Dual mTORC1/2 Inhibition as a Novel Strategy for the ... · al. These include mTOR regulation by...

Small Molecule Therapeutics

Dual mTORC1/2 Inhibition as a Novel Strategy forthe Resensitization and Treatment of Platinum-Resistant Ovarian CancerFernanda Musa1, Amandine Alard2, Gizelka David-West1, John P. Curtin1,3,Stephanie V. Blank1,3, and Robert J. Schneider2,3

Abstract

There is considerable interest in the clinical development ofinhibitors of mTOR complexes mTORC1 and 2. BecausemTORC1 and its downstream mRNA translation effectors mayprotect against genotoxic DNA damage, we investigated theinhibition of mTORC1 and mTORC1/2 in the ability to reverseplatinum resistance in tissue culture and in animal tumormodels of serous ovarian cancer. Cell survival, tumor growth,PI3K–AKT–mTOR pathway signaling, DNA damage and repairresponse (DDR) gene expression, and translational controlwere all investigated. We show that platinum-resistantOVCAR-3 ovarian cancer cells are resensitized to low levels ofcarboplatin in culture by mTOR inhibition, demonstratingreduced survival after treatment with either mTORC1 inhibitoreverolimus or mTORC1/2 inhibitor PP242. Platinum resistanceis shown to be associated with activating phosphorylation ofAKT and CHK1, inactivating phosphorylation of 4E-BP1, the

negative regulator of eIF4E, which promotes increased cap-dependent mRNA translation and increased levels of CHK1and BRCA1 proteins. Animals with platinum-resistant OVCAR-3 tumors treated with carboplatin plus mTORC1/2 inhibitionhad significantly longer median survival and strikingly reducedmetastasis compared with animals treated with carboplatinplus everolimus, which inhibits only mTORC1. Reduced tumorgrowth, metastasis, and increased survival by mTORC1/2 inhi-bition with carboplatin treatment was associated with reducedAKT-activating phosphorylation and increased 4E-BP1 hypo-phosphorylation (activation). We conclude that mTORC1/2inhibition is superior to mTORC1 inhibition in reversingplatinum resistance in tumors and strongly impairs AKT acti-vation, DNA repair responses, and translation, promotingimproved survival in the background of platinum resistance.Mol Cancer Ther; 15(7); 1–11. �2016 AACR.

IntroductionOvarian cancer is the leading cause of mortality in gynecologic

malignancies, responsible for over 14,000 deaths annually in theUnited States (1). Seventy-five percent of patients have advanceddisease at presentation and face an approximate 30% 5-yearsurvival despite surgery and chemotherapy (1). Response toplatinum-based chemotherapy, the standard of care for ovariancancer, is excellent upfront, even in high-grade serous cancer(HGSC) (reviewed in ref. 2), although most patients with HGSCwill ultimately relapse, with approximately 25% of patientsacquiring de novo resistance during primary treatment or relapsing

within 6 months (reviewed in ref. 3). In the recurrent setting,effective options are lacking for women with platinum-resistantdisease. Recent research efforts such as the Cancer Genome Atlas(TCGA) project have highlighted the need for better moleculartargets to improve treatment (4–6).

Resistance to platinum-based chemotherapy is thought toinvolve three mechanisms: reduced drug uptake or increasedexport, increased DNA repair mechanisms, or alterations in proa-poptotic mechanisms or genes (reviewed in ref. 7). A primarymechanism for platinum/carboplatin resistance is the inactiva-tion of the p53 gene (TP53; refs. 8, 9). Analysis of the TCGA dataindicates that over 95% of HGSC tumors contain mutations inTP53 (4–6, 10). HGSC ovarian cancers also harbor a variety ofextensive copy number changes, a variety of genetic alterationsthat disrupt homologous DNA recombination repair, and a highlevel of intratumoral heterogeneity in the absence of prevalentoncogenic driver mutations, apart from TP53mutation (4, 5, 11).One notable exception, however, is the high prevalence (15%–

20%) of germline or somatic mutation of the DNA repair BRCA-1or -2 genes (reviewed in refs. 3, 5, 6) or hypermethylation (down-regulation) of the BRCA-1 transcriptional promoter (5).However,whereas BRCA-1/2–mutated ovarian cancers are generally moremetastatic, they are also typically more responsive to platinum-based chemotherapy (12, 13). A recent whole-genome analysis ofHGSC ovarian cancers found that platinum resistance in a subsetof patients was associated with either reversion of BRCA-1 orBRCA-2 mutations or loss of BRCA-1 promoter methylation (5).Treatment failure was also associated with a higher level of

1Department Obstetrics and Gynecology, NYU School of Medicine,New York, New York. 2Department of Microbiology, NYU School ofMedicine, NewYork, NewYork. 3NYU Cancer Institute, NewYork, NewYork.

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Current address for F. Musa: Winthrop University Hospital, Mineola, New York,NY 11501; current address for G. David-West: New York Medical College,Westchester Medical Center, Hawthorne, NY 10532.

Corresponding Author: Robert J. Schneider, NYU School of Medicine,Alexandria Center for Life Sciences, 450 East 29th Street, 3rd Floor, New York,NY 10016. Phone: 212-263-6006; Fax: 646-501-4541; E-mail:[email protected]

doi: 10.1158/1535-7163.MCT-15-0926

�2016 American Association for Cancer Research.

MolecularCancerTherapeutics

www.aacrjournals.org OF1

on June 4, 2020. © 2016 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst May 16, 2016; DOI: 10.1158/1535-7163.MCT-15-0926

http://mct.aacrjournals.org/

CCNE1 (cyclin E) mutation (5, 14), implicating features of cell-cycle control. That said, at least one-third of tumors have not beenfound to have any discernable genetic pattern, mutation in drivergenes or tumor suppressor genes, or alterations in mRNA ormiRNA expression (5). Thus, there is a need to explore the roleof posttranscriptional control mechanisms (largely translationcontrol) in ovarian cancer platinum resistance, which has notbeen conducted by studies performed to date. Moreover, theimplication of survival and cell-cycle control pathways inHGSC platinum resistance aligns well with established rolesfor mRNA translational regulation in these pathways (reviewedin ref. 15).

Protein synthesis overall, and selective mRNA translationalregulation in particular, are highly targetable for cancer therapybecause they constitute a tightly regulated process that is oftenupregulated in human cancers with metastatic progression,accounting for over 50% of the energy requirements of a tumorcell (15). Most major oncogenic signaling pathways known to beinvolved in human cancers converge on protein synthesis at thelevel of the protein kinase mTOR, a key regulator of cellularmetabolism and mRNA translation. As has been seen in othersolid tumors, in ovarian cancer, the PI3K/AKT/mTOR pathway isoften hyperactivated and promotes translational reprograming ofthe ovarian cancer cell, resulting in selective translation of more"oncogenic" mRNAs that promote cell proliferation, survival,DDR, andother key pathways (15).Manyof thesemRNAs containlong and/or more highly structured 50 untranslated regions(50UTR) and/or common sequence motifs, and benefit the great-est from increased mTOR activity (16–20). In fact, the mTORpathway is hyperactivated in approximately 70% of epithelialovarian cancers, where it contributes to tumorigenesis and che-moresistance (21).

Cap-dependent mRNA translation is largely regulated by thecomplex mTORC1, consisting of mTOR, Raptor, and GbLproteins (22). mTORC1 is responsible for the phosphorylation(inactivation) of 4E-BP1, the negative regulator of cap-bindingprotein eIF4E. 4E-BP1 binds and blocks the activity of thetranslation initiation factor eIF4E by competing for interactionwith translation initiation factor eIF4G, a molecular scaffold forthe translation initiation machinery (23). eIF4E binds themethyl-7 GTP cap on mRNAs, recruiting the translation initi-ation complex consisting of eIF4G and the mRNA helicaseeIF4A, which are required for mRNA translation (15, 24–27). Weand others have shown that hyperactivated mTORC1 and itsdownstream effectors (eIF4E, eIF4G) drive the selective transla-tion of specific oncogenic mRNAs encoding proteins responsiblefor survival factors (survivin, Mcl1, XIAP), tumor angiogenesis(VEGF-A, FGF2), and the DDR (BRCA1, 53BP1, gH2AX, others),among other pathways (28).

Despite compelling molecular reasons for their clinical use,currently available mTORC1 rapamycin-based allosteric inhibi-tors (rapalogs) including everolimus, temsirolimus, and othershave shown limited anticancer activity in clinical trials,marked bypoor response or rapid development of tumor resistance. More-over, rapalogs such as everolimus only weakly inhibit mTORC1,which can increase AKT activity by positive feedback regulationfrom loss ofmTORC1 suppression of PI3K/AKT by S6K/IRS1, andfrom mTORC2 itself, that is not a target of rapalog inhibition(29, 30). mTORC2 consists of mTOR, Rictor, and GbL proteins.mTORC2 acts upstream from mTORC1 to phosphorylate andactivate pro-oncogenic AKT at serine 473, leading to increased

cancer cell proliferation and survival (31). In addition, rapalogsonly poorly block mTORC1 phosphorylation of certain down-stream targets, particularly 4E-BP1 (32, 33), and thus, havelimited activity as translation downregulators. Moreover, manycancers have intrinsic or acquired resistance to rapalogs, in manycases paralleling the acquisition of resistance to genotoxic ther-apies (34, 35). Ultimately, rapalog resistance reflects molecularand biochemical pathway alterations that maintain cancer cellproliferative pathways that are also targets of genotoxic therapies.

The recent development of catalytic (ATP active site) mTORinhibitors capable of blocking both mTORC1 and mTORC2activity has stimulated interest in this pathway as a means ofbypassingmechanisms of rapalog resistance and possibly increas-ing chemotherapeutic sensitivity. Emergingdata suggest that thesecompounds efficiently inhibit the feedback upregulation of AKTbymTORC2, and are also superior inhibitors of mTORC1 activityand its downstream inactivation of 4E-BP1 (32). In this regard,while the anticancer cell and anti-cell proliferation effects ofmTORC1 inhibition involve the downregulation of protein syn-thesis, possibly selectively, mTORC1 also impacts on other keypathways and proteins that regulate cell proliferation and surviv-al. These include mTOR regulation by phosphorylation of theautophagy pathway (36) through the ULK1–mAtg13–FIP200complex (37), the growth factor receptor bound protein 10(Grb10) protein, which blocks growth factor mitogenic signaling(38, 39), and the phosphorylation of mRNA-binding proteinssuch as LARP1 that regulates cell division (40), among others.Consequently, if chemotherapy resistance can be reversedthrough mTOR inhibition, it might or might not directly involvemTORC1 phosphorylation of protein synthesis effectors S6K and4E-BP1.

We therefore investigated the use of mTORC1-only andmTORC1/2 dual inhibitors in reversing platinum-resistant ovar-ian cancer and the role of selective cap-dependent mRNA trans-lation. Using tissue culture and animal tumor models of high-grade platinum-resistant serous ovarian cancer, we demonstratethat only mTORC1/2 inhibition significantly restores platinumsensitivity, which is associated with both inhibition of activatingAKT phosphorylation and downregulation of cap-dependentmRNA translation of key proteins involved in survival and DDRpathways. This results in reduced cell proliferation and cellsurvival responses, platinum resensitization, increased tumor cellkilling, improved tumor control, reduced metastasis, andincreased survival in an animal model. Several key mRNAs areinvolved in these responses and are sensitive to mTORC1/2inhibition, including CHK1 and BRCA-1. As hyperactivation ofmTOR is an integrative step between several oncogenic pathways(AKT, MAPK, RAS, and others) that are commonly altered inovarian cancer, we believe there is rationale to pursuemTORC1/2inhibition as a novel therapeutic target for this disease.

Materials and MethodsCell lines

Platinum-resistant OVCAR-3 cells were obtained from theATCC. OVCAR-3 cells were authenticated by the ATCC usingshort tandem repeat (STR) profiling and used within 6 monthsafter resuscitation. OVCAR-3 cells were derived from a patientwith a platinum-resistant recurrence of high-grade serous ovariancancer and shown to have acquired increased resistance to plat-inumdrugs (41) and at dose levels similar to those used here (42).

Musa et al.

Mol Cancer Ther; 15(7) July 2016 Molecular Cancer TherapeuticsOF2




An OVCAR-3 cell line engineered to express GFP and Fireflyluciferase (FFLuc) genes by retroviral transduction was kindlyprovided by Dr. Renier Brentjens (Memorial Sloan KetteringCancer Center, New York, NY; ref. 43). Cells were isolated byFACS and sorted for high GFP expression and firefly luciferaseactivity to enable animal studies using bioluminescent imaging(IVIS Lumina II). These cells were not independently validated byus. OVCAR-3 cells were cultured at 37�C in 5%CO2 in RPMI1640media supplemented with 20% FBS, penicillin/streptomycin,amphotericin B, and plasmodium prophylaxis.

Platinum-resistant SKOV-3 cells were obtained from the ATCC.SKOV-3 cells were authenticated by the ATCC using STR profilingand used within 6 months after resuscitation. SKOV-3 cells wereoriginally derived from the ascites of a patient with platinum-resistant epithelial ovarian cancer and shown to have acquiredincreased resistance to platinum drugs at dose levels similar tothose used here (44, 45). SKOV-3 cells were cultured at 37�C in5% CO2 in McCoy 5a medium modified (ATCC cat. #30-3007)supplemented with 10% FBS, penicillin/streptomycin, ampho-tericin B, and plasmodium prophylaxis.

Cells were routinely tested and found to be mycoplasma free.Prior to cell culture and tumor growth into animals, cells werecultured in media enriched with FBS but free of antibiotics,antifungals, or other additives. Cells were tested for mycoplasmaevery 2 months and found to be free of infection.

DrugsPP242 (Chemdea) is a dual mTORC1/2 inhibitor. PP242 is

highly selective for both mTOR complexes, and from 10-to 100-fold more specific than for other members of the PI3K family. Atdose levels used here, PP242 does not significantly target kinasesother thanmTORC1/2 (46). Everolimus (RAD001, BioTang) is anallosteric mTORC1 inhibitor. For tissue culture studies, PP242and everolimus were diluted into 10% DMSO to the indicatedconcentrations. For animal experiments, PP242was suspended in15% polyvinyl-pyrrolidone (PVP) þ 5% N-methylpyrrolidone.Everolimus was suspended in PBS with 10% DMSO. Controlgroups in tissue culture experiments were treated with PBSþ 10%DMSO. For animal studies, control groups were treated withsterile pH 7.2 buffered water in 15% PVP þ 5% N-methylpyrro-lidone. Carboplatin was chosen as the DNA-damaging agentgiven its conventional use in the treatment of ovarian cancer andfavorable toxicity profile. Carboplatin was used at a stock dose of10 mg/mL diluted in PBS to the desired concentration.

Clonogenic assaysCells (100,000) were plated into 6-well plates and allowed to

settle over 24 hours, studies were repeated five times. Cells weretreated as follows: (i) PP242 alone, (ii) everolimus alone, (iii)carboplatin alone, (iv) PP242 and carboplatin, (v) everolimusand carboplatin, (vi) control. Treatment included PP242 or ever-olimus over 24 hours versus control (PBS þ 10% DMSO). At theend of 24 hours, carboplatin-treated groups were treated for 5hours with carboplatin, and control groups weremock treated for5 hours with PBS. After treatment, cells were washed thoroughlywith PBS and the media replaced, then incubated for 10 daysduring which the media were changed every 3 days. Cells werefixed and stained with crystal violet dye, and colonies counted bylight microscopy from 6 randomly chosen fields twice and aver-aged. Colony diameters were assessed using ImageJ64 software asdescribed previously (47).

Dose finding and growth inhibitory 50% (GI50) experimentsCellswere plated into 96-well plates and treatedwith increasing

doses of PP242, everolimus, and carboplatin as single agents(Supplementary Fig. S1). Cells were treated for 24 hours witheverolimus or PP242, or 5 hours with carboplatin. After 72 hours,cells were harvested and growth quantified using MTT assays asdescribed by the manufacturer (Promega #G4000). GI50 doseswere established from these data as follows: tissue culture: PP242,2.5 mmol/L; everolimus, 20 nmol/L; and carboplatin, 1.0 mmol/L.The same doses were maintained for treatment combinations.

ImmunoblottingFor immunoblotting, the following antibodies from Cell Sig-

naling Technology were used unless otherwise indicated: rabbitanti-S6K (#2217), rabbit anti-PS6K (Thr389, #2211), rabbit anti-4E-BP1 (#9452), rabbit anti-P-4E-BP1 (Ser65, #9451), rabbitanti-AKT (#9272), rabbit anti-P-AKT (Ser473, #9271), all usedat 1:1,000. For DNA repair targets: mouse anti-CHK1 (#2360),rabbit anti-P-CHK1 (Ser345, #2348), rabbit anti-CHK2 (#2662),rabbit anti-P-CHK2 (Thr68, #2661), rabbit anti-BRCA1 (#9010),rabbit anti-P-BRCA1 (Ser1524, #9009), mouse anti-ATM(#2873), mouse anti-P-ATM (Ser1981, #4526), rabbit anti-ATR(#2790), rabbit anti-mTOR (#2983), rabbit anti-p-mTOR (Ser2448, #2448) rabbit anti-P-ATR (Ser428, #2853), and anti-PARP(#9542). Loading controls weremouse anti-GAPDH (#2118) andmouse anti-b tubulin (#2146). Other targets were survivin(#2802) and caspase 3 (#9662).

After treatments of tissue culture cells, cells were washed twicein ice-cold PBS and lysed in 0.5%NP-40 lysis buffer at 4�Cor0.5%SDS lysis buffer. NP-40 lysates were clarified by centrifugation at13,000� g for 10minutes andprotein concentrations determinedby Bradford method (Bio-Rad). To determine the total levels andphosphorylation status of specific proteins, equal amounts ofprotein were resolved by SDS-PAGE and transferred to polyviny-lidene difluoride membranes (Millipore). The phosphorylationstatus of proteins was determined either by immunoblotting themembrane first with P-specific antibody then stripping the mem-branes using Restore Western blot stripping buffer (Pierce), fol-lowed by reprobing the membranes with non-P–specific antibo-dies, or by running two separate gels. Tumor immunoblottingwasconducted as above by combining two tumors for each conditionsurgically removed at day 14 after treatment and soluble proteinsextracted using 0.5% SDS lysis buffer. Equal protein amountswere used for SDS-PAGE.

ImmunofluorescenceCells were grown and treated on chamber slides, fixed, and

immunostained for anti-53BP1 (Abcam, #ab36823) and gH2AX(Cell Signaling Technology, #9718). Secondary antibodies wereconjugated with fluorescent marker (FITC, Jackson) for 1 hour.After washing, slides were allowed to air dry. Fluorescent foci werecounted from 6 fields chosen at random as a surrogatemeasure ofDNA double-strand breaks.

Assessment of mRNA levels by qRT-PCRForward and reverse primers were designed to detect the

following mRNAs: 4E-BP1, AKT, BRCA1, BRCA2, ATM, ATR,CHK1, CHK2, rS6, GAPDH (control). RNA was extracted fromcells treated as follows: (i) PP242 alone, (ii) everolimus alone,(iii) carboplatin alone, (iv) PP242 and carboplatin, (v) ever-olimus and carboplatin. cDNAwas synthesized from the extracted

mTORC1/2 in Ovarian Cancer Platinum Resistance

www.aacrjournals.org Mol Cancer Ther; 15(7) July 2016 OF3




and quantified sample of RNA (GoScript, Promega). Real-timePCR was performed in triplicate using SYBR Green (Life Tech-nologies) on a 7500 Fast-Dx RT-PCR Instrument (Applied Bio-systems). The DCt was calculated using the 7500 software and thedata were further normalized to control (GAPDH). Fold changevalues relative to the GAPDH control were calculated andreported as graphs designed on GraphPad Prism.

Metabolic protein labeling and protein synthesis ratesThe effect of PP242 on mRNA translation was assessed using

[35S]-methionine metabolic labeling into nascent proteins.OVCAR-3 cells were cultured and treated for 24 hours with: (i)DMSO control; (ii) carboplatin at 10 mmol/L; (iii) PP242 at 2.5mmol/L; and (iv) carboplatin and PP242 in combination. Aftertreatment, cells were labeled with 25 mCi of [35S]-methionine/cysteine/mL (EasyTag Express Protein Labeling Mix, PerkinElmer) in Met/Cys-free DMEM supplemented with gentamicinat 0.04 mg/mL, 5% FBS, and bovine insulin at 0.01 mg/mL, andincubated at 37�C for 30 minutes. Lysates were prepared usingNP-40 buffer and specific activity of [35S]-methionine/cysteineincorporation into nascent protein was determined by trichlor-oacetic acid (TCA) precipitation onto GF/C filters and liquidscintillation counting. Studies were repeated three times and datawere presented as means, normalized to the control, with SEM.

Intraperitoneal animal modelAll studies were approved by the NYU School of Medicine

Institutional Animal Care and Use Committee (IACUC) andconducted in accordance with IACUC guidelines. Female SCIDbgmice, age 5 to 6 weeks (Taconic Farms, Inc.) were implantedintraperitoneally with 6 � 106 cells (OVCAR-3 expressing fireflyluciferase and GFP). Tumor growth was assayed by biolumines-cent imaging (IVIS Lumina II) every 2 weeks until flux over thedefined region of interest (ROI) of the entiremouse abdomenwasapproximately 108 photons/second. Treatment allocation wasmade randomly in groups of 10: (i) PP242, (ii) everolimus, (iii)carboplatin, (iv) PP242 and carboplatin, (v) everolimus andcarboplatin, and (vi) control vehicle. As the GI50 responses ofOVCAR-3 and SKOV cells were within the dosing range of manyother cell lines, for animal studies, established dosing regimensand doses were used consistent with the human equivalent drugexposure and near MTD levels: PP242 at 100 mg/kg daily by oralgavage, everolimus at 2.5mg/kg daily by oral gavage (46, 48), andcarboplatin at 30mg/kg every 4 days i.p. The same dose levels andschedules were used for combinatorial dosing.

Assessment of tumor treatment response and toxicityAnimals were imaged weekly by intraperitoneal injection of

200-mL Luciferin (diluted in PBS as per manufacturer's instruc-tions) and analyzed using Living Image software. The ROI wasdefined as an ellipse surrounding the mouse abdomen. Flux wascalculated within the ROI per mouse and averaged across all micewithin each treatment group. Average flux was compared acrosstreatment groups over time and normalized to baseline flux levelsprior to initiation of treatment to allow the assessment of relativetumor growth, independent of starting volume. Studies werecarried out independently in triplicate. Mice were sacrificed ifthey appeared ill as per Institutional Animal Care and UseCommittee (IACUC) protocol. Tumors were collected from rep-resentative mice within each treatment group for correlativestudies. Animals were inspected daily for overt signs of toxicity

or increased tumor burden as per IACUC protocol. Percentageweight change at every imaging time-point was used as a surrogatefor toxicity.

Assessment of sub-G0–G1 cell population by flow cytometryDNA fragmentationwas assessed bydetermining the fractionof

cell nuclei in the sub-G0–G1 phase by flow cytometry withpropidium iodide (PI) staining. After cell harvest, cells were fixedwith 2% formaldehyde for 30minutes on ice, incubatedwith 70%ethanol for 15minutes at 37�C, treatedwith 40mg/mLRNaseA for30 minutes at 37�C, washed in cold PBS, resuspended in PI at 50mg/mL, and then subjected to FACSCalibur flow cytometry (Bec-tonDickinson), scoring 10,000 events for each sample, carriedoutthree times.

Statistical analysesColony counts and colony diameterweremultiplied to develop

a colony burden score accounting for the total effect observed.Colony counts, diameters, and their products were comparedacross treatment groups using nonparametric ANOVA methods.Paired tests used Student t tests. Statistical analyses were per-formed on SPSS and GraphPad Prism. Median survival wascalculated for each treatment according to the Kaplan–Meiermethod (SPSS version 21).

ResultsProliferation of carboplatin-resistant ovarian cancer cells isonlymodestly inhibited by long-termor short-termcarboplatintreatment with mTORC1 or mTORC1/2 inhibition

We first determined the GI50 for OVCAR-3 cells using an MTTassay for single-agent everolimus, PP242, and carboplatin(Supplementary Fig. S1). The values established are in keepingwith the literature for OVCAR-3 cells (PP242, 2.5 mmol/L;everolimus, 20 nmol/L; carboplatin, 1.0 mmol/L; e.g., 49,50–52). Using the GI50 dose level for each agent, as well asa low dose of carboplatin (1.0 mmol/L) that does not affect cellproliferation or survival, we conducted a single-agent doseresponse 10-day clonogenic cell survival analysis. Carboplatinat subphysiologic (1.0 mmol/L) levels also established a base-line for combinatorial testing to determine whether synergisticcell killing can be established. Clonogenic analysis utilizes cellsurvival over an extended period time, providing a morerelevant and biologically functional determination of cell pro-liferation and survival. Cells were treated with each agent for 24hours prior to plating. Viable cell colonies�40 cells in size werecounted by light microscopy and total colony counts comparedacross all treatment groups. Inhibition of mTORC1 (everoli-mus) and mTORC1/2 (PP242) showed similar single-agentinhibition of cell proliferation and colony size at the physio-logic GI50 doses as first established by MTT assay (Fig. 1A).

We next determined whether the failure to observe a significantdifference between inhibition of mTORC1 and mTORC1/2 inresensitization to carboplatin in resistant ovarian cancer cells is aresult of toxicity from long-term (24-hour) exposure to carbo-platin. Clonogenic 10-day cell survival studies were performed asabove but after 24-hour exposure to mTOR inhibitors, with only5-hour carboplatin treatment. Results are shown for theplatinum-resistant ovarian cancer cell lines OVCAR-3 (Fig. 1B) and SKOV-3(Supplementary Fig. S2). Again there was a significant decrease inboth colony counts for both cell lineswith carboplatin at sub-GI50levels (1 mmol/L) when combined with PP242 or everolimus, but

Musa et al.





similar to carboplatin alone at its GI50 level (10 mmol/L). Toaccount for cytostatic cell growth–inhibitory effects of mTORinhibitors that may not necessarily lead to a decrease in colonynumber, the average colony diameter in two dimensions wasestimated using ImageJ64 software to develop a colony burdenscore, calculated by multiplying colony counts by colony diam-eter (47). Inhibition ofmTORC2provided no additional decreasein colony diameter per 40-cell number compared with inhibitionof mTORC1 only (Fig. 1B). The similar decrease in colony countsand average colony diameter score for both mTORC1 (everoli-mus) and mTORC1/2 (PP242) inhibition is understandablegiven the fact that mTORC1 inhibition by either agent impairscell proliferation,whichwouldbe expected tohave amuch greaterimpact on colony size (dominated by cell number) than inhibi-tion of cell growth (size) by mTORC2 inhibition (19). In sum-mary, the antiproliferative effects of the mTOR inhibitors onplatinum-resistant OVCAR-3 cells by clonogenic assay weremod-

est and just slightly less than single-agent carboplatin at its GI50dose.

Catalytic active site inhibition of mTORC1 is a more effectiveinhibitor of 4E-BP1 activation than allosteric everolimus

Only the catalytic (PP242) mTOR inhibitor was associatedwith decreased target ribosomal S6K (rS6K) and 4E-BP1 proteinphosphorylation at GI50 dose levels (2.5 mmol/L PP242) withno change in total protein levels (Fig. 2A). At the GI50 dose level(20 nmol/L), everolimus did not reduce phosphorylation ofS6K or 4E-BP1 but did so at 2.5 times the GI50 (50 nmol/L),suggesting that OVCAR-3 cells are either intrinsically resistantor have acquired resistance to rapalogs coincident with resis-tance to carboplatin. It is not surprising that inhibition of cellproliferation by everolimus can be uncoupled from mTORC1phosphorylation of S6K and 4E-BP1 targets. Studies haveshown that rapalog resistance can arise in a number of ways,

0

50

100

150

0

50

100

150

200

250

A

B

Colony counts

Cel

l col

onie

s (>

40 c

ells

)

Colony diameter (mean)

Mea

n co

lony

dia

met

er (p

ixel

s)

Contro

l

Carbop

latin

1 μmol/

L

Carbop

latin

1 μmol/

L

PP242 2

.5 μm

ol/L

PP242 2

.5 μm

ol/L

Everol

imus

20 nm

ol/L

Everol

imus

20 nm

ol/L

Carbo 1

μmol/

L + P

P242 2

.5 μm

ol/L

Carbo 1

μmol/

L + P

P242 2

.5 μm

ol/L

Carbo 1

μmol/

L + ev

erolim

us 20

nmol/

L

Carbo 1

μmol/

L + ev

erolim

us 20

nmol/

L 0

50

100

150

0

50

100

150

200

250

0

10,000

20,000

30,000

Colony diameter (mean)Colony counts

Cel

l col

onie

s (>

40 c

ells

)

Mea

n co

lony

dia

met

er (p

ixel

s) Colony burden score

Mea

n co

lony

cou

nt x

mea

n co

lony

dia

met

er

Contro

l

Contro

l

Carbop

latin

1 μmol/

L

PP242 2

.5 μm

ol/L

Everol

imus

20 nm

ol/L

Carbo 1

μmol/

L + P

P242 2

.5 μm

ol/L

Carbo 1

μmol/

L + e

verol

imus

20 nm

ol/L

* ** * * *

Contro

l

Carbop

latin

1 μmol/

L

PP242 2

.5 μm

ol/L

Everol

imus

20 nm

ol/L

Carbop

latin

10 μm

ol/L

Contro

l

Carbop

latin

1 μmol/

L

PP242 2

.5 μm

ol/L

Everol

imus

20 nm

ol/L

Carbop

latin

10 μm

ol/L

Carbop

latin

10 μm

ol/L

Carbop

latin

10 μm

ol/L

Carbop

latin

10 μm

ol/L

**

**

**

Figure 1.

mTORC1 and mTORC1/2 inhibition sensitizes platinum-resistant OVCAR-3 cells to carboplatin-mediated growth inhibition. A, 10-day clonal cell survival assays with24-hour drug treatments. Data demonstrate reduced colony number and diameter with 24-hour inhibition of mTORC1 (everolimus), mTORC1/2 (PP242), orcarboplatin at GI50 dose levels. Carboplatin also used a low (1 mmol/L) level. Studies were performed five times and are represented as mean � SEM. � , P < 0.001,by t test. Control, untreated (vehicle only). B, 10-day clonal cell survival assay with 24-hour mTOR inhibitor and 5-hour carboplatin (Carbo) treatment. Datademonstrate reduced number and diameter of colonies, and reduced colony burden (number x size) at GI50 dose levels. Studies were performed five times.� and �� , P < 0.001 by t test. Only colonies over 40 cells were scored. Colony diameters were scored by pixel breadth.






including mutation of mTOR and loss of the FKBP12 regulatorybinding site (53), and/or altered (typically reduced) expressionof rapamycin binding FK506-binding protein family members,which confer differential mTOR and other target specificities(54). Among these, for example, is the control of the PHLPP byFKBP51, which regulates proproliferative AKT activity (55).This is consistent with the observation that with increasedtransformation cells lose responsiveness to rapalog inhibitionof S6K phosphorylation but not cell-cycle inhibition (56). Thecharacteristic phosphorylation of AKT in cancer cells was alsohighly resistant to everolimus, but not to PP242 (Fig. 2A). Veryhigh doses (50 nmol/L) of everolimus were also necessary todownregulate activating phosphorylation of mTOR, whereas

GI50 levels of PP242 (2.5 mm) were sufficient. At 2.5 times theGI50 levels of everolimus, there was only modest evidence forcell killing, indicated by PARP cleavage, compared with cata-lytic mTOR inhibition. Consistent with the resistance to ever-olimus, OVCAR-3 cells were only very slightly reduced inprotein synthesis (�10%) by GI50 levels of everolimus, butmore strongly impaired (�30%) by PP242, regardless of car-boplatin treatment (Fig. 2B). Carboplatin treatment alone hadno effect on mTOR signaling or protein synthesis. These find-ings indicate that catalytic inhibition of mTOR is more effectivethan allosteric inhibition in reducing mTOR signaling andprotein synthesis, yet in tissue culture there is no gain inovarian cancer cell killing or resensitization to carboplatin bymTORC1/2 inhibition.

Inhibition of mTORC1/2 is superior to mTORC1 in treatmentand control of platinum-resistant ovarian cancer in axenotransplant model

We therefore examined the ability of allosteric mTORC1 andcatalytic mTORC1/2 inhibitors to control ovarian tumor devel-opment and metastasis in a xenotransplant model of OVCAR-3cells. We sought to determine whether the ability to block bothmTORC1 and mTORC2 as well as or in addition to a greaterability to block 4E-BP1 might provide for resensitization tocarboplatin and decreased tumor growth not observable in cellculture studies. Mice were injected intraperitoneally (i.p.) withOVCAR-3 cells that were previously engineered to constitutive-ly express firefly luciferase and a GFP reporter to enable bio-luminescence tracking (43) and imaged once a week. A strongmetastatic signal could be seen by bioluminescence imaging by80 days (Fig. 3A). Using Living Image software, the flux(photons/second) produced by tumor cell luciferase was cal-culated across the entire mouse. When the flux approximated108 photons/second, mice were randomly allocated into treat-ment groups, at which point treatment was initiated. Treatmentarms consisted of: (i) PP242, 100 mg/kg daily by oral gavage;(ii) everolimus, 2.5 mg/kg daily by oral gavage; (iii) carbopla-tin, 30 mg/kg every 4 days, i.p.; (iv) PP242 and carboplatin atsame doses; (v) everolimus and carboplatin at same doses; or(vi) control vehicle.

Bioluminescence imaging data taken at day 8 after treatmentand day 28 (Fig. 3A and B) showed high tumor burden andsignificant metastasis for animals in the everolimus and carbo-platin treatment groups, and a slight improvement by PP242alone. Combined treatment of carboplatin and everolimus(mTORC1 inhibition) was marginally better than carboplatinalone, whereas carboplatin and PP242 (mTORC1/2 inhibition)demonstrated superior tumor control and significantly reducedmetastasis compared with other treatments. To analyze the targetspecificity of treatments, two animals from each group werechosen at random at day 14 after treatment, tumors removed,combined, and extracted proteins analyzed by immunoblotting(Fig. 3C). Only the group treated with mTORC1/2 inhibitorPP242 demonstrated strong inhibition of P-S6K and 4E-BP1phosphorylation (the latter by electrophoretic mobility). More-over, inhibition of mTORC1/2 by PP242 but not mTORC1inhibition by everolimus blocked activating AKT phosphoryla-tion, which can promote tumor progression. Survival of animalsin each treatment armwas determined over a period of 50 days oftreatment. Superior survival was observed for the carboplatin andPP242 group, followed by PP242 alone and carboplatin alone

0 2.5 10 μmol/L 0 20 50 PP242Everolimus

β-tubulinS6KP-S6K (T389)

P-4E-BP1 (S65)4E-BP1

AKTP-AKT (S473)

mTORP-mTOR (S2448)PARPcl.-PARP

A

B

Contro

l

Carbop

latin

1 μmol/

L

PP242 2

.5 μm

ol/L

Carbop

latin

1 μmol/

L + P

P242 2

.5 μm

ol/L

0.0

0.5

1.0

1.5

Pro

tein

syn

thes

is a

ctiv

ity(%

of c

ontro

l)

Everol

imus

20 m

mol/L

Carbop

latin

1 μmol/

L + ev

erol. 2

0 mmol/

L

Figure 2.

Immunoblot analysis of mTORC1 and mTORC1/2 targets and effectors aftertreatment with everolimus or PP242. A, shown are representative immunoblotsfrom at least three independent experiments of equal amounts of whole-cell lysates of the identified proteins and phospho-proteins in control,everolimus (Everol), or PP242-treated platinum-resistant OVCAR-3 cells.Treatment was carried out at drug concentrations shown for 24 hours. Allantibodies were titrated to provide a nonsaturating response. B, effect of 20nmol/L everolimus, 2.5 mmol/L PP242, and/or 1 mmol/L carboplatin onglobal protein synthesis, measured by metabolic labeling with 35S-methionine.Cells were treated as in A, followed by 5-hour mock or 1 mmol/L carboplatintreatment, and then labeled with 35S-methionine. Equal protein amountswere analyzed by liquid scintillation counting. Average of three studies shownwith SEM.

Musa et al.





(Fig. 3D). Mice in the carboplatin plus PP242 (mTORC1/2inhibition) group showed a near doubling in median survivalcompared with vehicle alone, and three times better than ever-olimus plus carboplatin (Fig. 3E, P < 0.001). The surprisingworsening of survival mice in the everolimus plus carboplatingroup (Fig. 3C and D) may be due to increased toxicity by this

combination, and/or the inability of the rapalog to counteractivating phosphorylation of pro-oncogenic AKT. The poor sur-vival of animals in both the everolimus, and everolimus pluscarboplatin arms is consistentwith this possibility, demonstratingan increase in metastatic activity despite treatment, which is alsoevident in the bioluminescence results (Fig. 3A).We can conclude

Normalized tumor signal treatment day 28 (photons/sec)

Carbo.+Everol.

Carbo.+PP242

Everol.

PP242

Carbo.

Control 1.00

0.88

0.43

0.60

0.08

0.30

BA

C

Contro

l

Carbop

latin

PP242

Everol

imus

Carbo +

PP24

2

Carbo +

evero

limus

0

10

20

30

Med

ian

surv

ival

(day

s)

*

* *

40

Con

trol

Day 8 Day 28

Car

bopl

atin

PP

242

Eve

rolim

usE

vero

l. +

Car

boP

P24

2 +

Car

bo

Days posttreatment

DP

erce

nt s

urvi

val

0

50

100

Days after starting treatment0 20 40 60

Carbo + PP242

PP242Carbo

EverolCarbo + everolVehicle (control)

0.0 0.5 1.0

E

S6K

P-S6K

4E-BP1

Cont

rol

Ever

ol.PP

242

Carb

o.Ca

rbo.

+eve

rol.

Carb

o.+P

P242

P-AKT

AKT

Hyper-PHypo-P

Figure 3.

Inhibition of mTORC1/2 comparedwith mTORC1 alone more stronglysensitizes to carboplatin andincreased tumor control and survival.Female SCIDbg mice, age 5 to 6 weekswere implanted intraperitoneally with6 � 106 OVCAR-3 cells expressingFirefly luciferase and GFP. Tumorgrowth was assayed weekly bybioluminescent imaging. Treatmentwas conducted starting at 80 daysafter implantation on groups of 10animals per arm. A, representativeanimals are shown for control andeach treatment arm at day 8 and day28 after initiation of treatment.Animals were treated daily by oralgavage with 100 mg/kg PP242, 2.5mg/kg everolimus (Everol), or vehicle,�30 mg/kg carboplatin (Carbo) i.p.every 4 days, or vehicle control. B,shown is themean bioluminescent fluxof OVCAR cancers over the wholeanimal on the last day of treatment,normalized to untreated control.Smallest flux corresponds to strongesttreatment response. C, immunoblotanalysis of tumors harvested from 14-day treated animals. Equal amounts ofsoluble tumor–lysate proteins wereused. Each lane contains twocombined tumors. D, survival plotof animals throughout the durationof study, initiating with treatmentday 0 equal to day 80 of tumordevelopment. E, median survival (indays) of animals on the last day oftreatment in the study. � and�� , P < 0.001 by t test. SD for eachdata are shown.






that mTORC1/2 inhibition is superior to mTORC1 inhibitionwhen combined with carboplatin in a model of platinum-resis-tant ovarian cancer. We therefore examined whether in additionto inhibition of AKT activity, superior platinum resensitization,increased tumor control and superior inhibition of metastasis bymTORC1/2 blockade also involves selective suppression ofmRNA translation that cannot be achieved by everolimus.

Catalytic but not allosteric mTOR inhibition preventstranslation of DDR, cell proliferation, and cell survival mRNAs

We previously showed that high levels of mTORC1 activationand eIF4G1, a target and effector of mTORC1, promote theselectively increased translation of DDR and cell-cycle controlmRNAs (28). We therefore examined the expression of proteinsinvolved in the DDR and cell proliferation responses, includingATM, ATR, CHK1, CHK2, BRCA-1, BRCA-2, as well as AKT, rS6K,and their phosphorylated forms. We found that in addition to apreviously observed strong decrease in phosphorylated (active)AKT, and phosphorylated (inactive) 4E-BP1 by treatment withthe dual mTORC1/2 inhibitor, levels of key mediators of DNArepair responses were strongly reduced in platinum-resistantOVCAR-3 cells (Fig. 4A). These included total and phosphor-ylated forms of CHK1, ATR, and BRCA-1. There were no sig-nificant changes in the other proteins tested. Everolimus aloneor in combination with carboplatin had no such effect (Fig. 4Band C). Of note, the substantial increase in BRCA-1 protein inresponse to carboplatin-induced DNA damage was blocked by

dual mTORC1/2 inhibition with PP242, but not by mTORC1only inhibition by everolimus (Fig. 4C). Corresponding mRNAlevels were assessed by qRT-PCR for 4E-BP1, AKT, BRCA-1,BRCA-2, ATM, ATR, CHK1, CHK2, rS6K, and GAPDH (control;Supplementary Fig. S3). There were no reductions or only veryminor reduction in mRNA levels relative to control for all except4E-BP1, which was decreased by half with mTORC1/2 inhibi-tion, and BRCA-1, which was reduced by approximately 20%(Supplementary Fig. S3). Notably, reduced levels of 4E-BP1protein with mTORC1/2 inhibition is part of a describedhomeostatic mechanism for eIF4E/4E-BP1 protein interactionthat increases eIF4E availability (57). We therefore soughtevidence for reduced levels of CHK1, ATR, and BRCA-1 proteinsin tumors of animals treated with mTORC1/2 or mTORC1inhibitors. Tumor lysates corresponding to those shownin Fig. 3C were probed for CHK1, ATR, and BRCA-1 proteinlevels (Fig. 4D). Decreased levels were observed only inmTORC1/2 inhibited (PP242) and not in mTORC1-only inhib-ited (everolimus) tumors. While the downregulation of proteinlevels was not as strong as in tissue culture cells, although stillsubstantial, this is possibly due to more limited drug exposurein animals. The reduction in CHK1, ATR, and BRCA-1 proteinlevels with mTORC1/2 but not mTORC1 inhibition, in theabsence of equal reduction in mRNA levels, is particularlystriking and indicative of translation inhibition. We note thatthe CHK1 mRNA contains a very long (>800 nt) and highlystructured 50 untranslated region (50UTR, DG >�200 kcal/mol),

− + − + PP242− − + + Carbo.

β-Tubulin

AKT

P-AKT (S473)

ATM

P-ATM (S1981)

ATR

P-ATR (S428)

CHK1

P-CHK1 (S345)

CHK2

P-CHK2 (T68)

A

C− − + − + − PP242− + − − + + Carbo.

− − − + − + Everol.

BRCA1

P-BRCA1 (S1524)

PARPcl. PARP

B

- + - + Everolimus- - + + Carbo.

GAPDH

CHK1

ATR

ATM

CHK1

ATR

BRCA-1

β-Tubulin

D

Cont

rol

Ever

ol.PP

242

Carb

o.Ca

rbo.

+eve

rol.

Carb

o.+P

P242

Figure 4.

Immunoblot of AKT/mTOR pathway targets andDDR effectors after treatment with everolimus orPP242 without and with concurrent carboplatin. Aand B, shown are representative immunoblots fromat least three independent experiments of equalamounts of whole-cell lysates of the identifiedproteins and phospho-proteins. Cells were treatedwith 2.5 mmol/L PP242 or 20 nmol/L everolimus(Everol) for 24 hours, or in combinationwith 1mmol/Lcarboplatin (Carbo) for the last 5 hours, thenharvested for immunoblot analysis. Equal proteinamounts were used for immunoblots. A, PP242 and/or carboplatin-treated cells. B, everolimus and/orcarboplatin-treated cells. C, everolimus, PP242,carboplatin, or combination-treated OVCAR-3 cells.D, immunoblot analysis of tumors harvested from14-day treated animals. Equal amounts of solubletumor–lysate proteins were used. Each lane containstwo combined tumors.

Musa et al.





a hallmark of mRNAs that are typically poorly translated andhighly dependent on eIF4E (15, 23, 27).

We therefore investigated whether the reduction in key DDRproteins was of functional significance. We assessed the accu-mulation of unrepaired DNA damage foci by immunofluores-cence staining for 53BP1 and gH2AX, which decorate DNAdouble-strand breaks. We observed much higher accumulationof fluorescent foci in cells treated with carboplatin plus PP242compared with carboplatin alone at GI50 levels (Fig. 5A).Quantification of foci in single agent and combination treat-ments, scoring cells with a baseline of 3 foci per cell toeliminate background normal lesions, showed the highestnumber by far in the combination of PP242 and carboplatin(Fig. 5B). Independent evidence for increased DNA damage bycotreatment with PP242 and carboplatin was obtained byanalyzing the sub-G0–G1 fraction of cells. PI-stained cells weresubjected to flow cytometry at 24 hours after treatment tocapture the early stage of induced increased DNA fragmentation(Fig. 5C). There was a 9- to 10-fold increase in the fraction ofcells with DNA fragmentation in the sub-G0–G1 population in

the carboplatin plus mTORC1/2-inhibited group comparedwith carboplatin alone, compared with a 2-fold increase ineverolimus plus carboplatin-treated cells.

DiscussionThe mTOR pathway is a multifaceted integrator of extracel-

lular signaling that regulates cell growth, proliferation, migra-tion, angiogenesis, and the metabolism of cancer cells (22).There is considerable evidence that the hyperactivation ofmTOR signaling and the resulting aberrant upregulation ofmRNA translation, both overall and selective, are critical foroncogenic processes in ovarian and other cancers (15, 27). Ourwork suggests that selectively increased translation of specificmRNAs due to mTOR hyperactivation also plays an importantrole in platinum resistance in ovarian cancer. Thus, targetingthe mTOR/selective mRNA translation pathway may be anattractive novel strategy for therapy of platinum-resistant ovar-ian cancer.

Treatment of platinum-resistant OVCAR-3 cells with a dualmTORC1/2 inhibitor resulted in strong inhibition of cancer cellproliferation, improved tumor control, strikingly reduced metas-tasis, and significantly increased animalmedian survival.Weweresurprised by the exquisite sensitivity of ovarian cancer cells tomTORC1/2 inhibition in tumors, a reflection of the hyperacti-vated mTORC1 and mTORC2 state. In a growth factor–deprivedstate, hypophosphorylated 4E-BP1 is tightly bound to eIF4E,inhibiting from 15%- to 40% of cap-dependent mRNA transla-tion. Published work from our group and others has demonstrat-ed that hyperactivated mTORC1 drives the increased selectivetranslation of specific mRNAs involved in the oncogenic process(15, 25–28, 58, 59). These include mRNAs encoding proteinsresponsible for cell survival (survivin, Mcl1, XIAP), tumor angio-genesis (VEGF-A, FGF2), and the DNA repair response (BRCA1,53BP1, gH2AX) among others, in concert with an increase in thelevels of mTOR and/or the activity of mTOR-targeted translationfactors.

Platinum-resistant OVCAR-3 cells and tumors in animals trea-ted with carboplatin and a dual mTORC1/2 inhibitor results in anotable increase in tumor cell killing, inhibition of cell prolifer-ation, reduced metastasis, increased survival, and reducedmTORC1/2 andAKT activity (determined by surrogate phosphor-ylation), indicating a partial resensitization to platinum chemo-therapy. In addition to the expected decrease in phosphorylationof 4E-BP1, S6K, and AKT with mTORC1/2 inhibition, we unex-pectedly found decreased levels of CHK1, ATR, and BRCA-1proteins, consistent with the selective reduction observed in DDRpathway mRNA translation. It is likely that the reduction inactivating AKT phosphorylation, in concert with the higher levelof selective translational downregulation of DDR pathwaymRNAs, are in part responsible for the enhanced effects seen intumors with mTORC1/2 inhibition compared with inhibition ofmTORC1, and the reason that this was not detectable in 10-dayclonogenic cell survival studies. The extended duration of tumorgrowth and treatment, extending for several months in animalscompared with a single treatment in 10-day cell cultures, capturesthe increasedneed for higher level translationof thesemRNAs andthe impact of downregulation of AKT activity.

Multiple TCGA studies have now analyzed the exome of morethan 500 epithelial ovarian tumors (4–6, 10). Other than apredominance of p53 mutations, there are no major other

A

B

0

25

50

75

*100

% C

ells

with

>3

fpc

Cont

rol

Ever

ol. 2

0 m

mol

/L

Ever

ol. 2

0mm

ol/L

+

Carb

o. 1

μm

ol/L

Carb

o. 1

μm

ol/L

PP

242

2.5

μmol

/L

C

01020

30

*

40

% C

ells

in s

ub-G

0–G

1

Cont

rol

Carb

o. 1

μm

ol/L

5060

PP24

2 2.

5 μm

ol/L

+

Carb

o. 1

μm

ol/L

PP24

2 2.

5 μm

ol/L

+

Carb

o. 1

μm

ol/L

PP24

2 2.

5 μm

ol/L

Ever

ol. 2

0 m

mol

/L +

Carb

o. 1

μm

ol/L

Ev

erol

. 20

mm

ol/L

Carboplatin

Untreated

PP242 + carboplatin

Figure 5.

Immunofluorescence analysis of double-strandDNA (dsDNA) damage responsein platinum-resistant OVCAR-3 cellswithmTORC1/2 inhibition and/or treatmentwith carboplatin. A, cells were grown and treated on chamber slides, fixed, andimmunostained for anti-53BP1 and gH2AX, then stained with secondaryantibodies conjugated to FITC. Fluorescent foci were counted from 6 fieldschosen at random under 63� power. Representative images from threeindependent studies of untreated, carboplatin (Carbo)-treated, and PP242 pluscarboplatin-treated cells are shown. B, quantification of 6 fields for DNAdamagefoci. Cellswith >3 foci per cell (fpc)were scored. � ,P <0.001, by t test. C, sub-G0–

G1 cell population from cell-cycle distribution analysis. Cells treated as in B werelabeled with PI and subjected to flow cytometry to quantify to sub-G0–G1

population, representative of dying cells with DNA fragmentation. The results ofthree studies were averaged þ SEM. � , P < 0.001, by t test. Everol, everolimus.






mutations or mRNA expression profiles that have emerged toindicate a "typical" molecular signature of high-grade serousovarian cancer, the conclusion being that ovarian cancer is veryheterogeneous. Mutations (somatic or germline) found acrossmany signaling and biochemical pathways leads to changes inovarian cancer cell DNA repair capacity, tumor angiogenesis, andmetastatic potential. However, mutational and gene expressionanalyses do not capture the vast posttranscriptional and transla-tional alterations that are also involved in ovarian cancer growthand drug resistance. It is not a coincidence that most majoroncogenic signalingpathways that are involved inhuman cancers,including ovarian cancer, converge to hyperactivate mTOR, amajor regulator of mRNA translation, responsible for theincreased synthesis of proteins directly involved in tumor cellgrowth, angiogenesis, and survival (15, 27). Analysis of altera-tions in DNA repair pathways of ovarian tumors in the TCGAreveals that the majority did not have alterations in BRCA-1/2genes (4–6, 10), as noted previously.Moreover, if they did, BRCA-1/2 mutations actually predispose to increased sensitivity toplatinum-based therapy. Collectively, these data support theconclusion that there is an over-reliance of ovarian cancers onincreased DDR pathway activity, which may partly explain theinitial sensitivity to killing by platinum compounds. It alsosuggests the importance of incorporating an understanding ofpostmutational and posttranscriptional mechanisms such asmTOR hyperactivation, selective mRNA translation, and inhibi-tion of AKT activity in overcoming platinum resistance. This issupported by our work. Specifically, ovarian cancers often harboractivating mutations in PIK3R1, PIK3C2G, and the MAPK path-way, which may be involved in platinum resistance (and arepresent in OVCAR-3 and SKOV-3 cells; ref. 60). These pathwayshyperactivate mTOR and increase both overall and selective

mRNA translation (15) and may inform a strategic direction inthe development of new therapeutic approaches for this disease.

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

Authors' ContributionsConception and design: F. Musa, A. Alard, R.J. SchneiderDevelopment of methodology: F. Musa, A. Alard, R.J. SchneiderAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): F. Musa, A. Alard, G. David-WestAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): F. Musa, G. David-West, S.V. Blank, R.J. SchneiderWriting, review, and/or revision of the manuscript: F. Musa, J.P. Curtin,S.V. Blank, R.J. SchneiderStudy supervision: J.P. Curtin, S.V. Blank, R.J. Schneider

AcknowledgmentsThe authors thank Dr. Renier Brentjens (Memorial Sloan Kettering Cancer

Center, New York, NY) for the GFP/Firefly luciferase OVCAR-3 cells and B.Dinardo and R. Arzu for additional studies.

Grant SupportThis work was supported by grants from the U.S. Department of Health and

Human Services, NIH 1R24OD018339-01 (to R.J. Schneider); Avon Founda-tion for Women 02-2014-075 (to R.J. Schneider); U.S. Department of Healthand Human Services, NIH, and National Center for Advancing TranslationalSciences (NCATS) TR000038 (to B. Cronstein).

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

ReceivedDecember 10, 2015; revised April 12, 2016; accepted April 23, 2016;published OnlineFirst May 16, 2016.

References1. Siegel R, NaishadhamD, Jemal A. Cancer statistics, 2013. CA Cancer J Clin

2013;63:11–30.2. Pignata S, Cannella L, Leopardo D, Pisano C, Bruni GS, Facchini G.

Chemotherapy in epithelial ovarian cancer. Cancer Lett 2011;303:73–83.

3. Davis A, Tinker AV, Friedlander M. "Platinum resistant" ovarian cancer:what is it, who to treat and how to measure benefit? Gynecol Oncol2014;133:624–31.

4. TheCancerGenomeAtlasNetwork. Integrated genomic analyses of ovariancarcinoma. Nature 2011;474:609–15.

5. PatchAM,Christie EL, EtemadmoghadamD,GarsedDW,George J, FeredayS, et al. Whole-genome characterization of chemoresistant ovarian cancer.Nature 2015;521:489–94.

6. Verhaak RG, Tamayo P, Yang JY, Hubbard D, Zhang H, Creighton CJ, et al.Prognostically relevant gene signatures of high-grade serous ovarian car-cinoma. J Clin Invest 2013;123:517–25.

7. Galluzzi L, Senovilla L, Vitale I,Michels J,Martins I, KeppO, et al.Molecularmechanisms of cisplatin resistance. Oncogene 2012;31:1869–83.

8. Feldman DR, Bosl GJ, Sheinfeld J, Motzer RJ. Medical treatment ofadvanced testicular cancer. JAMA 2008;299:672–84.

9. Gadducci A, Cosio S, Muraca S, Genazzani AR. Molecular mechanisms ofapoptosis and chemosensitivity to platinum and paclitaxel in ovariancancer: biological data and clinical implications. Eur J Gynaecol Oncol2002;23:390–6.

10. Ahmed AA, EtemadmoghadamD, Temple J, Lynch AG, Riad M, Sharma R,et al. Driver mutations in TP53 are ubiquitous in high grade serouscarcinoma of the ovary. J Pathol 2010;221:49–56.

11. Ciriello G, Miller ML, Aksoy BA, Senbabaoglu Y, Schultz N, Sander C.Emerging landscape of oncogenic signatures across human cancers. NatGenet 2013;45:1127–33.

12. Ben David Y, Chetrit A, Hirsh-Yechezkel G, Friedman E, Beck BD, Beller U,et al. Effect of BRCA mutations on the length of survival in epithelialovarian tumors. J Clin Oncol 2002;20:463–6.

13. Chen KG, Sikic BI. Molecular pathways: regulation and therapeutic impli-cations of multidrug resistance. Clin Cancer Res 2012;18:1863–9.

14. Etemadmoghadam D, deFazio A, Beroukhim R, Mermel C, George J, GetzG, et al. Integrated genome-wide DNA copy number and expressionanalysis identifies distinct mechanisms of primary chemoresistance inovarian carcinomas. Clin Cancer Res 2009;15:1417–27.

15. Silvera D, Formenti SC, Schneider RJ. Translational control in cancer. NatRev Cancer 2010;10:254–66.

16. Koromilas AE, Lazaris-Karatzas A, Sonenberg N. mRNAs containingextensive secondary structure in their 50 non-coding region translateefficiently in cells overexpressing initiation factor eIF-4E. EMBO J 1992;11:4153–8.

17. Mamane Y, Petroulakis E,Martineau Y, Sato T-A, LarssonO, Rajasekhar VK,et al. Epigenetic activation of a subset ofmRNAsby eIF4E explains its effectson cell proliferation. PLoS ONE 2007;2:e242.

18. Lazaris-Karatzas A, Montine KS, Sonenberg N. Malignant transformationby a eukaryotic initiation factor subunit that binds tomRNA 50 cap. Nature1990;345:544–7.

19. Dowling RJ, Topisirovic I, Alain T, Bidinosti M, Fonseca BD, Petroulakis E,et al. mTORC1-mediated cell proliferation, but not cell growth, controlledby the 4E-BPs. Science 2010;328:1172–6.

20. Shatsky IN, Dmitriev SE, Andreev DE, Terenin IM. Transcriptome-widestudies uncover the diversity of modes of mRNA recruitment to eukaryoticribosomes. Crit Rev Biochem Mol Biol 2014;49:164–77.

21. Shayesteh L, Lu YL, Kuo WL, Baldocchi R, Godfrey T, Collins C, et al.PIK3CA is implicated as an oncogene in ovarian cancer. Nat Genet 1999;21:99–102.

Musa et al.





22. Sabatini DM. mTOR and cancer: insights into a complex relationship. NatRev Cancer 2006;6:729–34.

23. Sonenberg N, Hinnebusch AG. Regulation of translation initiation ineukaryotes: mechanisms and biological targets. Cell 2009;136:731–45.

24. Borden KL. Targeting the oncogene eIF4E in cancer: from the bench toclinical trials. Clin Invest Med 2011;34:E315.

25. Topisirovic I, Ruiz-GutierrezM, BordenKL. Phosphorylation of the eukary-otic translation initiation factor eIF4E contributes to its transformation andmRNA transport activities. Cancer Res 2004;64:8639–42.

26. Graff JR, Konicek BW, Vincent TM, Lynch RL, Monteith D, Weir SN, et al.Therapeutic suppression of translation initiation factor eIF4E expres-sion reduces tumor growth without toxicity. J Clin Invest 2007;117:2638–48.

27. Hsieh AC, Ruggero D. Targeting Eukaryotic Translation Initiation Factor 4E(eIF4E) in cancer. Clin Cancer Res 2010;16:4914–20.

28. Badura M, Braunstein S, Zavadil J, Schneider RJ. DNA damage and eIF4G1in breast cancer cells reprogram translation for survival and DNA repairmRNAs. Proc Natl Acad Sci U S A 2012;109:18767–72.

29. O'Reilly KE, Rojo F, She QB, Solit D, Mills GB, Smith D, et al. mTORinhibition induces upstream receptor tyrosine kinase signaling and acti-vates Akt. Cancer Res 2006;66:1500–8.

30. Carracedo A, Ma L, Teruya-Feldstein J, Rojo F, Salmena L, Alimonti A,et al. Inhibition of mTORC1 leads to MAPK pathway activation througha PI3K-dependent feedback loop in human cancer. J Clin Invest 2008;118:3065–74.

31. Sarbassov DD, Guertin DA, Ali SM, Sabatini DM. Phosphorylation andregulation of Akt/PKB by the rictor-mTOR complex. Science 2005;307:1098–101.

32. Korets SB, Musa F, Curtin J, Blank SV, Schneider RJ. Dual mTORC1/2inhibition in a preclinical xenograft tumor model of endometrial cancer.Gynecol Oncol 2014;132:468–73.

33. Kang SA, PacoldME,Cervantes CL, LimD, LouHJ,Ottina K, et al.mTORC1phosphorylation sites encode their sensitivity to starvation and rapamycin.Science 2013;341:1236566.

34. Gruppuso PA, Boylan JM, Sanders JA. The physiology and pathophys-iology of rapamycin resistance: implications for cancer. Cell Cycle2011;10:1050–8.

35. LaMonica S, Galetti M, Alfieri RR, Cavazzoni A, Ardizzoni A, TiseoM, et al.Everolimus restores gefitinib sensitivity in resistant non-small cell lungcancer cell lines. Biochem Pharmacol 2009;78:460–8.

36. Nyfeler B, Bergman P, Triantafellow E, Wilson CJ, Zhu Y, Radetich B, et al.Relieving autophagy and 4EBP1 from rapamycin resistance. Mol Cell Biol2011;31:2867–76.

37. Ganley IG, Lam du H, Wang J, Ding X, Chen S, Jiang X. ULK1.ATG13.FIP200 complexmediatesmTOR signaling and is essential for autophagy. JBiol Chem 2009;284:12297–305.

38. Yu Y, Yoon SO, Poulogiannis G, Yang Q, Ma XM, Villen J, et al. Phospho-proteomic analysis identifies Grb10 as an mTORC1 substrate that nega-tively regulates insulin signaling. Science 2011;332:1322–6.

39. HsuPP, Kang SA, Rameseder J, Zhang Y,OttinaKA, LimD, et al. ThemTOR-regulated phosphoproteome reveals a mechanism of mTORC1-mediatedinhibition of growth factor signaling. Science 2011;332:1317–22.

40. Deragon JM, Bousquet-Antonelli C. The role of LARP1 in translation andbeyond. Wiley Interdiscip Rev RNA 2015;6:399–417.

41. Hamilton TC, Young RC, McKoy WM, Grotzinger KR, Green JA, Chu EW,et al. Characterization of a human ovarian carcinoma cell line (NIH:OVCAR-3) with androgen and estrogen receptors. Cancer Res 1983;43:5379–89.

42. Godwin AK, Meister A, O'Dwyer PJ, Huang CS, Hamilton TC, AndersonME. High resistance to cisplatin in human ovarian cancer cell lines isassociated with marked increase of glutathione synthesis. Proc Natl AcadSci U S A 1992;89:3070–4.

43. Chekmasova AA, Rao TD, Nikhamin Y, Park KJ, Levine DA, Spriggs DR,et al. Successful eradication of established peritoneal ovarian tumors inSCID-Beige mice following adoptive transfer of T cells genetically targetedto the MUC16 antigen. Clin Cancer Res 2010;16:3594–606.

44. Morimoto H, Safrit JT, Bonavida B. Synergistic effect of tumor necrosisfactor-alpha- and diphtheria toxin-mediated cytotoxicity in sensitiveand resistant human ovarian tumor cell lines. J Immunol 1991;147:2609–16.

45. Morimoto H, Yonehara S, Bonavida B. Overcoming tumor necrosis factorand drug resistance of human tumor cell lines by combination treatmentwith anti-Fas antibody and drugs or toxins. Cancer Res 1993;53:2591–6.

46. Feldman ME, Apsel B, Uotila A, Loewith R, Knight ZA, Ruggero D, et al.Active-site inhibitors of mTOR target rapamycin-resistant outputs ofmTORC1 and mTORC2. PLoS Biol 2009;7:e38.

47. Cai Z, Chattopadhyay N, Liu WJ, Chan C, Pignol JP, Reilly RM. Optimizeddigital counting colonies of clonogenic assays using ImageJ software andcustomized macros: comparison with manual counting. Int J Rad Biol2011;87:1135–46.

48. Tanaka C, O'Reilly T, Kovarik JM, Shand N, Hazell K, Judson I, et al.Identifying optimal biologic doses of everolimus (RAD001) inpatients with cancer based on the modeling of preclinical and clinicalpharmacokinetic and pharmacodynamic data. J Clin Oncol 2008;26:1596–602.

49. Penzvalto Z, Lanczky A, Lenart J, Meggyeshazi N, Krenacs T, Szoboszlai N,et al. MEK1 is associated with carboplatin resistance and is a prognosticbiomarker in epithelial ovarian cancer. BMC Cancer 2014;14:837.

50. Van DN, Roberts CF, Marion JD, Lepine S, Harikumar KB, Schreiter J, et al.Innate immune agonist, dsRNA, induces apoptosis in ovarian cancer cellsand enhances the potency of cytotoxic chemotherapeutics. FASEB J2012;26:3188–98.

51. Smith JA, Ngo H, Martin MC, Wolf JK. An evaluation of cytotoxicity of thetaxane and platinum agents combination treatment in a panel of humanovarian carcinoma cell lines. Gynecol Oncol 2005;98:141–5.

52. Thomadaki H, Scorilas A. Molecular profile of breast versus ovariancancer cells in response to treatment with the anticancer drugs cisplatin,carboplatin, doxorubicin, etoposide and taxol. Biol Chem 2008;389:1427–34.

53. Hassan B, Akcakanat A, Sangai T, Evans KW, Adkins F, Eterovic AK, et al.Catalytic mTOR inhibitors can overcome intrinsic and acquired resistanceto allosteric mTOR inhibitors. Oncotarget 2014;5:8544–57.

54. Marz AM, Fabian AK, Kozany C, Bracher A, Hausch F. Large FK506-bindingproteins shape the pharmacology of rapamycin. Mol Cell Biol 2013;33:1357–67.

55. PeiH, Li L, Fridley BL, Jenkins GD, Kalari KR, LingleW, et al. FKBP51 affectscancer cell response to chemotherapy by negatively regulating Akt. CancerCell 2009;16:259–66.

56. Kim SH, Zukowski K, Novak RF. Rapamycin effects on mTOR signaling inbenign, premalignant and malignant human breast epithelial cells. Anti-cancer Res 2009;29:1143–50.

57. Yanagiya A, Suyama E, Adachi H, Svitkin YV, Aza-Blanc P, Imataka H, et al.Translational homeostasis via the mRNA cap-binding protein, eIF4E. MolCell 2012;46:847–58.

58. Braunstein S, Karpisheva K, Pola C, Goldberg J, Hochman T, Yee H, et al. Ahypoxia-controlled cap-dependent to cap-independent translation switchin breast cancer. Mol Cell 2007;28:501–12.

59. Decarlo L, Mestel C, Barcellos-Hoff MH, Schneider RJ. Eukaryotic transla-tion initiation factor 4E is a feed-forward translational coactivator oftransforming growth factor beta early protransforming events in breastepithelial cells. Mol Cell Biol 2015;35:2597–609.

60. Barretina J, Caponigro G, Stransky N, Venkatesan K, Margolin AA, Kim S,et al. The Cancer Cell Line Encyclopedia enables predictive modelling ofanticancer drug sensitivity. Nature 2012;483:603–7.






Published OnlineFirst May 16, 2016.Mol Cancer Ther Fernanda Musa, Amandine Alard, Gizelka David-West, et al. Cancer

OvarianResensitization and Treatment of Platinum-Resistant Dual mTORC1/2 Inhibition as a Novel Strategy for the

Updated version

10.1158/1535-7163.MCT-15-0926doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2016/05/14/1535-7163.MCT-15-0926.DC1

Access the most recent supplemental material at:

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

Subscriptions

Reprints and

[email protected] at

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

Permissions

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

.http://mct.aacrjournals.org/content/early/2016/06/23/1535-7163.MCT-15-0926To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-15-0926

http://mct.aacrjournals.org/content/suppl/2016/05/14/1535-7163.MCT-15-0926.DC1

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

mailto:[email protected]

http://mct.aacrjournals.org/content/early/2016/06/23/1535-7163.MCT-15-0926


Dual mTORC1/2 Inhibition as a Novel Strategy for the ... · al. These include mTOR regulation by...

Documents

Transcript of Dual mTORC1/2 Inhibition as a Novel Strategy for the ... · al. These include mTOR regulation by...