Cancer Research WNT11ExpressionIsInducedbyEstrogen...

Ther

WNandIncr

MaryDmitr

Abst

Intro

Thephantranscoutcomwhethnonicaofmetsion isliver (4that reexprescycle afines oa reguenerg

AuthorDuke U

Note:Resear

CorrescologyResear681-71

doi: 10

©2010

Cance9298

Published OnlineFirst September 24, 2010; DOI: 10.1158/0008-5472.CAN-10-0226

Canceresearch

apeutics, Targets, and Chemical Biology

T11 Expression Is Induced by Estrogen-Related Receptor αβ-Catenin and Acts in an Autocrine Manner to

R

ease Cancer Cell Migration

A. Dwyer, James D. Joseph, Hilary E. Wade, Matthew L. Eaton, Rebecca S. Kunder,
i Kazmin, Ching-yi Chang, and Donald P. McDonnell
ractElev

with apathopapprohibitedbiochebindinβ-catregulamaticamigrattioned

lator of my demand

s' Affiliationniversity Me

Supplemench Online (h

ponding Auand Cancerch Drive, D39; E-mail:

.1158/0008-

American A

r Res; 70(

Downloa

ated expression of the orphan nuclear receptor estrogen-related receptor α (ERRα) has been associatednegative outcome in several cancers, although the mechanism(s) by which this receptor influences thehysiology of this disease and how its activity is regulated remain unknown. Using a chemical biologyach, it was determined that compounds, previously shown to inhibit canonical Wnt signaling, also in-the transcriptional activity of ERRα. The significance of this association was revealed in a series of

mical and genetic experiments that show that (a) ERRα, β-catenin (β-cat), and lymphoid enhancer-g factor-1 form macromolecular complexes in cells, (b) ERRα transcriptional activity is enhanced byexpression and vice versa, and (c) there is a high level of overlap among genes previously shown to beted by ERRα or β-cat. Furthermore, silencing of ERRα and β-cat expression individually or together dra-lly reduced the migratory capacity of breast, prostate, and colon cancer cells in vitro. This increasedion could be attributed to the ERRα/β-cat–dependent induction of WNT11. Specifically, using (a) condi-medium from cells overexpressing recombinant WNT11 or (b) WNT11 neutralizing antibodies, we wereshow that this protein was the keymediator of the promigratory activities of ERRα/β-cat. Together, these
able to
data provide evidence for an autocrine regulatory loop involving transcriptional upregulation of WNT11 byERRα and β-cat that influences the migratory capacity of cancer cells. Cancer Res; 70(22); 9298–308. ©2010 AACR.

metabcanceIn a

ERRαat sombetwein thethat thwouldpathwminedare cothis crERRαin vitro

duction

estrogen-related receptor α (ERRα; NR3B1) is an or-member of the nuclear receptor (NR) superfamily ofription factors whose expression tracks with a negativee in breast and ovarian cancers (1–3). It is unclear

er this ubiquitously expressed receptor requires a ca-l small molecule ligand. However, under conditionsabolic stress, such as fasting, exercise, or cold, its expres-rapidly induced in tissues such as the heart, muscle, and, 5). In these tissues, it directs a gene expression programsults in increased mitochondrial number and increasedsion of key enzymes required for the tricarboxylic acidnd β-oxidation of fatty acids (6–8). Thus, within the con-f normal physiology, ERRα seems to function primarily as

etabolic function under conditions of high-(9, 10). The extent to which its role as a

an indCry

bers obindinas to cdock sof ERRbindinphenyable ina reguthe st

: Department of Pharmacology and Cancer Biology,dical Center, Durham, North Carolina

tary data for this article are available at Cancerttp://cancerres.aacrjournals.org/).

thor: Donald P. McDonnell, Department of Pharma-Biology, Duke University Medical Center, C259 LSRC,urham, NC 27710. Phone: 919-684-6035; Fax: [email protected].

5472.CAN-10-0226

ssociation for Cancer Research.

22) November 15, 2010

Research. on January 4, 202cancerres.aacrjournals.org ded from

olic regulator is involved in the pathophysiology ofr, however, remains to be determined (1–3).ddition to its important role in oxidative metabolism,can modulate the activity of the estrogen receptor (ER)e target genes (11). Indeed, the amino acid homologyen ERRα and the classic ERs (ERα and ERβ), particularlyir respective DNA-binding domains, initially suggestede primary effect of ERRα in hormone-dependent cancersbe to modulate or interface with estrogen signalingays in cells. However, we and others have recently deter-that only a relatively small percentage of ER target genesregulated by ERRα (12, 13). Although the significance ofoss talk remains to be established, the observation thatknockdown dramatically affected the in vivo growth andmigration of ER-negativeMDA-MB231 cells highlightedependent role for this receptor in breast cancer (13).stallographic analysis of the structure of several mem-f the NR superfamily has indicated that the hormone-g domain of these proteins is configured in such a wayreate a cavity of between 360 and 1,400 A3 that serves tomall molecule agonists or antagonists (14). In the caseα, however, it has been shown that its potential ligandg cleft is occupied by the bulky side chains of fourlalanines and that the remaining space (100 A3) avail-the pocket is likely to be too small to accommodate
latory ligand (15). Furthermore, by comparison withructures of other agonist-activated NRs, the apoERRα
0. © 2010 American Association for Cancer

http://cancerres.aacrjournals.org/

proteifindinactivitand pmanipand awhichshownERRαthe exreceptIndeedare regto actiextentcancerwhat pcofactpathwtherefodownsmay b

Mate

PlasmThe

2X9 wpcDNAchasedDuke UeratedpENTR

Cell cCel

Collecand ccells oby moand M[InvitPC-3 (gen; Mbovineacids,transfeeraseusing

CoimWh

sis bufglycerhibitoantibolymph

technoantiboA/G P4°C), w2× samtransf[ERRαtechn(Jacks

ImmuWh

tion aNaCl,1 mmby 10brane(21),βtein 5ogy), o

ERRα and β-cat Enhance Cancer Cell Migration

www.a


n seems to be in the “active” conformation (15, 16). Thisg has raised the question as to how the transcriptionaly of this receptor is regulated andwhether the processesathways that impinge on and activate ERRα can beulated for therapeutic advantage. Cofactor availabilityctivity are likely to be the primary mechanisms byERRα activity is regulated (5, 17 18). It has been, for instance, that the generally low basal activity ofin cells can be dramatically upregulated by increasingpression of either peroxisome proliferator-activatedor-γ coactivator-1 (PGC-1) isoform PGC-1α or PGC-1β., the expression level and activity of these coactivatorsulated by the physiologic stresses that have been shownvate ERRα transcriptional activity (19, 20). However, theto which these proteins regulate ERRα activity in breastremains to be determined. Furthermore, it is unclearathologic signals regulate the activity/expression of theseors and whether or not there are cofactor-independentays that modulate ERRα activity. The goal of this study,re, was to identify pathways and processes upstreamand
tream of ERRα that effect tumor pathophysiology and Santa
AdenoAde

PGC-1(21). Mtion o

GeneChe

were uHSS10WNT1glucocHSS10HSS11mentasix-we(Dhar

RNA pTot

rificatiScrip[0.25 μwith icalcul

ChromChr

formementalisted

WNT1pM

e amenable to therapeutic manipulation.

rials and Methods

ids3X-ERE-tata-luciferase reporter and pcDNA3-PGC-1αere previously described (21). pCMV-β-gal (Clontech),(Invitrogen), and pBlueScriptII (Statagene) were pur-. pCMX-ΔN89 and TOP-Flash were gifts (Dr. B. Hogan,niversity). The pMSCV-GFP-hWNT11 plasmid was gen-by subcloning WNT11 cDNA (MGC:141946) into3c (Invitrogen) and recombining into pMSCV-IRES-GFP.

ulturel lines were obtained from American Type Culturetion (ATCC; 2007–2009), expanded for two passages,ryopreserved. All experiments were performed withf passage of <25. These cell lines were authenticatedrphologic inspection, short tandem repeat profiling,ycoplasma testing by ATCC and cultured in RPMIrogen; MDA-MB 436 (HTB-130), SKBR3 (HTB-30),CRL-1435), and HCT-116 (CCL-247)] or DMEM (Invitro-DA-MB 231 (HTB-26)] supplemented with 8.5% fetalserum (FBS; Sigma), 0.1 mmol/L nonessential amino

and 1 mmol/L sodium pyruvate (Invitrogen). Transientctions were performed as described previously (22). Lucif-and β-galactosidase (β-gal) activities were measureda Perkin-Elmer Fusion Instrument (22).

munoprecipitationole-cell extracts were prepared using nondenaturing ly-fer [20 mmol/L Tris-HCl (pH 8), 137 mmol/L NaCl, 10%ol, 1% Nonidet P-40, 2 mmol/L EDTA, and protease in-rs (Sigma)]. Proteins were immunoprecipitated using
dies to ERRα (21), β-catenin (β-cat; BD Biosciences),oid enhancer-binding factor-1 (LEF-1; Santa Cruz Bio-
DBD (Scienc

acrjournals.org

Research. on January 4, 202cancerres.aacrjournals.org Downloaded from

logy), and mouse IgG (Santa Cruz Biotechnology; 5 μgdy/500 μg whole-cell extract, 16 hours, 4°C) and protein-LUS-Agarose beads (Santa Cruz Biotechnology; 4 hours,ashed using lysis buffer three times, and heat eluted inple buffer. Proteins were separated by 10% SDS-PAGE,

erred to nitrocellulose, and detected byWestern blotting(21), β-cat (EMD Biosciences), LEF-1 (Santa Cruz Bio-

ology), and a light chain–specific secondary antibodyon Immunoresearch)].

noblottingole-cell extracts prepared using radioimmunoprecipita-ssay buffer [50 mmol/L Tris-HCl (pH 7.5), 150 mmol/L1% Nonidet P-40, 0.5% sodium deoxycholate, 0.05% SDS,ol/L EDTA, protease inhibitors (Sigma)] were separated% SDS-PAGE, transferred onto nitrocellulose mem-s, and detected using the following antibodies: ERRα-cat, (BD Biosciences),WNT11 (Abcam), autophagy pro-(ATG5; Cell Signaling), lamin A (Santa Cruz Biotechnol-r glyceraldehyde-3-phosphate dehydrogenase (GAPDH;Cruz Biotechnology).

viral transductionnoviruses expressing β-gal, PGC-1α, PGC-1α 2×9, orα L2L3M were generated as described previouslyDA-MB 231 cells were infected at multiplicity of infec-f 30 for 48 hours.

silencingmical small interfering RNAs (siRNA; Invitrogen, Qiagen)sed to silence ERRα (siERRα A, HSS103381; siERRα B,3382), β-cat (siβ-cat A, HSS102460; siβ-cat B, HSS102461),1 (siWNT11 A, SI00763378; siWNT11 B, SI03148719), serumorticoid kinase 1 (SGK1; siSGK1 A, HSS109684; siSGK1 B,9685), or ATG5 (siATG5 A, HSS114103; siATG5 B,4104). Control siRNA sequences are listed in Supple-ry Table S2. MDA-MB 231 cells were seeded (250,000 perll plate), and siRNAs were transfected using Dharmafect1macon; 100 nmol/L, 48 hours).

reparation and analysisal RNA was isolated using the Bio-Rad Aurum RNA pu-ion kit. cDNA was synthesized from 1μg total RNA usingt (Bio-Rad). Quantitative PCR (qPCR) was performedL cDNA, 0.3 μmol/L primers (Supplementary Table S1)Q SYBRGreen supermix (Bio-Rad)], and results wereated using the 2-ΔΔCt method.

atin immunoprecipitation assaysomatin immunoprecipitation (ChIP) assays were per-d as described previously (additional details in Supple-ry Materials and Methods) using primer sequencesin Supplementary Table S3.

1 retroviral overexpressionSCV-IRES-GFP-hWNT11 and pMSCV-IRES-GFP-Gal4-
control) were cotransfected (FuGene, Roche Appliede) with the pCL10A1 packaging vector (Imigenex) into
Cancer Res; 70(22) November 15, 2010 9299



293TSbreneMB 23were ssion, y231/W

MigraFor

DMEMHCT-1HEPESPC-3)Biocoacoated(MDA116). Tmethatranswwere cformeTiter B100 μLfollow12 howas mfluorefollowthe metransfbeforetion aslines,studieMDA-

ImmuCM

harvesproteibit IgG

StatisTra

meanevaluafor mi

Resu

CrossA ch

engagelogic icell-balate thin MDing co

(harmwhichon a shas beical Wsomeadipoglished(24, 25plicatethat b

Figureβ-cat pSKBR3measuractivatiof endowhole-cell extracts was followed by Western blot analysis for theindicate

Dwyer et al.

Cance9300


cells. These viral supernatants were clarified, poly-supplemented (8 μg/mL), and used to infect MDA-1 cells for two serial 24-hour infections. Positive cellselected by three rounds of cell sorting for GFP expres-ielding the MDA-MB 231/Control (C) and MDA-MBNT11 cell lines.

tion and viability assaysmigration assays, cells were serum starved [18 hours,(MDA-MB 231) or RPMI (MDA-MB 436, PC-3, and

16) with 0.1% bovine serum albumin and 10 mmol/L], and 2 × 104 cells (MDA-MB 231, MDA-MB 436, andor 7.5 × 104 cells (HCT-116) in 100 μL were plated (BDt Control Inserts 8.0 micron, BD Biosciences), collagen(HCT-116), and migrated toward 8% FBS for 4 hours

-MB 231, MDA-MB 436, and PC-3) or 16 hours (HCT-he membrane was stained (5% crystal violet in 20%nol), and cells that migrated were counted. Duplicateells were used, and three high-powered fields (200×)ounted per membrane. Cell viability assays were per-d in parallel with the migration studies using the Celllue Assay (data not shown; Promega). Cells (2 × 104 in) were seeded in triplicate on a 96-well plate for 4 hoursed by the addition of Cell Titer Blue dye (20 μL) forurs. Resultant fluorescence at (535 nmEx, 620 nmEm)easured using a Perkin-Elmer Fusion Instrument. Thescence was calculated as the triplicate average (±SEM)ed by subtraction of background fluorescence fromdium. For ERRα, β-cat, andWNT11 silencing, cells wereectedwith siRNAs as described (Gene Silencing) 48 hoursthemigration assay. ForWNT11 overexpression, migra-says were performed using the stable MDA-MB 231 cellcontrol, and WNT11. For conditioned medium (CM)s, MDA-MB 231 cells migrated toward CM from theMB 231/C or MDA-MB 231/WNT11 cells.

nodepletion of WNT11from MDA-MB 231/C and MDA-MB 231/WNT11 wasted at 80% confluence and was depleted of WNT11n by incubation with WNT11 antibody (Abcam) or rab-(Santa Cruz Biotechnology; 10μg/mL, 4°C for 16 hours).

tical analysesnsfection, qPCR, and migration data are represented as± SEM for three biological replicates. Significance wasted by ANOVA and the Neumann-Keul's post hoc testgration studies (GraphPad).

lts

talk between ERRα and Wnt signaling pathwaysemical biology approachwas used to define how ERRα isd in the regulation of pathways and processes of patho-mportance in cancer. Specifically, a high-throughputsed assay was used to identify compounds that modu-e transcriptional activity of the PGC-1α/ERRα complex
A-MB 436 cells (ERα negative). Among the most interest-mpounds identified in this manner were the carbolines
Asof ER

r Res; 70(22) November 15, 2010


ol, harmine, harmane, and 6-methoxyharmalan), all ofinhibited ERRα transcriptional activity when assayedimple reporter (Supplementary Fig. S1). Previously, iten shown that these carbolines could inhibit the canon-nt signaling pathway and, in doing so, enhance peroxi-proliferator-activated receptor γ (PPARγ) activity inenesis assays in vitro (23). Given these data and the estab-role(s) of Wnt signaling in cell migration and invasion), together with our previously published data, which im-s ERRα in this biological process (13), it seemed likelyoth effectors were components of the same pathway.

d proteins. IP Ab: I, input; NS, IgG; E, ERRα; B, β-cat; L, Lef-1.

1. Cross talk of ERRα and Wnt signaling pathways. A, ERRα andotentiate each other's transcriptional activity when assessed incells using the TOP-FLASH and 3X-ERE-tata-luc reporters toe β-cat and ERRα activity, respectively. Similar transcriptionalon was observed in MDA-MB 436 cells. B, coimmunoprecipitationgenous ERRα, β-cat, and LEF-1 from SKBR3 and MDA-MB 231

an initial step in our analysis, we evaluated the effectRα expression on the transcriptional activity of a

Cancer Research



constiFlashFig. 1Aexpresof β-cERRαenhantata-lubasal a

activitcells. Tassaysies, winteramunopresse

Figurecancerof ERRMDA-MPC-3, aCells wdifferen(A andscrambfor 18 hof migrviabilityMock (siβ-lactcells exsiC-treaFig. S7XCT790MDA-MmigratioXCT790for 30 hwith coanotherviabilityperformdenote


www.a


tutively active β-cat mutant when assayed on the TOP-[T-cell factor (TCF)/LEF] luc-reporter. As shown in, ERRα expression had a minimal effect on the basalsion of the reporter. However, a robust enhancementat–dependent transcriptional activity was observed onexpression. Similarly, expression of β-cat significantlyced the activity of ERRα when assayed on the 3X-ERE-
c reporter (Fig. 1A). As expected, β-cat increased thectivity of this reporter by enhancing the transcriptional
(datawith b

significance (P < 0.05).

acrjournals.org


y of the endogeneous ERRα levels expressed in SKBR3he pathway cross talk revealed in these transcriptionalwas further reinforced in a series of biochemical stud-hich indicated that ERRα, β-cat, and LEF-1 physicallyct (Fig. 1C). Specifically, we were able to show by coim-precipitation studies performed with endogenously ex-d proteins from SKBR3, MDA-MB 231, or MDA-MB 436
not shown) breast cancer cells that ERRα interactedoth β-cat and LEF-1. We confirmed the direct nature
2. ERRα and β-cat promotecell migration. A, silencingα and/or β-cat reducedB 231, MDA-MB 436,nd HCT-116 migration.ere transfected with twot sequences of siRNAsB) for ERRα, β-cat,le (siC), and serum starvedfollowed by assessment

atory capacity and(data not shown).

transfection reagent),amase, and siATG5-treatedhibited similar migration toted cells (Supplementary). B, ERRα degradation byimpedes MDA-MB 231,B 436, PC-3, and HCT-116n. Cells were treated with(0, 2.5, 5, and 10 μmol/L)and then serum starved

ntinued drug treatment for18 h. Migration and cell(data not shown) wereed as in A. Different letters




of thedownical anWnt/βof like

BothinflueERR

studie(13, 25and exit seemtion ouatedβ-cat,multipMDA-ing a Btion wby eithand thboth pditionsignifiponen(Suppbility w(data nassayspressiimmutreatmdepen231, MSuppleviabili

that bof sev

ERRαencodIn l

nalingteinsmigraclassicwere etheretomeable to(a) wecell mmanngenesThe

evaluaERRαlatedsuch aPPARγmakesusingproblethat incan be(6). Inusingan inaMD

pressiquant

urelveivatiindT11NAse inress, orqPCmaltive to β-gal.

Dwyer et al.

Cance9302


se interactions using glutathione S-transferase pull-assays (Supplementary Fig. S2). Thus, using both chem-d biochemical approaches, we determined that the-cat and ERRα signaling pathways converge, a findingly importance in cancer pathogenesis.

ERRα- and β-cat–regulated signaling pathwaysnce the migratory capacity of cancer cellsα and β-cat have both been shown in independents to function as regulators of cancer cell migration). Thus, given that these proteins physically interacthibit functional cross talk at the level of transcription,ed likely that they may also cooperate in the regula-

f pathways that regulate cell migration. Thus, we eval-the effect of siRNA-mediated knockdown of ERRα andindividually or combined, on the migratory capacity ofle cancer cell lines including breast (MDA-MB 231 andMB 436), prostate (PC-3), and colon (HCT-116) cells us-oyden chamber migration assay. Decreased cell migra-as observed after silencing of ERRα or β-cat expressioner of two distinct siRNAs directed against each target,is activity was further reduced when the expression ofroteins was knocked down simultaneously (Fig. 2A). Ad-ally, migration of MDA-MB 231 and PC-3 cells was notcantly affected by silencing of SGK1 or ATG5, a key com-t of autophagy with no known function in migrationlementary Fig. S3). No significant differences in cell via-ere observed in cells treated with the selected siRNAsot shown) under the conditions used for the migration. The efficacy of each of the siRNAs in reducing the ex-on of their respective targets was confirmed by Westernnoblot analysis (Supplementary Fig. S4A). Furthermore,ent with the inverse agonist XCT790 resulted in a dose-dent degradation of ERRα and an inhibition ofMDA-MBDA-MB 436, PC-3, and HCT-116 cell migration (Fig. 2B;
mentary Fig. S4B) with no significant changes in cellty (data not shown). Taken together, these data indicate
mRNAthat E

r Res; 70(22) November 15, 2010


oth ERRα andβ-cat can promote themigratory capacityeral types of cancer cells.

regulates the expression of target genesing proteins with promigratory activitiesight of the cross talk between the ERRα and β-cat sig-pathways observed, we next asked whether these pro-cooperated in the regulation of genes involved in thetory response. In previously published work, we usedmicroarray analysis to identify ERRα target genes thatxpressed in HepG2 andMCF-7 cells (6, 13, 26). Likewise,are published studies that describe the β-cat transcrip-in 293T cells (26). With these data sets in hand, we wereperform a comparative analysis and identify genes thatre regulated in both data sets and (b) were important forigration as shown in previous studies (26–31). In thiser, WNT11, MSX1, and N-cadherin were identified asthat are likely to be coregulated by ERRα and β-cat.expression of WNT11, MSX1, and N-cadherin was nextted in MDA-MB 231 cells following the activation of. Whereas the transcriptional activity of ERRα is regu-by the relative expression and/or activity of cofactorss PGC-1α and PGC-1β (19, 20), other NRs, includingand HNF-4, can also be coactivated by PGC-1α, whichit difficult to study the ERRα signaling axis in isolationthis cofactor as an activator (19, 32). To circumvent thism, we developed a variant of PGC-1α (PGC-1α 2 × 9)teracts in a highly selective manner with ERRα and thusused to specifically regulate the activity of this receptoraddition, when analyzing ERRα target gene expressionPGC-1α as an activator, we also evaluated the effect ofctive variant of PGC-1α (PGC-1α L2L3M) in parallel.A-MB 231 cells were transduced with adenoviruses ex-ng βgal, PGC-1α, PGC-1α 2 × 9, or PGC-1α L2L3M, anditative PCR was used to assess the resulting changes in

Figinvoact2X9WNmRwerexp2X9bynorrela

expression of target genes (FigRRα mRNA levels are increase

0. © 2010 American Associa

3. Novel ERRα target genesd in migration. ERRαon by PGC-1α and PGC-1αuces the expression of, MSX1, and N-cadherinin MDA-MB 231 cells. Cellsfected with adenovirusesing β-gal, PGC-1α, PGC-1αPGC-1α L2L3M, followedR analysis of mRNA levelsized to 36B4 expression and

. 3). It was determinedd on overexpression of

Cancer Research

tion for Cancer


Figure231 celalone) aβ-cat, aDiagramby XV irecruitm


www.a


4. ERRα and β-cat regulate genes involved in cell migration. A, ERRα and β-cat were silenced individually or in combination using siRNA in MDA-MBls. Expression of the indicated genes was measured by qPCR and normalized to 36B4 expression and relative to siC. Mock (transfection reagentnd two siRNA sequences for ERRα and β-cat were used. B, Western blot analysis of siRNA-treated MDA-MB 231 whole-cell extract for ERRα,nd WNT11 to confirm knockdown with GAPDH as loading control. C, ChIP of ERRα and β-cat to WNT11 genomic sequence in MDA-MB 231 cells.of putative TCF-4 REs and ERR REs in WNT11 genomic sequence. ERRα and β-cat recruitment was tested by qPCR. Inhibition of GSK-3β kinase

ncreased β-cat levels and enhanced recruitment of β-cat to both the WNT11 P6 and D5. Downregulation of ERRα and/or β-cat reducesent to the putative ERRα and TCF-4 P6 site. Black columns, ERRα antibody; striped columns, β-cat antibody; white columns, IgG.

Cancer Res; 70(22) November 15, 2010acrjournals.org 9303

Research. on January 4, 2020. © 2010 American Association for Cancercancerres.aacrjournals.org Downloaded from


PGC-1regula(12), wlevelsWNT1by theSimilaHCT-1

CoregERRαThe

tion o

Specifcontroof siRent siReitherWNT1neoussultedmRNAdid nothat n

Dwyer et al.

Cance9304


α (wild-type or 2 × 9), consistent with it being an auto-ted gene (33). Similarly, GSTM1, an ERRα target geneas also regulated by PGC-1α, whereas β-cat mRNAwere not significantly affected (Fig. 3). Interestingly,1, MSX1, and N-cadherin were all significantly inducedexpression of PGC-1α in MDA-MB 231 cells (Fig. 3).r results were observed in MDA-MB 436, PC-3, and16 cells (Supplementary Fig. S5).

ulation of WNT11, MSX1, and N-cadherin byand β-cat
relative importance of ERRα and β-cat in the regula- Discus
r Res; 70(22) November 15, 2010


ically, the expression of each mRNA and appropriatels were measured in cells following the introductionNAs directed against ERRα or β-cat. Using two differ-NAs against each target, we observed that silencing ofERRα or β-cat led to a significant diminution of1, MSX1, and N-cadherin mRNA expression. Simulta-knockdown of both mRNAs in MDA-MB 231 cells re-in a further decrease in the expression level of theses (Fig. 4A). Importantly, silencing of β-cat expressiont affect the mRNA level of GSTM1 (Fig. 4A), suggestingot all ERRα target genes are regulated by β-cat (see
sion). Notably, ERRα and β-cat downregulation, either
n, resulted in a reduction of WNT11

ureratiA-MNAformanculatT11ractressretiot anrexpnificance (P < 0.05).

f WNT11, MSX1, and N-cadherin was next evaluated. singly or in combinatio

FigmigMDsiRperenhpopWNchaexpsecbloovesig

0. © 2010 America

5. WNT11 promotes MDA-MB 231on. A, WNT11 downregulation reducesB 231 migration. WNT11 or control (siC)transfections and migration assays wereed as in Fig. 2. B, WNT11 overexpressiones MDA-MB 231 migration. Stableions ofMDA-MB231 cells overexpressingor empty vector (C) were generated anderized for migration, viability, and proteinion as previously described. Enhancedn of WNT11 was observed by Westernalysis from MDA-MB 231 cellsressing WNT11. Different letters denote

Cancer Research

n Association for Cancer


proteiof WNviouslwe fomechacancerThe

nism(sTo thithe WsponsMDA-of thesignifithe foSupplknockbindinbindinnationand β

WNT1cancePrev

ty of Wintesti(28–30WNT1we usMDA-defineshownwere aprote∼60%effectviabiliThis ethat siof MDsamedata pgratiothesetives oproducwith athesetargetmigraGive

bindintestedreduceERRαexperitainin

overexchambwith cto theCM frincreaconcecells.overexthese(Fig. 6partiaknockpletiobut nocapacemptymentaevidenWNT11 that influences the migratory capacity in this modelof bre

Discu

Croway holism,Wnt/βresultscriptias anendog

FiguresilencinsiC or sMigratiderivatiof WNTimmun(P < 0.0nonspe


www.a


n levels (Fig. 4B). Given the extremely robust regulationT11 mRNA expression by ERRα and β-cat and its pre-y described activity as a promigratory factor (28–30),cused the remainder of the studies on defining thenism by which this gene is regulated and how it affectscell biology.next step in these studies was to define the mecha-) by which ERRα andβ-cat regulateWNT11 expression.s end, we scanned the genomic sequence surroundingNT11 gene using Consite for putative TCF and ERRα re-e elements (Supplementary Fig. S6A). ChIP assays inMB 231 cells were then used to confirm the functionalityputative ERRα and/or TCF binding sites. In this manner,cant binding of both ERRα and β-cat was detected atllowing two sites: P6 (-3042) and D5 (+27794; Fig. 4C;ementary Fig. S6B). Importantly, siRNA-mediateddown of ERRα expression significantly reduced β-catg at the P6 site, and knockdown of β-cat reduced ERRαg (Fig. 4C). These ChIP data provide a molecular expla-for the observed cross talk that occurs between ERRα-cat on the WNT11 gene.

1 acts in an autocrine manner to promoter cell migrationious studies have highlighted the promigratory activi-NT11 in biological contexts, including migration of

nal epithelial cells, neural crest cells, and gastrulation); however, little has been done to define the role of1 in cellular models of cancer. To address this issue,ed siRNAs to knock down WNT11 expression inMB 231 cells and used transwell migration assays tothe effect of this manipulation on cell migration. Asin Fig. 5A, using either of two different siRNAs, weble to achieve a quantitative knockdown of WNT11in expression, a manipulation that resulted in anreduction in cell migration. We did not observe anyof WNT11 knockdown on total cell number and/or cellty under the same experimental conditions (Fig. 5A).ffect was not restricted toMDA-231 cells, as we observedlencing of WNT11 expression also decreased migrationA-MB 436, PC-3, and HCT-116 cells when assayed in themanner (Supplementary Fig. S7). These overexpressionrovide a strong support that WNT11 is promoting mi-n, as this could not be a survival effect. To complementexperiments, we created WNT11-overexpressing deriva-f MDA-MB 231 cells and determined that the increasedtion in both total and secretedWNT11 protein correlatedsignificant increase in cell migration (Fig. 5B). Together,experiments suggest that the ERRα/β-cat–dependentWNT11 acts in an autocrine manner to increase thetion of cancer cells.n that WNT11 is a secreted protein that functions byg to the extracellular frizzled receptor 7 (30), we alsowhether exogenously added WNT11 could restore thed migratory phenotype resulting from the silencing ofand β-cat in the MDA-MB 231 parent cell line. For these
ments, we isolated the CM fromMDA-MB 231 cells (con-g vector alone) or from cells that were engineered to
tweenand s

acrjournals.org


pressWNT11. Subsequently, CMwas added to the lowerer of a transwell plate and MDA-MB 231 cells treatedontrol siRNA (siC) or siERRα and siβ-cat were addedupper chamber. In cells treated with siC, addition ofom WNT11-overexpressing cells did not significantlyse cell migration, probably due to the sufficiently highntration of WNT11 already in the medium in controlHowever, immunodepletion of WNT11 from WNT11-pressing CM significantly reduced the migration ofcells, indicating that WNT11 is a promigratory factor). Importantly, CM from WNT11-overexpressing cellslly reversed the decreased migration resulting fromdown of ERRα and β-cat. Furthermore, immunode-n of WNT11 from this CM using a specific antibody,t with an irrelevant antibody, blocked this restorativeity. Similar results were obtained using WNT11 andvector CM from mouse L-cell fibroblasts (Supple-ry Fig. S8). Together, these data provide compellingce for a regulatory loop involving ERRα, β-cat, and

cific immunodepleted.

ast cancer.

ssion

ss talk between NRs and the Wnt/β-cat signaling path-as been shown previously (34). In the realm of metab-it has been shown that cross talk between PPARγ and-cat signaling enables precise control of adipogenesising from a reciprocal inhibition of each other's tran-onal activities (35). It has been shown that β-cat servesandrogen receptor (AR) coactivator when assessed onenous AR target genes such as PSA (36). Cross talk be-

6. WNT11 partially restores reduced MDA-MB 231 migration byg of ERRα and β-cat. MDA-MB 231 cells were transfected withiERRα and β-cat and serum starved as previously described.on assays were then performed with CM from the MDA-MB 231ve cell lines overexpressingWNT11 or empty vector. Immunodepletion11 from CM reduced migration compared with mock

odepletion with rabbit IgG. Different letters denote significance5). NT, no treatment; ID, WNT11 immunodepleted; NSID,

the Wnt pathway and ERα, LRH-1, retinoid X receptor,everal other NRs has also been shown to occur in a




varietycomperegulagenesmigracell mseque(28–30WNT1and itWn

pathwIn thein disinhibiincreahas twcan trexpresof tranE-cadhactionto-meswith tthat inβ-cat,genesβ-cat,migraproteiin addregulaby regcancerE-cadhthat thsequenDefiniversuscontinIn t

of theregulatheseregionadjacethat oit is likare usinary esets toscribefoundfoundbreastfoundcells bFurth

detecstudy,β-cat–(Suppanalysgenesof thesite wfor β-The sidentiThe

targetto us aassociis upreand bmatioidentinicalWcreasecalciuand pshown(28–30actin cmigraactivacanonE-cadthe crnalingin thisTak

crineof WNthe mdata ppoundcompl

Discl

No p

Ackn

WeDr. B.McDon

Grant

NIHThe

of pageaccorda

Dwyer et al.

Cance9306


of cell-based assays (34, 37). In this study, we providelling evidence that ERRα and β-cat are involved in thetion of WNT11, MSX1, and N-cadherin expression,implicated previously in processes that regulate celltion. Given the wealth of literature linking WNT11 toigration, we focused on evaluating the functional con-nces of ERRα/β-cat–mediated regulation of WNT11). In this manner, we were able to determine that1 is important for migration in breast cancer cells,s expression is influenced by both ERRα and β-cat.t signaling occurs via a canonical or a noncanonicalay, both of which may interface with ERRα/β-cat (38, 39).canonical pathway, Wnt activation of Fzd results

heveled-mediated degradation of axin and GSK-3βtion. Inhibition of this destruction complex leads to anse in the intracellular pool of β-cat. The stabilized β-cato nonexclusive fates within the cell. Firstly, the proteinanslocate to the nucleus, where it regulates target genesion through its interaction with the TCF/LEF familyscription factors. Alternatively, β-cat can interact witherin in adherens junctions and stabilize cell-cell inter-s. Furthermore, β-cat expression promotes epithelial-enchymal transition, a process that is frequently associatedhe loss of E-cadherin. Importantly, it has been shownhibition of E-cadherin expression increases nuclearresulting in the increased expression of promigratory(40). Thus, depending on the relative partitioning ofit can have either a positive or a negative effect on celltion. We have established that the ERRα, a karyophilicn, associates directly with β-cat; thus, it is possible that,ition to cooperating with β-cat in the transcriptionaltion of promigratory genes, ERRαmay affect migrationulating the cellular partitioning of β-cat. In many breastcells, including the MDA-MB 231 cells studied herein,erin expression is extremely low, and thus, it is likelye effects we have observed onmigration are a direct con-ce of the nuclear action of the ERRα/β-cat complex (41).ng the processes that impinge on and regulate the nuclearcytoplasmic actions of the ERRα/β-cat is a focus of ourued efforts in this area.his study, we have performed a comprehensive analysismechanisms underlying the ERRα/β-cat–dependenttion of WNT11. Whereas we can show that both ofproteins interact in cells and can bind to the samein the WNT11 gene, we do not know if they bind tont sites or if a tethering mechanism is involved. Giventher NRs use both mechanisms to engage target genes,ely that both types of interaction, direct and tethering,ed. To address this issue, we have performed a prelim-xamination of available ChIP-ChIP and ChIP-seq dataevaluate the extent to which the binding sites de-

d for ERRα and β-cat converge. In this manner, wea small but significant overlap in target genes that wereto be enriched for ERRα binding in MCF-7 or SKBR3cancer cells by ChIP-ChIP analysis and those that wereto be enriched for β-cat binding in HCT116 colon cancer
y ChIP-seq analysis (Supplementary Fig. S9A; refs. 12, 42).ermore, out of the 547 unique ERRα-regulated genes
ReceOnlineF

r Res; 70(22) November 15, 2010


ted in our previously published MCF-7 microarray39 of these genes were found to contain at least oneenriched region in the HCT116 ChIP-seq experimentlementary Fig. S9B; refs. 13, 42). Finally, bioinformatices using Patser revealed that of the 988 β-cat targetidentified in the HCT116 ChIP-seq experiment, ∼17%genes also contain at least one putative ERR bindingithin the same 600-bp region of DNA that is enrichedcat binding (Supplementary Fig. S9C; refs. 42–44).ignificance and functionality of the convergent sitesfied in this manner are currently under investigation.identification of WNT11 as a direct transcriptionalof the ERRα/β-cat complex was of particular interests (a) the expression of this protein has previously beenated with increased cell migration (28–30, 45); (b) WNT11gulated in several cancers, including colorectal, prostate,reast cancer (46, 47); and (c) WNT11 induces transfor-n of mammary epithelial cells (48). WNT11, initiallyfied as a noncanonical Wnt, can activate the noncano-nt/Ca2+ pathway, resulting in G protein–dependent in-s in intracellular calcium and subsequent activation ofm/calmodulin-dependent protein kinase II (CAMKII)rotein kinase C (PKC; ref. 28). This activity has beento increase intestinal epithelial cellular migration, 46, 48). Interestingly, both CAMKII and PKC facilitateytoskeleton rearrangements that are critical for cellulartion (49). Furthermore, WNT11 has been shown to alsote the canonical Wnt pathway (50). The fact that non-ical pathway activation of PKC can result in decreasedherin function further underscores the significance ofoss talk between canonical and noncanonical Wnt sig-and the likely importance of the ERRα/β-cat complexprocess.en together, our findings provide evidence for an auto-regulatory loop involving transcriptional upregulationT11 by ERRα and β-cat, an activity that influencesigratory capacity of cancer cells. Furthermore, theserovide a strong rationale for the development of com-s that inhibit ERRα or the activity of the ERRα/β-catex as cancer therapeutics.

osure of Potential Conflicts of Interest

otential conflicts of interest were disclosed.

owledgments

thank Dr. V. Giguere (McGill University) for the ERRα antibody,Hogan (Duke University) for plasmids, and the members of thenell laboratory for insightful discussions.

Support

grant DK074652.costs of publication of this article were defrayed in part by the paymentcharges. This article must therefore be hereby marked advertisement innce with 18 U.S.C. Section 1734 solely to indicate this fact.

ived 01/21/2010; revised 09/08/2010; accepted 09/20/2010; publishedirst 09/24/2010.

Cancer Research



Refe1. Ari

relares

2. FuCliin

3. Suma200

4. CasioJ P

5. SctionorpCh

6. Gativapa

7. MoPGis a657

8. Sctormi647

9. VillTre

10. Gigest

11. Bome20

12. Dedirter61

13. StecritCa

14. Mc20

15. Katra(ERwitCh

16. Greviest

17. Hurecnaland

18. KalighigSc

19. Yogen413

20. Linrat200

21. GasisMo

22. No


www.a


Alutra

23. WaanMe

24. Deinvβ-c87

25. Bitinh

26. ChVacaEM

27. Ishbinan49

28. Ouingint

29. Ulrcetio

30. WiWNfriz

31. SuingCa

32. Rhbyar40

33. LaGiinpeof

34. Munuyo

35. Liuproaddead

36. Tatiomi

37. Boan20

38. Lodis

39. Veme5:3

40. CoBecaMA

41. Kecapro

rencesazi EA,ClarkGM,Mertz JE. Estrogen-related receptorαand estrogen-ted receptor γ associate with unfavorable and favorable biomarkers,pectively, in human breast cancer. Cancer Res 2002;62:6510–8.jimoto J, Alam SM, Jahan I, Sato E, Sakaguchi H, Tamaya T.nical implication of estrogen-related receptor (ERR) expressionovarian cancers. J Steroid Biochem Mol Biol 2007;104:301–4.zuki T, Mikki Y, Moriya T, et al. Estrogen-related receptor α in hu-n breast carcinoma as a potent prognostic factor. Cancer Res4;64:4670–6.rtoni R, Leger B, Hock MB, et al. Mitofusins 1/2 and ERRα expres-n are increased in human skeletal muscle after physical exercise.hysiol 2005;567:349–58.hreiber SN, Knutti D, Brogli K, Uhlmann T, Kralli A. The transcrip-al coactivator PGC-1 regulates the expression and activity of thehan nuclear receptor estrogen-related receptor α (ERRα). J Biolem 2003;278:9013–8.illard S, Grasfeder LL, Haeffele CL, et al. Receptor-selective coac-tors as tools to define the biology of specific receptor-coactivatorirs. Mol Cell 2006;24:797–803.otha VK, Handschin C, Arlow D, et al. Errα and Gabpa/b specifyC-1α-dependent oxidative phosphorylation gene expression thatltered in diabetic muscle. Proc Natl Acad Sci U S A 2004;101:0–5.hreiber SN, Emter R, Hock MB, et al. The estrogen-related recep-α (ERRα) functions in PPARγ coactivator 1α (PGC-1α)-inducedtochondrial biogenesis. Proc Natl Acad Sci U S A 2004;101:2–7.ena JA, Kralli A. ERRα: a metabolic function for the oldest orphan.nds Endocrinol Metab 2008;19:269–76.uere V. Transcriptional control of energy homeostasis by therogen-related receptors. Endocr Rev 2008;29:677–96.nnelye E, Aubin JE. Estrogen receptor-related receptor α: adiator of estrogen response in bone. J Clin Endocrinol Metab05;90:3115–21.blois G, Hall JA, Perry MC, et al. Genome-wide identification ofect target genes implicates estrogen-related receptor α as a de-minant of breast cancer heterogeneity. Cancer Res 2009;69:49–57.in RA, Chang CY, Kazmin DA, et al. Estrogen-related receptor α isical for the growth of estrogen receptor-negative breast cancer.ncer Res 2008;68:8805–12.Ewan IJ. Nuclear receptors: one big family. Methods Mol Biol09;505:3–18.llen J, Schlaeppi JM,BitschF, et al. Evidence for ligand-independentnscriptional activation of the human estrogen-related receptor αRα): crystal structure of ERRα ligand binding domain in complexh peroxisome proliferator-activated receptor coactivator-1α. J Biolem 2004;279:49330–7.eschik H, Wurtz JM, Sanglier S, et al. Structural and functionaldence for ligand-independent transcriptional activation by therogen-related receptor 3. Mol Cell 2002;9:303–13.ss JM, Torra IP, Staels B, Giguere V, Kelly DP. Estrogen-relatedeptor α directs peroxisome proliferator-activated receptor α sig-ing in the transcriptional control of energy metabolism in cardiacskeletal muscle. Mol Cell Biol 2004;24:9079–91.

mei Y, Ohizumi H, Fujitani Y, et al. PPARγ coactivator 1β/ERRand 1 is an ERR protein ligand, whose expression induces ah-energy expenditure and antagonizes obesity. Proc Natl Acadi U S A 2003;100:12378–83.on JC, Puigserver P, Chen G, et al. Control of hepatic gluconeo-esis through the transcriptional coactivator PGC-1. Nature 2001;:131–8.J, Yang R, Tarr PT, et al. Hyperlipidemic effects of dietary satu-ed fats mediated through PGC-1β coactivation of SREBP. Cell5;120:261–73.illard S, Dwyer MA, McDonnell DP. Definition of the molecular ba-
for estrogen receptor-related receptor-α-cofactor interactions.l Endocrinol 2007;21:62–76.rris J, Fan D, Aleman C, et al. Identification of a new subclass of
42. Bo{β}Nu

acrjournals.org


DNA repeats which can function as estrogen receptor-dependentnscriptional enhancers. J Biol Chem 1995;270:22777–82.ki H, Park KW, Mitro N, et al. The small molecule harmine is antidiabetic cell-type-specific regulator of PPARγ expression. Celltab 2007;5:357–70.mir R, Dimmler A, Naschberger E, et al. Malignant progression ofasive tumour cells seen in hypoxia present an accumulation ofatenin in the nucleus at the tumour front. Exp Mol Pathol 2009;:109–16.ler BG, Menzl I, Huerta CL, et al. Intracellular MUC1 peptidesibit cancer progression. Clin Cancer Res 2009;15:100–9.amorro MN, Schwartz DR, Vonica A, Brivanlou AH, Cho KR,rmus HE. FGF-20 and DKK1 are transcriptional targets of β-tenin and FGF-20 is implicated in cancer and development.BO J 2005;24:73–84.ii M, Han J, Yen HY, Sucov HM, Chai Y, Maxson RE, Jr. Com-ed deficiencies of Msx1 and Msx2 cause impaired patterningd survival of the cranial neural crest. Development 2005;132:37–50.ko L, Ziegler TR, Gu LH, Eisenberg LM, Yang VW. WNT11 signal-promotes proliferation, transformation, and migration of IEC6

estinal epithelial cells. J Biol Chem 2004;279:26707–15.ich F, Concha M, Heid PJ, et al. Slb/WNT11 controls hypoblastll migration and morphogenesis at the onset of zebrafish gastrula-n. Development 2003;130:5375–84.tzel S, Zimyanin V, Carreira-Barbosa F, Tada M, Heisenberg CP.T11 controls cell contact persistence by local accumulation ofzled 7 at the plasma membrane. J Cell Biol 2006;175:791–802.yama K, Shapiro I, GuttmanM, Hazan RB. A signaling pathway lead-to metastasis is controlled by N-cadherin and the FGF receptor.ncer Cell 2002;2:301–14.ee J, InoueY, Yoon JC, et al. Regulation of hepatic fasting responsePPARγ coactivator-1α (PGC-1): requirement for hepatocyte nucle-factor 4α in gluconeogenesis. Proc Natl Acad Sci U S A 2003;100:12–7.ganière J, Tremblay GB, Dufour CR, Giroux S, Rousseau F,guère V. A polymorphic autoregulatory hormone response elementthe human estrogen-related receptor α (ERRα) promoter dictatesroxisome proliferator-activated receptor γ coactivator-1α controlERRα expression. J Biol Chem 2004;279:18504–10.lholland DJ, Dedhar S, Coetzee GA, Nelson CC. Interaction ofclear receptors with the Wnt/β-catenin/Tcf signaling axis: Wntu like to know? Endocr Rev 2005;26:898–915.J, Farmer SR. Regulating the balance between peroxisomeliferator-activated receptor γ and β-catenin signaling duringipogenesis. A glycogen synthase kinase 3β phosphorylation-fective mutant of β-catenin inhibits expression of a subset ofipogenic genes. J Biol Chem 2004;279:45020–7.plin ME, Rajeshkumar B, Halabi S, et al. Androgen receptor muta-ns in androgen-independent prostate cancer: Cancer and Leuke-a Group B Study 9663. J Clin Oncol 2003;21:2673–8.trugno OA, Fayard E, Annicotte JS, et al. Synergy between LRH-1d β-catenin induces G1 cyclin-mediated cell proliferation. Mol Cell04;15:499–509.gan CY, Nusse R. The Wnt signaling pathway in development andease. Annu Rev Cell Dev Biol 2004;20:781–810.eman MT, Axelrod JD, Moon RT. A second canon. Functions andchanisms of β-catenin-independent Wnt signaling. Dev Cell 2003;67–77.nacci-Sorrell M, Simcha I, Ben-Yedidia T, Blechman J, Savagner P,n-Ze'ev A. Autoregulation of E-cadherin expression by cadherin-dherin interactions: the roles of β-catenin signaling, Slug, andPK. J Cell Biol 2003;163:847–57.nny PA, Lee GY, Myers CA, et al. The morphologies of breastncer cell lines in three-dimensional assays correlate with theirfiles of gene expression. Mol Oncol 2007;1:84–96.
ttomly D, Kyler SL, McWeeney SK, Yochum GS. Identification of-catenin binding regions in colon cancer cells using ChIP-Seq.cleic Acids Res 2010;38:5735–45.



43. Hetist199

44. Slarelacha540

45. Uynetat

46. Kiriza

47. Zhprorec

48. Chseep

49. Szthe20

Dwyer et al.

Cance9308


rtz GZ, Stormo GD. Identifying DNA and protein patterns with sta-ically significant alignments of multiple sequences. Bioinformatics9;15:563–77.dek R, Bader JA, Giguere V. The orphan nuclear receptor estrogen-ted receptor α is a transcriptional regulator of the human medium-in acyl coenzyme A dehydrogenase gene. Mol Cell Biol 1997;17:0–9.sal-Onganer P, Kawano Y, Caro M, et al. Wnt-11 promotesuroendocrine-like differentiation, survival and migration of pros-
e cancer cells. Mol Cancer 2010;9:55.ikoshi H, Sekihara H, Katoh M. Molecular cloning and character-tion of human WNT11. Int J Mol Med 2001;8:651–6.
50. Tanoem

r Res; 70(22) November 15, 2010


u H, Mazor M, Kawano Y, et al. Analysis of Wnt gene expression instate cancer: mutual inhibition by WNT11 and the androgeneptor. Cancer Res 2004;64:7918–26.ristiansen JH, Monkley SJ, Wainwright BJ. Murine WNT11 is acreted glycoprotein that morphologically transforms mammaryithelial cells. Oncogene 1996;12:2705–11.alay J, Bruno P, Bhati R, et al. Associations of PKC isoforms withcytoskeleton of B16F10 melanoma cells. J Histochem Cytochem

01;49:49–66.o Q, Yokota C, Puck H, et al. Maternal WNT11 activates the ca-
nical wnt signaling pathway required for axis formation in Xenopusbryos. Cell 2005;120:857–71.
Cancer Research



2010;70:9298-9308. Published OnlineFirst September 24, 2010.Cancer Res Mary A. Dwyer, James D. Joseph, Hilary E. Wade, et al. Cancer Cell Migration

-Catenin and Acts in an Autocrine Manner to Increaseβand αWNT11 Expression Is Induced by Estrogen-Related Receptor

Updated version

10.1158/0008-5472.CAN-10-0226doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2010/09/24/0008-5472.CAN-10-0226.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/70/22/9298.full#ref-list-1

This article cites 50 articles, 24 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/70/22/9298.full#related-urls

This article has been cited by 11 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] at

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

Permissions

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

.http://cancerres.aacrjournals.org/content/70/22/9298To request permission to re-use all or part of this article, use this link

Research. on January 4, 2020. © 2010 American Association for Cancercancerres.aacrjournals.org Downloaded from


http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-10-0226

http://cancerres.aacrjournals.org/content/suppl/2010/09/24/0008-5472.CAN-10-0226.DC1

http://cancerres.aacrjournals.org/content/70/22/9298.full#ref-list-1

http://cancerres.aacrjournals.org/content/70/22/9298.full#related-urls

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

mailto:[email protected]

http://cancerres.aacrjournals.org/content/70/22/9298


Cancer Research WNT11ExpressionIsInducedbyEstrogen...

Documents

Transcript of Cancer Research WNT11ExpressionIsInducedbyEstrogen...