stat3-mediated autophagy dependence identiﬁ subtypes of ...therapeutics, targets, and chemical...

Therapeutics, Targets, and Chemical Biology

STAT3-Mediated Autophagy Dependence IdentifiesSubtypes of Breast Cancer Where Autophagy Inhibition CanBe Efficacious

Paola Maycotte1, Christy M. Gearheart2, Rebecca Barnard3, Suraj Aryal1, Jean M. Mulcahy Levy2,Susan P. Fosmire2, Ryan J. Hansen3, Michael J. Morgan1, Christopher C. Porter2, Daniel L. Gustafson3, andAndrew Thorburn1

AbstractAutophagy is a protein andorganelle degradation pathway that is involved in diverse diseases, including cancer.

Recent evidence suggests that autophagy is a cell survival mechanism in tumor cells and that its inhibition,especially in combination with other therapy, could be beneficial but it remains unclear if all cancer cells behavethe same way when autophagy is inhibited. We inhibited autophagy in a panel of breast cancer cell lines andfound that some of them are dependent on autophagy for survival even in nutrient rich conditions without anyadditional stress, whereas others need autophagy only when stressed. Survival under unstressed conditions is dueto cell type–specific autophagy regulation of STAT3 activity and this phenotype is enriched in triple-negativecell lines. This autophagy-dependency affects response to therapy because autophagy inhibition reduced tumorgrowth in vivo in autophagy-dependent but not in autophagy-independent breast tumors, whereas combinationtreatment with autophagy inhibitors and other agent was preferentially synergistic in autophagy-dependent cells.These results imply that autophagy-dependence represents a tumor cell–specific characteristic where autophagyinhibition will be more effective. Moreover, our results suggest that autophagy inhibition might be a potentialtherapeutic strategy for triple-negative breast cancers, which currently lack an effective targeted treatment.Cancer Res; 74(9); 2579–90. �2014 AACR.

IntroductionAutophagy is a long-lived protein and organelle degradation

pathway in which cytoplasmic material is engulfed in a doublemembrane structure known as the autophagosome and laterdelivered to the lysosome for degradation. Macroautophagy(referred herein as autophagy) is thought to be the major typeof autophagy (1). It has been extensively studied in recent years,increasing our understanding of the major constituents of theautophagic pathway, encoded by the ATG genes, and unravel-ing many of its roles in homeostasis and development. Mostimportantly, defects in the autophagic pathway have beeninvolved in diverse diseases, including cancer (1–4).Under normal conditions, basal autophagy has been pro-

posed to function as a tumor suppressor mechanism byreducing oxidative stress, inflammation, and genome instabil-

ity (2). However, autophagy has also been suggested to act as asurvival mechanism in established tumors. It is well knownthat in cells under stress—including starvation, growth factordeprivation, hypoxia, radio- and chemotherapy—autophagy isupregulated to recycle cytoplasmic components and providethe cell with amino acids and ATP to support essentialmetabolic pathways and protein synthesis (2). This is criticalin tumor cells, which are constantly exposed to bothmetabolicstress via hypoxia, inadequate glucose supply and increasedenergetic demands of rapidly proliferating cells as well asproteotoxic stress induced by high levels of genomic instabilityfound in cancers. Although this autophagy requirement couldbe a generally important mechanism of survival in tumor cells,recent evidence suggests that certain tumors are "autophagyaddicted." In this regard, RAS transformation is known toinduce high levels of basal autophagy in cancer cell lines;autophagy is required for efficient RAS-induced tumorigenesis,and many, but not all, RAS-transformed cell lines are highlydependent on autophagy (5–7).

Despite implementation of prevention programs and noveltherapeutic strategies, breast cancer remains the second lead-ing cause of cancer-related death among women in the UnitedStates (8). One of the biggest recognized barriers to progress inprevention, diagnosis, and treatment is the clinical and geneticheterogeneity of the disease (9). In this regard, gene expressionanalyses have led to the definition of five molecular "intrinsic"subtypes of breast cancer (luminal A, luminal B, HER2-

Authors' Affiliations: Departments of 1Pharmacology, and 2Pediatrics,University of Colorado School of Medicine, Aurora; and 3Department ofClinical Sciences, Colorado State University, Fort Collins, Colorado

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

Corresponding Author: Andrew Thorburn, Department of Pharmacology,University of Colorado School of Medicine, Mail Stop 8303, 12801 East17th Avenue, Aurora, CO 80045. Phone: 303-724-3290; Fax. 303-724-3664; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-13-3470

�2014 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 2579

on September 5, 2021. © 2014 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst March 3, 2014; DOI: 10.1158/0008-5472.CAN-13-3470

http://cancerres.aacrjournals.org/

enriched, basal-like, and claudin-low), which have differencesin incidence, survival, and response to treatment (10). Basal-like and claudin-low tumors comprise the majority of triple-negative breast cancers (TNBC; ref. 10), a subgroup of tumorsthat do not express clinically significant levels of estrogenreceptor, progesterone receptor, and HER2 and thus cannot betreated with endocrine or anti–HER2-targeted therapies. Theyinclude 10% to 24% of invasive breast cancers and have a worseprognosis when compared with other tumor subtypes. Impor-tantly, although patients with TNBC do benefit from chemo-therapy, there are no known targeted agents for this type ofcancer and after therapy, they tend to relapse with distantmetastases, resulting in a worse overall survival (11) andunderscoring the need to develop better, less toxic treatmentapproaches.

Among many other distinctive characteristics, TNBC cellsare known to have high levels of activation of the JAK–STATpathway (12). STATs are transcription factors that are acti-vated in the cytoplasm by tyrosine phosphorylation inresponse to cytokine receptor activation (IFN or interleukin-6) and their associated Janus-activated kinase (JAK) or togrowth factor receptor signaling (EGF and platelet–derivedgrowth factor), either directly or through recruitment of asso-ciated proteins (13). Cytoplasmic kinases such as SRC andABL1 can also phosphorylate and activate STATs (14). Phos-phorylated STATs dimerize and translocate to the nucleus-activating transcription. Ligand-dependent activation ofSTATs is associated with cellular differentiation and growthregulation. Conversely, their constitutive activation is associ-ated with tumorigenesis by inducing the transcription of genesthat promote cell-cycle progression, prevention of apoptosis(BCL2L1, CCND1, and MYC), cancer inflammation, and inhi-bition of antitumor immunity (14–16).

We found that autophagy inhibition is much more impor-tant in some breast lines than others and that cancer cell lineswith constitutively high levels of STAT3 activity were moresensitive to autophagy inhibition than those with low levels ofbasal STAT3 activity. The most autophagy-dependent linesrequire autophagy to survive even under conditions with noadded metabolic stress and are enriched in the basal-like andclaudin-low subtypes of breast cancer (10). These effects weredue to autophagy-dependent STAT3 activation and were alsoassociated with increased ability to synergize when autophagywas inhibited in combination with another anticancer drug.Our results suggest that autophagy inhibition might provebeneficial especially for "basal-like" and "claudin-low" breastcancers, which mostly constitute the TNBC subgroup and thatcurrently lack effective treatment strategies.

Materials and MethodsCell culture

All cell lines were acquired from the University of ColoradoTissue Culture Core in 2012 and were authenticated by alleletesting (date of authentication is as follows: MCF10A, MCF7,T47D, HCC1937, MDA-MB-468, and MDA-MB-231: October2011; MCF12A and BT549: May 2012). After acquisition, cellswere grown, frozen, and each time theywere thawed, theywere

not used for longer than 6 months. Cell culture media used areshown in Supplementary Materials and Methods. STAT3Cplasmid and its empty vector (Addgene plasmids 24983 and14883) were a gift from Linzhao Cheng (17) and David Balti-more (18), respectively. MDA-MB-231 and MDA-MB-468 cellswere transduced with lentiviruses expressing these plasmidsand sorted for a high green fluorescent population.

The autophagy short hairpin RNA libraryWe designed a lentiviral short hairpin RNA (shRNA) library

with 629 shRNAs targeting autophagy proteins from theautophagic core machinery, proteins described in the autop-hagy interactome (19) and apoptosis-related proteins. A pool ofshRNA-expressing plasmids fromThe RNAi Consortium (TRC)library (pLKO.1 vector) was prepared at the Functional Geno-mics Core of the University of Colorado Cancer Center (Boul-der, CO). Of note, three to five shRNAs per gene target wereused. Lentiviruses were generated according to protocols fromTRC (http://www.broadinstitute.org/rnai/trc/lib) and cellswere transduced at a multiplicity of infection (MOI) of 0.3and selected with puromycin. Posttransduction samples werecollected 18 hours after the addition of virus in quintuplicate.Another sample of transduced cells was replated with puro-mycin and allowed to grow for 15 days, after which the growthsample was collected in quintuplicate. Samples were barcodedas described previously (20) and analyzed in an IlluminaGenome Analyzer IIX.

Functional genetic screening data were analyzed using ourin-house bioinformatics pipeline (20–24), and as described inSupplementary Materials and Methods.

shRNA lentiviral transductionLentiviruses were prepared according to protocols published

at TRC webpage (http://www.broadinstitute.org/rnai/trc/lib).Viruses (containing either nonsilencing, ATG5TRCN0000151474,ATG7 TRCN0000007587, BECN1 TRCN0000033549, STAT3TRCN0000020840, or STAT3 TRCN0000020842 pLKO.1 vectors)were added to achieve anMOI of 5 to10. Cells were selected withpuromycin according to a dose–response curve and with theconcentrations shownonSupplementaryMaterials andMethodsfor 2 to 3 days, then tripsinized and used for experiments.

Viability assaysFor long-term clonogenic assays, cells were plated in 12-well

plates and allowed to grow for 8 to 10 days. They were fixed andstained with crystal violet (Becton, Dickinson and Company).Stain was solubilized and absorbance was measured at 540 nm.For proliferation curves, cells were counted at 3, 6, 8, or 10 daysafter plating. Trypan blue–negative cells were considered viable.

For MTS assays, cells were plated in 96-well plates andallowed to grow for 2 to 6 days. Cells were treated with MTSreagent (Promega) according to the manufacturer's instruc-tions. For cell death assessment, media were collected andevaluated for lactate dehydrogenase (LDH) activity with theLDH Cytoscan Cytotoxicity Kit (G-Biosciences) according tothe manufacturer's instructions. For time-lapse movies, cellswere plated at equal densities and imaged for 48 hours. Forpropidium iodide (PI) staining, cells transduced with the

Maycotte et al.

Cancer Res; 74(9) May 1, 2014 Cancer Research2580




different shRNAs were plated, allowed to grow for 48 hours,stained with PI, and imaged.

Protein isolation and Western blotsOf note, 2 to 3 days after plating, cells were lysed with RIPA

buffer containing protease inhibitor (Roche) and Halt phos-phatase inhibitor cocktail (Thermo Scientific). Protein wasquantitated using Bradford reagent. Antibodies used are listedin Supplementary Materials and Methods. Western blot anal-ysis figures show a portion of the membranes containing thecorresponding bands cropped for clarity.

Animal studiesAll animal studies were performed in accordance with the

Colorado State University Animal Care and Use Committee.Female nude nu/nu mice were purchased from the NationalCancer Institute (Frederick, MD) and challenged with 5 � 106

MDA-MB-231 cells. Female nude nu/nu mice were first ovari-ectomized and received a subcutaneous, 60 day release, 0.25-mg estradiol implant (Innovative Research). Mice were thenchallenged, 7 days later, with 5� 106MCF7 cells. Cells (100 mL)in 50% serum-free media and 50% Matrigel (BD Biosciences)were injected into the fourthmammary fat pad. Upon reachinga tumor volume of 100 mm3, mice received either 60 mg/kgchloroquine diphosphate salt or 0.9% saline given by intraper-itoneal injection, once daily for the duration of the study. Thestudy was followed until tumors reached four times initialvolume (TV�4; ref. 25).

Statistical analysisStatistical differences were determined using a two-sample

equal variance Student t test. Log-rank analysis for theKaplan–Meier curves was done in Prism v5.0a. Three MCF7 xenograftsdid not quite reach four times initial volume (TV�4), so time toreach TV�4 was extrapolated using calculated tumor doublingtime. The combination index (CI) was calculated using Com-puSyn (ComboSyn, Inc.).A more detailed description of the Materials and Methods

section was included in Supplementary Materials andMethods.

ResultsThe autophagy-focused shRNA lentiviral libraryWe designed an autophagy-specific shRNA library, targeting

the core components of the autophagic pathway (ATG genes)as well as some of their interacting proteins (19) and used thislibrary to evaluate "autophagy dependence" in a panel of breastcancer cell lines. Because functions independent from autop-hagy have been described for ATG genes (26–28), we reasonedthat if autophagy is important for tumor cell survival andproliferation to different extents in different cell lines, thiswould be manifested by differential selection for or againstshRNAs that target the autophagy pathway during cell growth,leading to overall differences in the pattern of shRNA repre-sentation of the library in different breast cell populations, andnot to changes in the shRNAof a particular ATGprotein. On theother hand, if autophagy is not important for the growth or

survival of cells, shRNA representation after growth selectionshould be similar to the starting population. Because theapproach examines large numbers of shRNAs whose onlysimilarity is their connection to autophagy, differences in theoverall representation of the whole library should representthe importance of autophagy itself rather than any otherfunction for individual autophagy regulators (Fig. 1A). Weused a nontumorigenic cell line (MCF12A), two luminal (MCF7and T47D), and two triple-negative cell lines, one basal (MDA-MB-468) and one claudin-low (MDA-MB-231), according to theclassification by Prat and colleagues (10). OneDNA samplewascollected 18 hours after transduction (posttransduction) andanother sample collected 15 days following cell growth incomplete medium, both in quintuplicate. Samples were bar-coded and analyzed by next-generation sequencing asdescribed previously (20). Hierarchical clustering of the shRNAreadouts produced twomain clusters (Fig. 1B). The first clustercontained all the posttransduction samples combinedwith theMCF7 and MCF12A growth samples, indicating a similarshRNA representation among these conditions. The secondcluster included MDA-MB-231, T47D, and MDA-MB-468growth samples, suggesting differences in shRNA representa-tion between this group of samples and the other cluster, but asimilar representation among them. Those shRNAs "lost" in thegrowth sample but present in the posttransduction representproteins essential for growth, because their knockdown causesdecreased proliferation or cell death and a lack of represen-tation of the shRNA in the growth sample (as in Fig. 1A,panel 1). One condition cluster in the heatmap showed shRNAsthat had less reads in growth samples of MDA-MB-231, T47D,and MDA-MB-468 cell lines when compared with the post-transduction cluster (Supplementary Fig. S1). This clusterincluded mostly shRNAs that target autophagy at the nucle-ation and expansion steps, indicating that MDA-MB-231,T47D, and MDA-MB-468 had a decreased proliferation ordeath induction when transduced with these shRNAs, suggest-ing these cells were sensitive to autophagy inhibition. On theother hand, MCF7 and MCF12A cell lines clustered with theposttransduction samples with no significant selection for oragainst the shRNAs during the 15-day growth period, indicat-ing they proliferate normally despite having shRNAs that targetthe autophagic pathway. These results suggest that breastcancer cell lines behave differently when autophagy is mod-ulated and that some breast cancer cell lines are dependent onautophagy for proliferation or survival, whereas others aremuch less dependent on it. A complete list of the shRNAs in thelibrary and their representation per cell line is included inSupplementary Table S1.

Breast cancer cell lines have a differential sensitivity toautophagy inhibition

We validated our shRNA library results with individualshRNAs targeting the core autophagy proteins ATG5, ATG7,and BECN1 that were each independently validated as inhibit-ing autophagy. Figure 2A shows a panel of breast cell linesorganized in increasing sensitivity to autophagy inhibition incomplete medium when measured by a clonogenic survivalassay. These cell lines were differentially affected in their

Autophagy and STAT3 Control Breast Cancer Survival

www.aacrjournals.org Cancer Res; 74(9) May 1, 2014 2581




growth/survival when autophagy was inhibited. MCF10A cellswere the least affected by autophagy inhibition, whereas MDA-MB-468 were the most affected. To test how inhibition of

autophagy was affecting cell number, we performed a growthcurve (Fig. 2B). MCF10A cells were resistant to autophagyinhibition; in fact, they proliferated faster. Although therewas a

Figure 1. Theautophagy-focused shRNA library. A, experimental outline. Cellswere transducedwith a shRNA library containing autophagy-focused shRNAs. Aftertransduction, posttransduction (PT) samples were collected. Growth samples (G) were collected after 15 days in culture, both in quintuplicate. Hypotheticalheatmaps for the three possible outcomes are shown in the figure, showing different representations of shRNA tag counts. If most shRNAs targeted a processimportant for cell survival (yellow cell) or proliferation (green cell), they would be underrepresented in the growth sample when compared with posttransductionand the overall representation of shRNAs would look as in panel 1. The dendogram would produce two different clusters with quintuplicates grouping together. IfshRNAs in this cluster represent shRNAs that target the autophagic pathway, then autophagy would be important for proliferation/survival in these cells. Ifthe shRNAshad noeffect on proliferation/survival (black cells), theheatmapwould look as in panel 2 andsampleswould cluster without a clear distinction betweenposttransduction and growth. If this cluster was enriched in shRNAs targeting the autophagic pathway, it would mean that autophagy has no effect on growth/proliferation. Finally, if the shRNAs gave a proliferative advantage (red cells), they would be overrepresented in the growth sample when compared with theposttransduction and representation would look as in panel 3. The dendogram would produce two well-defined groups representing posttransduction andgrowthsamples. If thesewereshRNAs inhibitingautophagy, itwouldmeanthat autophagydecreasesproliferationor inducescell death in thesecells.B,hierarchicalclustering of breast cancer cell lines transduced with an autophagy-focused shRNA library produced two main clusters. Posttransduction samples from allcell lines are labeled in black. First cluster included all the posttransduction samples as well as MCF7 and MCF12A growth samples. Second cluster includedMDA-MB-231, T47D, and MDA-MB-468 growth samples, showing a differential response to autophagy manipulation among breast cancer cell lines.

Maycotte et al.





decrease in cell number inMCF7,MCF12A, andT47D cells withautophagy inhibition, the fact that cell number continued toincrease over time in the knockdown cells suggested that theirproliferation rate was only decreased. The most affected celllines were MDA-MB-231, HCC1937, BT549, and MDA-MB-468(Fig. 2A and B), in which autophagy inhibition greatlydecreased proliferation, or completely inhibited it. Important-ly, the three shRNAs showed a similar response in each cell line.

These results, together with the autophagy library screen,suggest that inhibition of the autophagic pathway and not anunspecific effect of the inhibition of a certain ATG genedifferentially affects breast cancer cell proliferation or survival.

Analysis of clonogenic experiments and growth curvesfrom all the cell lines indicated that TNBC cell lines (basaland claudin-low) were the most sensitive to autophagyinhibition (Fig. 2A and B). Conversely, in nontumorigenic

Figure 2. Breast cancer cell lines have a differential sensitivity to autophagy inhibition. A panel of breast cancer cell lines, including two nontumorigenic(MCF10A and MCF12A), two luminal (MCF7 and T47D), and four TNBC cell lines (two basal, HCC1937 and MDA-MB-468; and two claudin-low, MDA-MB-231 and BT549) were transduced with shRNAs against ATG5, ATG7, or BECN1 and tested for clonogenic proliferation 8 days after plating. A,cell lines were ordered according to their increasing sensitivity to autophagy inhibition as measured by crystal violet quantitation. Sample clonogenicimages are shown. Alternatively, in B, live (Trypan blue negative) cells were counted at different time points after plating and graphed in a proliferation curve.Graph in A, mean � SE of three independent experiments performed in triplicate. �, different from its respective nonsilencing control; P < 0.01. B,graphs show one representative experiment performed in triplicate.






and luminal cells, proliferation was induced (MCF10A) orwas decreased to a much lesser extent (MCF12A, MCF7, andT47D) than TNBC cell lines, suggesting a delay in cellproliferation rather than cell death. The shRNAs were effec-tive at knocking down their target proteins in all cell lines,showing markedly decreased starvation-induced autophagyas measured by LC3-II accumulation when autophagic fluxwas blocked (Supplementary Fig. S2).

Decreased clonogenic growth could be caused bydecreased proliferation or cell death. To test these possibil-ities, we chose a luminal line with only a modest effect ofautophagy inhibition (MCF7) and two TNBC cell lines(MDA-MB-231 and MDA-MB-468) in which autophagy inhi-bition had a significant effect and imaged them by time-lapse microscopy after transduction with nonsilencing,ATG7 or BECN1 shRNAs (Fig. 3A and Supplementary Movies1–6). The same number of cells were plated for all conditionsand Fig. 3A shows the cells after 48 hours. ATG7 and BECN1knockdown resulted in a decreased number of cells whencompared with the nonsilencing controls in the three celllines. MCF7 cells after ATG7 or BECN1 knockdown lookednormal and only proliferated at a somewhat slower ratecompared with the nonsilencing cells, resulting in smallercolonies. Conversely, MDA-MB-231 cells showed significant-ly decreased proliferation when compared with the nonsi-lencing controls and some cells died. The most affected cellswere MDA-MB-468s, in which ATG7 and BECN1 knockdowncells did not proliferate and most of these cells died duringthe 48 hours period. In agreement with these observations,MCF7 cells did not show an increase in LDH release in a 6-day period when autophagy was inhibited (Fig. 3B). MDA-MB-231 cells showed increased LDH release after 6 days andMDA-MB-468 cells showed a maximum LDH release sinceday 3. All three cell lines had a similar knockdown whenmeasured by Western blot analysis (Fig. 3C). Similar resultswere found in a PI staining experiment to measure celldeath, in which the highest increase in PI staining was seenon the MDA-MB-468 cells in which autophagy had beeninhibited (Supplementary Fig. S3).

Because autophagy inhibition did not induce cell death inMCF7 cells, we tested the functionality of the autophagicpathway in these cells by starving them with Earle's BalancedSalt Solution, a standard treatment for autophagy induction.Pharmacologic inhibition of autophagywith chloroquine treat-ment (Fig. 3D) or inhibition of autophagy with ATG7 or BECN1shRNAs (Supplementary Fig. S3B) greatly decreased cell via-bility in a clonogenic assay, suggesting that MCF7 cells needautophagy for survival but only under stressed conditions.

Chloroquine treatment had a similar effect as ATG7 orBECN1 knockdown when cells were grown in full medium(Fig. 4). A differential sensitivity to chloroquine was foundamong all the breast cancer cell lines, with the TNBC cellsagain being the most sensitive. An increase in LC3II wasobserved in all the cell lines with chloroquine treatment (Fig.4). However, this increase in LC3II did not correlate withsensitivity to autophagy inhibition, suggesting that differ-ences in basal autophagy among the cell lines are not thecause of autophagy dependence. Together, these data show

that breast cancer cell lines differ greatly in their depen-dency on autophagy, with some lines barely affected whenautophagy is inhibited, some showing only growth inhibi-tion, and some dying. The most autophagy-dependent cellstended to be the triple negatives, which require autophagyfor their survival even under conditions of no added stress.

Autophagy regulates STAT3 phosphorylation and cellsurvival

Autophagy has recently been linked to JAK/STAT pathwayactivation (29–31) and some breast cancer cell lines areknown to have high levels of constitutively activated STAT3,particularly those belonging to the TNBC subgroup (12, 32).

Figure 3. Autophagy inhibition induces cell death in somecell lines but notothers. MCF7, MDA-MB-231, and MDA-MB-468 cells were transducedwith lentiviruses expressing nonsilencing, ATG7, or BECN1 shRNAs,plated and evaluated for cell death in a time-lapse movie (A andSupplementary Movies). Frames show a representative image taken 48hours after plating; bar, 200 mm. B, cell death was also evaluated by LDHrelease. C, cell lysates were evaluated for protein knockdown (NS,nonsilencing; A7, ATG7; B1, BECN1 shRNAs). D, MCF7 cells werestarved in Earle's Balanced Salt Solution for 24 hours � 10 mmol/Lchloroquine (CQ). Pictures show a representative image taken 5 daysafter the removal of the starvation medium. Graphs in B showmean�SEof three independent experiments performed in triplicate. �, different fromits respective nonsilencing control (P < 0.05).

Maycotte et al.





Importantly, basal and claudin-low subtypes of breast can-cer comprise most (70%–80%) of TNBC (10) and the basaland claudin-low cell lines used in this study are known to betriple negative (33).In agreement with previous reports (12), TNBC cell lines

(MDA-MB-231, HCC1937, BT549, and MDA-MB-468) showedsubstantially higher levels of activated STAT3 by Tyr phos-phorylation (Fig. 5A) than the luminal cell lines. These cellswere determined to be STAT3 dependent, because geneticinhibition of STAT3 by gene knockdown showed a signi-ficant increase in LDH release as well as decreased MTSactivity (Fig. 5B) and clonogenic growth (Fig. 5C) in bothMDA-MB-231 and MDA-MB-468. Although MCF7 cellsshowed a decrease in MTS activity and clonogenic growth,

STAT3 knockdown did not induce a substantial amount ofLDH release, suggesting that proliferation was impaired butthat MCF7 cells are not STAT3 dependent. Moreover, phar-macologic inhibition of STAT3 with Stattic, a STAT3 small-molecule inhibitor (34), killed breast cancer cell lines in asimilar manner and with the same relative specificity toautophagy inhibition by chloroquine or ATG gene knock-down (Fig. 5D and Supplementary Fig. S4A).

Tyrosine-phosphorylated STAT3 could not be detected inMCF7 cells in basal conditions (Fig. 5A) or upon autophagyinhibition (Fig. 5E). On the other hand, ATG gene knockdowndecreased tyrosine-phosphorylated STAT3 in both the MDA-MB-231 and MDA-MB-468 cells and led to a reduction in thelevels of transcriptional targets of STAT3-like cyclin D andBCL2L1 (also known as BCL-XL, Fig. 5E). Similar results wereobserved in the other triple-negative cell lines (HCC1937 andBT549; Supplementary Fig. S4B). Pharmacologic inhibitionof autophagy by chloroquine treatment (Fig. 5F) alsodecreased phosphorylation of STAT3 in MDA-MB-231 andMDA-MB-468 cell lines. Together, these data suggest thatautophagy inhibition decreases STAT3 phosphorylation inTNBC cell lines and that this reduction in STAT3 activityinduces cell death. To test this, we overexpressed a consti-tutively active form of STAT3 (STAT3C), which partiallyrestored phospho-Tyr STAT3 levels (Fig. 6A) and decreaseddeath induced by autophagy inhibition in both MDA-MB-231and MDA-MB-468 cell lines (Fig. 6B).

Chloroquine increases the median survival of MDA-MB-231 mouse xenografts

To evaluate the effect of autophagy inhibition in vivo, weperformed mouse xenograft studies with the MCF7 and MDA-MB-231 cell lines. Tumor growth was followed until tumorsreached four times initial volume (TV�4) as suggested byTeicher (25). Chloroquine treatment significantly delayed themedian time to reach TV�4 ofmice injected withMDA-MB-231cells and not in those having MCF7 xenografts (Fig. 7A). LC3staining revealed more LC3-positive puncta in the chloro-quine-treated tumors when compared with their respectivevehicle-treated controls, indicating inhibition of autophagy inboth MCF7 and MDA-MB-231 chloroquine-treated tumors.These results suggest that autophagy inhibition has differentialeffects on breast cancer tumors in vivo and that TNBC tumorssimilar to the MDA-MB-231 xenografts are more likely tobenefit from autophagy inhibition even in the absence oftreatment with other agents.

Autophagy inhibition with chloroquine synergizes withchemotherapy primarily in autophagy-dependent cells

Because MDA-MB-231 and MDA-MB-468 cells were themost affected by autophagy inhibition, we hypothesized thatchloroquine treatment would synergize with chemotherapy inthese cells. Figure 7B shows the CI for treatment with doxo-rubicin and chloroquine at different concentrations.MDA-MB-231 and MDA-MB-468 cells showed the highest synergy (log CI< 0), whereas the combined treatment in MCF7 cells was onlyweakly synergistic in most concentrations tested and evenantagonistic in others (log CI > 0).

Figure 4. Breast cancer cell lines have a differential sensitivity topharmacologic inhibition of autophagy. A, cells were treated withdifferent concentrations of chloroquine (CQ; mmol/L) and evaluated fordeath/survival byLDH releaseorMTS reagent after 48hours of treatment.B, LC3II accumulation was evaluated after a 4-hour chloroquinetreatment in all cell lines. MCF7 cells were included in both gels forcomparison. Graphs, mean � SE of three independent experimentsperformed in triplicate. �, different from its respective untreatedcontrol (P < 0.05).






DiscussionAutophagy is frequently increased in tumors, particularly in

areas of low oxygen and blood supply, where it serves as asurvival mechanism that helps the cells survive periods ofmetabolic stress (35). Although autophagy could, in principle,serve as a general survival mechanism in all tumor cells, recentstudies suggest that certain types of cancers are particularlysensitive to autophagy inhibition. In this regard, pancreaticand RAS-transformed cancers seem to have high energeticrequirements that are supported by autophagic mitigation ofreactive oxygen species (6), and autophagic maintenance ofmitochondrial (5) and glucose metabolism (7). However, evenwith RAS-driven cancers, the role of autophagy is context

dependent because a recent study suggests that autophagyinhibition prevents Kras-driven tumor development when p53is wild-type but promotes tumorigenesis and progressionwhen p53 is deleted (36).

In this work, we studied autophagy dependence in a panelof breast cancer cell lines representing different intrinsicsubtypes of the disease. Importantly, breast cancer has notpreviously been considered particularly addicted to autop-hagy. Only two previous studies have suggested that autop-hagy could be a therapeutic target in breast cancer: oneusing the MDA-MB-231 cell line, which is KRAS transformed(7), and another one using the MMTV (mouse mammarytumor virus)–PyMT mouse model of breast cancer, which

Figure 5. TNBC cell lines haveconstitutively activated the JAK/STAT pathway and autophagyregulates their STAT3phosphorylation status. A, a panelof breast cancer cell lines wastested for STAT3 Tyr-phosphorylation by Western blotanalysis. B and C, MCF7, MDA-MB-231, and MDA-MB-468cells were tested for STAT3dependence by transduction withSTAT3 (S3) shRNAs (#40 and #42)and evaluated for death/survivalwith an LDH release or an MTSassay (B), or with a clonogenicgrowth assay (C). B, cellswere collected 48 hoursposttransduction and evaluated forSTAT3 knockdown. D, the samecell lines were tested for theirsensitivity to pharmacologicinhibition of STAT3 with Stattictreatment at the indicatedconcentrations (72 hours; mmol/L).E and F, autophagy was inhibitedby transduction with a shRNA forATG7 (A7) or BECN1 (B1) or treatedwith chloroquine (CQ), and STAT3phosphorylation was evaluated byWestern blot analysis either 48hours after plating (E) or 24 hoursafter chloroquine treatment (F). F,chloroquine concentration was asfollows: MCF7 30 mmol/L, MDA-MB-231 25 mmol/L, and MDA-MB-468 10 mmol/L, which were thehighest concentrations that did notkill a significant amount of cells atthis time point and if anything,underestimates the effect ofchloroquine treatment on MDA-MB-231 and MDA-MB-468 cells.Graphs in B and D, mean � SE ofthree independent experimentsperformed in triplicate. �, differentfrom its respective nonsilencingcontrol (P < 0.01).

Maycotte et al.





induces Ras, Src, and phosphoinositide 3-kinase activation(37). Importantly, RAS has not been found to be frequentlyactivated in breast cancer (38).Here, we found that different subtypes of breast cancer

vary markedly in their dependence on autophagy, withTNBC cell lines displaying particular sensitivity to autop-hagy inhibition by ATG gene knockdown or chloroquinetreatment when compared with luminal and nontrans-formed cell lines (Figs 1–3). These cells were dependent onautophagy for survival even in complete media, i.e., in theabsence of stresses, which are known to stimulate autop-hagy to maintain cellular homeostasis. Conversely in autop-hagy-independent MCF7 cells autophagy was needed forsurvival upon starvation but not in unstressed conditions.Autophagy dependence was also independent of RAS path-way status because the only TNBC cell line that we used thatis known to be RAS mutated is the MDA-MB-231 cell line(39). Instead, we found that STAT3 constitutive activity isregulated by autophagy only in some cells and that this isrequired and sufficient for autophagy dependence for sur-vival (Figs 5 and 6). This differential requirement for autop-hagy has effects on tumor treatment because mice withchloroquine-treated MDA-MB-231 xenografts had a betterevent-free survival (when compared with their matchedcontrols) than those with MCF7 tumors (Fig. 7A) andautophagy inhibition with chloroquine treatment synergizedwith chemotherapy in cell lines that were autophagy-depen-dent and not in the ones that were autophagy-independent(Fig. 7B).The effect of autophagy inhibition during cancer treatment

remains a controversial issue (1, 4, 40). Although numerousstudies have shown a significant decrease in cell number withautophagy inhibition in combination with radio- or chemo-therapy in breast cancer cell lines (41, 42) and in tumorxenografts (43–45); other studies show a limited or null effect(46). Although it makes sense that autophagy inhibition may

be more beneficial when used with particular cancer thera-pies (e.g., those that themselves directly activate the autop-hagy pathway), our data suggest that in addition, only somecancer cells will really benefit from its suppression duringtreatment. Particularly, we propose that TNBCs with highconstitutive levels of STAT3 phosphorylation will be autop-hagy dependent and will, therefore, respond the most toautophagy inhibition alone or in combination with chemo-therapy. Therefore, synergy with chemotherapy and autop-hagy inhibition will work best in such tumor cells. Themechanism by which autophagy promotes STAT3 phosphor-ylation remains to be determined.

Recent studies have suggested a correlation between theautophagic marker LC3B and patient outcome in breast can-cer, in which breast cancers with the highest LC3B expressionwere found to show increased markers for cell proliferation,aggressiveness, and had the worst outcome (47, 48). Aboutbreast cancer subtypes, Lazova and colleagues found that onlyluminal A tumors correlated between LC3B expression andpoor outcome, whereas Chen and colleagues found this cor-relation only in TNBC. Although both studies suggest thatincreased LC3B total staining could be used as a marker ofincreased autophagy, this is not necessarily true becauseincreased total LC3 or autophagosome number could alsoreflect a block in autophagic flux (49). Our results suggest thatregardless of the levels of autophagy in breast cancer cells,molecular markers like activated STAT3, could be used todetermine which breast cancers would benefit the most fromautophagy inhibition.

This work has potential therapeutic implications becauseTNBC has the worst prognosis among all breast cancers. Thereare currently no specific therapies for patients with TNBC and,although they can be treated with chemotherapy, they tend torelapse andmetastasize early (11). Our results suggest that thissubtype of the disease may be most responsive to autophagyinhibition and that screening for high levels of constitutive

Figure 6. Constitutively activeSTAT3 decreases cell deathinduced by autophagy inhibition.MDA-MB-231 and MDA-MB-468cells were transduced with alentivirus expressing aconstitutively active form of STAT3(STAT3C). A, as shown byWesternblot analysis, STAT3C increased p-Tyr STAT3 levels after autophagyinhibition with ATG7 (A7) or BECN1(B1) shRNAs. B, cells were thenevaluated for death with an LDHassay 3 days (MDA-MB-231) or 1day (MDA-MB-468) after plating.Graphs, mean � SE of threeindependent experimentsperformed in triplicate. �, differentto its empty vector control,P < 0.05.






STAT3 activation in tumors may serve as a selection strategyfor treatment with autophagy inhibitors.

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

Authors' ContributionsConception and design: P.Maycotte, M.J. Morgan, D.L. Gustafson, A. ThorburnDevelopment of methodology: P. Maycotte, J.M. Mulcahy Levy, D.L.GustafsonAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): P. Maycotte, R. Barnard, S. Aryal, J.M. Mulcahy Levy,R.J. Hansen, D.L. Gustafson

Analysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): P.Maycotte, C.M. Gearheart, R. Barnard, R.J. Hansen,C.C. Porter, D.L. Gustafson, A. ThorburnWriting, review, and/or revision of the manuscript: P. Maycotte, C.M.Gearheart, R. Barnard, J.M. Mulcahy Levy, R.J. Hansen, M.J. Morgan, C.C. PorterAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): S. Aryal, S.P. FosmireStudy supervision: P. Maycotte, A. Thorburn

AcknowledgmentsThe authors are thankful for the contribution to this research made by the

following University of Colorado Cancer Center Research Cores: ProteinProduction/Mab/Tissue Culture, Flow Cytometry, Functional Genomics, andResearch Histology Core (E. Erin Smith, April Otero, and Jenna Van DerVolgen).

Figure 7. Autophagy inhibition withchloroquine (CQ) increases themedian event-free survival ofMDA-MB-231 but not MCF7 breasttumor xenografts and synergizeswith chemotherapy primarily inautophagy-dependent cells. A,female nu/nu mice were injectedwith 5 � 106 MDA-MB-231 orMCF7 cells into the fourthmammary fat pad. Upon reaching atumor volume of 118 mm3 in MCF7and 126 mm3 in MDA-MB-231,mice were treated with chloroquine(60 mg/kg/d) or 0.9% saline,intraperitoneally. Data arepresented as time to reach fourtimes initial volume (TV�4). Log-rank (Mantel–Cox) analysis of thecurves shows significant differencefor the MDA-MB-231 tumors andno significant difference for theMCF7 tumors. Tumor sectionswere stained for LC3 and picturesshow a representative image. Bar,200 mm. B, MCF7, MDA-MB-231,and MDA-MB-468 cell lines weretreated with doxorubicin (DOX) atdifferent doses (mmol/L) incombination with chloroquine (CQ;mmol/L). The CI was determinedwhere log CI < 0 is synergistic,CI ¼ 0 additive, and CI > 0 isantagonistic. Graph, onerepresentative experiment of twoindependent experiments.

Maycotte et al.





Grant SupportThis work was supported by NIH grant CA150925 and the following shared

resources supported by P30 CA046934 Protein Production/Mab/Tissue Culture,Flow Cytometry, Functional Genomics, and Histology. This work was alsosupported (in part) by a research grant from the Cancer League of Colorado,Inc. (P. Maycotte); Elope, Inc. St. Baldrick Foundation Scholar Award (C.M.Gearheart and J.M. Mulcahy Levy); and NIH/NICHD Child Health ResearchCareer Development Award (K12 HD068372; J.M. Mulcahy Levy).

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

Received December 2, 2013; revised February 18, 2014; accepted February 19,2014; published OnlineFirst March 3, 2014.

References1. White E. Deconvoluting the context-dependent role for autophagy in

cancer. Nat Rev Cancer 2012;12:401–10.2. Debnath J. The multifaceted roles of autophagy in tumors implica-

tions for breast cancer. J Mammary Gland Biol Neoplasia 2011;16:173–87.

3. Mizushima N, KomatsuM. Autophagy: renovation of cells and tissues.Cell 2011;147:728–41.

4. Rubinsztein DC, Codogno P, Levine B. Autophagy modulation as apotential therapeutic target for diverse diseases. Nat Rev Drug Discov2012;11:709–30.

5. GuoJY,ChenHY,MathewR, Fan J,StroheckerAM,Karsli-UzunbasG,et al. Activated Ras requires autophagy to maintain oxidative metab-olism and tumorigenesis. Genes Dev 2011;25:460–70.

6. Yang S,Wang X, Contino G, LiesaM, Sahin E, Ying H, et al. Pancreaticcancers require autophagy for tumor growth. Genes Dev 2011;25:717–29.

7. LockR, RoyS, KenificCM,Su JS, Salas E, RonenSM, et al. Autophagyfacilitates glycolysis during Ras-mediated oncogenic transformation.Mol Biol Cell 2011;22:165–78.

8. DeSantis C, Siegel R, Bandi P, Jemal A. Breast cancer statistics, 2011.CA Cancer J Clin 2011;61:409–18.

9. Prat A, Parker JS, Fan C, Cheang MC, Miller LD, Bergh J, et al.Concordance among gene expression-based predictors for ER-pos-itive breast cancer treated with adjuvant tamoxifen. Ann Oncol 2012;23:2866–73.

10. Prat A, Perou CM. Deconstructing the molecular portraits of breastcancer. Mol Oncol 2011;5:5–23.

11. Carey L, Winer E, Viale G, Cameron D, Gianni L. Triple-negative breastcancer: disease entity or title of convenience? Nat Rev Clin Oncol2010;7:683–92.

12. Marotta LL, Almendro V, Marusyk A, Shipitsin M, Schemme J, WalkerSR, et al. The JAK2/STAT3 signaling pathway is required for growth ofCD44(þ)CD24(�) stem cell-like breast cancer cells in human tumors. JClin Invest 2011;121:2723–35.

13. Levy DE, Darnell JE Jr. Stats: transcriptional control and biologicalimpact. Nat Rev Mol Cell Biol 2002;3:651–62.

14. Bowman T, Garcia R, Turkson J, Jove R. STATs in oncogenesis.Oncogene 2000;19:2474–88.

15. Bromberg JF, Wrzeszczynska MH, Devgan G, Zhao Y, PestellRG, Albanese C, et al. Stat3 as an oncogene. Cell 1999;98:295–303.

16. YuH, Pardoll D, Jove R. STATs in cancer inflammation and immunity: aleading role for STAT3. Nat Rev Cancer 2009;9:798–809.

17. Hillion J, Dhara S, Sumter TF, Mukherjee M, Di Cello F, Belton A, et al.The high-mobility group A1a/signal transducer and activator of tran-scription-3 axis: an achilles heel for hematopoietic malignancies?Cancer Res 2008;68:10121–7.

18. Lois C, Hong EJ, Pease S, Brown EJ, Baltimore D. Germline trans-mission and tissue-specific expression of transgenes delivered bylentiviral vectors. Science 2002;295:868–72.

19. Behrends C, Sowa ME, Gygi SP, Harper JW. Network organization ofthe human autophagy system. Nature 2010;466:68–76.

20. Porter CC, Kim J, Fosmire S, Gearheart CM, van Linden A, Baturin D,et al. Integrated genomic analyses identify WEE1 as a critical mediatorof cell fate and a novel therapeutic target in acute myeloid leukemia.Leukemia 2012;26:1266–76.

21. Kim J, Tan AC. BiNGS!SL-seq: a bioinformatics pipeline for theanalysis and interpretation of deep sequencing genome-widesynthetic lethal screen. Methods Mol Biol 2012;802:389–98.

22. Langmead B, Trapnell C, Pop M, Salzberg S. Ultrafast and memory-efficient alignment of short DNA sequences to the human genome.Genome Biology 2009;10:R25.

23. Kolde R. pheatmap: Pretty heatmaps. R package version 0.7.4, 2012.24. R Development Core Team. R: a language and environment for

statistical computing. Vienna, Austria: R Foundation for StatisticalComputing; 2012.

25. Teicher B. In vivo tumor response end points. In:Teicher B, editor.Tumor models in cancer research. Totowa, NJ: Humana Press; 2002.p. 593–616.

26. Lee IH, Kawai Y, Fergusson MM, Rovira II, Bishop AJ, Motoyama N,et al. Atg7 modulates p53 activity to regulate cell cycle and survivalduring metabolic stress. Science 2012;336:225–8.

27. Radoshevich L, Murrow L, Chen N, Fernandez E, Roy S, Fung C, et al.ATG12 conjugation to ATG3 regulatesmitochondrial homeostasis andcell death. Cell 2010;142:590–600.

28. Liu J, Xia H, KimM, Xu L, Li Y, Zhang L, et al. Beclin1 controls the levelsof p53by regulating the deubiquitination activity of USP10 andUSP13.Cell 2011;147:223–34.

29. Chang YP, Tsai CC, Huang WC, Wang CY, Chen CL, Lin YS, et al.Autophagy facilitates IFN-gamma-induced Jak2-STAT1 activationand cellular inflammation. J Biol Chem 2010;285:28715–22.

30. NomanMZ, Janji B, KaminskaB, VanMoerK, PiersonS,Przanowski P,et al. Blocking hypoxia-induced autophagy in tumors restores cyto-toxic T-cell activity and promotes regression. Cancer Res 2011;71:5976–86.

31. Kang R, Loux T, TangD, Schapiro NE, Vernon P, Livesey KM, et al. Theexpression of the receptor for advanced glycation endproducts(RAGE) is permissive for early pancreatic neoplasia. Proc Natl AcadSci U S A 2012;109:7031–6.

32. Garcia R, Bowman TL, Niu G, Yu H, Minton S, Muro-Cacho CA, et al.Constitutive activation of Stat3 by the Src and JAK tyrosine kinasesparticipates in growth regulation of human breast carcinoma cells.Oncogene 2001;20:2499–513.

33. Neve RM, Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T, et al. Acollectionof breast cancer cell lines for the studyof functionally distinctcancer subtypes. Cancer Cell 2006;10:515–27.

34. Schust J, Sperl B, Hollis A,Mayer TU, Berg T. Stattic: a small-moleculeinhibitor of STAT3 activation and dimerization. Chem Biol 2006;13:1235–42.

35. MaesH, RubioN,Garg AD, Agostinis P. Autophagy: shaping the tumormicroenvironment and therapeutic response. Trends Mol Med2013;19:428–46.

36. Rosenfeldt MT, O'Prey J, Morton JP, Nixon C, Mackay G, MrowinskaA, et al. p53 status determines the role of autophagy in pancreatictumour development. Nature 2013;504:296–300.

37. Wei H, Wei S, Gan B, Peng X, Zou W, Guan JL. Suppression ofautophagy by FIP200 deletion inhibits mammary tumorigenesis.Genes Dev 2011;25:1510–27.

38. Koboldt DC, Fulton RS, McLellan MD, Schmidt F, Kalicki-Veizer J,McMichael JF, et al. Comprehensive molecular portraits of humanbreast tumours. Nature 2012;490:61–70.

39. Forbes SA, Bindal N, Bamford S, Cole C, Kok CY, Beare D, et al.COSMIC: mining complete cancer genomes in the Catalogue ofSomatic Mutations in Cancer. Nucleic Acids Res 2011;39:D945–50.

40. Maycotte P, Thorburn A. Autophagy and cancer therapy. Cancer BiolTher 2011;11:127–37.

41. BristolML,Di X, BeckmanMJ,Wilson EN,HendersonSC,Maiti A, et al.Dual functions of autophagy in the response of breast tumor cells to






radiation: cytoprotective autophagy with radiation alone and cytotoxicautophagy in radiosensitization by vitamin D 3. Autophagy 2012;8:739–53.

42. Milani M, Rzymski T, Mellor HR, Pike L, Bottini A, Generali D, et al.The role of ATF4 stabilization and autophagy in resistance ofbreast cancer cells treated with Bortezomib. Cancer Res 2009;69:4415–23.

43. Rao R, Balusu R, FiskusW, Mudunuru U, Venkannagari S, Chauhan L,et al. Combination of pan-histone deacetylase inhibitor and autophagyinhibitor exerts superior efficacy against triple-negative human breastcancer cells. Mol Cancer Ther 2012;11:973–83.

44. Cufi S, Vazquez-Martin A, Oliveras-Ferraros C, Corominas-Faja B,Cuyas E, Lopez-Bonet E, et al. The anti-malarial chloroquine over-comes primary resistance and restores sensitivity to trastuzumab inHER2-positive breast cancer. Sci Rep 2013;3:2469.

45. Lee KH, Hsu EC, Guh JH, Yang HC, Wang D, Kulp SK, et al. Targetingenergy metabolic and oncogenic signaling pathways in triple-negative

breast cancer by a novel adenosinemonophosphate-activated proteinkinase (AMPK) activator. J Biol Chem 2011;286:39247–58.

46. Bristol ML, Emery SM, Maycotte P, Thorburn A, Chakradeo S, GewirtzDA. Autophagy inhibition for chemosensitization and radiosensitiza-tion in cancer: do thepreclinical data support this therapeutic strategy?J Pharmacol Exp Ther 2013;344:544–52.

47. Lazova R, Camp RL, Klump V, Siddiqui SF, Amaravadi RK, PawelekJM. Punctate LC3B expression is a common feature of solid tumorsand associated with proliferation, metastasis, and poor outcome. ClinCancer Res 2012;18:370–9.

48. Chen S, Jiang YZ, Huang L, Zhou RJ, Yu KD, Liu Y, et al. The residualtumor autophagy marker LC3B serves as a prognostic marker in localadvanced breast cancer after neoadjuvant chemotherapy. Clin CancerRes 2013;19:6853–62.

49. Klionsky DJ, Abdalla FC, Abeliovich H, Abraham RT, Acevedo-Arozena A, Adeli K, et al. Guidelines for the use and interpretationof assays for monitoring autophagy. Autophagy 2012;8:445–544.


Maycotte et al.




2014;74:2579-2590. Published OnlineFirst March 3, 2014.Cancer Res Paola Maycotte, Christy M. Gearheart, Rebecca Barnard, et al. Breast Cancer Where Autophagy Inhibition Can Be EfficaciousSTAT3-Mediated Autophagy Dependence Identifies Subtypes of

Updated version

10.1158/0008-5472.CAN-13-3470doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2014/03/04/0008-5472.CAN-13-3470.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/74/9/2579.full#ref-list-1

This article cites 46 articles, 16 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/74/9/2579.full#related-urls

This article has been cited by 19 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://cancerres.aacrjournals.org/content/74/9/2579To 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-13-3470

http://cancerres.aacrjournals.org/content/suppl/2014/03/04/0008-5472.CAN-13-3470.DC1

http://cancerres.aacrjournals.org/content/74/9/2579.full#ref-list-1

http://cancerres.aacrjournals.org/content/74/9/2579.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/74/9/2579


stat3-mediated autophagy dependence identiﬁ subtypes of ...therapeutics, targets, and chemical...

Documents