anewroleforera:silencingviadnamethylation of basal, stem ...presence of eradrives the luminal...

Chromatin, Epigenetics, and RNA Regulation

ANewRole forERa: SilencingviaDNAMethylationof Basal, Stem Cell, and EMT GenesEric A. Ariazi1, John C. Taylor1, Michael A. Black2, Emmanuelle Nicolas1, Michael J. Slifker1,Diana J. Azzam3, and Jeff Boyd1,3

Abstract

Resistance to hormonal therapies is a major clinical problemin the treatment of estrogen receptor a–positive (ERaþ) breastcancers. Epigenetic marks, namely DNA methylation of cyto-sine at specific CpG sites (5mCpG), are frequently associatedwith ERaþ status in human breast cancers. Therefore, ERa mayregulate gene expression in part via DNA methylation. Thishypothesis was evaluated using a panel of breast cancer cell linemodels of antiestrogen resistance. Microarray gene expressionprofiling was used to identify genes normally silenced in ERaþ

cells but derepressed upon exposure to the demethylating agentdecitabine, derepressed upon long-term loss of ERa expression,and resuppressed by gain of ERa activity/expression. ERa-dependent DNA methylation targets (n ¼ 39) were enrichedfor ERa-binding sites, basal-up/luminal-down markers, cancer

stem cell, epithelial–mesenchymal transition, and inflamma-tory and tumor suppressor genes. Kaplan–Meier survival curveand Cox proportional hazards regression analyses indicatedthat these targets predicted poor distant metastasis–free surviv-al among a large cohort of breast cancer patients. The basalbreast cancer subtype markers LCN2 and IFI27 showed thegreatest inverse relationship with ERa expression/activity andcontain ERa-binding sites. Thus, genes that are methylated inan ERa-dependent manner may serve as predictive biomarkersin breast cancer.

Implications: ERa directs DNA methylation–mediated silencingof specific genes that have biomarker potential in breast cancersubtypes. Mol Cancer Res; 15(2); 152–64. �2016 AACR.

IntroductionEstrogen receptor a (ERa, ESR1) has proven to be the single

most important target in breast cancer. Approximately 70% to80% of breast cancers are ERaþ, for which routine testing is usedto predict response to antihormonal therapy (1). As demonstratedby genome-wide studies, ERa is a global regulator of gene tran-scription in breast cancer that orchestrates well-integrated hor-monal responses that promote proliferation and survival andinhibit apoptosis (2–5).

As a result of regulating expression of thousands of genes, thepresence of ERa drives the luminal classification of breast cancer.There are five intrinsic tumor subtypes, luminal A, luminal B,HER2-enriched, claudin-low, and basal-like, as well as a normalbreast-like group. Patients with either luminal B, HER2-enriched,basal-like, or claudin-low tumors experience worse clinical out-come than patients with luminal A tumors (6–8).

ERahas been shown to negatively regulate gene expression, butnot much is currently known on how it can achieve this. Epige-netic marks, namely DNAmethylation of cytosine at specific CpGsites (5mCpG), are frequently associated with ERaþ status inhuman breast cancers. ERa may play a role in directing DNAmethylation to target genes, as specific 5mCpG marks associatewith ERa status in human breast cancer and predict risk of tumorrecurrence (9–12).

Methylation of cytosine at CpG dinucleotide sites (5mCpG) byDNA methyltransferases (DNMT) in transcriptional regulatoryregions mediates stable epigenetic gene silencing. In cancer cells,DNA methylation is highly correlated with repressive chromatinmarks, such as trimethylated H3K27 (H3K27me3; ref. 13).H3K27 trimethylation is catalyzed by EZH2, the histone methyl-transferase enzymatic subunit of the polycomb repressor complex2 (PRC2; ref. 14). Together, EZH2 and PRC2 then recruit DNMTs(13, 15). Methylated CpG sites near transcriptional start sites(TSS) can silence gene expression by interacting with effectors,such as methyl-CpG–binding domain proteins, that impedebinding of transcription factors, block transcriptional initiation,and recruit histone deacetylases (HDAC) to promote chromatincompaction (16).

The relationship between ERa and DNA methylation pattern-ing in breast cancer has been reported. In a comprehensivebioinformatics study,methylation of CpG sites near ERa-bindingregions tended to be lower in ERaþ tumors than ERa� tumors.This indicated a passive role for ERa in preventing gene silencing.The methylation status of DNA sequences at ERa-binding sites istightly coupled with ERa activity (12). Differentially methylatedgenes have also been identified in antihormone-resistant versuswild-type MCF-7 cells (17, 18), and in ERa RNAi-depleted versusnondepletedMCF-7 cells (19). Consistentwith this notion, loss of

1Fox Chase Cancer Center, Temple University Health System, Philadelphia,Pennsylvania. 2Department of Biochemistry, University of Otago, Dunedin, NewZealand. 3Department of Human and Molecular Genetics, Herbert WertheimCollege of Medicine, Florida International University, Miami, Florida.

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

Corresponding Authors: Jeff Boyd, Department of Human and MolecularGenetics, Florida International University HerbertWertheimCollegeofMedicine,11200 SW 8th Street, AHC2-693, Miami, FL 33199. Phone: 305-348-0646; Fax:305-348-0651; E-mail: [email protected]; and Eric A. Ariazi, Freenome, Inc., 201Gateway Blvd., South San Francisco, CA 94080. Phone: 650-446-6630; E-mail:[email protected]

doi: 10.1158/1541-7786.MCR-16-0283

�2016 American Association for Cancer Research.

MolecularCancerResearch

Mol Cancer Res; 15(2) February 2017152

on February 11, 2021. © 2017 American Association for Cancer Research. mcr.aacrjournals.org Downloaded from

Published OnlineFirst November 15, 2016; DOI: 10.1158/1541-7786.MCR-16-0283

http://mcr.aacrjournals.org/

ERa activity leads to silencing of estrogen-responsive genes, suchas PgR (18, 19). Yet, ERamayalso play an active role in promotingsilencing. A functional link between ERa and DNA hypermethy-lation has been demonstrated at the CYP1A1 locus, whose geneproduct converts 17b-estradiol (E2) into ametabolite that inhibitsproliferation; ERa silenced CYP1A1 by recruiting DNMT3B (20).

We sought to identify ERa targets for CpG methylation–medi-ated silencingby selecting the intersectionof (i) genes upregulated(i.e., derepressed) by the demethylating agent decitabine (DAC);(ii) genes upregulated by loss of ERa expression in a series ofantihormone-resistant T47D and MCF7 cell lines; and (iii) genesdownregulated by E2 reexposure or increased ERa expression inantihormone-resistant T47D and MCF7 cells. Additional experi-ments verified the functional dependence on ERa for silencingand DNAmethylation of the basal breast cancer subtype markersLCN2 and IFI27 in wild-type and antihormone-resistant T47D-based cell lines. Therefore, we show that ERa targets genes forDNA methylation–mediated silencing that may potentially bepredictive biomarkers of breast cancer subtypes.

Materials and MethodsCell lines

Sources and culture conditions of cell lines generated in thisstudy are provided in Supplementary Materials and Methods. Aschema representing the derivation of antihormone-resistant cellsis shown in Supplementary Fig. S1. The fulvestrant (FUL)–resis-tant cell lines (T47D/FUL, MCF7/FUL) and the estrogen depriva-tion (ED)–resistant cell lines (T47D/ED1, T47D/ED2) were gen-erated by continuous culture (8 weeks to >1 year) of wild-typeT47D and MCF-7 cells in estrogenized media (RPMI1640 plus10%whole FBS) supplemented with 100 nmol/L fulvestrant or inestrogen-free media (phenol red-free RPMI1640 plus 10% dex-tran-coated charcoal-stripped FBS), as appropriate. Antihormone-resistant cells were maintained as polyclonal populations. All celllines were authenticated by gene expression microarrays, mor-phology, andby verifying ERa, PgR,HER2, LCN2, and IFI27 levelsand cell line growth responses to estrogen, ED, and fulvestrant.

The lentiviral cell lines, T47D/ED1/VC, T47D/ED1/VCþE2,T47D/ED1/ERa, and T47D/ED1/ERaþE2 were generated byinfecting ERa� T47D/ED1 cells with an ERa-expressing lentivirusor an empty vector control (VC) lentivirus, as appropriate.Infected cells were maintained in estrogen-free or in 1 nmol/LE2–supplemented medium for 12 weeks. After initial recoveryfrom infection and again 4 weeks later, infected cells were sortedfor the lentiviral ZsGreen fluorescent marker using a BectonDickinson FACS-VantageSE/DiVa cell sorter. To produce the len-tiviral vectors, ERa's coding region was excised from pHEGOusing EcoRI and inserted into the EcoRI site of the lentiviral vectorpLVX-EF1a-IRES-ZsGreen1 (Clontech Laboratories).

RNA isolationRNA was purified using Qiagen's RNeasy Plus Kits. RNA sam-

ples were required to exhibit an RNA integrity number of 9.8 to10.0 on an Agilent 2100 Bioanalyzer.

qRT-PCR assaysqRT-PCR assays were carried out as described previously (21)

but using AMV First-Strand cDNA Kit, predesigned TaqManassays, TaqMan Universal PCR Master Mix, and a 7900HT FastReal-Time PCR system (Thermo Fisher Scientific). Data were

analyzed by comparison with a serial dilution series of cDNA.All qPCR data represent the mean and SDs of three independentbiological replicates and two technical replicates per biologicalreplicate.

Agilent gene expression microarraysGenome-wide RNA profiling was carried out by the Genomics

Facility at Fox Chase Cancer Center (Philadelphia, PA) usingAgilent's Human Gene Expression 4 � 44 K v2 oligonucleotidemicroarrays. RNA labeling (one-color cyanine 3-CTP), hybridiza-tion to the arrays, and quality assessment of hybridizations wereperformed according to the manufacturer's instructions.

Immunoblot analysesImmunoblotswere done as described previously (21)but using

RIPA buffer and 40 mg protein per lane. Antibodies used are listedin Supplementary Materials and Methods. Blots were visualizedusing the Odyssey Infrared Imaging System (Li-Cor Biosciences).

DNA methylation analysis by pyrosequencingGenomicDNAwas isolated using theDNeasy Blood and Tissue

Kit (Qiagen) and treated with bisulfite [EpiTect Bisulfite Conver-sion Kit (Qiagen)] to change unmethylated cytosine nucleotidesto thymines. Pyrosequencing reactions were carried out at Epi-genDx as a service using their predesigned assays. In pyrograms,the ratios of methylated cytosines to thymines (which representunmethylated cytosines) are internally normalized values. Allpyrosequencing data represent the mean and SD of 4 replicates.

Human breast cancer cohortsBreast cancer data from The Cancer Genome Atlas (TCGA)

project were downloaded via the International Cancer GenomeConsortium (ICGC) data portal (https://dcc.icgc.org/releases/release_18/Projects/BRCA-US). Methylation data were retrievedfor 1,013 patients, 967 of whom also had ERa status available.CpG differential methylation by ERa status was assessed asdescribed in Supplementary Materials and Methods.

The gene expressionmicroarray cohort comprising 2,116 breastcancers was previously constructed in ref. 22 from publiclyavailable datasets and is described in Supplementary Materialsand Methods. Breast cancer–intrinsic subtypes were previouslyclassified as described in ref. 22.

MetagenesTo analyze the composite expression level of gene sets in a

tumor, gene sets were represented as metagenes and metagenescores, or single number summary values were determined acrossthe expression array breast cancer cohort. These scores represent alinear combination of expression values of each gene in the geneset in individual tumors. Metagene scores were generated bydetermining the first "principle component" or "eigenvector" ofeach gene set in each tumor using singular value decomposition(SVD). The eigenvector produced by SVD was rescaled to a rank-based score between 0 and 1, with 0 relating to the lowestcomposite expression value for a gene set, and 1 relating to thehighest. Thus,metagene scores capture themajority of variation ingene expression that is common to themajority of genes in a geneset across a population of samples.

To construct the ERaDNAmethylation metagene, Entrez geneidentifiers were used to match the Agilent probes from theexpression microarrays used in this study to the Affymetrix probe

ERa-Dependent Gene Silencing via DNA Methylation

www.aacrjournals.org Mol Cancer Res; 15(2) February 2017 153




sets used in the combined breast cancer cohort. This resulted inmatching 34 of 39 ERa DNA methylation genes (SupplementaryTable S1). TheERa status–associatedmetagene consists of the 100most differentially expressed genes between ERþ and ERa�

tumors in the 2,116 breast cancer cohort as determined usingthe "limma" package (23) for R software (www.r-project.org). Thespecific genes comprising each metagene are provided in Supple-mentary Excel File S1.

Accession numbersMicroarray data are deposited in theNCBIGEO repository with

accession number GSE85536.

Statistical analysesExpression array data were log2 transformed for all compar-

isons. Differentially expressed genes were identified by serialpairwise comparisons using SAM (24) at an FDR <5% and a 2-fold cutoff, except a 1.5-fold cutoff was used when comparingT47D/ED2/E2 versus T47D/ED2 cells because the ERa levels inthese cells were <5% that of wild-type T47D cells (Fig. 1A). Geneenrichment in Supplementary Excel Files S4, S5, and S10 wasassessed by one-way Fisher exact tests using the R softwareapplication. Associations between ERa DNA methylation meta-gene scores and distant metastasis–free patient survival (DMFS)were evaluated by Kaplan–Meier analysis and log-rank tests, andby univariate andmultivariable Cox proportional hazards regres-sion models as described previously (22). The additional covari-ates used were age at diagnosis, intrinsic subtype, ERa status,tumor size, and tumor grade. Statistical tests used in Figs. 5 and 6are specified in thefigure legends andwere carried out using Prismv4.03 (GraphPad Software). Where specified, one-way tests wereemployed because gene expression and CpG methylation wasassumed a priori to be inversely related.

ResultsIdentification of genes inversely correlated with ERaexpression/activity

To identify ERa targets for DNA methylation–mediatedsilencing, we sought to find the intersection of genes thatfulfilled three conditions: (i) those genes derepressed by lossof ERa expression; (ii) those genes resilenced by increased ERaactivity or expression; and (iii) those genes derepressed by lossof DNA methylation.

To begin, breast cancer cell line models were developed thatexhibited loss of ERa to enable subsequent identificationof genes,which inversely correlated with ERa expression/activity. Weelected not to use RNAi-based methods, as acute depletion ofERa in estrogen-dependent cells leads to widespread cell death(unpublished observation). Therefore, starting with wild-typeERaþ T47D and MCF-7 luminal breast cancer cells, a panel ofERa-low/negative T47D andMCF7breast cancer cells was derivedby long-term selection of cells in 100 nmol/L fulvestrant or inestrogen-free media for 8 weeks to greater than 1 year (schema inSupplementary Fig. S1). Thus fulvestrant-resistant (T47D/FUL,MCF7/FUL) and ED-resistant (T47D/ED1, T47D/ED2) cell lineswere derived. ERamRNA levels were measured by qRT-PCR (Fig.1A). T47D/ED1 cells lost 99.9%, T47D/FUL and T47D/ED2 cellslost �95%, and MCF7/FUL cells (at week 8 of derivation) lost90% of ERa mRNA compared with respective wild-type parentalcells. Immunoblotting also demonstrated similar ERa proteinlosses (Fig. 4).

To determine global changes in gene expression, whichcorrelated with loss of ERa expression, transcriptional profilingwas performed using Agilent 4 � 44 K v2 oligonucleotidemicroarrays. ERa-low/negative cell lines, T47D/FUL, ED1, ED2,and MCF7/FUL (week 8), were compared against their respec-tive wild-type parental T47D or MCF-7 cell line (four separatepairwise comparisons). Supplementary Excel Files S2 and S3list the 324 and 153 significantly up- and downregulated genes,respectively. These differentially regulated genes were examinedfor enrichment of functional gene groups consistent withacquired antihormone resistance using one-way Fisher exacttests (tables and P values in Supplementary Excel Files S4 andS5; Supplementary Fig. S2). As expected, genes with ERa-binding sites were overrepresented. ERE-regulated genes weretaken from previously published datasets (see SupplementaryFig. S3 and its legend for a list of genes and references).Importantly, basal markers were very significantly enrichedamong the upregulated genes, whereas luminal markers werevery significantly enriched among the downregulated genes.Also observed was enrichment of cancer stem cell (CSC),epithelial–mesenchymal transition (EMT), and tumor suppres-sor genes (TSG; see the legend of Supplementary Fig. S3 forreferences used). Examples of key genes in these functionalgroups are shown in Supplementary Fig. S3. This indicates thatthe antihormone-resistant ERa-low/negative T47D and MCF-7cells transitioned to a differentiation state similar to the basal-like and claudin-low breast cancer subtype. Such a change indifferentiation has previously been observed in T47D tumors invivo following antiestrogen treatment or estrogen withdrawaland termed "luminobasal" (25).

To further refine the list of ERa inversely correlated genes,T47D/ED2 cells were reexposed to E2 for 38 weeks, resulting inT47D/ED2/E2 cells. Interestingly, ERa RNA (Fig. 1) and proteinlevels (Fig. 4) never rebounded, indicating permanent ERa silenc-ing as observed elsewhere (26). In fact, ERa RNA levels actuallydecreased �50% more; this likely reflected a known E2–ERanegative feedback regulatory loop indicative of ERa transcrip-tional activity (27). MCF/FUL cells were also further selected.These cells at week 8 of derivation showed 90% loss of ERa, butafter 13 weeks of additional exposure to fulvestrant (total 21weeks), ERa levels rebounded to wild-type cell levels (Figs. 1Aand 4E). Transcriptional profiling showed increased expression ofwell-known E2-stimulated genes in T47D/ED2/E2 versus T47D/ED2 cells (e.g., PGR, CA12, ERBB4) and in MCF7/FUL week 21versus week 8 cells (e.g., CXCL12, GREB1, ERBB4), as well asdecreased expression of E2-repressed genes (e.g., OASL, C3; bothcell lines). Furthermore, the expression pattern of many (but notall) basal and luminal, CSC, EMT, and TSGs reversed upon E2reexposure in T47D/ED2/E2 or increased ERa expression inMCF7/FUL (week 21) cells compared with respective parentalcells (Supplementary Fig. S3).

Taking into account all cell line transcriptional profiles, 161genes were identified that consistently inversely related with ERaexpression/activity, whereas only 9 genes were directly related(Supplementary Excel Files S6 and S7, respectively).

Candidate ERa DNA methylation targetsAs ERa inversely related genes whose expression was regu-

lated by DNA methylation were sought, genes upregulated bythe DNA demethylating agent DAC were identified. Wild-typeT47D cells were treated with 1 mmol/L DAC or control (CON)-

Ariazi et al.

Mol Cancer Res; 15(2) February 2017 Molecular Cancer Research154




A

B

Figure 1.

Candidate ERa targets for DNA methylation and ERa mRNA levels in the cell lines used to identify the targets. A, ERa mRNA levels in matched wild-type (wt)fulvestrant (FUL)-resistant, ED-resistant, and ED-resistant reexposed to E2 (ED/E2) cell line models at the indicated weeks (wk) of derivation. The selectionprocess schema is shown in Supplementary Fig. S1. ERa mRNA levels normalized to TBP mRNA were measured by qRT-PCR. B, The 39 candidate ERa DNAmethylation targets. Cell lines were transcriptionally profiled using Agilent Human Gene Expression 4 � 44 K v2 microarrays. Shown is the intersection ofDAC-regulated genes and genes whose expression consistently showed an inverse relationship to ERa expression/activity across all wild-type and antihormone-resistant cell lines. Genes are ranked by their average fold increase in expression in T47D/FUL, T47D/ED1, and T47D/ED2 versus wild-type T47D cells. Note,profiles of T47D/ED2/E2 week 38 and not week 24 cells were compared against T47D/ED2 cells for significantly differentially expressed genes. Basal-up/luminal-down genes were established according to references in Supplementary Excel File S10.






treated for 96 hours and then transcriptionally profiled. Thisresulted in the identification of 1,049 genes (SupplementaryExcel File S8).

Subsequently, the intersection of ERa inversely related genesand DAC-induced genes was determined. This intersection repre-sented the set of genes that fulfilled the following criteria: (i) genesinduced by DAC versus CON-treated wild-type T47D cells; (ii)genes upregulated in each of the ERa-low/negative cell lines, that

is, T47D/FUL, T47D/ED1, and T47D/ED2 cells, versus wild-typeT47D cells; (iii) genes downregulated by E2 in T47D/ED2/E2versus T47D/ED2 cells; (iv) genes upregulated in ERa-lowMCF7/FUL week 8 versus wild-type MCF-7 cells; and (v) genes down-regulated in ERaþ MCF-7/FUL week 21 versus ERa-low MCF-7/FULweek8 cells. These selection criteria pinpointed 39high-valuecandidates for ERa-mediated silencing viaDNAmethylation (Fig.1B; Supplementary Excel File S9).

(1,904) (1,057)(1,059) (1,905)

A B

D

F

E

C

Figure 2.

Evaluation of the ERa DNA methylation target metagene in a 2,116 sample breast cancer cohort. A–E, ERa status and ERa status–associated metagene scores(A), intrinsic subtype (B), luminobasal metagene scores (C), CD44þ/CD24�/low metagene scores (D), and EMT metagene scores (E) were plotted againstERa DNA methylation target metagene scores in the 2,116 sample breast cancer cohort. Metagene scores were used to divide the breast cancer samples into low,medium, and high tertiles. F, Kaplan–Meier survival curves of ERa DNA methylation metagene scores versus DMFS. Patients were grouped on the basis ofmetagene scores split at the 10th, 50th, and 90th percentile. Significance was assessed by log-rank tests. The number of patients in each group is shown in thelegends; numbers of patients are less than the total cohort size of 2,116 due to missing survival data.

Ariazi et al.





Initially, these 39 candidate genes were evaluated for methyl-ation in human breast cancer (Supplementary Table S1). UsingTCGA-processed breast cancer methylation data, a set of 1,996CpGs associated with these genes was identified. These CpG siteswere assessed for differential methylation between ERaþ andERa� breast cancers using one-sided Wilcoxon rank sum testsadjusted for FDR. Using a permutation analysis to determinewhether similar results could be achieved using 1,000 sets of 39random genes, it was concluded that the candidate ERa DNAmethylation targets tended to display highermethylation levels inERaþ comparedwith ERa� tumors thanwould be expected for anidentically sized set of randomly selected genes (permutation-based P ¼ 0.011).

Next, the 39 candidate ERa DNA methylation targets wereanalyzed for enrichment of the same gene groups as the ERainversely related genes (tables and P values in SupplementaryExcel File S10; Supplementary Fig. S2). Similar to the earlierresults, the candidate methylation targets were enriched for geneswith ERa-binding sites, basal markers, CSC-upregulated genes,EMT-upregulated genes, and TSGs. The candidate methylationtargets were also enriched for EMT-downregulated genes, butthere were almost twice as many EMT-upregulated genes thandownregulated (13 vs. 7, respectively).

Expression analysis of the candidate ERa DNA methylationgene set in breast cancer

The candidate ERa DNA methylation target gene set wasanalyzed relative to other important tumor-related gene setsand clinical variables in a cohort of 2,116 breast cancers. Genesets were represented as a composite entity termed an "expres-sion metagene" and a single value summary of the gene set'sexpression level in an individual tumor as a "metagene score."To enable evaluation of the distribution of gene set expressionlevels, metagene scores were used to divide the breast cancers inthe cohort into "tertiles" (lowest 33%, middle 33%, highest33%).

ERa DNA methylation metagene scores were plotted versusERa status, an ERa status–associated metagene, breast cancerintrinsic subtypes, luminobasal signature metagenes, EMT meta-genes and CD44þ/CD24�/low CSC metagenes (Fig. 2). The ERastatus–associated metagene encapsulated the 100 most differen-tially expressed probe sets between ERa-positive and -negativetumors in the 2,116 breast cancer cohort. Congruent with theoriginal selection criteria, the ERa DNA methylation metageneshowed a clear negative associationwith ERa status and 100 otherERa status–associated genes, with ERaþ tumors tending to havelower scores (and thus indicating lower levels of geneexpression; Fig. 2A). With regard to intrinsic subtype, luminalA and B subtypes displayed the lowest ERa DNA methylationmetagene scores, whereas the basal-like subtype exhibited thehighest scores (Fig. 2B). This was consistent with enrichment ofbasal-up/luminal-down genes as previously noted. Furthermore,the ERaDNAmethylationmetagene clearly directly related to theluminobasal signature metagenes (Fig. 2C), suggesting the ERaDNA methylation targets program this type of change in differ-entiation. Again as expected from the enrichment analysis, ERaDNA methylation metagene scores were associated with CD44þ/CD24�/low metagenes (Fig. 2D) and selectively with the EMT-upregulated metagene (Fig. 2E). This helps explain why the ERaDNA methylation metagene scores also associated with the clau-din-low breast cancer subtype (Fig. 2B).

The ERa DNA methylation metagene was next evaluated forpredicting DMFS in the breast cancer cohort by Kaplan–Meiersurvival curves (Fig. 2F) and Cox proportional hazards regressionmodels (Supplementary Table S4). In each analysis, patients wereseparated into two groups according to metagene scores split atthe 10th, 50th, or 90th percentiles, and then the proportion ofpatients exhibiting DMFS in each group was plotted against time.Patients with metagene scores in the bottom 10th or top 90thpercentile experienced significantly decreased DMFS. Likewise,univariate Cox proportional hazards regression analysis demon-strated that metagene scores split at the 10th and 90th percentilesassociatedwithDMFS (P¼ 0.00003 and 0.035, respectively). ERaDNA methylation metagene scores split at the 10th percentileremained significantly associated with DMFS in a multivariableCox proportional hazards regression model (P ¼ 0.026), but notwhen split at the 90th percentile. These results suggested thatsome genes in the ERa DNA methylation metagene whenexpressed at low levels promoted poor DMFS, whereas othersdid so when expressed at high levels.

To determine which of the genes of the candidate ERa DNAmethylation metagene when expressed at low or high levels maypromote poor DMFS, Kaplan–Meier survival curves and univar-iate Cox proportional hazards regression analysis was conductedfor each gene. On the basis of these analyses, the candidate ERamethylation targets were separated into low- and high-expressionmetagenes (defined in Supplementary Excel File S11). The low-andhigh-expressionmetagenes poorly correlatedwith each other,indicating they indeed likely represented different biologicalprocesses (Supplementary Table S5). Patients were then dividedaccording to their tumor's low- and high-expression metagenescores split at the 50th percentile and evaluated for DMFS asbefore (Fig. 3A). The Kaplan–Meier plots showed clear separa-tions of survival curves in which patients in the low expressionmetagene's bottom 50% group and patients in the high expres-sion metagene's top 50% group displayed poor DMFS.

Low- (Fig. 3B) and high-expression ERa DNA methylationmetagenes (Fig. 3C) were next assessed for associations withvarious tumor-related metagenes in the breast cancer cohort.These results indicated that the low-expression metagene associ-atedwith tumor suppressor and focal adhesion gene expression inbreast cancers. Accordingly, low levels of these types of geneswould be predicted to promote metastasis. Conversely, the high-expressionmetagene associatedwithhigh-grade tumors, aswell asproliferation and proinflammatory Th1 immune response geneexpression in breast cancers; this would also promote poorDMFS.Supplementary Excel File S11 contains references that help pro-vide a rationale for the segregation of genes into either the low- orhigh-expression metagenes.

Inverse relationship between LCN2 and IFI27 expression andERa

IFI27 and LCN2were the top two genes inversely related to ERaexpression/activity in the T47D-based cell lines (Fig. 1B). BothLCN2 (5) and IFI27 (2) contain ERa-binding sites. Also, both arebasal markers (7, 28, 29) and promote EMT (30, 31). Hence,LCN2 and IFI27 were selected for validation of ERa-dependentchanges in expression and 5mCpG levels.

LCN2 mRNA and protein levels dramatically increased in atime-dependent manner after precipitous drops in ERa levelsacross all antihormone-resistant models (Fig. 4A–C, and E).Furthermore, LCN2 expression decreased in a time-dependent






manner after extended E2 reexposure in T47D/ED2/E2 cells (Fig.4D) and once ERa expression rebounded in MCF/FUL cells (Fig.4E). LCN2 has previously been reported to upregulate the key

EMT transcription factor slug (SNAI2; ref. 30); therefore, slugexpression was examined. Across all antihormone-resistant mod-els, changes in slug expression followed similar changes in LCN2,

(1,057)(1,059)

(1,057)(1,059)

A

B

C

Figure 3.

Evaluation of the low- and high-expression ERa DNA methylation target metagenes in the combined breast cancer cohort. A, Kaplan–Meier survival curvesof the low- and high-expression ERa DNAmethylation metagenes. Genes were parsed into the low versus high metagene according to how each gene's expressionlevel tended to associate with poor DMFS. In the survival curves, patients were grouped on the basis of metagene scores split at the 50th percentile. Thelog-rank test P value and the number of patients in each group are shown in the legends.B andC, Low-expression (B) and high-expression (C) ERaDNAmethylationmetagene scores plotted against tumor grade as well as proliferation, tumor suppressor, focal adhesion, and Th1 immune response metagene tertiles inthe breast cancer cohort.

Ariazi et al.





although in MCF7/FUL cells, slug induction was delayed untilafter LCN2was upregulated from8 to12weeks and silenced againdue to ERa reexpression (Fig. 4B–E). Together, these results areconsistent with LCN2 regulating slug expression and demonstratethat LCN2 may have promoted EMT via slug.

Like LCN2, IFI27 mRNA expression was strikingly upregulated270- to 1,900-fold across the ERa-low/negative T47D-basedantihormone-resistant compared with wild-type cell lines andwas dramatically repressed again upon E2 reexposure in T47D/ED2/E2 versus T47D/ED2 cells (Fig. 4F).

Direct relationship between methylation of LCN2 and IFI27and ERa

Levels of selected 5mCpG sites near the TSSs of LCN2 and IFI27were quantitated across the T47D-based models by pyrosequen-cing of bisulfite-treated gDNA. This analysis found LCN2 andIFI27 CpG sites to be significantly hypomethylated in ERa-low/negative T47D/FUL, T47D/ED1, and T47D/ED2 cells versus wild-type ERaþT47Dcells (Fig. 5A), and significantly hypermethylatedupon E2 reexposure in T47D/ED2/E2 cells compared with paren-tal T47D/ED2 cells (Fig. 5B). Therefore, LCN2 and IFI27 CpGmethylation levels directly associated with ERa expression/activity.

Next, a causal relationship between ERa expression and CpGmethylation of LCN2 and IFI27was tested. ERa� T47D/ED1 cellswere infected with an ERa-expressing lentivirus or an empty VC

lentivirus generating T47D/ED1/ERa and T47D/ED1/VC cells,respectively. These infected cells were maintained with and with-out E2 for 12weeks and subjected to two rounds of cell sorting forthe lentiviral ZsGreen fluorescent marker. Characterization ofthese lentiviral cells lines demonstrated functional ERa signalingas ERa and PgR mRNAs were expressed at high levels in ERa-infected cells. Also, LCN2 and IFI27 mRNA levels were down-regulated in an ERa-dependent manner (plus E2 for IFI27; Fig.5C). Next, LCN2 and IFI27 CpG methylation levels were quan-titated by pyrosequencing and found to be significantly increasedin lentiviral ERa plus E2 (LCN2) or just ERa (IFI27) comparedwith VC cells. (Fig. 5D). Therefore, increased CpGmethylation ofLCN2 and IFI27was dependent onERaplus E2 stimulation. In thecase of IFI27, repression of its expression did not occur until itsCpGmethylation levels weremaximally increased by the presenceof E2, indicating that perhaps amethylation thresholdwas neededto cause its repression.

LCN2 and IFI27 expression and CpG methylation in breastcancer cell lines

LCN2 and IFI27 were examined in a panel of 11 breast cancercell lines. Initial characterization showed 4 cell lines were ERaþ

and 4wereHER2þ, where only BT-474 cells were positive for bothprognosticators (Fig. 6A). Then, LCN2 protein (Fig. 6B) and IFI27mRNA levels (Fig. 6C) were measured and found to be signifi-cantly lower in ERa-positive compared with ERa� cells. Next,

A

C D

E F

B

Figure 4.

LCN2 and IFI27 expression inverselyrelates to ERa expression/activity.A–E,Effect of long-term ED, fulvestrant(FUL), and E2 reexposure onERa, LCN2,and slug expression. In T47D-based celllines, both ED and fulvestrant led tosilencing of ERamRNA (A) and proteinlevels (B and C), followed by dramaticinduction of LCN2 mRNA (A) andprotein (B and C). In T47D/ED2/E2versus T47D/ED2 cells, long-term E2reexposure repressed LCN2 (D). InMCF7-based cell lines, fulvestrantexposure up to 12 weeks repressed ERaand induced LCN2 expression, but by 16weeks of fulvestrant, ERa reboundedand LCN2 was again silenced (E). Slugprotein levels increased or decreasedafter similar changeswere seen in LCN2expression (B–E). F, Effect of long-termED and fulvestrant on IFI27 expression.IFI27 RNA was silenced in ERaþ wild-type T47D cells, highly upregulated inERa-low/negative T47D/FUL andT47D/ED cells, and again silenced byreexposure to E2 in T47D/ED/E2 cells.RNA levels weremeasured by qRT-PCRand proteins levels by immunoblotting.






A

B

C

D

Figure 5.

LCN2 and IFI27 CpGmethylation levels are directly related to ERa expression/activity.A,Decreased LCN2 and IFI27 CpGmethylation in ERa-low/negative cell linescompared with wild-type T47D cells. B, Increased LCN2 and IFI27 CpG methylation in E2 reexposed T47D/ED2/E2 compared with T47D/ED2 cells. C, ERa,PgR, LCN2, and IFI27mRNA expression in lentiviral VC and ERa-infected cells. ERa and the ERa-responsive gene PgRwere substantially upregulated, whereas LCN2and IFI27 were downregulated in cells expressing lentiviral ERa and maintained in E2. RNA levels normalized to TBP were measured by qRT-PCR. D, IncreasedCpG methylation levels of LCN2 and IFI27 in lentiviral ERa compared with VC cells. A and C, Significance was assessed by repeated measures one-wayANOVA, followed by Dunnett multiple comparison tests for subgroup analysis. B, Significance was assessed by one-tailed paired t tests. Genomic DNAwas bisulfitetreated and methylation was quantitated by pyrosequencing. TSS, transcriptional start site.

Ariazi et al.





A C

B

D

E

Figure 6.

LCN2 and IFI27 expression inversely associates, whereas CpGmethylation directly associates with ERa status in breast cancer cell lines.A, Characterization of HER2and ERa protein expression. B and C, LCN2 protein (B) and IFI27 RNA (C) expression levels. For both LCN2 and IFI27, expression levels were scaled relativeto their median value (ZR751 cells for LCN2, and T47D cells for IFI27). LCN2 and IFI27 expression inversely associated with ERaþ status. For both genes, expressionvalues were log2 transformed because their variances were significantly different between ERaþ and ERa� cell lines (both P < 0.0001, F test). Significancewas assessed using two-tailed unpaired t tests. HER2, ERa, and LCN2 protein levels were measured by immunoblotting and IFI27 mRNA levels by qRT-PCR. D,Correlation between CpGmethylation and expression in breast cancer cell lines. Correlationswere determined using Spearman r coefficient and a one-tailedP value.E, LCN2 and IFI27 CpG methylation levels positively associated with ERa status. Only those CpG sites that showed a significant inverse correlation betweenmethylation and gene expression by Spearman r were evaluated for an association with ERa status. Significance was assessed considering all tested CpGsites together using two-tailed paired t tests in which CpG methylation levels were paired by site location. Individual CpG sites are presented to show pairings.The line in the ERa-positive and -negative subgroups represents the mean methylation value. Methylation levels were quantitated by pyrosequencing ofbisulfite-treated gDNA.






methylation levels of CpG sites near the TSSs of LCN2 and IFI27were quantitated by pyrosequencing of bisulfite-treated gDNA.Correlations between expression and 5mCpG levels were deter-mined by Spearman r (Fig. 6D). Methylation of all 5 of LCN2'stested CpG sites and 4 of IFI27's 9 tested sites (CpG sites þ404,þ438, þ508, and þ550) significantly correlated with eachrespective gene's expression levels. Finally, an associationbetween CpG methylation levels and ERa status was evaluated.Although all 5 of LCN2's CpGs were evaluated, only those 4CpGs of IFI27 that significantly correlated with expression wereconsidered. This analysis showed for both LCN2 and IFI27, that5mCpG levels were significantly higher in ERaþ than ERa� cells(Fig. 6E).

LCN2 and IFI27 expression as predictors of DMFS in humanbreast cancer

LCN2 and IFI27 RNA expression levels were examined in thebreast cancer cohort with respect to DMFS by Cox proportionalhazards regression analysis (Supplementary Table S6). In univar-iate models, LCN and IFI27 both significantly associated withDMFS (P¼ 0.040 and 0.0023, respectively), but this did not holdin multivariable models.

DiscussionWehypothesized that ERamay regulate gene expression in part

via DNA methylation as methylation of specific CpG sites associ-ateswith ERaþ status in humanbreast cancer. This hypothesiswastested by identifying genes normally silenced in ERaþ breastcancer cell lines but which were derepressed upon exposure to

the demethylating agentDAC, derepressed upon long-term loss ofERa expression, and resuppressed by gain of ERa activity/expres-sion. On the basis of these criteria, 39 candidate ERa DNAmethylation targets were found. These 39 targets were used toconstruct an ERa DNA methylation metagene that inverselyassociated with ERa status in human breast cancers and directlyassociated with expression signatures of basal-like and claudin-low breast cancer subtypes (25). Congruent with these associa-tions, the candidate ERa DNA methylation targets were enrichedfor basal markers, CSC, and EMT genes.

LCN2 and IFI27 were the top two ERa inversely related genesidentified and were selected for validation. Both LCN2 (5) andIFI27 contain ERa-binding sites (2), are basal markers (7, 28, 29),and involved in EMT (30, 31). First, LCN2 and IFI27 wereoriginally silenced in wild-type T47D and MCF-7 cells, but theirexpression dramatically increased upon loss of ERa, while their5mCpG levels significantly decreased in all antihormone-resistantT47D cell lines. Second, LCN2 and IFI27 were resilenced upon E2reexposure in T47D/ED2 cells while their 5mCpG levelsincreased. Third, lentiviral ERa plus E2 in T47D/ED1 cells alsorepressed LCN2 and IFI27 expression while increasing their5mCpG levels. Fourth, LCN2's and IFI27's 5mCpG levels posi-tively associated with ERa status but inversely correlated withexpression in a panel of 11 breast cancer cell lines. Together, theseresults provide correlative and functional evidence that ERadirected DNA methylation–mediated silencing of LCN2 andIFI27.

As ERa plays such a pivotal role in a more favorable outcomein breast cancer, genes targeted by ERa for DNA methylation–mediated silencing likely play important roles in disease pro-gression. In addition to the CSC and EMT genes, Kaplan–Meiersurvival curve analyses indicated that the candidate ERa DNAmethylation targets consisted of two classes of genes thatpredicted poor DMFS, one when expressed at low levels anda second when expressed at high levels. The low expressionclass associated with tumor suppressor and focal adhesion geneexpression in breast cancer. Conversely, the high-expressionclass associated with proliferation and inflammatory responsegene expression in breast cancer. In addition, the two validatedtargets for methylation, LCN2 and IFI27, predicted DMFS inunivariate Cox proportional hazards models. Thus, genes meth-ylated and silenced in an ERa-dependent manner may be goodtargets for therapeutic intervention in ERa� breast cancer wherethey are expressed.

How might ERa direct DNA methylation to specific genes? Wepropose it may begin with transcriptional repression. E2 actuallyrepresses transcription of more genes than it stimulates (3, 4).Studies on E2-dependent transcriptional repression have demon-strated that ERa recruits coregulators (corepressors; refs. 32–36)and coactivators that act as corepressors (36, 37). The coregulatorsserve as scaffolds to interact with HDACs and a host of additionalcofactors (32, 33, 38, 39), such as EZH2 (34, 40), that togetherremove activating histone marks, add repressive marks, andrestructure chromatin structure (38, 39, 41–43). We hypothesizethat not only does ERa direct epigenetic silencing via histonemodification, but also via cytosine methylation at CpG sites.

Genome-wide kinetics of DAC-induced DNA demethylationand subsequent remethylation after drug withdrawal in breastcancer cells showed that CpGs differ in both their susceptibility todemethylation and propensity for remethylation after drugremoval (44).

Figure 7.

Model of ERa-mediated silencing via DNA methylation. ERa initiallyrepresses transcription by recruiting corepressors (CoR) and HDACs that inturn recruit EZH2 to modify histones with repressive H3K27me3 marks. EZH2, acomponent of PRC2, then tethers DNMTs to catalyze C methylation in CpGdinucleotides to maintain long-term silencing. Gene silencing of basalmarkers, CSC, and EMT genes may in part program ERa breast cancers asthe luminal subtype. See Discussion for additional details.


Ariazi et al.




This is also plausible as EZH2 recruits DNMTs directly andindirectly through PRC2 (13, 15). Other protein–protein inter-actions exist as well that could support formation of a multi-component complex containing ERa and DNMTs, such as thosebetween ERa and EZH2 (45, 46), between HDACs and DNMTs(47–49), and between corepressors and DNMTs (40, 49).Evidence for such a complex exists at least at the CYP1A1promoter, where it was demonstrated that ERa and DNMT3Binteracted (20). Thus, it is possible that ERa could silencetargeted genes via DNA methylation by directly and indirectlyrecruiting corepressors, HDACs, EZH2 in PRC2, and DNMTs(model shown in Fig. 7).

Taken together, our data indicate that ERa can silence genes viaDNA methylation, such as LCN2 and IFI27. Moreover, ERa maydirect DNA methylation–mediated silencing of a subpopulationof basal markers, CSC, and EMT genes that may potentiallyenforce luminal differentiation of breast cancer cells.

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

Authors' ContributionsConception and design: E.A. Ariazi, J. BoydDevelopment of methodology: E.A. Ariazi, J.C. Taylor

Acquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): E.A. Ariazi, J.C. Taylor, E. Nicolas, J. BoydAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): E.A. Ariazi, M.A. Black, M.J. SlifkerWriting, review, and/or revision of the manuscript: E.A. Ariazi, M.A. Black,M.J. Slifker, D.J. Azzam, J. BoydAdministrative, technical, or material support (i.e., reporting or organizingdata, constructing databases): E.A. Ariazi, J.C. Taylor, M.J. SlifkerStudy supervision: E.A. Ariazi, J. Boyd

AcknowledgmentsThe authors thank Dennis DeSimone and Trung Nguyen, clinical fellows in

the laboratory for technical support. The authors also thank the ExpressionMicroarray facility, the Genotyping and Real-Time PCR facility, and the FlowCytometry facility at Fox Chase Cancer Center for technical support.

Grant SupportThis work was supported by the Commonwealth Universal Research

Enhancement (CURE) Program Award from the Pennsylvania Department ofHealth (to J. Boyd) and NIHP30 CA006927 (Fox Chase Cancer Center CoreGrant).

The costs of publication of this articlewere defrayed inpart by the payment ofpage charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received August 19, 2016; revised October 10, 2016; accepted October 18,2016; published OnlineFirst November 15, 2016.

References1. Ariazi EA, Jordan VC. Estrogen receptors as therapeutic targets in breast

cancer. In:Ottow E,WeinmannH, editors. Nuclear receptors as drug targets.Methods and principles in medicinal chemistry. M€orlenbach, Germany:Wiley-VCH; 2008. p.127–99.

2. Fullwood MJ, Liu MH, Pan YF, Liu J, Xu H, Mohamed YB, et al. Anoestrogen-receptor-alpha-bound human chromatin interactome. Nature2009;462:58–64.

3. Lin CY, Vega VB, Thomsen JS, Zhang T, Kong SL, Xie M, et al. Whole-genome cartography of estrogen receptor alpha binding sites. PLoS Genet2007;3:e87.

4. Frasor J, Danes JM, Komm B, Chang KC, Lyttle CR, Katzenellenbogen BS.Profiling of estrogen up- and down-regulated gene expression in humanbreast cancer cells: insights into gene networks and pathways underlyingestrogenic control of proliferation and cell phenotype. Endocrinology2003;144:4562–74.

5. Seth P, Porter D, Lahti-Domenici J, Geng Y, Richardson A, Polyak K.Cellular and molecular targets of estrogen in normal human breast tissue.Cancer Res 2002;62:4540–4.

6. Perou CM, Borresen-Dale AL. Systems biology and genomics of breastcancer. Cold Spring Harb Perspect Biol 2011;3:pii:a003293.

7. Prat A, Parker JS, Karginova O, Fan C, Livasy C, Herschkowitz JI, et al.Phenotypic and molecular characterization of the claudin-low intrinsicsubtype of breast cancer. Breast Cancer Res 2010;12:R68.

8. Prat A, Perou CM. Deconstructing the molecular portraits of breast cancer.Mol Oncol 2011;5:5–23.

9. Ronneberg JA, Fleischer T, Solvang HK, Nordgard SH, Edvardsen H,Potapenko I, et al. Methylation profiling with a panel of cancer relatedgenes: association with estrogen receptor, TP53 mutation status andexpression subtypes in sporadic breast cancer. Mol Oncol 2011;5:61–76.

10. Fackler MJ, Umbricht CB, Williams D, Argani P, Cruz LA, Merino VF, et al.Genome-wide methylation analysis identifies genes specific to breastcancer hormone receptor status and risk of recurrence. Cancer Res 2011;71:6195–207.

11. Hill VK, Ricketts C, Bieche I, Vacher S, Gentle D, Lewis C, et al. Genome-wide DNAmethylation profiling of CpG islands in breast cancer identifiesnovel genes associated with tumorigenicity. Cancer Res 2011;71:2988–99.

12. Ung M, Ma X, Johnson KC, Christensen BC, Cheng C. Effect of estrogenreceptor alpha binding on functional DNA methylation in breast cancer.Epigenetics 2014;9:523–32.

13. Schlesinger Y, Straussman R, Keshet I, Farkash S, Hecht M, ZimmermanJ, et al. Polycomb-mediated methylation on Lys27 of histone H3 pre-marks genes for de novo methylation in cancer. Nat Genet 2007;39:232–6.

14. KimKH, Roberts CW. Targeting EZH2 in cancer. NatMed2016;22:128–34.15. Vire E, Brenner C, Deplus R, Blanchon L, Fraga M, Didelot C, et al. The

Polycomb group protein EZH2 directly controls DNAmethylation. Nature2006;439:871–4.

16. Kulis M, Esteller M. DNA methylation and cancer. Adv Genet 2010;70:27–56.

17. FanM, Yan PS,Hartman-Frey C, Chen L, PaikH,Oyer SL, et al. Diverse geneexpression and DNAmethylation profiles correlate with differential adap-tation of breast cancer cells to the antiestrogens tamoxifen and fulvestrant.Cancer Res 2006;66:11954–66.

18. Stone A, Valdes-Mora F, Gee JM, Farrow L, McClelland RA, Fiegl H, et al.Tamoxifen-induced epigenetic silencing of oestrogen-regulated genes inanti-hormone resistant breast cancer. PLoS One 2012;7:e40466.

19. Leu YW, Yan PS, Fan M, Jin VX, Liu JC, Curran EM, et al. Loss of estrogenreceptor signaling triggers epigenetic silencing of downstream targets inbreast cancer. Cancer Res 2004;64:8184–92.

20. Marques M, Laflamme L, Gaudreau L. Estrogen receptor alpha can selec-tively repress dioxin receptor-mediated gene expression by targeting DNAmethylation. Nucleic Acids Res 2013;41:8094–106.

21. Ariazi EA, Kraus RJ, Farrell ML, Jordan VC, Mertz JE. Estrogen-relatedreceptor alpha1 transcriptional activities are regulated in part via theErbB2/HER2 signaling pathway. Mol Cancer Res 2007;5:71–85.

22. Nagalla S, Chou JW, Willingham MC, Ruiz J, Vaughn JP, Dubey P, et al.Interactions between immunity, proliferation and molecular subtype inbreast cancer prognosis. Genome Biol 2013;14:R34.

23. Ritchie ME, Phipson B, Wu D, Hu Y, Law CW, Shi W, et al. limma powersdifferential expression analyses for RNA-sequencing and microarray stud-ies. Nucleic Acids Res 2015;43:e47.

24. Tusher VG, Tibshirani R, Chu G. Significance analysis of microarraysapplied to the ionizing radiation response. Proc Natl Acad Sci U S A 2001;98:5116–21.

25. Haughian JM, Pinto MP, Harrell JC, Bliesner BS, Joensuu KM, Dye WW,et al. Maintenance of hormone responsiveness in luminal breastcancers by suppression of Notch. Proc Natl Acad Sci U S A 2012;109:2742–7.






26. Pink JJ, Bilimoria MM, Assikis J, Jordan VC. Irreversible loss of the oestro-gen receptor in T47D breast cancer cells following prolonged oestrogendeprivation. Br J Cancer 1996;74:1227–36.

27. Ellison-Zelski SJ, Solodin NM, Alarid ET. Repression of ESR1 throughactions of estrogen receptor alpha and Sin3A at the proximal promoter.Mol Cell Biol 2009;29:4949–58.

28. Neve RM,Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T, et al. A collection ofbreast cancer cell lines for the study of functionally distinct cancer subtypes.Cancer Cell 2006;10:515–27.

29. Charafe-Jauffret E, Ginestier C, Monville F, Finetti P, Adelaide J, Cervera N,et al. Gene expression profiling of breast cell lines identifies potential newbasal markers. Oncogene 2006;25:2273–84.

30. Yang J, Bielenberg DR, Rodig SJ, Doiron R, Clifton MC, Kung AL, et al.Lipocalin 2 promotes breast cancer progression. Proc Natl Acad Sci U S A2009;106:3913–8.

31. Li S, Xie Y, Zhang W, Gao J, Wang M, Zheng G, et al. Interferon alpha-inducible protein 27 promotes epithelial-mesenchymal transition andinduces ovarian tumorigenicity and stemness. J SurgRes 2015;193:255–64.

32. Augereau P, Badia E, Balaguer P, Carascossa S, Castet A, Jalaguier S, et al.Negative regulation of hormone signaling by RIP140. J Steroid BiochemMol Biol 2006;102:51–9.

33. Gurevich I, Flores AM, Aneskievich BJ. Corepressors of agonist-boundnuclear receptors. Toxicol Appl Pharmacol 2007;223:288–98.

34. Hwang C, Giri VN, Wilkinson JC, Wright CW, Wilkinson AS, Cooney KA,et al. EZH2 regulates the transcription of estrogen-responsive genesthrough association with REA, an estrogen receptor corepressor. BreastCancer Res Treat 2008;107:235–42.

35. Ye Y, Xiao Y, Wang W, Yearsley K, Gao JX, Barsky SH. ERalpha suppressesslug expression directly by transcriptional repression. Biochem J 2008;416:179–87.

36. Stossi F, Madak-Erdogan Z, Katzenellenbogen BS. Estrogen receptor alpharepresses transcription of early target genes via p300 and CtBP1. Mol CellBiol 2009;29:1749–59.

37. Cvoro A, Tzagarakis-Foster C, Tatomer D, Paruthiyil S, Fox MS, LeitmanDC. Distinct roles of unliganded and liganded estrogen receptors intranscriptional repression. Mol Cell 2006;21:555–64.

38. Green KA, Carroll JS. Oestrogen-receptor-mediated transcription andthe influence of co-factors and chromatin state. Nat Rev Cancer 2007;7:713–22.

39. Manavathi B, Dey O, Gajulapalli VN, Bhatia RS, Bugide S, Kumar R.Derailed estrogen signaling and breast cancer: an authentic couple. EndocrRev 2013;34:1–32.

40. Kiskinis E, Hallberg M, Christian M, Olofsson M, Dilworth SM, White R,et al. RIP140 directs histone and DNA methylation to silence Ucp1expression in white adipocytes. EMBO J 2007;26:4831–40.

41. Mann M, Cortez V, Vadlamudi RK. Epigenetics of estrogen receptorsignaling: role in hormonal cancer progression and therapy. Cancers 2011;3:1691–707.

42. Magnani L, Lupien M. Chromatin and epigenetic determinants ofestrogen receptor alpha (ESR1) signaling. Mol Cell Endocrinol 2014;382:633–41.

43. Osmanbeyoglu HU, Lu KN, Oesterreich S, Day RS, Benos PV, CoronnelloC, et al. Estrogen represses gene expression through reconfiguring chro-matin structures. Nucleic Acids Res 2013;41:8061–71.

44. Bell JS, Kagey JD, Barwick BG, Dwivedi B, McCabe MT, Kowalski J, et al.Factors affecting the persistence of drug-induced reprogramming of thecancer methylome. Epigenetics 2016;11:273–87.

45. Shi B, Liang J, Yang X, Wang Y, Zhao Y, WuH, et al. Integration of estrogenand Wnt signaling circuits by the polycomb group protein EZH2 in breastcancer cells. Mol Cell Biol 2007;27:5105–19.

46. Lee ST, Li Z, Wu Z, Aau M, Guan P, Karuturi RK, et al. Context-specificregulation of NF-kappaB target gene expression by EZH2 in breast cancers.Mol Cell 2011;43:798–810.

47. Fuks F, Burgers WA, Brehm A, Hughes-Davies L, Kouzarides T. DNAmethyltransferase Dnmt1 associates with histone deacetylase activity. NatGenet 2000;24:88–91.

48. Fuks F, Burgers WA, Godin N, Kasai M, Kouzarides T. Dnmt3a bindsdeacetylases and is recruited by a sequence-specific repressor to silencetranscription. EMBO J 2001;20:2536–44.

49. Cai Y, Geutjes EJ, de Lint K, Roepman P, Bruurs L, Yu LR, et al. The NuRDcomplex cooperates with DNMTs to maintain silencing of key colorectaltumor suppressor genes. Oncogene 2014;33:2157–68.


Ariazi et al.




2017;15:152-164. Published OnlineFirst November 15, 2016.Mol Cancer Res Eric A. Ariazi, John C. Taylor, Michael A. Black, et al. Cell, and EMT Genes

: Silencing via DNA Methylation of Basal, StemαA New Role for ER

Updated version

10.1158/1541-7786.MCR-16-0283doi:

Access the most recent version of this article at:

Material

Supplementary

http://mcr.aacrjournals.org/content/suppl/2017/01/20/1541-7786.MCR-16-0283.DC1

Access the most recent supplemental material at:

Cited articles

http://mcr.aacrjournals.org/content/15/2/152.full#ref-list-1

This article cites 48 articles, 15 of which you can access for free at:

Citing articles

http://mcr.aacrjournals.org/content/15/2/152.full#related-urls

This article has been cited by 1 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://mcr.aacrjournals.org/content/15/2/152To request permission to re-use all or part of this article, use this link



http://mcr.aacrjournals.org/lookup/doi/10.1158/1541-7786.MCR-16-0283

http://mcr.aacrjournals.org/content/suppl/2017/01/20/1541-7786.MCR-16-0283.DC1

http://mcr.aacrjournals.org/content/15/2/152.full#ref-list-1

http://mcr.aacrjournals.org/content/15/2/152.full#related-urls

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

mailto:[email protected]

http://mcr.aacrjournals.org/content/15/2/152


anewroleforera:silencingviadnamethylation of basal, stem ...presence of eradrives the luminal...

Documents