein levels (Fig. 4D), and knockdown of Sirt1suppressed Pgc-1 protein levels (Fig. 4E). To confirm that Pgc-1αplays an important role in regulating Nrf2 expression, MCF-7 cells

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 3. Metformin-mediated downregulation of Sirt1 inhibits Nrf2 expression and increases susceptibility of wild-type p53 cancer cells to oxidative stress. (A) Effects ofmetformin on Nrf2 protein levels in MCF-7 cells after 24 h of treatment (top) and after 48 and 72 h of treatment (bottom). Nrf2 and β-actin protein levels in cell lysates wereassayed by Western blotting. (B) Effects of metformin on Nrf2 mRNA expression. QRT-PCR demonstrated the effect of metformin on Nrf2 mRNA levels in MCF-7 cells after 24h of treatment. All experiments were performed in triplicate. Bars indicate mean 7 SD; np o 0.05 vs control. (C) Role of Sirt1 in regulating Nrf2 expression in cancer cells.MCF-7 cells were transfected with Sirt1 siRNA, a Sirt1 expression control siRNA, or the pCMV6 vector for 48 h. Cell lysates were subjected to Western blotting usingantibodies specific for Sirt1, Nrf2, and β-actin. (D) Effects of metformin on HO-1, SOD2, and catalase expression in MCF-7 cells. HO-1, SOD2, catalase, and β-actin proteinlevels in cell lysates were assayed by Western blotting. (E) Effects of metformin and H2O2 on intracellular ROS production in MCF-7 cells. Intracellular ROS production wasdetermined using the fluorescent probe H2DCFDA. Bars represent mean 7 SD. np o 0.05 vs control, n ¼ 6. (F and G) Effects of metformin on H2O2-induced survivalinhibition in wild-type p53 MCF-7 cells as measured by MTT assay. (H) Effects of metformin on H2O2-induced cytotoxicity in wild-type p53 MCF-7 cells as measured by LDHassay. All experiments were performed in quadruplicate. Bars represent mean 7 SD; np o 0.05 vs control; #p o 0.05 vs cells treated with H2O2 and metformin alone.(I) Effects of metformin on H2O2-induced apoptosis. MCF-7 cells were pretreated with 5 mM metformin for 24 h, followed by stimulation with 100 μM H2O2 for an additional12 h. PARP and β-actin in cell lysates were analyzed by Western blotting. (J) Effects of metformin and H2O2 on cell viability of p53-mutated MDA-MB-231 cells as measuredby MTT assay. (K) Effects of metformin and H2O2 on cytotoxicity of p53-mutated MDA-MB-231 cells as measured by LDH assay. All experiments were performed inquadruplicate. Bars represent mean 7 SD.






were transfected with a Pgc-1α expression vector for 48 h and thetransfected cells were harvested for Western blot analysis. Asshown in Fig. 4F, overexpression of Pgc-1α induced Nrf2 proteinlevels in MCF-7 cells. These results suggest that downregulation ofNrf2 expression by metformin-mediated Sirt1 downregulation ismediated through the Pgc-1α reduction.

Metformin suppresses Nrf2 expression by inhibiting PPARγtranscriptional activity and attenuating PPARγ binding to the Nrf2promoter

Previous studies demonstrated that PPARγ upregulates Nrf2expression by activating a PPRE binding sequence in the Nrf2promoter [26,27], whereas Sirt1 induced PPARγ transcriptionalactivity by increasing Pgc-1α expression [28,29]. To determine theeffects of metformin on PPARγ transcriptional activity, MCF-7 cellswere transiently transfected with a PPRE-luciferase vector andthen treated with metformin (1–5 mM) for 24 h. Metforminsignificantly inhibited PPRE-luciferase activity (Fig. 5A). However,Western blot data indicated that metformin did not affect PPARγprotein levels (Fig. 5B), although nuclear protein extraction indi-cated that metformin reduced Sirt1, Pgc-1, and PPARγ protein

levels (Fig. 5C). PPARγ overexpression by gene transfer increasedNrf2 protein levels in transfected MCF-7 cells (Fig. 5D). Further-more, MCF-7 cells were treated with metformin for 24 h, and thecells were harvested for nuclear protein extraction and IP assay.IP results indicated that metformin reduced the Pgc-1 and PPARγcomplex (Fig. 5E). A bioinformatics search indicated that there is aputative PPRE of the AGACCT core-binding sequence at �363 to�358 located on the human Nrf2 promoter (GenBank AccessionNo. NM_006164.4). A ChIP assay was performed with Nrf2-pro-moter-specific primers that contained a putative PPRE bindingsequence. Metformin inhibited the binding of PPARγ to a PPREbinding sequence that exists on the Nrf2 promoter construct(Fig. 5F). These results demonstrate that metformin inhibitedPPARγ transcriptional activity, attenuating the binding of PPARγto the Nrf2 promoter and reducing Nrf2 expression.

Upregulation of DR5 expression by metformin sensitizes wild-typep53 cancer cells to TRAIL-induced apoptosis

Previous studies demonstrated that Sirt1 inhibition sensitizedcancer cells to TRAIL-induced apoptosis through induction ofCHOP and DR5 expression [10] and Nrf2 expression has been

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 3. (continued)






associated with TRAIL resistance in cancer cells [21]. We hypothe-sized that the downregulation of Sirt1 and Nrf2 expression bymetformin would upregulate CHOP and DR5 expression andsensitize TRAIL-resistant MCF-7 cells to TRAIL-induced apoptosis.MCF-7 cells were treated with 1 mM metformin for 24 and 48 h,followed by qRT-PCR assessment of HOP and DR5 mRNA levels.Metformin significantly induced CHOP and DR5 mRNA levels(Figs. 6A and B). Western blot analysis indicated that metformininduced DR5 protein levels after 24 and 48 h of treatment (Fig. 6C).To elucidate the role of Sirt1 in the regulation of DR5 expression,MCF-7 cells were transfected with siRNA Sirt1, and the cells wereharvested for analysis of DR5 protein expression by Westernblotting. As shown in Fig. 6D, the deletion of Sirt1 clearly inducedDR5 protein expression. Notably, when wild-type p53 MCF-7 cellswere pretreated with 1 mM metformin for 24 h, followed bytreatment with 25 ng/ml TRAIL for an additional 48 h, metforminclearly potentiated TRAIL-mediated cell death (Fig. 6E). Interest-ingly, pretreatment with even a low concentration of metforminsignificantly enhanced TRAIL-mediated survival inhibition in MCF-7 cancer cells (Fig. 6F). To investigate whether the death caused bymetformin and TRAIL treatment was related to apoptosis, weanalyzed PARP protein expression as a marker of apoptosis. MCF-7 cells were pretreated with 1 mM metformin for 24 h and thentreated with 25 ng/ml TRAIL for an additional 24 h. PARP protein

levels in cell lysates were analyzed by Western blotting. Pretreat-ment with metformin potentially enhanced TRAIL-induced cleavedPARP protein levels (Fig. 6G). Confirming that the upregulation ofDR5 expression by metformin relates to TRAIL sensitization,knockdown of DR5 by using siRNA significantly restored cellviability in response to metformin and TRAIL treatment (Fig. 6H).Moreover, the knockdown of DR5 also suppressed the metformin-mediated sensitization to TRAIL-induced apoptosis (Fig. 6I). How-ever, metformin did not increase TRAIL sensitivity in p53-mutatedMDA-MB-231 cells (Fig. 6J). These results indicate that metforminsensitizes wild-type p53 cancer cells to TRAIL-induced apoptosisthrough DR5 upregulation.

Discussion

Cancer is a leading cause of death worldwide. Lung, stomach,liver, colon, and breast cancers cause the most cancer deaths eachyear, although the prevalence differs between males and females[30]. Pharmacological intervention is an attractive approach forcancer prevention and treatment and the antidiabetic drug met-formin is rapidly emerging as a potential antitumor agent thatreduces the risk of breast, colon, lung, liver, pancreatic, andprostate cancer [31]. In this study, we report that by upregulating

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 4. Metformin-inhibited Nrf2 expression is mediated by Pgc-1α suppression. (A) Metformin reduced Pgc-1 protein levels in MCF-7 cells after 24 h of treatment.(B) Metformin reduced Pgc-1 protein levels in MCF-7 cells after 48 and 72 h of treatment. Pgc-1 and β-actin protein levels in cell lysates were assayed by Western blotting.(C) qRT-PCR demonstrated the effect of metformin on Pgc-1αmRNA levels in MCF-7 cells after 24 h of treatment. All experiments were performed in triplicate. Bars indicatemean 7 SD; np o 0.05 vs control. (D) Sirt1 overexpression upregulated Pgc-1 protein levels in MCF-7 cells. (E) Sirt1 depletion repressed Pgc-1 protein levels in MCF-7 cells.MCF-7 cells were transfected with a Sirt1 expression vector or the pCMV6 vector or with Sirt1 siRNA or control siRNA for 48 h. Cell lysates were analyzed byWestern blottingusing antibodies specific for Pgc-1 and β-actin. (F) Pgc-1α overexpression induced Nrf2 protein levels in cancer cells. MCF-7 cells were transfected with a Pgc-1α expressionvector or the pcDNA4 vector for 48 h. Cell lysates were subjected to Western blotting using antibodies specific for Pgc-1, Nrf2, and β-actin.






p53, metformin induced miR-34a to downregulate Sirt1 expres-sion, subsequently reducing Nrf2 and enhancing the susceptibilityof wild-type p53 cancer cells to oxidative stress. Metformin-mediated downregulation of Sirt1 induced DR5 expression, sensi-tizing wild-type p53 cancer cells to TRAIL-induced apoptosis.

Previous studies demonstrated that metformin downregulatedvarious oncogenes such as HER2 or mTOR/p70S6K signaling incancer cells [32,33]. We found that metformin clearly down-regulated Sirt1 protein levels in cancer cell lines harboring wild-type p53 (MCF-7, HCT 116, and A549), but did not affect Sirt1protein levels in cancer cells harboring a p53 mutant or p53knockout (MDA-MB-231, HCT 116 p53� /� , and null-p53 SKOV3).Notably, accumulating evidence suggests that Sirt1 has an onco-genic function in wild-type p53 cancer cells [2,8]. Therefore, thedownregulation of Sirt1 expression, in part, contributes to theanticancer activities of metformin in cancer cells expressing wild-type p53. The tumor suppressor p53 is a direct substrate of Sirt1,and p53 transcriptional activity is regulated by the deacetylaseactivity of Sirt1 [7,8]. However, p53 also regulates Sirt1 expression,which is mediated by induction of miR-34a via a positive-feedback

loop [23]. Previous study indicated that nicotinamide inhibitedproliferation and induced apoptosis through activation of the p53/miR-34a signaling pathway and downregulation of Sirt1 expres-sion in chronic lymphocytic leukemia cells [15]. A recent studydemonstrated that by activating p53, deoxycholic acid increasedmiR-34a expression to suppress Sirt1 expression, leading toincreased cellular caspase activation and apoptosis in primary rathepatocytes and rat liver in vivo [24]. Moreover, miR-34a expres-sion is associated with p53 status in acute myeloid leukemia(AML) patients in which miR-34a was on average 22.8-fold down-regulated in altered p53 compared to wild-type p53 AML cases,and the wild-type p53-induced upregulation of miR-34a leads tocell cycle arrest and apoptosis through downregulation of miR-34atarget genes, including Sirt1 [34]. In our study, metforminincreased p53 protein levels and its activated form, acetyl-p53,in MCF-7 cells, whereas metformin did not change p53 levels inp53-mutant MDA-MB-231 cells. Consistent with elevated p53protein levels, metformin induced miR-34a levels in MCF-7 cells,but not in MDA-MB-231 breast cancer cells. HCT 116 cells harbor-ing wild-type p53 and p53-knockout HCT 116 cells were chosen to

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 5. Metformin inhibits Nrf2 expression mediated by suppressing transcriptional activity of PPARγ. (A) Metformin inhibits PPRE-luciferase promoter activity in MCF-7cells. All experiments were performed in triplicate. Bars indicate mean 7 SD; np o 0.05 vs control. (B) Effects of metformin on PPARγ protein expression in MCF-7 cells.PPARγ and β-actin protein levels in whole-cell lysates were assayed by Western blotting. (C) Metformin inhibits Sirt1, Pgc-1, and PPARγ proteins in the nucleus of MCF-7cells. Sirt1, Pgc-1, PPARγ, and β-actin protein levels in the nuclear protein fraction were assayed by Western blotting. (D) PPARγ overexpression increased Nrf2 protein levelsin cancer cells. MCF-7 cells were transfected with a PPARγ expression vector or the pCMV6 vector for 48 h. Cell lysates were subjected to Western blotting using antibodiesspecific for PPARγ, Nrf2, and β-actin. (E) Metformin reduced the Pgc-1 and PPARγ complex in the nucleus of MCF-7 cells. Nuclear protein fractions were used for IP assay asdescribed under Materials and methods. PPARγ was detected in bound proteins by Western blotting. (F) Metformin suppressed the binding of PPARγ to a PPRE-bindingsequence in the Nrf2 promoter. The ChIP assay was performed as described under Materials and methods.






confirm the critical role of p53-mediated miR-34a inductionbecause these cell lines differ only in their p53 levels. Corroborat-ing results obtained indicated that metformin-mediated elevationof p53 protein levels in wild-type p53 HCT 116 cancer cells wasrequired for the induction of miR-34a, which is in agreement withother reports that p53 plays an important role in regulating miR-34a levels [23,35]. In addition, inhibition of miR-34a recovered theinhibitory effect of metformin on Sirt1 protein levels, whereasmiR-34a overexpression in MCF-7 cells dramatically reduced Sirt1protein levels. Thus, p53 activation is necessary for miR-34ainduction, which leads to the downregulation of Sirt1 expressionby metformin in wild-type p53 cancer cells.

Our results indicated that the activation of the p53/miR-34asignaling pathway and downregulation of Sirt1 expression playimportant roles in enhancing sensitivity of wild-type p53 MCF-7cells to oxidative stress or TRAIL-induced apoptosis. These results

could not be observed in p53-mutant MDA-MB-231 cells in whichSirt1 expression was not affected by metformin. MDA-MB-231cells were more resistant to H2O2-induced cytotoxicity that couldbe caused by the high constitutive SOD2 expression in MDA-MB-231 cells compared with MCF-7 cells [36]. Notably, metforminpretreatment did not enhance H2O2-induced cytotoxicity in MDA-MB-231 cells. On the other hand, because of lower cell surfaceexpression of DR4 and DR5 [37] and higher expression of decoyreceptor-4 [38], MCF-7 cells are resistant to TRAIL compared withMDA-MB-231 cells. Therefore, treatment with TRAIL alone wasmore cytotoxic to MDA-MB-231 cells than to MCF-7 cells. Simi-larly, metformin pretreatment did not enhance TRAIL-inducedcytotoxicity in MDA-MB-231 cells. These results indicated thatthe activation of the p53/miR-34a signaling pathway and thedownregulation of Sirt1 expression by metformin are responsiblefor enhancing cytotoxicity of H2O2 and TRAIL in wild-type p53

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 6. Upregulation of DR5 expression by metformin sensitizes wild-type p53 cancer cells to TRAIL-induced apoptosis. (A and B) Metformin induced CHOP and DR5 mRNAexpression in MCF-7 cells after 24 and 48 h of treatment. CHOP and DR5 mRNA levels were analyzed by qRT-PCR. All experiments were performed in triplicate. Barsrepresent mean 7 SD; np o 0.05 vs control. (C) Metformin induced DR5 protein expression in MCF-7 cells after 24 and 48 h of treatment. DR5 and β-actin protein levels incell lysates were analyzed by Western blotting. (D) Sirt1 depletion induced DR5 protein levels in cancer cells. MCF-7 cells were transfected with Sirt1 or control siRNA for 48h. Cell lysates were subjected to Western blotting using antibodies specific for Sirt1, DR5, and β-actin. (E) Metformin sensitized MCF-7 breast cancer cells to TRAIL-inducedcell death. (F) The effect of metformin on TRAIL-induced survival inhibition in MCF-7 cells as measured by MTT assay. All experiments were performed in quadruplicate. Barsrepresent mean 7 SD; np o 0.05 vs control; #p o 0.05 vs cells treated with TRAIL and metformin alone. (G) Metformin enhanced TRAIL-induced apoptosis in MCF-7 breastcancer cells. MCF-7 cells were pretreated with 1 mM metformin for 24 h and then treated with 25 ng/ml TRAIL for an additional 24 h. PARP and β-actin protein levels in celllysates were analyzed by Western blotting. (H) Effects of knockdown of DR5 expression on the suppression of cell viability caused by pretreatment with metformin andfollowed by TRAIL. MCF-7 cells were transfected with DR5 siRNA or a nonspecific control siRNA for 24 h. The siRNA-transfected cells were pretreated with metformin for 24 hand then with TRAIL for an additional 48 h. Cell viability was measured by MTT assay. All experiments were performed in quadruplicate. Bars represent mean 7 SD. np o0.05 vs control; #p o 0.05 vs cells treated with metformin or TRAIL alone; ¥p o 0.05 vs cells treated with both metformin and TRAIL. (I) Effects of knockdown of DR5expression on the metformin-mediated sensitization to TRAIL-induced apoptosis. MCF-7 cells were transfected with DR5 siRNA or a nonspecific control siRNA for 48 h. ThesiRNA-transfected cells were pretreated with metformin for 24 h and then with TRAIL for an additional 12 h. DR5, PARP, and β-actin protein levels in cell lysates wereanalyzed by Western blotting. (J) Effect of metformin on TRAIL-induced survival inhibition in p53-mutated MDA-MB-231 cells as measured by MTT assay. All experimentswere performed in quadruplicate. Bars represent mean 7 SD; np o 0.05 vs control.






breast cancer cells. Our result agrees with a recent study demon-strating that p53 mutations increase apoptotic resistance inkeratinocytes by interfering with miR-34a-mediated regulationof Sirt1 expression [39].

The transcription factor Nrf2 is emerging as a master regulatorof antioxidant defense systems. In cancer, a gain of Nrf2 functionenhances the resistance of cancer cells to chemotherapeutic agents[20,21]. Here, metformin-downregulated Nrf2 expression resultedin suppressed expression of the Nrf2-regulated genes HO-1 andSOD2, which play an important role in regulating cellular redoxhomeostasis and the balance of generation and elimination of ROS.Notably, catalase expression is mainly controlled by the Akt/mTORsignaling pathway in MCF-7 cells [40]. Metformin strongly inhib-ited the Akt/mTOR signaling pathway [41], which may explain whymetformin upregulated catalase expression in MCF-7 breast cancercells. Our results also suggest that the upregulation of catalaseexpression by metformin may represent a compensatory mechan-ism by which metformin-treated MCF-7 cells show no increase in

basal ROS production. Interestingly, previous studies indicated thatoverexpression of SOD2 completely protected MCF-7 cells fromtumor necrosis factor (TNF)-induced cytotoxicity. Cell viability wasrecovered more than 60% by SOD2 overexpression [42], whereasonly 20% cell viability was recovered by catalase overexpression inTNF-treated MCF-7 cells [43]. Similarly, the overexpression ofSOD2 exhibited more potential for protecting MCF-7 cells againstH2O2-induced cytotoxicity compared to catalase overexpression[42,43]. Additionally, the overexpression of HO-1 also partlycontributed to protecting cancer cells against H2O2-induced cyto-toxicity [44]. The present results indicated that metformin sup-pressed SOD2 and HO-1 expression in MCF-7 cells, and metformindownregulated Sirt1 expression, and Sirt1 participated in protec-tion against H2O2-induced cancer cell death [45]. These observa-tions explain why metformin pretreatment significantly sensitizedwild-type p53 cancer cells to H2O2- or TRAIL-induced cytotoxicityeven though the catalase expression was induced by metformin.Our results agree with a previous study that reported that Nrf2� /�

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 6. (continued)






hematopoietic stem progenitor cells did not display elevated basallevels of ROS but were more sensitive to apoptosis by inducedoxidative stress [46].

Pgc-1α is a regulator of mitochondrial biogenesis genes thatsimultaneously upregulates many genes known to protect againstoxidative stress and lipogenesis. A previous study demonstratedthat loss of Pgc-1α protects against tumorigenesis and that Pgc-1αcoordinately regulates mitochondrial and fatty acid metabolism topromote tumor growth [47]. Moreover, increased Pgc-1α expres-sion and activation of Pgc-1α target genes enhanced the ability ofbreast cancer cells to metastasize to brains of mice [48], and Pgc-1α promoted the growth of HER2-induced mammary tumors byregulating nutrient supply [49]. Previous studies demonstratedthat the Sirt1/Pgc-1α/Nrf2 signaling pathway exists in both normaland cancer cells, which relates to diverse cellular responses tooxidative and inflammatory stress or cellular homeostasis [7,25].For instance, depletion of Sirt1 downregulated Pgc-1α and Nrf2expression in the rat Kupffer cell line RKC1 [25], and Sirt1 over-expression strongly induced Pgc-1α and Nrf2 expression in ahuman cancer cell line [7]. Additionally, Sirt1 reduction wascaused by deoxycholic acid-mediated induction of p53/miR-34a-increased acetylated Pgc-1α (a transcriptionally inactive form) inprimary rat hepatocytes [24]. Sirt1 regulates PPARγ transcriptionalactivity by increasing Pgc-1α expression [28,29]. A recent reportsuggested that PPARγ could be the transcription factor activatedby Pgc-1α that binds to the promoter region on the Nrf2 gene,stimulating its activation/transcription [50]. Moreover, Nrf2expression was suppressed in the lung with decreased PPARγlevels by siRNA silencing, indicating that PPARγ could act onupstream signal pathways for Nrf2 activation [27]. Our resultsindicate that metformin downregulates Sirt1, Pgc-1α, and Nrf2expression in human breast cancer cells. The knockdown of Sirt1by siRNA markedly reduced Pgc-1 and Nrf2 protein levels, and theforced expression of Sirt1 by a Sirt1 gene transfer induced Pgc-1and Nrf2 protein levels in MCF-7 breast cancer cells. Previousstudies demonstrated that metformin suppressed PPARγ tran-scription activity in hepatoma cells [51], and the effect of metfor-min on PPARγ was not mediated by binding to the ligand-bindingdomain of PPARγ [52]. In this study, inhibition of PPARγ transcrip-tion activity was also observed in metformin-treated MCF-7 breastcancer cells. Using an IP assay, we found that metformin reducednuclear Pgc-1 and PPARγ transcription complex in breast cancercells. The complex dissociation of Pgc-1 and PPARγ proteins maybe caused by a reduction in Sirt1 leading to increased acetylatedPgc-1α, which is the inactive form [53], and decreased Pgc-1

protein levels by metformin. ChIP data confirmed that metforminsuppressed the binding of PPARγ to a PPRE binding sequence thatexists on the Nrf2 promoter construct. Our results indicate that areduction in Sirt1 and Pgc-1α by metformin caused Nrf2 down-regulation via inhibition of PPARγ transcriptional activity in breastcancer cells.

Downregulation of Sirt1 strongly induced CHOP and DR5expression, and the upregulation of DR5 potentiates TRAIL-induced apoptosis in cancer cells [10]. Additionally, Nrf2-inducedS5a and α-5 expression resulted in TRAIL-resistant colon epithelialcells through increased NF-κB activation and expression of anti-apoptotic target genes [21]. Moreover, the loss or gain of Nrf2expression using Nrf2 siRNA or Nrf2 gene transfer significantly up-or downregulated CHOP expression in human thyroid cancer cells,suggesting that Nrf2 acts as a direct transcriptional repressor orthat Nrf2 inhibited the binding of other transcription factors to theCHOP promoter [54]. Additionally, the HO-1/carbon mono-xide system suppressed CHOP expression under homeostatic andunfolded protein response conditions [55]. Therefore, metformin-mediated HO-1 reduction was able to induce CHOP expression,followed by upregulation of DR5 in MCF-7 breast cancer cells.Notably, the depletion of Sirt1 expression using Sirt1 siRNAincreased DR5 protein levels in MCF-7 cells. These results suggestthat the downregulation of the Sirt1/Pgc-1α/Nrf2 axis by metfor-min may induce CHOP and DR5 expression. The molecularmechanism by which CHOP induces DR5 expression is known[56]. The important role of DR5 upregulation by metformin inincreased sensitivity of cancer cells to TRAIL-induced apoptosiswas confirmed by using DR5 siRNA. Notably, metformin, even at alow concentration (20–100 μM), significantly sensitized MCF-7cancer cells to TRAIL-induced cell survival inhibition.

In conclusion, we report for the first time that metformininduces miR-34a to suppress the Sirt1/Pgc-1α/Nrf2 axis andincreases H2O2-induced cancer cell death. The plausible mechan-ism by which metformin downregulates Nrf2 expression bysuppressing Pgc-1α-mediated PPARγ transcriptional activity,attenuating PPARγ binding to the Nrf2 promoter, is also elucidated.Upregulation of DR5 by metformin-mediated Sirt1 downregula-tion enhances the sensitivity of cancer cells to TRAIL-inducedapoptosis. Our work identifies the downregulation of the Sirt1/Pgc-1α/Nrf2 pathway as being responsible for the enhancedsusceptibility of cancer cells to oxidative stress and therapeuticagents; the proposed signaling pathways are shown in Fig. 7.

Acknowledgments

This research was supported by the Basic Science ResearchProgram through the National Research Foundation of Koreafunded by the Ministry of Science, ICT, and Future Planning(NRF-2013R1A2A2A01067892) and by the Priority Research Cen-ters Program through the National Research Foundation of Koreafunded by the Ministry of Education, Science, and Technology(2009-0093815), Republic of Korea.

References

[1] Radak, Z.; Koltai, E.; Taylor, A. W.; Higuchi, M.; Kumagai, S.; Ohno, H.; Goto, S.;Boldogh, I. Redox-regulating sirtuins in aging, caloric restriction, and exercise.Free Radic. Biol. Med. 58:87–97; 2013.

[2] Brooks, C. L.; Gu, W. How does SIRT1 affect metabolism, senescence andcancer? Nat. Rev. Cancer 9:123–128; 2009.

[3] Luo, J.; Nikolaev, A. Y.; Imai, S.; Chen, D.; Su, F.; Shiloh, A.; Guarente, L.; Gu, W.Negative control of p53 by Sir2alpha promotes cell survival under stress. Cell107:137–148; 2001.

[4] Vaziri, H.; Dessain, S. K.; Ng Eaton, E.; Imai, S. I.; Frye, R. A.; Pandita, T. K.;Guarente, L.; Weinberg, R. A. hSIR2 (SIRT1) functions as an NAD-dependentp53 deacetylase. Cell 107:149–159; 2001.

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 7. Proposed signaling pathways underlying the effects of metformin on theSirt1/Pgc-1α/Nrf2 pathway leading to increased susceptibility of wild-type p53cancer cells to oxidative stress and TRAIL-induced apoptosis.






[5] Amat, R.; Planavila, A.; Chen, S. L.; Iglesias, R.; Giralt, M.; Villarroya, F. SIRT1controls the transcription of the peroxisome proliferator-activated receptor-gamma co-activator-1alpha (PGC-1alpha) gene in skeletal muscle through thePGC-1alpha autoregulatory loop and interaction with MyoD. J. Biol. Chem.284:21872–21880; 2009.

[6] Frazzi, R.; Valli, R.; Tamagnini, I.; Casali, B.; Latruffe, N.; Merli, F. Resveratrol-mediated apoptosis of Hodgkin lymphoma cells involves SIRT1 inhibition andFOXO3a hyperacetylation. Int. J. Cancer 132:1013–1021; 2013.

[7] Kulkarni, S. R.; Donepudi, A. C.; Xu, J.; Wei, W.; Cheng, Q. C.; Driscoll, M. V.;Johnson, D. A.; Johnson, J. A.; Li, X.; Slitt, A. L. Fasting induces nuclear factor e2-related factor 2 and ATP-binding cassette transporters via protein kinase A andsirtuin-1 in mouse and human. Antioxid. Redox Signaling 20:15–30; 2014.

[8] Liu, T.; Liu, P. Y.; Marshall, G. M. The critical role of the class III histonedeacetylase SIRT1 in cancer. Cancer Res. 69:1702–1705; 2009.

[9] Jung-Hynes, B.; Nihal, M.; Zhong, W.; Ahmad, N. Role of sirtuin histonedeacetylase SIRT1 in prostate cancer: a target for prostate cancer managementvia its inhibition? J. Biol. Chem 284:3823–3832; 2009.

[10] Kim, M. J.; Hong, K. S.; Kim, H. B.; Lee, S. H.; Bae, J. H.; Kim, D. W.; Dao, T. T.; Oh,W. K.; Kang, C. D.; Kim, S. H. Ku70 acetylation and modulation of c-Myc/ATF4/CHOP signaling axis by SIRT1 inhibition lead to sensitization of HepG2 cells toTRAIL through induction of DR5 and down-regulation of c-FLIP. Int. J. Biochem.Cell Biol. 45:711–723; 2013.

[11] Huffman, D. M.; Grizzle, W. E.; Bamman, M. M.; Kim, J. S.; Eltoum, I. A.;Elgavish, A.; Nagy, T. R. SIRT1 is significantly elevated in mouse and humanprostate cancer. Cancer Res. 67:6612–6618; 2007.

[12] Ota, H.; Tokunaga, E.; Chang, K.; Hikasa, M.; Iijima, K.; Eto, M.; Kozaki, K.;Akishita, M.; Ouchi, Y.; Kaneki, M. Sirt1 inhibitor, Sirtinol, induces senescence-like growth arrest with attenuated Ras–MAPK signaling in human cancer cells.Oncogene 25:176–185; 2006.

[13] Sun, Y.; Sun, D.; Li, F.; Tian, L.; Li, C.; Li, L.; Lin, R.; Wang, S. Downregulation ofSirt1 by antisense oligonucleotides induces apoptosis and enhances radiationsensitization in A549 lung cancer cells. Lung Cancer 58:21–29; 2007.

[14] Li, L.; Wang, L.; Wang, Z.; Ho, Y.; McDonald, T.; Holyoake, T. L.; Chen, W.;Bhatia, R. Activation of p53 by SIRT1 inhibition enhances elimination of CMLleukemia stem cells in combination with imatinib. Cancer Cell 21:266–281;2012.

[15] Audrito, V.; Vaisitti, T.; Rossi, D.; Gottardi, D.; D'Arena, G.; Laurenti, L.;Gaidano, G.; Malavasi, F.; Deaglio, S. Nicotinamide blocks proliferation andinduces apoptosis of chronic lymphocytic leukemia cells through activation ofthe p53/miR-34a/SIRT1 tumor suppressor network. Cancer Res. 71:4473–4483;2011.

[16] Cheng, H. L.; Mostoslavsky, R.; Saito, S.; Manis, J. P.; Gu, Y.; Patel, P.; Bronson, R.;Appella, E.; Alt, F. W.; Chua, K. F. Developmental defects and p53 hyperacetyla-tion in Sir2 homolog (SIRT1)-deficient mice. Proc. Natl. Acad. Sci. USA100:10794–10799; 2003.

[17] Wang, R. H.; Zheng, Y.; Kim, H. S.; Xu, X.; Cao, L.; Luhasen, T.; Lee, M. H.; Xiao, C.;Vassilopoulos, A.; Chen, W.; Gardner, K.; Man, Y. G.; Hung, M. C.; Finkel, T.;Deng, C. X. Interplay among BRCA1, SIRT1, and Survivin during BRCA1-associated tumorigenesis. Mol. Cell 32:11–20; 2008.

[18] Wang, R. H.; Sengupta, K.; Li, C.; Kim, H. S.; Cao, L.; Xiao, C.; Kim, S.; Xu, X.;Zheng, Y.; Chilton, B.; Jia, R.; Zheng, Z. M.; Appella, E.; Wang, X. W.; Ried, T.;Deng, C. X. Impaired DNA damage response, genome instability, and tumor-igenesis in SIRT1 mutant mice. Cancer Cell 14:312–323; 2008.

[19] Cherry, A. D.; Suliman, H. B.; Bartz, R. R.; Piantadosi, C. A. Peroxisomeproliferator-activated receptor gamma co-activator 1-alpha as a critical co-activator of the murine hepatic oxidative stress response and mitochondrialbiogenesis in Staphylococcus aureus sepsis. J. Biol. Chem. 289:41–52; 2014.

[20] Wang, X. J.; Sun, Z.; Villeneuve, N. F.; Zhang, S.; Zhao, F.; Li, Y.; Chen, W.; Yi, X.;Zheng, W.; Wondrak, G. T.; Wong, P. K.; Zhang, D. D. Nrf2 enhances resistanceof cancer cells to chemotherapeutic drugs, the dark side of Nrf2. Carcinogen-esis 29:1235–1243; 2008.

[21] Arlt, A.; Bauer, I.; Schafmayer, C.; Tepel, J.; Muerkoster, S. S.; Brosch, M.; Roder, C.;Kalthoff, H.; Hampe, J.; Moyer, M. P.; Folsch, U. R.; Schafer, H. Increasedproteasome subunit protein expression and proteasome activity in colon cancerrelate to an enhanced activation of nuclear factor E2-related factor 2 (Nrf2).Oncogene 28:3983–3996; 2009.

[22] Do, M. T.; Kim, H. G.; Khanal, T.; Choi, J. H.; Kim, D. H.; Jeong, T. C.; Jeong, H. G.Metformin inhibits heme oxygenase-1 expression in cancer cells throughinactivation of Raf–ERK–Nrf2 signaling and AMPK-independent pathways.Toxicol. Appl. Pharmacol. 271:229–238; 2013.

[23] Yamakuchi, M.; Ferlito, M.; Lowenstein, C. J. miR-34a repression of SIRT1regulates apoptosis. Proc. Natl. Acad. Sci. USA 105:13421–13426; 2008.

[24] Ferreira, D. M.; Afonso, M. B.; Rodrigues, P. M.; Simao, A.; Pereira, D. M.;Borralho, P. M.; Rodrigues, C. M.; Castro, R. E. JNK1/c-Jun-activation of the p53/miRNA-34a/Sirtuin1 pathway contributes to apoptosis induced by deoxycholicacid in rat liver. Mol. Cell Biol. 34:1100–1120; 2014.

Q2 [25] Peng, Y.; Lee, J. Z.; Murr, M. S., R. Tu1758 SIRT1/Pgc-1�/NRF2 pathwaysmediate improvements in oxidative stress in rat liver after RYGB. Gastroenter-ology 144:S-1134–S-1135; 2013.

[26] Park, E. Y.; Cho, I. J.; Kim, S. G. Transactivation of the PPAR-responsiveenhancer module in chemopreventive glutathione S-transferase gene by theperoxisome proliferator-activated receptor-gamma and retinoid X receptorheterodimer. Cancer Res. 64:3701–3713; 2004.

[27] Cho, H. Y.; Gladwell, W.; Wang, X.; Chorley, B.; Bell, D.; Reddy, S. P.; Kleeberger,S. R. Nrf2-regulated PPAR{gamma} expression is critical to protection againstacute lung injury in mice. Am. J. Respir. Crit. Care Med. 182:170–182; 2010.

[28] Ulrich, S.; Loitsch, S. M.; Rau, O.; von Knethen, A.; Brune, B.; Schubert-Zsilavecz, M.; Stein, J. M. Peroxisome proliferator-activated receptor gamma asa molecular target of resveratrol-induced modulation of polyamine metabo-lism. Cancer Res. 66:7348–7354; 2006.

[29] Wang, R.; Li, J. J.; Diao, S.; Kwak, Y. D.; Liu, L.; Zhi, L.; Bueler, H.; Bhat, N. R.;Williams, R. W.; Park, E. A.; Liao, F. F. Metabolic stress modulates Alzheimer'sbeta-secretase gene transcription via SIRT1–PPARgamma–PGC-1 in neurons.Cell Metab. 17:685–694; 2013.

[30] Jemal, A.; Bray, F.; Center, M. M.; Ferlay, J.; Ward, E.; Forman, D. Global cancerstatistics. CA Cancer J. Clin 61:69–90; 2011.

[31] Pierotti, M. A.; Berrino, F.; Gariboldi, M.; Melani, C.; Mogavero, A.; Negri, T.;Pasanisi, P.; Pilotti, S. Targeting metabolism for cancer treatment and preven-tion: metformin, an old drug with multi-faceted effects. Oncogene 32:1475–-1487; 2013.

[32] Zakikhani, M.; Dowling, R.; Fantus, I. G.; Sonenberg, N.; Pollak, M. Metforminis an AMP kinase-dependent growth inhibitor for breast cancer cells. CancerRes. 66:10269–10273; 2006.

[33] Del Barco, S.; Vazquez-Martin, A.; Cufi, S.; Oliveras-Ferraros, C.; Bosch-Barrera,J.; Joven, J.; Martin-Castillo, B.; Menendez, J. A. Metformin: multi-facetedprotection against cancer. Oncotarget 2:896–917; 2011.

[34] Rucker, F. G.; Russ, A. C.; Cocciardi, S.; Kett, H.; Schlenk, R. F.; Botzenhardt, U.;Langer, C.; Krauter, J.; Frohling, S.; Schlegelberger, B.; Ganser, A.; Lichter, P.;Zenz, T.; Dohner, H.; Dohner, K.; Bullinger, L. Altered miRNA and geneexpression in acute myeloid leukemia with complex karyotype identifynetworks of prognostic relevance. Leukemia 27:353–361; 2013.

[35] Boysen, J.; Sinha, S.; Price-Troska, T.; Warner, S. L.; Bearss, D. J.; Viswanatha, D.;Shanafelt, T. D.; Kay, N. E.; Ghosh, A. K. The tumor suppressor axis p53/miR-34a regulates Axl expression in B-cell chronic lymphocytic leukemia: implica-tions for therapy in p53-defective CLL patients. Leukemia 28:451–455; 2014.

[36] Kattan, Z.; Minig, V.; Leroy, P.; Dauca, M.; Becuwe, P. Role of manganesesuperoxide dismutase on growth and invasive properties of human estrogen-independent breast cancer cells. Breast Cancer Res. Treat. 108:203–215; 2008.

[37] Zhang, Y.; Zhang, B. TRAIL resistance of breast cancer cells is associated withconstitutive endocytosis of death receptors 4 and 5. Mol. Cancer Res. 6:1861–-1871; 2008.

[38] Sanlioglu, A. D.; Dirice, E.; Aydin, C.; Erin, N.; Koksoy, S.; Sanlioglu, S. SurfaceTRAIL decoy receptor-4 expression is correlated with TRAIL resistance in MCF7breast cancer cells. BMC Cancer 5:54; 2005.

[39] Herbert, K. J.; Cook, A. L.; Snow, E. T. SIRT1 inhibition restores apoptoticsensitivity in p53-mutated human keratinocytes. Toxicol. Appl. Pharmacol.277:288–297; 2014.

[40] Glorieux, C.; Auquier, J.; Dejeans, N.; Sid, B.; Demoulin, J. B.; Bertrand, L.;Verrax, J.; Calderon, P. B. Catalase expression in MCF-7 breast cancer cells ismainly controlled by PI3K/Akt/mTor signaling pathway. Biochem. Pharmacol.89:217–223; 2014.

[41] Zakikhani, M.; Blouin, M. J.; Piura, E.; Pollak, M. N. Metformin and rapamycinhave distinct effects on the AKT pathway and proliferation in breast cancercells. Breast Cancer Res. Treat. 123:271–279; 2010.

[42] Manna, S. K.; Zhang, H. J.; Yan, T.; Oberley, L. W.; Aggarwal, B. B. Over-expression of manganese superoxide dismutase suppresses tumor necrosisfactor-induced apoptosis and activation of nuclear transcription factor-kappaBand activated protein-1. J. Biol. Chem. 273:13245–13254; 1998.

[43] Lupertz, R.; Chovolou, Y.; Kampkotter, A.; Watjen, W.; Kahl, R. Catalaseoverexpression impairs TNF-alpha induced NF-kappaB activation and sensi-tizes MCF-7 cells against TNF-alpha. J. Cell. Biochem. 103:1497–1511; 2008.

[44] Was, H.; Cichon, T.; Smolarczyk, R.; Rudnicka, D.; Stopa, M.; Chevalier, C.;Leger, J. J.; Lackowska, B.; Grochot, A.; Bojkowska, K.; Ratajska, A.; Kieda, C.;Szala, S.; Dulak, J.; Jozkowicz, A. Overexpression of heme oxygenase-1 inmurine melanoma: increased proliferation and viability of tumor cells,decreased survival of mice. Am. J. Pathol. 169:2181–2198; 2006.

[45] Cao, C.; Lu, S.; Kivlin, R.; Wallin, B.; Card, E.; Bagdasarian, A.; Tamakloe, T.;Wang, W. J.; Song, X.; Chu, W. M.; Kouttab, N.; Xu, A.; Wan, Y. SIRT1 confersprotection against UVB- and H2O2-induced cell death via modulation of p53and JNK in cultured skin keratinocytes. J. Cell. Mol. Med. 13:3632–3643; 2009.

[46] Merchant, A. A.; Singh, A.; Matsui, W.; Biswal, S. The redox-sensitivetranscription factor Nrf2 regulates murine hematopoietic stem cell survivalindependently of ROS levels. Blood 118:6572–6579; 2011.

[47] Bhalla, K.; Hwang, B. J.; Dewi, R. E.; Ou, L.; Twaddel, W.; Fang, H. B.; Vafai, S. B.;Vazquez, F.; Puigserver, P.; Boros, L.; Girnun, G. D. PGC1alpha promotes tumorgrowth by inducing gene expression programs supporting lipogenesis. CancerRes. 71:6888–6898; 2011.

[48] Chen, E. I.; Hewel, J.; Krueger, J. S.; Tiraby, C.; Weber, M. R.; Kralli, A.; Becker, K.;Yates, J. R. 3rd; Felding-Habermann, B. Adaptation of energy metabolism inbreast cancer brain metastases. Cancer Res. 67:1472–1486; 2007.

[49] Klimcakova, E.; Chenard, V.; McGuirk, S.; Germain, D.; Avizonis, D.; Muller, W.J.; St-Pierre, J. PGC-1alpha promotes the growth of ErbB2/Neu-inducedmammary tumors by regulating nutrient supply. Cancer Res. 72:1538–1546;2012.

[50] Baldelli, S.; Aquilano, K.; Ciriolo, M. R. Punctum on two different transcriptionfactors regulated by PGC-1alpha: nuclear factor erythroid-derived 2-like 2 andnuclear respiratory factor 2. Biochim. Biophys. Acta 1830:4137–4146; 2013.

[51] Sozio, M. S.; Lu, C.; Zeng, Y.; Liangpunsakul, S.; Crabb, D. W. Activated AMPKinhibits PPAR-{alpha} and PPAR-{gamma} transcriptional activity in hepatomacells. Am. J. Physiol. Gastrointest. Liver Physiol. 301:G739–G747; 2011.

[52] Lenhard, J. M.; Kliewer, S. A.; Paulik, M. A.; Plunket, K. D.; Lehmann, J. M.;Weiel, J. E. Effects of troglitazone and metformin on glucose and lipid

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132






metabolism: alterations of two distinct molecular pathways. Biochem. Phar-macol. 54:801–808; 1997.

[53] Fernandez-Marcos, P. J.; Auwerx, J. Regulation of PGC-1alpha, a nodalregulator of mitochondrial biogenesis. Am. J. Clin. Nutr. 93:884S–890; 2011.

[54] Zong, Z. H.; Du, Z. X.; Li, N.; Li, C.; Zhang, Q.; Liu, B. Q.; Guan, Y.; Wang, H. Q.Implication of Nrf2 and ATF4 in differential induction of CHOP by proteasomeinhibition in thyroid cancer cells. Biochim. Biophys. Acta 1823:1395–1404;2012.

[55] Kim, K. M.; Pae, H. O.; Zheng, M.; Park, R.; Kim, Y. M.; Chung, H. T. Carbonmonoxide induces heme oxygenase-1 via activation of protein kinase R-likeendoplasmic reticulum kinase and inhibits endothelial cell apoptosis triggeredby endoplasmic reticulum stress. Circ. Res. 101:919–927; 2007.

[56] Yamaguchi, H.; Wang, H. G. CHOP is involved in endoplasmic reticulum stress-induced apoptosis by enhancing DR5 expression in human carcinoma cells. J.Biol. Chem. 279:45495–45502; 2004.

12345678

9101112131415






Metformin induces microRNA-34a to downregulate the Sirt1/Pgc-1α/Nrf2 pathway, leading to increased...

Documents

Transcript of Metformin induces microRNA-34a to downregulate the Sirt1/Pgc-1α/Nrf2 pathway, leading to increased...