KEAP1-Dependent Synthetic Lethality Induced by AKT...

Therapeutics, Targets, and Chemical Biology

KEAP1-Dependent Synthetic Lethality Induced by AKT andTXNRD1 Inhibitors in Lung Cancer

Bingbing Dai1, Suk-Young Yoo2, Geoffrey Bartholomeusz3, Ryan A. Graham3, Mourad Majidi1, Shaoyu Yan1,Jieru Meng1, Lin Ji1, Kevin Coombes2, John D. Minna4, Bingliang Fang1, and Jack A. Roth1

AbstractIntrinsic resistance to agents targeting phosphoinositide 3-kinase (PI3K)/AKT pathway is one of the major

challenges in cancer treatment with such agents. The objective of this study is to identify the genes or pathwaysthat can be targeted to overcome the resistance of non–small cell lung carcinoma (NSCLC) to the AKT inhibitorMK2206, which is currently being evaluated in phase I and II clinical trials. Using a genome-wide siRNA libraryscreening and biologic characterization, we identified that inhibition of thioredoxin reductase-1 (TXNRD1), oneof the key antioxidant enzymes, with siRNAs or its inhibitor, auranofin, sensitized NSCLC cells to MK2206treatment in vitro and in vivo. We found that simultaneous inhibition of TXNRD1 and AKT pathways inducedrobust reactive oxygen species production, which was involved in c-jun-NH2-kinase (JNK; MAPK8) activationand cell apoptosis. Furthermore, we found that the synthetic lethality interaction between the TXNRD1 and AKTpathways occurred through the KEAP1/NRF2 cellular antioxidant pathway. Finally, we found that syntheticlethality induced by TXNRD1 and AKT inhibitors relied on wild-type KEAP1 function. Our study indicatesthat targeting the interaction between AKT and TXNRD1 antioxidant pathways with MK2206 and auranofin, aU.S. Food and Drug Administration-approved drug, is a rational strategy to treat lung cancer and that KEAP1mutation statusmay offer a predicative biomarker for such combination approaches. Cancer Res; 73(17); 5532–43.�2013 AACR.

IntroductionPrevious studies have established the critical role of the

phosphoinositide 3-kinase (PI3K)/AKT pathway in cancer cellgrowth and survival; therefore, strategies that target thispathway have been developed and tested for cancer treatment,including lung cancer (1). MK2206 (Merck, Inc.), a small-molecule inhibitor that targets the AKT signaling pathway, iscurrently being tested in phase I and II clinical trials and hasshown promising results in terms of tumor shrinkage, diseasestabilization, and patient tolerability in patients with advancedsolid tumor (2). We previously reported that a large subset ofhuman non–small cell lung carcinoma (NSCLC) cells do notrespond to MK2206 (3). We and others have also reported thatcombination with other agents, such as AZD6244 (an inhibitor

of mitogen-activated protein kinase kinase) and erlotinib [anEGF receptor (EGFR) inhibitor], sensitized some lung cancercells to MK2206 (3, 4). Those findings indicate that identifi-cation and inhibition of additional targets are needed toimprove the efficacy of targeted therapeutic agents, such asMK2206.

Thioredoxin reductase and thioredoxin (TXNRD1/TXN) isone of the major thiol-dependent electron donor systems incells; it plays a critical role in regulating the cellular redoxenvironment and in a wide range of cellular activities, such asmaintenance of viability and proliferation (5). Preclinical stud-ies indicated that auranofin, an orally active gold compoundand a TXNRD1 inhibitor, induced apoptosis in cancer cells andenhanced the antitumor activities of chemotherapeutic agents(6–10). It is currently being tested in a phase II clinical trial fortreatment of chronic lymphocytic leukemia (11), and wasrecently repurposed as an orphan drug for human amebiasistreatment (12). Other TXNRD1/TXN inhibitors, such as PG12,have also been tested clinically in solid tumors (13, 14). Kelch-like ECH-associated protein (KEAP1) and nuclear factor-ery-throid-derived-related factor 2 (NRF2) pathways is anothermajor cellular antioxidant pathways. Under normal conditions,KEAP1 binds to NRF2 and promotes degradation by cullin3(Cul3)-based E3 ligase system (15). Activated NRF2 due tooxidative stress can stimulate multiple antioxidant moleculesby binding to antioxidant response elements (ARE; refs. 16, 17).Loss-of-function mutations of KEAP1 have been reported inmany cancer types including NSCLC with frequency of about19% (18–20). KEAP1mutations cause constitutive activation of

Authors' Affiliations: Departments of 1Thoracic and CardiovascularSurgery, 2Bioinformatics and Computational Biology, and 3ExperimentalTherapeutics, The University of Texas MD Anderson Cancer Center,Houston; and 4Hamon Center for Therapeutic Oncology Research, TheUniversity of Texas Southwestern Medical Center, Dallas, Texas

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

Corresponding Author: Bingbing Dai, Department of Thoracic and Car-diovascular Surgery, Unit 023, The University of Texas MD AndersonCancer Center, 1515 Holcombe Blvd., Houston, TX 77030. Phone: 713-792-5998; Fax: 713-794-4669; E-mail: [email protected]; and Jack A.Roth, [email protected]

doi: 10.1158/0008-5472.CAN-13-0712

�2013 American Association for Cancer Research.

CancerResearch

Cancer Res; 73(17) September 1, 20135532

on July 13, 2018. © 2013 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst July 3, 2013; DOI: 10.1158/0008-5472.CAN-13-0712

http://cancerres.aacrjournals.org/

NRF2, which is associated with cancer progression, treatmentresistance, and poor patient survival (21–23).Genome-wide siRNA library screening has been used to

identify targets or networks that can be targeted for cancertherapy. siRNA library screening has also been used to identifysynthetic lethality loci with specific drugs (24–27). Usinggenome-wide siRNA library screening and biochemical andbiologic methods, we found that inhibition TXNRD1 withsiRNAs or inhibitor, auranofin sensitized lung cancer cells toMK2206 in vitro and in vivo. Our study indicates that auranofin,a TXNRD1 inhibitor and clinically proven drug that has beenused to treat rheumatoid arthritis for 25 years, could berepurposed for lung cancer therapy in combination withMK2206.

Materials and MethodsCell linesHuman lung cancer cell lines HCC193, H1993, H460, H1299,

H2126, H292, H1648, H838, H1395, H23, H358, H1703, H1666,and A549 were from the Hamon Center for TherapeuticOncology Research at The University of Texas SouthwesternMedical Center (Dallas, TX). Cell linesweremaintained in high-glucose RPMI-1640 supplemented with 5% FBS and cultured at37�C in a humidified atmosphere containing 5% CO2 and 95%air. Cell lines have been authenticated regularly with short-tandem repeat (STR) fingerprinting method.

siRNA library screeningWe used the NSCLC cell line HCC193 for genome-wide

siRNA screening, because it has a high level of p-AKT and theIC50 for MK2206 is 10 mmol/L (3) The screening concentrationof MK2206 is 1 mmol/L, a dose that can sufficiently inhibitphosphorylated AKT (p-AKT) in vitro and in vivo (2, 3). Thescreening was conducted using a Dharmacon siRNA siGEN-OME (Dharmacon) library containing 84,508 siRNAs with fourunique siRNA duplexes that target each of 21,127 uniquehuman genes. On day 1, 2,000 cells were reverse transfectedwith 0.07 mL DharmaFECT 4 in a 384-well plate, to a finalconcentration of 40 nmol/L siRNAperwell.MK2206was addedon day 3 at a 1 mmol/L final concentration. Cell growth wasanalyzed on day 5 using the CellTiter Blue method (Promega).Raw data were log2 transformed. Data were normalized using16 negative controls in each plate. The t statistics were con-ducted to compare the significant differences between thevalues of siRNA transfection and the negative control andsiRNA transfection plus MK2206 treatment. P values wereobtained using computed t statistics. b-Uniform mixturemethods were used to account for multiple tests and estimatethe false discovery rate (FDR).

Proliferation assayThe drug treatment protocol was reported in our previous

publications (3, 28). In brief, 2,000 cells was plated in 96-wellplates and treated with the indicated doses of auranofin (0–0.5mmol/L), MK2206 (0–2.5 mmol/L), or both for 72 hours. Asulforhodamine B assay (SRB) was conducted to determinecell viability. IC50 values of auranofin and MK2206 in each cellline were calculated using Curve Expert 1.3 software and

plotted in dose–response curves. Chou–Talalay method wasused to calculate the two-drug combination index (CI) usingCalcuSyn software (Biosoft; CI, 0.2–0.9, synergism; CI < 0.2,strong synergism; ref. 29).

Western blot analysisWhole-cell lysates were collected using radioimmunopreci-

pitation assay (RIPA) reagent. Western blot analysis was con-ducted with 40 mg of total cell proteins and antibodies againstphosphorylated-c-jun-NH2-kinase (p-JNK), JNK, PARP, p-AKT,AKT (Cell Signaling Technology), KEAP1, NRF2 (Santa CruzBiotechnology), and TXNRD1 (Abcam).

Gene knockdown and overexpressionCells were transfected in a 96- or 6-well platewith siRNAs (40

nmol/L) using DharmaFECT 4, and 48 hours posttransfection,cells were treated with dimethyl sulfoxide (DMSO) or MK2206for additional 72 hours. For stable gene knockdown of KEAP1,short hairpin RNAs (shRNA) in lentivirus plasmids were pur-chased from Thermo Fisher Scientific, and lentiviruses werepackaged using a kit from System Biosciences. Cells wereinfected with lentiviruses containing specific shRNAs, followedby selection with puromycin (1 mg/mL) for 4 weeks. For over-expression of dominant-negative AKT (AKT-DN) or wild-typeKEAP1 (Dsred-KEAP1), plasmids of interest were transfectedinto HCC193, H1993, and H460 cells, respectively, with Lipo-fectamine 2000 followed by selection with G418 (500 mg/mL)for 4 weeks.

Measuring reactive oxygen species productionThe reactive oxygen species (ROS) measurement method

was described in our previous publication (30). In brief, cells(1 � 106/mL) were incubated in Hank's Balanced Saline Solu-tion containing 50 mmol/L 20,70-dichlorfluorescein-diacetate(DCFHDA) for 30 minutes. The cell population fluorescence isproportional to the levels of intracellular ROS generated andwas measured using fluorescence-activated cell sorting (FACS;Becton Dickinson) at 588 nm emission.

Real-time PCRTheNRF2 downstreammolecules glutamate–cysteine ligase

catalytic (GCLC) subunit was analyzed using real-time PCR,as described previously (28). The primers for GCLC are for-ward, 50-ctgttgcaggaaggcattgat-30 and reverse, 50-ttcaaacagtgt-cagtgggtctct-30. Primers for glyceraldehyde-3-phosphate de-hydrogenase (GAPDH) are, forward, 50-atcccatcaccatcttccag-30 and reverse, 50-atgagtccttccacgatacc-30.

Glutathione assayGlutathione (GSH) levels were tested using the GSH Assay

Kit (Sigma). After the indicated treatments, GSH levels weredetermined using the GSH-reductase recycling assay with 5,50-dithiobis-(2-nitrobenzoic acid; DNTB) as substrate in micro-titer plates, following the manufacturer's instructions.

TXNRD1 activity assayCells (0.5 � 108) were washed with PBS and extracted with

CelLytic M. The lysates were centrifuged at 10,000 � g for 10

Synthetic Lethal Interaction between AKT and TXNRD1

www.aacrjournals.org Cancer Res; 73(17) September 1, 2013 5533




minutes, and supernatants were used as the enzyme sample.Tumor tissues (frommice)were extractedwith four volumes of0.25� assay buffer containing protease inhibitor cocktail usinga Potter-Elvehjem homogenizer. The samples were centrifugedfor 15 minutes at 10,000 � g. TXNRD1 activity was measuredwith a TXNRD1 activity assay kit (Sigma) using DNTB as asubstrate.

Animal experimentsAnimal experiments were carried out according to the

approved protocol (10-11-11531) by The University of TexasMD Anderson Cancer Center (Houston, TX) InstitutionalReview Board and in accordance with the Guidelines for theCare and Use of Laboratory Animals published by the NIH(Bethesda, MD). A total of 2 � 106 H1993 cells were inoc-ulated subcutaneously into the right dorsal flanks of nudemice. Treatment started when tumors reached a meanvolume of approximately 40 mm3. Animals bearing xenografttumors derived from the H1993 cell line were randomlyassigned to four treatment groups, with 5 mice in eachgroup: (i) MK2206 only (25 mg/kg; by oral), (ii) auranofinonly (5 mg/kg; by intraperitoneal), (iii) MK2206 (25 mg/kg)and auranofin (5 mg/kg), and (iv) solvent only (control).Animals were treated daily, and tumor volumes were mea-sured blindly every 3 days. The tumor volumes were calcu-lated using the formula length � width2 � 0.52. Mice wereeuthanized when their tumor volume was larger than 2,000mm3. Mean tumor growth was analyzed using the Kruskal–Wallis test for overall significance among different treat-ment groups and Mann–Whitney U test for significancebetween two treatment groups (two-tailed). Kaplan–Meiermethod was used to analyze the animal survival.

Examination of NRF2 subcellular localizationImmunofluorescence staining was conducted to determine

the subcellular localization of NRF2 in H1993 cells, before andafter treatment as described previously (31). NRF2 subcellularlocalization was also examined with enhanced GFP (EGFP)fusion protein (32). Images were obtained under a Carl ZeissDeconvolution AxioVision 4 microscope (Carl Zeiss). Nuclearand cytoplasmic fractions were prepared with hypotonic lysisbuffer (20 mmol/L Tris pH 7.5, 5 mmol/L MgCl2, 5 mmol/LCaCl2, 1 mmol/L dithiothreitol (DTT), 1 mmol/L EDTA, andprotease inhibitor cocktail). Cells were passed through 27-gauge needle on ice for 30 times followed by centrifugationat 250 � g for 5 minutes at 4�C. The supernatant was usedas cytoplasmic fraction. The nuclear pellet was thoroughlywashed in PBS and nuclear protein was extracted with RIPA.

Luciferase reporter assayCells (2 � 104) were transfected with ARE-Luc plasmid

(provided by Dr. Donna Zhang, Arizona University, Tucson,AZ) and a Renilla-luciferase control plasmid in a 12-well plateusing Lipofectamine 2000. Forty-eight hours after transfection,cells were treated with DMSO or MK2206 (1 mmol/L) for 1, 3, 6,or 24 hours. Cells were then collected for luciferase activityassay. Luciferase activity was normalized using a Renilla-luciferase control.

Statistical analysisThe t statistics were used to determine the significant dif-

ferences between the survival values of cells treated withnegative control siRNA, targeted siRNA, and targeted siRNAplus MK2206. P values were obtained using computed t sta-tistics. b-Uniform mixture methods were used to account formultiple tests and estimate the FDR. The correlation betweenKEAP1 mutation and sensitivity to MK2206 and auranofinwas determined using the Fisher exact test with two-sidedP values at 95% confidence interval. Online Ingenuity Path-way Analysis software was used to conduct a signaling path-way analysis. The significance of the in vitro cell proliferationbetween and within different treatment groups were deter-mined using Student t test (two-tailed). The significance ofthe in vivo animal study data was determined using theKruskal–Wallis test for overall significance among differenttreatment groups and Mann–Whitney U test for significancebetween two treatment groups (two-tailed). Kaplan–Meiermethod was used to analyze the animal survival.

ResultsSynthetic lethality screening identifiedMK2206 syntheticlethal hits

To identify gene candidates that could be targeted toovercome the intrinsic resistance to MK2206 in NSCLC cells,genome-wide siRNA screening was conducted with theDharmacon siRNA siGENOME library in the NSCLC cell lineHCC193. The FDR of 0.1 was used as the cutoff to define thesignificant hits. Using a P value of 0.007542048 and at least50% of growth inhibition as the cutoff, we identified 156 hits.We analyzed the significant hits using Ingenuity PathwayAnalysis software and found that targets in the NF-kB,autophagy, and cellular redox regulation pathways weresynthetically lethal with MK2206 (Fig. 1A and Supplemen-tary Table S1). Using cell proliferation assay and siRNAknockdown, we further validated the top 10 hits, includingTXNRD1, CFLAR, ATP5J, SLC9A10, and CLN10. The resultsshowed that knockdown of TXNRD1, CFLAR, SCLC9A10,AHCYL1, and CLN10 sensitized both HCC193 and H1993cells to 1 mmol/L MK2206 (Fig. 1B and C).

TXNRD1 knockdown sensitized cells to MK2206treatment by inducing cell apoptosis

To determine whether TXNRD1 expression levels correlatewith the sensitivity to MK2206, we examined the TXNRD1 andP-AKT expression in 4 MK2206-sensitive and 10 MK2206-resistant NSCLC cell lines based on our previous publisheddata (3). We found that the expression levels of TXNRD1 arelow in all 4 MK2206-sensitive cell lines, whereas the expres-sions of TXNRD1 are significantly higher in 7 of 10 resistant celllines (P< 0.01), and there is a correlation of TXNRD1 levels withMK2206 resistance (r¼ 0.66; Fig. 2A). Knockdown of TXNRD1with three different siRNAs significantly sensitized theHCC193and H1993 cells (Fig. 2B and C). Cell apoptosis analysis by flowcytometry confirmed that TXNRD1 knockdown combinedwith MK2206 induced 10% to 30% apoptosis in HCC193 andH1993 cells within 48 hours treatment, whereas no apoptosiswas induced in the cells treated with MK2206 or siRNA only

Dai et al.

Cancer Res; 73(17) September 1, 2013 Cancer Research5534




(Fig. 2D). PARP cleavage, an apoptosis marker, was alsoobserved in HCC193 and H1993 cells treated with TXNRD1siRNA and MK2206 (Fig. 2E). Those results indicate thatTXNRD1 knockdown sensitizes NSCLC cells to AKT inhibitionby inducing cell apoptosis.

TXNRD1 inhibition with auranofin sensitized cells toMK2206 treatment in vitro and in vivoTo develop a treatment strategy that would be useful

clinically, we tested the combination ofMK2206 and auranofin,a TXNRD1 inhibitor that is used for rheumatoid arthritistreatment. Inhibition of TXNRD1 by different doses of aura-nofin was first confirmed in HCC193 cells. Approximately, 80%of TXNRD1 activity was inhibited by 0.1 mmol/L auranofin (Fig.3A), which is consistent with a previous report (33). At 0.1mmol/L, auranofin did not affect the viability of HCC193 andH1993 cells as was shown in Fig. 3B. We used a dose of 0.1mmol/L, which can inhibit TXNRD1 activity but not the cellproliferation for the combination with MK2206. Combinationof MK2206 (1 or 3 mmol/L) and 0.1 mmol/L auranofin inducedsignificant growth inhibition in HCC193 and H1993 cells,P < 0.01 (Fig. 3C). Liquid colonogenic assay further confirmedthat combination of 0.1 mmol/L auranofin and MK2206 sig-nificantly reduced the colony formation in HCC193 and

H1993 cells more than 80% compared with vehicle or eachsingle treatment alone (P < 0.01; Fig. 3D and E).

To determine the synthetic lethality induced byMK2206 andauranofin in vivo, xenograft tumors were established withH1993 cells. Single-drug treatment with MK2206 or auranofinhad no effect on tumor growth, but the combination signifi-cantly inhibited tumor growth by 85% in reduction of tumorvolumes compared with each drug alone and 90% comparedwith vehicle treatment (P ¼ 0.009 in all three comparisons).Tumor growth inhibition by auranofin (AUR) or MK2206 (MK)alone was significant as compared with vehicle only at a latetime point (day 28; AUR vs. vehicle, P ¼ 0.023; MK vs. vehicle,P ¼ 0.047), but the tumor volume reduction was very minimal(�5%; Fig. 3F). This combination also significantly prolongedanimal survival (median, 76 days) compared with vehicle(median, 31 days) or single-drug treatment (median, 34 daysfor both single treatments; P < 0.001; Fig. 3G). Significantinhibition of TXNRD1 by auranofin and auranofin andMK2206 combination treatment was confirmed in animal tu-mor tissues (Fig. 3H). Immunohistochemical staining indi-cated that p-AKT was inhibited in the tumor tissue treatedwith MK2206 or the combination (Fig. 3I). These resultsindicate that the combination of MK2206 and auranofininduces synthetic lethality in vivo.

Figure 1. RNAi library screeningidentified synthetic lethality hits. A,the network of synthetic lethalityhits with AKT inhibition wasanalyzed using Ingenuity PathwayAnalysis. B and C, validations ofthe synthetic lethality hits; cellswere transfected with nontargetingcontrol siRNAs and siRNAs fromDharmacon. Forty-eight hoursafter transfection, cells weretreated with vehicle (DMSO) orMK2206 at a final concentration of1 mmol/L for another 48 hours. Cellproliferation was analyzed using anSRB colorimetric assay. Valuesare the mean � SD of 3 assays;�, P < 0.05.






Combination of auranofin and MK2206 induced ROSproduction, which mediated cell apoptosis

Because TXNRD1/TXN is one of the major cellular redoxregulatory systems that regulate cellular ROS, we hypothe-sized that combination of auranofin and MK2206 may dis-rupt the cellular ROS balance. Indeed, we found that MK2206alone induced ROS production in HCC193 and H1993 cellsby approximately 2-fold (Fig. 4A). The combination of aur-anofin andMK2206 induced robust ROS production by 5-fold(P < 0.05; Fig. 4B). To determine whether ROS induction isessential for cell apoptosis induced by MK2206 and aura-nofin, we treated cells with the ROS blocker nordihydro-guaiaretic acid (NDGA; 20 mmol/L). NDGA blocked the ROSproduction in HCC193 and H1993 cells induced by thecombination (Fig. 4C), and reversed cell growth inhibition

in both HCC193 and H1993 cells (Fig. 4D). High level of ROScan activate JNK pathway, which mediated cell apoptosis(34). Indeed, we observed that combination of MK2206 andauranofin induced activation and PARP cleavage in H1993and HCC193 cells (Fig. 4E). In addition, JNK activation andPARP cleavage were blocked by ROS inhibitor NDGA (Fig.4F). Those results indicated that combination of MK2206and auranofin induced ROS production, which mediated thecell apoptosis.

Synthetic lethality induced by auranofin and MK2206 isthrough the KEAP1/NRF2 pathway

In addition to TXNRD1, KEAP1/NRF2 is another majorcellular antioxidant system. Activated NRF2 translocatesto the nucleus and stimulates gene expression, such as

Figure 2. TXNRD1 knockdown sensitized cells to MK2206 and induced cell apoptosis. A, TXNRD1 expression level correlates with sensitivity to MK2206.Western blotting was conducted to examine TXNRD1 and p-AKT levels in 4 MK2206-sensitive and 10 MK2206-resistant cell lines (left); relative expressionlevels of TXNRD1 were calculated by normalization with loading control (right). B, knockdown of TXNRD1 with three individual siRNAs targetingTXNRD1, siRNA-T1, siRNA-T2, and siRNA-T3 sensitized the cells to MK2206 treatment compared with two nontargeting siRNA controls, siRNA-C1and siRNA-C2. C, Western blot analysis was conducted to confirm the TXNRD1 knockdown. D, FACS was conducted to analyze cell apoptosis in thecells treated with TXNRD1 siRNA and different doses of MK2206 as indicated. �, P < 0.05; ��, P < 0.01. E, Western blot analysis was conducted to examinePARP cleavage and p-AKT inhibition in the cells treated with TXNRD1 siRNA and different doses of MK2206.

Dai et al.





GCLC and NAD(P)H dehydrogenase, quinone1 (NQO1),which mediate cellular antioxidant responses (35–37). Todetermine whether AKT inhibition–induced ROS produc-

tion was due to NRF2 inhibition, subcellular localizationof NRF2 was examined using immunofluorescence stain-ing. The results showed NRF2 was excluded from the nuclei

Figure 3. MK2206 and auranofin inhibited cancer cell growth in vitro and in vivo. A, inhibition of TXNRD1 activity by auranofin. B, dose responses of HCC193and H1993 cells to auranofin. C, auranofin sensitized the HCC193 (left) and H1993 (right) cells to MK2206. D, combination of auranofin and MK2206inhibited colony formation of HCC193 (left) and H1993 (right) cells. Cells (2,000) were plated in 60-mm dishes and treated with DMSOor 0.1 mmol/L auranofinand different doses of MK2206 as indicated for 7 days. Cell colonies were stained with crystal violet and counted. E, relative colony formation wascalculated by normalization with control (DMSO) treatment as 100%. Values are the mean � SD of 3 assays; ��, P < 0.01. F, combination of auranofinandMK2206 inhibited tumor growth in vivo. Values are themean� SD, n¼ 5. AUR-MK versus vehicle: P¼ 0.009; AUR-MK versus AUR: P¼ 0.009; AUR-MKversus MK: P ¼ 0.009. G, combination of auranofin and MK2206 prolonged animal survival compared with vehicle, auranofin, or MK2206 alone, P < 0.001.H, auranofin inhibited TXNRD1 activity in vivo. Fresh tumor samples were obtained after 7 days of treatment, and tumor tissues were minced and assayedfor TXNRD1 activity. I, inhibition of p-AKT in vivo; formalin-fixed, paraffin-embedded tissue sections of animal tumor tissue specimens were used forimmunostaining with antibodies against p-AKT (Y473).






in the H1993 cells treated with 1 mmol/L MK2206 (Fig. 5A).Luciferase reporter assay confirmed that MK22006 inhib-ited GCLC promoter activity in HCC193 and H1993 cells(Fig. 5B), indicating that the transcriptional activity ofNRF2 was inhibited by MK2206. Similarly, AKT-DN inhib-ited the transcriptional activity of NRF2 in HCC193 andH1993 cells in a luciferase reporter assay (Fig. 5C and D),and reduced NRF2 in the nuclear fractions of H1993cells (Fig. 5E). Downregulation of GCLC mRNA was alsoobserved in both HCC193 and H1993 cells with a timepoint–dependent manner (Fig. 5F). Because GCLC is arate-limiting molecule for GSH synthesis and GSH is themajor ROS scavenger, we further determined GSH produc-tion in cells treated with MK2206. The GSH level wasdramatically reduced in cells treated with MK2206 orcombination in HCC193 and H1993 cells (Fig. 5G). Theseresults indicate that MK2206 inhibited NRF2 transcriptio-

nal activity and its downstream molecules, such as GCLC,which subsequently reduced the GSH level and resulted inROS induction.

Synthetic lethality induced by MK2206 and auranofindepends on the genetic status of KEAP1

The above results indicate that the synthetic lethalityinduced byMK2206 and auranofin ismediated by ROS throughconcurrent inhibition ofNRF2 andTXNRD1, twomajor cellularredox regulators. Because KEAP1mutations cause constitutiveactivation of NRF2 (38, 39), which we expect will not beinhibited by MK2206 and thereby cause resistance to MK2206and auranofin combination. To test this hypothesis, we tested 5additional KEAP1 wild-type cell lines (H1299, H23, H358, H292,and H1703), 5 KEAP1-inactive mutant cell lines (H460, A549,H2126, H1648, and H838), and 2 heterozygous mutant celllines (H1993 and H1395) for their sensitivity to treatment with

Figure 4. Auranofin- andMK2206-induced cell apoptosis was mediated by ROS induction and JNK activation. A, MK2206 induced ROS production in H1993and HCC193 cells. Cells were treated with 1 mmol/L MK2206 for different time periods, and ROS levels were analyzed by staining with H2DCFDA. B,combination of auranofin and MK2206 induced dramatic ROS production. Values are the mean� SD of 3 assays. �, P < 0.05; ��, P < 0.01. C, NDGA blockedROS production mediated by auranofin–MK2206 combination. Values are mean � SD of 3 assays. �, P < 0.05; ��, P < 0.01. D, blocking ROS reducedproliferation inhibition induced by auranofin and MK2206 combination in HCC193 and H1993 cells. ��, P < 0.01. E, combination of auranofin and MK2206induced JNK activation and PARP cleavage in H1993 and HCC193 cells. F, blocking ROS inhibited JNK activation and PARP cleavage induced by auranofinandMK2206 combination. Cells were treatedwith DMSOor 0.1 mmol/L auranofin and different doses of MK2206, as indicated, for 48 hours. Cell lysates werecollected for Western blot analysis.

Dai et al.





MK2206, auranofin, or both (18, 19). All KEAP1 wild-type celllines and heterozygous mutant cell lines were sensitive toMK2206–auranofin combination based on the IC50s ofMK2206, which are below or close to 1 mmol/L, a clinicalachievable dose (Fig. 6A and Supplementary Table S1). Incontrast, all KEAP1-mutant cells were not responsive to thecombination treatment (IC50s of MK2206 are more than the 3mmol/L; Fig. 6B and Table 1). Statistical analysis revealed thatthe cell lines' sensitivity to the combination of MK2206 andauranofin was significantly correlated with the KEAP1 geneticstatus (P¼ 0.0013); no correlationwas found between cell lines'sensitivity and other mutations, including KRAS and LKB1mutations (Table 1). Combination of auranofin and MK2206induced PARP cleavage in KEAP1 wild-type H1299 but not in

KEAP1-mutant H460 cell lines (Fig. 6C). Our results indicatethat synthetic lethality induced by MK2206 and auranofindepends on the genetic status of KEAP1. Using an EGFP–NRF2fusion protein, we confirmed that MK2206 induced NRF2nuclear exclusion in KEAP1 wild-type H1299 cells but not inKEAP1-mutant H460 cells (Fig. 6D). Moreover, MK2206 couldnot inhibit NRF2 transcriptional activity byMK2206 in KEAP1-mutant H460 and A549 cells in a Luciferase reporter assay (Fig.6E). These results indicate that constitutive activation ofNRF2 due to inactive KEAP1 mutations cannot be overcomeby MK2206, thus causing resistance to the auranofin andMK2206 combination. To further confirm this, KEAP1wasstably knocked down in KEAP1 wild-type H1299 cells (Fig.6F, left). KEAP1 knockdown rendered cells' resistance to the

Figure 5. MK2206 inhibited KEAP1/NRF2 pathway. A, MK2206 excluded NRF2 nuclear localization. H1993 cells were treated with 1 mmol/L MK2206 for theindicated time periods and fixed and stained with NRF2 antibody and fluorescein isothiocyanate (FITC)–conjugated secondary antibody. B, MK2206inhibited NRF2 transcriptional activity. C, Western blotting confirmed the overexpression of AKT-DN in H1993 and HCC193 cells. D, AKT-DN inhibitedNRF2 transcriptional activity. E, overexpression of AKT-DN reduced nuclear NRF2 in H1993 cells. F, MK2206 reduced GCLC mRNA in H1993 andHCC193 cells. Real-time PCR was conducted to determine GCLC expression in cells treated with 1 mmol/L MK2206 for the indicated time periods. G, totalGSH level was reduced by MK2206 or auranofin–MK2206 combination. HCC193 and H1993 cells were treated with auranofin, with or without differentdoses of MK2206, for 24 hours. GSH levels were analyzed using the GSH Kit. All values are the mean � SD of 3 assays; �, P < 0.05; ��, P < 0.01; DAPI,40,6-diamidino-2-phenylindole.






combination treatment (Fig. 6F, middle and right). Whereasoverexpression of wild-type KEAP1 (32), DsRed-KEAP1 (D-KEAP1) fusion gene in KEAP1-mutant H460 cells sensitized

the cells to this combination (Fig. 6G). The apoptosis marker,PARP cleavage was also induced by the combination treat-ment in H460 cells with overexpression of a wild-type KEAP

Figure 6. Synthetic lethality induced by auranofin andMK2206 depends onKEAP1 genetic status. A, combination of auranofin andMK2206 induced syntheticlethality in KEAP1 wild-type NSCLC cell lines. B, combination of auranofin andMK2206 did not induce synthetic lethality in KEAP1-mutant NSCLC cell lines.C, combination of auranofin and MK2206 induced PARP cleavage, an apoptosis marker, in KEAP1 wild-type H1299 but not in KEAP1-mutant H460 cells.D, MK2206 induced NRF2 nuclear exclusion in H1299 (KEAP1 wild-type) but not in H460 (KEAP1-mutant) cells. E, MK2206 could not inhibit NRF2transcriptional activity in KEAP1-mutant H460 and A549 cells. F, KEAP1 knockdown in H1299 cells was validated usingWestern blotting (left); knockdown ofKEAP1 renderedH1299 cells resistant to auranofin andMK2206 combination (middle and right). G, overexpression of DsRed-KEAP1 (D-KEAP1) inH460 cellswas confirmed byWestern blotting (left); overexpression of functional KEAP1 sensitized the cells toMK2206 and auranofin combination (middle and right). Allvalues are the mean � SD of 3 assays; ��, P < 0.01; NS, not significant. H, proposed working model of synthetic lethality induced by auranofin and MK2206.

Dai et al.





(Supplementary Fig. S1). These results confirmed that synthe-tic lethality induced by MK2206 and auranofin depends on afunctional KEAP1 gene. Collectively, our results indicate thatcombination of auranofin and MK2206 concurrently inhibitsTXNRD1 and NRF2, two major cellular antioxidant systems inKEAP1 wild-type NSCLC cells, and induces ROS productionand cell apoptosis. A model of synthetic lethality induced byTXNRD1 and AKT inhibitor was shown in Fig. 6H.

DiscussionIn this study, we conducted a genome-wide siRNA library

screening with NSCLC cell line HCC193 to identify the genesthat can be targeted to sensitize the cells to AKT inhibitorMK2206. This screening revealed 156 MK2206 syntheticlethality hits (genes) with a FDR of 0.1, and a minimum50% inhibition with combination of siRNA and 1 mmol/LMK2206. Those hits include previously reported autophagy-related genes, such as MAP1LC3C, cathepsin E, and cathep-sin H, and components in the NF-kB pathway, such asCASP8 and CFLAR. Inhibition of autophagy could sensitizethe cells to PI3K/AKT inhibitor treatment, which has beenreported previously (40, 41). Other hits include ATPases andATP synthase members, such as ATP5J and ATP6V1B1, bothof which are Hþ transporters. The synthetic lethality inter-action between AKT and ATPases may occur through theautophagy pathway as previous studies reported that thevacuolar membrane Hþ-ATPase plays an important role inautophagy (42, 43). The synthetic lethality hits are related todifferent pathways, and many of those hits or pathways havecross-talk (44, 45). This screening result confirmed someprevious finding such as synthetic lethality induced byautophagy and AKT inhibition but also revealed some noveltargets, such as TXNRD1, the molecules in redox balance,suggesting potential synthetic lethal interactions betweenthe redox regulation system and AKT pathways.

Previous studies indicate that other pathways, such asKEAP1/NRF2, may compensate some functions of theTXNRD1/TXN in cancer cells especially in vivo. In addition,oncogenic pathways such as KRAS andNF-kB also regulate thecellular redox balance through interaction with TXNRD1 orKEAP1/NRF2 (46–49), suggesting complicated interactionbetween oncogenic signaling pathways and cellular redoxbalance system. Here, we showed that concurrent inhibitionof TXNRD1 and AKT pathways with auranofin and MK2206,respectively, induced robust ROS production and NSCLC cellapoptosis in vitro and tumor growth inhibition in vivo, whichwas not achieved by each drug alone.More importantly, NDGA,a ROS blocker, rescued the cell growth inhibition, JNK activa-tion, and PARP cleavage induced by the auranofin andMK2206combination. Our results indicate that interruption of thecellular redox balance plays an important role in the syntheticlethality induced by simultaneous inhibition of TXNRD1 andAKT. We found that MK2206 inhibited the NRF2 activity andreduced the GSH level through downregulating GCLC expres-sion, which is consistent with a recent study that inhibition ofmTOR reduced GSH level and induced ROS in KRAS-mutantlung cancer (50). Because mTOR pathway is regulated by AKT,it is also possible that MK2206-induced ROS production maybemediated by both NRF2 andmTOR inhibition. Nevertheless,our results indicate that targeting the interactions between thePI3K/AKT oncogenic pathway and cellular antioxidant system,TXNRD1 and KEAP1/NRF2, will disrupt cancer redox homeo-stasis and might be a promising cancer therapeutic approach.

Loss-of-function mutations of KEAP1 occur in many cancertypes and KEAP1 mutations cause constitutive activation ofNRF2 (18–20). Because the combination of auranofin andMK2206 induced synthetic lethality by simultaneous inhibitionof the TXNRD1 and NRF2 pathways, we hypothesized thatKEAP1 mutations might render cancer cells resistant to thecombination of MK2006 and auranofin. Our results showed

Table 1. Mutation statuses of KEAP1 correlate with the sensitivity to MK2206 and auranofin combination

Cell line KEAP1 KRAS/NRAS LKB1 EGFR P53IC50 ofMK2206 Sensitivity

H460 Mut Mut Mut WT WT 4.04 ResistantA549 Mut Mut Mut WT WT 4.35 ResistantH838 Mut WT DL WT Mut 3.70 ResistantH2126 Mut WT Mut WT Mut 6.19 ResistantH1648 Mut WT WT WT Mut 4.96 ResistantH1395 WT/mut WT DL WT WT 0.59 SensitiveH1993 WT/mut WT Mut WT Mut 0.81 SensitiveH1299 WT Mut WT WT WT 1.38 SensitiveH1703 WT WT WT WT Mut 1.26 SensitiveH292 WT WT WT WT WT 0.76 SensitiveH23 WT Mut Mut WT Mut 0.29 SensitiveH358 WT Mut WT WT WT 1.33 SensitiveP 0.0013 1.0 0.293 1.0

NOTE: Fisher exact test was used to determine the significant correlation between mutations status of KEAP1 gene and cell lines'sensitivity to auranofin and MK2206 combination. P < 0.05 indicates significance.Abbreviations: Mut, mutant; WT, wild-type; DL, deletion.






that lung cancer cells with wild-type KEAP1 were particularlysensitive to MK2206 and auranofin, whereas cells with mutantKEAP1 were not as sensitive. This finding was further sup-ported by evidence that MK2206 induced NRF2 nuclear exclu-sion and transcriptional activity in KEAP1 wild-type cells butnot in KEAP1-mutant cells. In addition, KEAP1 knockdownin KEAP1 wild-type cells or reexpression of functional KEAP1in KEAP1-mutant cells reversed the sensitivity to auranofinand MK2206 treatment. These findings indicate that thesynthetically lethal activity induced by MK2206 and auranofindepends on a functional KEAP1 gene. Further study of thegenetic mutations in this pathway and their association withdrug sensitivity may be useful for the identification of predic-tive biomarkers.

In summary, we identified a set of synthetic loci thatcould be targeted to improve the antitumor activity of AKTinhibitor MK2206. We discovered that the synthetic lethalinteraction between the AKT pathway and TXNRD1 occursthrough the KEAP1/NRF2 antioxidant system, and targetingsuch interaction with a combination of AKT inhibitor andauranofin is a promising approach for lung cancer therapy,with KEAP1 mutation status as a genetic biomarker of drugsensitivity.

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

Authors' ContributionsConception and design: B. Dai, M. Majidi, J.D. Minna, B. Fang, J.A. RothDevelopment of methodology: B. Dai, G. Bartholomeusz, R.A. Graham,M. Majidi, S. Yan, J. Meng, J.A. RothAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): B. Dai, G. Bartholomeusz, R.A. Graham, S. YanAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis):B.Dai, S.-Y. Yoo,M.Majidi, L. Ji, K. Coombes, J.A. RothWriting, review, and/or revision of the manuscript: B. Dai, M. Majidi,K. Coombes, J.D. Minna, B. Fang, J.A. RothAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): B. Dai, M. Majidi, S. Yan, J. Meng,J.A. RothStudy supervision: B. Dai, M. Majidi, L. Ji, J.A. Roth

AcknowledgmentsThe authors thank Drs. Donna Zhang (Arizona University, Tucson, AZ) for

providing the ARE-LUC plasmid; Yue Xiong (University of North Carolina atChapel Hill, Chapel Hill, NC) for providing DesRed-KEAP1 and EGFP–NRF2plasmids; and Michael Peyton, Wei Guo, Xiaoying Liu, and Kai Xu for techniquehelp. The authors also thank Ann M. Sutton for scientific editing.

Grant SupportThis work was supported in part by the NIH through MD Anderson Cancer

Center Core Grant CA-016672—Lung Program, the Research Animal SupportFacility, NIH/NCI Specialized Program of Research Excellence (SPORE) grantCA-070907 (to J.D. Minna and J.A. Roth), SPORE Development Award (to B. Dai),and NIH/NCI R01 grant CA-124951 (to B. Fang).

The costs of publication of this article were defrayed in part by the paymentof page charges. This article must therefore be hereby marked advertisementin accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received March 11, 2013; revised May 22, 2013; accepted June 4, 2013;published OnlineFirst July 3, 2013.

References1. Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB. Exploiting the PI3K/

AKT pathway for cancer drug discovery. Nat Rev Drug Discov2005;4:988–1004.

2. Yap TA, Yan L, Patnaik A, Fearen I, Olmos D, Papadopoulos K, et al.First-in-man clinical trial of the oral pan-AKT inhibitor MK-2206in patients with advanced solid tumors. J Clin Oncol 2011;29:4688–95.

3. Meng J, Dai B, Fang B, Bekele BN, Bornmann WG, Sun D, et al.Combination treatment with MEK and AKT inhibitors is more effectivethaneachdrugalone inhumannon–small cell lungcancer in vitroand invivo. PLoS ONE 2010;5:e14124.

4. Jiang J, Mo ZC, Yin K, Zhao GJ, Lv YC, Ouyang XP, et al. Epigallo-catechin-3-gallate prevents TNF-alpha-induced NF-kappaB activa-tion thereby upregulating ABCA1 via the Nrf2/Keap1 pathway inmacrophage foam cells. Int J Mol Med 2012;29:946–56.

5. Powis G, Kirkpatrick DL. Thioredoxin signaling as a target for cancertherapy. Curr Opin Pharmacol 2007;7:392–7.

6. CoxAG,BrownKK,Arner ES,HamptonMB.The thioredoxin reductaseinhibitor auranofin triggers apoptosis through a Bax/Bak-dependentprocess that involves peroxiredoxin 3 oxidation. Biochem Pharmacol2008;76:1097–109.

7. Hedstrom E, Eriksson S, Zawacka-Pankau J, Arner ES, Seliva-nova G. p53-dependent inhibition of TrxR1 contributes to thetumor-specific induction of apoptosis by RITA. Cell Cycle 2009;8:3576–83.

8. Madeira JM, GibsonDL, KeanWF, Klegeris A. The biological activity ofauranofin: implications for novel treatment of diseases. Inflammophar-macology 2012;20:297–306.

9. Marzano C, Gandin V, Folda A, Scutari G, Bindoli A, Rigobello MP.Inhibition of thioredoxin reductase by auranofin induces apoptosis incisplatin-resistant human ovarian cancer cells. Free Radic Biol Med2007;42:872–81.

10. Mirabelli CK, Johnson RK, Sung CM, Faucette L, Muirhead K, CrookeST. Evaluation of the in vivo antitumor activity and in vitro cytotoxic

properties of auranofin, a coordinated gold compound, in murinetumor models. Cancer Res 1985;45:32–9.

11. Weir SJ, DeGennaro LJ, Austin CP. Repurposing approved and aban-doned drugs for the treatment and prevention of cancer throughpublic–private partnership. Cancer Res 2012;72:1055–8.

12. Debnath A, Parsonage D, Andrade RM, He C, Cobo ER, Hirata K, et al.A high-throughput drug screen for Entamoeba histolytica identifies anew lead and target. Nat Med 2012;18:956–60.

13. Ramanathan RK, Stephenson JJ, Weiss GJ, Pestano LA, Lowe A,Hiscox A, et al. A phase I trial of PX-12, a small-molecule inhibitor ofthioredoxin-1, administered as a 72-hour infusion every 21 days inpatients with advanced cancers refractory to standard therapy. InvestNew Drugs 2012;30:1591–6.

14. Jordan BF, Runquist M, Raghunand N, Gillies RJ, Tate WR, Powis G,et al. The thioredoxin-1 inhibitor 1-methylpropyl 2-imidazolyl disulfide(PX-12) decreases vascular permeability in tumor xenografts moni-tored by dynamic contrast enhanced magnetic resonance imaging.Clin Cancer Res 2005;11:529–36.

15. Kobayashi A, Kang MI, Okawa H, Ohtsuji M, Zenke Y, Chiba T, et al.Oxidative stress sensor Keap1 functions as an adaptor for Cul3-basedE3 ligase to regulate proteasomal degradation of Nrf2. Mol Cell Biol2004;24:7130–9.

16. Sykiotis GP, Bohmann D. Keap1/Nrf2 signaling regulates oxidativestress tolerance and lifespan in Drosophila. Dev Cell 2008;14:76–85.

17. Sekhar KR, Crooks PA, Sonar VN, Friedman DB, Chan JY, MeredithMJ, et al. NADPHoxidase activity is essential for Keap1/Nrf2-mediatedinduction of GCLC in response to 2-indol-3-yl-methylenequinuclidin-3-ols. Cancer Res 2003;63:5636–45.

18. Ohta T, Iijima K, Miyamoto M, Nakahara I, Tanaka H, Ohtsuji M, et al.Loss of Keap1 function activates Nrf2 and provides advantages forlung cancer cell growth. Cancer Res 2008;68:1303–9.

19. Singh A, Misra V, Thimmulappa RK, Lee H, Ames S, Hoque MO, et al.Dysfunctional KEAP1-NRF2 interaction in non–small-cell lung cancer.PLoS Med 2006;3:e420.

Dai et al.





20. Hammerman PS, Hayes DN,WilkersonMD, Schultz N, Bose R, Chu A,et al. Comprehensive genomic characterization of squamous cell lungcancers. Nature 2012;489:519–25.

21. Shibata T, Saito S, Kokubu A, Suzuki T, Yamamoto M, Hirohashi S.Global downstream pathway analysis reveals a dependence of onco-genic NF-E2-related factor 2 mutation on the mTOR growth signalingpathway. Cancer Res 2010;70:9095–105.

22. Zhang P, Singh A, Yegnasubramanian S, Esopi D, Kombairaju P,Bodas M, et al. Loss of Kelch-like ECH-associated protein 1 functionin prostate cancer cells causes chemoresistance and radioresistanceand promotes tumor growth. Mol Cancer Ther 2010;9:336–46.

23. Shibata T, Ohta T, Tong KI, Kokubu A, Odogawa R, Tsuta K, et al.Cancer related mutations in NRF2 impair its recognition by Keap1-Cul3 E3 ligase and promote malignancy. Proc Natl Acad Sci U S A2008;105:13568–73.

24. Astsaturov I, Ratushny V, Sukhanova A, Einarson MB, Bagnyukova T,Zhou Y, et al. Synthetic lethal screen of an EGFR-centered network toimprove targeted therapies. Sci Signal 2010;3:ra67.

25. Luo J, Emanuele MJ, Li D, Creighton CJ, Schlabach MR, WestbrookTF, et al. A genome-wide RNAi screen identifies multiple syntheticlethal interactions with the Ras oncogene. Cell 2009;137:835–48.

26. Turner NC, Lord CJ, Iorns E, Brough R, Swift S, Elliott R, et al. Asynthetic lethal siRNA screen identifying genesmediating sensitivity toa PARP inhibitor. EMBO J 2008;27:1368–77.

27. Whitehurst AW, Bodemann BO, Cardenas J, Ferguson D, Girard L,Peyton M, et al. Synthetic lethal screen identification of chemosensi-tizer loci in cancer cells. Nature 2007;446:815–9.

28. Dai B, Meng J, Peyton M, Girard L, Bornmann WG, Ji L, et al. STAT3mediates resistance to MEK inhibitor through microRNA miR-17.Cancer Res 2011;71:3658–68.

29. Chou TC, Talalay P. Quantitative analysis of dose-effect relationships:the combined effects of multiple drugs or enzyme inhibitors. AdvEnzyme Regul 1984;22:27–55.

30. Wei X, Guo W, Wu S, Wang L, Huang P, Liu J, et al. Oxidative stress inNSC-741909-induced apoptosis of cancer cells. J Transl Med 2010;8:37.

31. Na HK, Kim EH, Jung JH, Lee HH, Hyun JW, Surh YJ. (�)-Epigallo-catechin gallate induces Nrf2-mediated antioxidant enzyme expres-sion via activation of PI3K and ERK in humanmammary epithelial cells.Arch Biochem Biophys 2008;476:171–7.

32. Furukawa M, Xiong Y. BTB protein Keap1 targets antioxidant tran-scription factor Nrf2 for ubiquitination by the Cullin 3-Roc1 ligase. MolCell Biol 2005;25:162–71.

33. Omata Y, FolanM, ShawM,Messer RL, Lockwood PE, Hobbs D, et al.Sublethal concentrations of diverse gold compounds inhibit mamma-lian cytosolic thioredoxin reductase (TrxR1). Toxicol In Vitro 2006;20:882–90.

34. Pham CG, Bubici C, Zazzeroni F, Papa S, Jones J, Alvarez K, et al.Ferritin heavy chain upregulation by NF-kappaB inhibits TNFalpha-induced apoptosis by suppressing reactive oxygen species. Cell 2004;119:529–42.

35. Kobayashi A,Ohta T, YamamotoM.Unique function of theNrf2-Keap1pathway in the inducible expression of antioxidant and detoxifyingenzymes. Methods Enzymol 2004;378:273–86.

36. Ishii T, Itoh K, Yamamoto M. Roles of Nrf2 in activation of antioxidantenzyme genes via antioxidant responsive elements. Methods Enzymol2002;348:182–90.

37. Itoh K, Chiba T, Takahashi S, Ishii T, Igarashi K, Katoh Y, et al. An Nrf2/small Maf heterodimer mediates the induction of phase II detoxifyingenzyme genes through antioxidant response elements. Biochem Bio-phys Res Commun 1997;236:313–22.

38. Nioi P, Nguyen T. A mutation of Keap1 found in breast cancer impairsits ability to repress Nrf2 activity. Biochem Biophys Res Commun2007;362:816–21.

39. Wakabayashi N, Itoh K, Wakabayashi J, Motohashi H, Noda S, Taka-hashi S, et al. Keap1-null mutation leads to postnatal lethality due toconstitutive Nrf2 activation. Nat Genet 2003;35:238–45.

40. Fan QW, Cheng C, Hackett C, Feldman M, Houseman BT, NicolaidesT, et al. Akt and autophagy cooperate to promote survival of drug-resistant glioma. Sci Signal 2010;3:ra81.

41. Degtyarev M, De Maziere A, Orr C, Lin J, Lee BB, Tien JY, et al. Aktinhibition promotes autophagy and sensitizes PTEN-null tumors tolysosomotropic agents. J Cell Biol 2008;183:101–16.

42. Ramachandran N, Munteanu I, Wang P, Aubourg P, Rilstone JJ,Israelian N, et al. VMA21 deficiency causes an autophagic myopathyby compromising V-ATPase activity and lysosomal acidification. Cell2009;137:235–46.

43. Meyer H, Bug M, Bremer S. Emerging functions of the VCP/p97 AAA-ATPase in the ubiquitin system. Nat Cell Biol 2012;14:117–23.

44. Gloire G, Piette J. Redox regulation of nuclear post-translationalmodifications during NF-kappaB activation. Antioxid Redox Signal2009;11:2209–22.

45. Reynaert NL, van der Vliet A, Guala AS, McGovern T, Hristova M,Pantano C, et al. Dynamic redox control of NF-kappaB throughglutaredoxin-regulatedS-glutathionylation of inhibitory kappaBkinasebeta. Proc Natl Acad Sci U S A 2006;103:13086–91.

46. HuangHC,Nguyen T, Pickett CB. Phosphorylation ofNrf2 at Ser-40 byprotein kinase C regulates antioxidant response element-mediatedtranscription. J Biol Chem 2002;277:42769–74.

47. Huang HC, Nguyen T, Pickett CB. Regulation of the antioxidantresponse element by protein kinase C-mediated phosphorylation ofNF-E2-related factor 2. Proc Natl Acad Sci U S A 2000;97:12475–80.

48. Thu KL, Pikor LA, Chari R, Wilson IM, Macaulay CE, English JC, et al.Genetic disruption of KEAP1/CUL3 E3 ubiquitin ligase complex com-ponents is a key mechanism of NF-kappaB pathway activation in lungcancer. J Thorac Oncol 2011;6:1521–9.

49. Lee DF, Kuo HP, Liu M, Chou CK, Xia W, Du Y, et al. KEAP1 E3 ligase-mediated downregulation of NF-kappaB signaling by targeting IKK-beta. Mol Cell 2009;36:131–40.

50. De Raedt T, Walton Z, Yecies JL, Li D, Chen Y, Malone CF, et al.Exploiting cancer cell vulnerabilities to develop a combination therapyfor ras-driven tumors. Cancer Cell 2011;20:400–13.






2013;73:5532-5543. Published OnlineFirst July 3, 2013.Cancer Res Bingbing Dai, Suk-Young Yoo, Geoffrey Bartholomeusz, et al. Inhibitors in Lung CancerKEAP1-Dependent Synthetic Lethality Induced by AKT and TXNRD1

Updated version

10.1158/0008-5472.CAN-13-0712doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2013/07/09/0008-5472.CAN-13-0712.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/73/17/5532.full#ref-list-1

This article cites 50 articles, 18 of which you can access for free 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/73/17/5532To 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-13-0712

http://cancerres.aacrjournals.org/content/suppl/2013/07/09/0008-5472.CAN-13-0712.DC1

http://cancerres.aacrjournals.org/content/73/17/5532.full#ref-list-1

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/73/17/5532


KEAP1-Dependent Synthetic Lethality Induced by AKT...

Documents

Transcript of KEAP1-Dependent Synthetic Lethality Induced by AKT...