RESEARCH ARTICLE

Estrogen receptor beta as

rdhlm

) a-aa

superfamily of nuclear steroid receptors that, functioning cer cells, at least in part due to ER betas inhibition of ER

Rizza et al. Breast Cancer Research 2014, 16:R21http://breast-cancer-research.com/content/16/1/R21biological aggressiveness [6-9].Calabria, Via P. Bucci, Arcavacata di Rende (CS), Cosenza 87036, ItalyFull list of author information is available at the end of the articleas ligand-activated transcription factors, are able to inter-act with a host of different coregulators to regulate genetranscription.The roles of estrogen receptor (ER) alpha and beta

in breast cancer pathogenesis are becoming increasingly

alpha selective target gene expression, and can be consid-ered as an endogenous partial dominant negative receptor[2,3]. Indeed, the progression of breast cancer is associatedwith a change in the expression ratio of the isoforms ofER, with ER alpha the predominant isoform expressed [4].Moreover, compared with tumors expressing ER alphaalone, the co-expression of ER beta has been correlatedwith a more favorable prognosis [5] and decreased

* Correspondence: sebastiano.ando@unical.itEqual contributors1Department of Pharmacy, Health and Nutritional Sciences, University ofare transduced via the action of specific members of abreast cancer progression mainly by antagonizing ER-alpha signaling. However, to date no studies have specificallyevaluated a potential involvement of ER-beta in the inhibitory effects of androgens.

Methods: ER-beta expression was examined in human breast cancer cell lines using real-time PCR, Western blottingand small interfering RNA (siRNA) assays. Mutagenesis studies, electromobility shift assay (EMSA) and chromatinimmunoprecipitation (ChIP) analysis were performed to assess the effects of mibolerone/AR on ER-beta promoteractivity and binding.

Results: In this study, we demonstrate that mibolerone, a synthetic androgen ligand, up-regulates ER-beta mRNAand protein levels in ER-positive breast cancer cells. Transient transfection experiments, using a vector containingthe human ER-beta promoter region, show that mibolerone increases basal ER-beta promoter activity. Site-directedmutagenesis and deletion analysis reveal that an androgen response element (ARE), TGTTCT motif located atpositions 383 and 377, is critical for mibolerone-induced ER-beta up-regulation in breast cancer cells. Thisoccurs through an increased recruitment of AR to the ARE site within the ER-beta promoter region, along withan enhanced occupancy of RNA polymerase II. Finally, silencing of ER-beta gene expression by RNA interference isable to partially reverse the effects of mibolerone on cell proliferation, p21 and cyclin D1 expression.

Conclusions: Collectively, these data provide evidence for a novel mechanism by which activated AR, through anup-regulation of ER-beta gene expression, inhibits breast cancer cell growth.

IntroductionSex steroid hormones are critical for the development andprogression of endocrine-dependent diseases, includingbreast cancers. Estrogen and androgen hormone signals

elucidated by several clinical and in vitro studies. ERalpha mediates cancer-promoting effects of estrogen andhas been shown to be an effective therapeutic target fordecades [1]. In contrast, ER beta has a well known growthand invasion inhibitory activity in ER-positive breast can-androgen receptor actionPietro Rizza1, Ines Barone1, Domenico Zito1, Francesca GioLoredana Mauro1, Diego Sisci1, Stefania Catalano1, Karin Da

Abstract

Introduction: The two isoforms of estrogen receptor (ERproliferation and differentiation of breast cancers, with ERas a tumor suppressor. Emerging data have reported that 2014 Rizza et al.; licensee BioMed Central LtCommons Attribution License (http://creativecreproduction in any medium, provided the orOpen Access

a novel target ofin breast cancer cell linesano1, Marilena Lanzino1, Francesca De Amicis1,an Wright2, Jan-ake Gustafsson3 and Sebastiano And1*

lpha and beta play opposite roles in regulatinglpha mediating mitogenic effects and ER-beta actingndrogen receptor (AR) activation inhibits ER-positived. This is an Open Access article distributed under the terms of the Creativeommons.org/licenses/by/2.0), which permits unrestricted use, distribution, andiginal work is properly credited.

from UBI (Chicago, IL, USA), and triazol, SYBR Green

dem repeat analysis at our Sequencing Core; morphology,

Rizza et al. Breast Cancer Research 2014, 16:R21 Page 2 of 13http://breast-cancer-research.com/content/16/1/R21Androgen actions and androgen receptors (ARs) havebeen described in human breast cancers both in vivoand in vitro, but considerably less is known on theirimpact on this disease. Emerging evidence indicates thatthe androgen signaling pathway mainly exerts inhibitoryeffects on the growth of normal mammary epithelialcells and plays a protective role in the pathogenesis ofbreast cancer [10-13]. In particular, AR is present inapproximately 70% to 90% of invasive breast carcinomas,a percentage equal to or higher than that of ER (70% to80%) and its expression shows significant associationwith favorable clinicopathological characteristics, such aslower tumor grade, smaller tumor size, better outcomesand improved response to hormone therapy in ER-positive breast cancers [14-18]. In vitro studies havedemonstrated that androgen signaling may counteractthe proliferative effect of estrogens in AR-positive breastcancer cells [19], while over-expression of the AR mark-edly decreases ER alpha transcriptional activity in ER-positive breast cancer cells [20,21]. In these lattermodels, basal as well as estradiol-induced proliferationare inhibited by the non-aromatizable androgen 5-a-dihydrotestosterone (DHT) [21-23], through an in-crease in AR protein cell content [21], along with ablock in G1 to S transition of the cell cycle [21,22].AR-induced apoptosis has also been reported inthese cell lines [24]. Recently, we have shown a directdown-regulation of the cyclin D1 gene expression byAR activation via interaction with the orphan nuclearreceptor DAX1, as an additional mechanism involvedin the androgen-dependent inhibition of ER-positivebreast cancer cell growth [25].Given the ability of AR to function as an anti-pro-

liferative effecter by antagonizing ER signaling [26],the aim of this study was to investigate whether ERbeta may also be a target of androgen actions. Here,we demonstrate that mibolerone, a synthetic non-metabolizable androgen, up-regulates ER beta expres-sion and gene promoter activity, through a directbinding of AR to a newly identified androgen responseelement (ARE) located in the human ER beta genepromoter.

MethodsChemicals and reagentsDulbeccos Modified Eagles Medium/Nutrient MixtureF-12 Ham, DMEM, 100 bp DNA ladder, l-glutamine,penicillin, streptomycin, bovine serum albumin andphosphate-buffered saline were purchased from Invitrogen(Carlsbad, CA, USA), and Sephadex G50 spin columnsand poly (dI-dC) from Roche (Indianapolis, IN, USA).GoTaq DNA polymerase, T4 polynucleotide kinase, dual

luciferase kit, FuGENE 6 and CMV renilla luciferaseplasmid were provided by Promega (Madison, WI, USA).doubling times, estrogen sensitivity, and mycoplasma negativ-ity were tested (MycoAlert, Lonza, Walkersville, MD, USA).

PlasmidsThe plasmids containing the human ER beta pro-moter 0 N region or its deletions (A: p 1568/+315;B: p 1283/+315, D: p 355/+315) were a gift fromProf. Karin Dahlman-Wright (Department of Biosci-ences and Nutrition, Karolinska Institute, Sweden). Cplasmid (p-400/+315) and mutated on ARE site Cplasmid (M ARE mut) were generated by PCR usingthe following primers: forward 5- ATATACGCGTAATCAGACATCTGTTCTGAATGACACTTATGTGAG 3and reverse 5- ATATCTCGAGCGAAGGGGCGCTTACC 3; forward 5- ATATACGCGTAATCAGACATCTACCCTGAATGACACTTATGTG-3. The amplifiedDNA fragments were digested with Mlu I and Xho I andUniversal PCR Master Mix was from Biosystems (ForsterCity, CA, USA).

Cell culturesThe three human breast cancer cell lines MCF-7, ZR-75and MDA-MD 231 were acquired in 2010 from AmericanType Culture Collection (ATCC, Manassas, VA, USA)where they were authenticated, stored according to thesuppliers instructions, and used within a month after fro-zen aliquots were resuscitated. MCF-7 cells were culturedin DMEM-F12 containing 5% fetal bovine serum (FBS),1% l-glutamine and 1 mg/ml penicillinstreptomycin.ZR-75 cells were maintained in DMEM supplementedwith 10% FBS, 1% l-glutamine and 1 mg/ml penicil-linstreptomycin. MDA-MB- 231 cells were cultured inDMEM containing 5% FBS, 1% l-glutamine and 1 mg/mlpenicillinstreptomycin.Every four months cells were authenticated by single tan-The RETROscript kit and DNase I were purchased fromAmbion (Austin, TX, USA). Aprotinin, leupeptin, phenyl-methylsulfonyl fluoride, sodium orthovanadate, formal-dehyde, NP-40, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), dimethyl sulfoxide, protein-ase K, tRNA, IGF-1, DHT, bicatulamide(Cx) were fromSigma (Milan, Italy). Antibodies against cyclin D1, p21,GAPDH, and polymerase II (N20) were provided by SantaCruz Biotechnology (Santa Cruz, CA, USA). Antibody anti-ER beta (clone 644, rabbit monoclonal immuno-globulin G (IgG), 1:1000 dilution) was purchased fromMillipore. The ECL System, 3H thymidine and [32P]ATP were purchased from PerkinElmer (Wellesley,MA, USA), salmon sperm DNA/protein A agarose wasligated into pGL3-basic vector. The sequences were con-firmed by nucleotide sequence analysis.

Rizza et al. Breast Cancer Research 2014, 16:R21 Page 3 of 13http://breast-cancer-research.com/content/16/1/R21Western blot analysisCells were treated as indicated before lysis. Equal amountsof cell extracts were subjected to SDS-PAGE, as describedpreviously [27,28]. Blots are representative of at least threeindependent experiments.

Real-time RT-PCRThe gene expression of ER beta, cyclin D1, p21 and GAPDHwas evaluated by real-time RT-PCR. Total RNA was reversetranscribed with RETROscript kit. Diluted (1:3) cDNA (5 l)was analyzed in triplicate by real-time PCR in an iCycler iQDetection System (Bio-Rad, Hercules, California, USA)using SYBR Green Universal PCR Master Mix, followingthe manufacturers recommendations. Each sample wasnormalized on its GAPDH mRNA content. Primers usedfor the amplification were: forward 5-CCCTGCTGTGATGAATTACAG 3 and reverse 5- TCGGTTCCCACTAACCTTCC-3 (ER beta); forward 5GCATGACAGATTTCTACCACTCC-3 and reverse 5-AAGATGTAGAGCGGGCCTTT-3 (p21); forward 5-CTGGAGGTCTGCGAGGAA-3 and reverse 5-GGGGATGGTCTCCTTCATCT-3 (CYCD1); forward 5-CCCACTCCTCCACCTTTGAC-3 and reverse 5-TGTTGCTGTAGCCAAATTCGTT-3 (GAPDH). The relative gene expression levels werecalculated as described [29].

Transient transfection assaysMCF-7 cells were transiently transfected using theFuGENE 6 reagent with ER beta gene promoter, differ-ent deleted segments, and M ARE mut and treated asindicated. Empty vectors were used to ensure that DNAconcentrations were constant in each transfection. Lucif-erase activities were assayed using the Dual Luciferaseassay system (Promega) [30,31].

Electrophoretic mobility shift assay (EMSA)Nuclear extracts were prepared from MCF-7 cellstreated with vehicle or mibolerone 10 nM for 16 hoursas previously described [32,33]. The DNA sequencesused as probe or as cold competitor are the following(mutations are shown as lowercase letters): 5- AGAATCAGACATCTGTTCTGA ATGACA 3, 5 T GTCATTCAGAACAGATGTCTGATTCT-3 (ARE); 5- AGA ATCAGACATCTaccCTGAATGACA-3, 5- TGTCATTCAGggtAGATGTCTGATTCT 3 (mutated ARE). Probegeneration and the protein-binding reactions were carriedout as described [34]. For experiments involving anti-bodies, the reaction mixture was incubated with AR orIgG antibodies at 4C for 12 hours before addition oflabeled probe.

RNA interference (RNAi)

MCF-7 cells were transfected with RNA duplex of stealthRNA interference (RNAi)-targeted for human ER betamRNA sequence 5-CACUUCUGCGCUGUCUGCAGCGAUU 3 (Life Science, Hercules, California, USA) orwith a stealth RNAi-negative control (Life Science) to afinal concentration of 100 nM using Lipofectamine 2000as recommended by the manufacturer [35,36].

Chromatin immunoprecipitation assaysCells were treated with vehicle or mibolerone 10 nMand after 16 hours the DNA/protein complexes wereextracted as previously described [37]. The preclearedchromatin was immunoprecipitated with anti-AR andanti-poymerase II antibodies. A normal mouse serumIgG was used as negative control. For each sample andinput DNA, 5 l were used for PCR amplification withthe following primers flanking ARE sequence present inthe ER beta promoter region: 5- GGTTTCACCACTGGCTCCTT-3 (forward) and 5-ACTGATACAGCCAGTCTGGG-3 (reverse). Final results were calculatedusing the Ct method, using input Ct values insteadof the GAPDH mRNA. The basal sample was used ascalibrator.

Cell proliferation assaysCell proliferation was determined by using the MTTassay as previously described [38] Data are representa-tive of three independent experiments, each performedin triplicate. For [3H]thymidine incorporation, a total of1 105 cells was seeded onto 12-well plates in completemedium and then treated as indicated. Control () cellswere treated with the same amount of vehicle alone(dimethylsulfoxide (DMSO)) that never exceeded a con-centration of 0.01% (v/v). [3H]Thymidine incorporationwas evaluated after a 24-hour incubation period with1 Ci of [3H]thymidine per well. Cells were washed oncewith 10% trichloroacetic acid, twice with 5% trichlo-roacetic acid, and lysed in 1 ml of 0.1 M NaOH at 37Cfor 30 minutes. The total suspension was added to 10 mlof Optifluor fluid and was counted in a scintillationcounter.

Statistical analysisData were analyzed for statistical significance using a two-tailed Students t-test, performed by Graph Pad Prism 4.Standard deviations (S.D.) are shown.

ResultsMibolerone increases ER beta expression in breast cancercellsWe have previously demonstrated that the non-aromatiz-able androgen DHT decreased cell proliferation in a dose-dependent manner in ER-positive MCF-7 breast cancercells [21,26]. In this study, to minimize the metabolic con-

version of androgen to estrogenic compounds by culturedcells, we used the synthetic non-metabolizable androgen

mibolerone to test its effects on cell proliferation by meas-uring changes in the rate of DNA synthesis (3H thymidineincorporation). We found similar inhibitory effects on theproliferation of ER-positive MCF-7 and ZR 75 breastcancer cells after two, four and six days of miboleronetreatment (Figure 1A). Given the known tumor repressiverole of ER beta in breast cancer cell lines, we wonderedwhether AR also functions as an anti-proliferative effectorin ER-positive breast cancer by affecting ER beta expres-sion. First, MCF-7 and ZR75 breast cancer cells weretreated with mibolerone for 24 and 48 hours and ER betamRNA and protein levels were evaluated by real timeRT-PCR and western blotting analysis. As shown inFigure 1B and C, mibolerone treatment increased ERbeta expression at all times investigated in both MCF-7and ZR75 cells.Clear evidence of the crucial role of AR in mediating

these effects has been pointed out by the fact that theAR inhibitor hydroxyflutamide (OH-Fl) completely re-versed the up-regulation of ER beta mRNA and proteincontent induced by mibolerone (Figure 2A and B). Itis important to underline that using another androgen

ligand, DHT, and an androgen antagonist, bicatulamide,we observed similar results on ER beta expression(Additional file 1: Figure S1B and S1C). Accordingly,bicalutamide reversed the inhibition mediated by bothmibolerone and DHT on MCF-7 and ZR-75 cell growth(Additional file 1: Figure S1A).

Activated AR up-regulates ER beta via an ARE site of itspromoterTo analyze if mibolerone might positively modulate ERbeta gene transcription, MCF-7 and ZR75 breast cancercell lines were transiently transfected with a luciferasereporter plasmid containing the human ER beta promoterregion spanning from 1568 bp to +315 bp. As shown inFigure 3A, a significant increase in ER beta promoteractivity was observed in cells treated with mibolerone,while this induction was abrogated in the presence of theAR inhibitor OH-Fl (Figure 3A). The human ER betapromoter contains multiple consensus sites for severaltranscription factors, including an AP-1 box, OCT-1,GATA, and ARE [39]. To identify the regions within theER beta promoter responsible for mibolerone-mediated

MCF-7 ZR-75A

60

80

100

120

60

80

100

120

idin

e in

corp

orat

ion

of c

ontro

l

mid

ine

inco

rpor

atio

n o

f con

trol

(-)Mib 10 nM

(-)Mib 10 nM

cedaasd tm

blors.

Rizza et al. Breast Cancer Research 2014, 16:R21 Page 4 of 13http://breast-cancer-research.com/content/16/1/R21- 2 4 6 (days)

0

50

100

150

200

250

0

50

100

150

200

250

24 48Mib 10 nM

-(h) 24 48Mib 10 nM

-(h)

MCF-7 ZR-75

B

tra

nscr

ipt l

evels

(% of

contr

ol )

ER

/GAPD

H m

RNA

tra

nscr

ipt l

evels

(% of

contr

ol )

ER

/GAPD

H m

RNA

0

20

40

3H T

hym

%

Figure 1 Mibolerone up-regulates ER beta expression in breast cantreated with vehicle () or mibolerone (Mib) 10 nM for two, four and sixwith triplicate samples expressed as percent of control. (B) Total RNA w10 nM for 24 and 48 hours, and reverse transcribed. cDNA was subjectesample was normalized to its GAPDH mRNA content. Data represent thepercentage of control assumed to be 100%. (C) Bottom panel, Westernand ZR75 cells, treated with vehicle () or Mib 10 nM for 24 and 48 hou

represent the mean S.D. of three independent experiments in which banand expressed as the percentage of the control assumed to be 100%. *, P0

50

100

150

200

250

0

50

100

150

200

250

MCF-7 ZR-75

24 48Mib 10 nM

-(h)GAPDH

ER

C

(Opti

cal d

ensit

y ar

bitra

ry u

nits

%)

24 48Mib 10 nM

-(h)GAPDH

ER

ER

/GAPD

H

(Opti

cal d

ensit

y ar

bitra

ry u

nits

%)

ER

/GAPD

H

0

20

40

3H T

hy %

- 2 4 6 (days)

r cells. (A) 3H Thymidine incorporation assays in MCF-7 and ZR-75 cellsys. Columns, mean of three independent experiments, each performedisolated from MCF-7 and ZR75 cells treated with vehicle () or Mibo real time PCR using specific primers for ER beta and GAPDH. Eachean S.D. of values from three separate RNA samples expressed as at analysis of ER beta expression in total protein extracts from MCF-7GAPDH was used as loading control. Upper panel, the histograms

d intensities were evaluated in terms of optical density arbitrary units

Rizza et al. Breast Cancer Research 2014, 16:R21 Page 5 of 13http://breast-cancer-research.com/content/16/1/R21100

150

200

250

0

50

100

150

200

250

MCF-7

A MCF-7

Mib 10 nM

OH-Fl 1 M

+

+

++

B

(Opti

cal d

ensit

y ar

bitra

ry u

nits

%)

ER

/GAPD

H

tra

nscr

ipt l

evels

(% of

contr

ol )

ER

/GAPD

H m

RNA stimulatory effects, we transiently transfected both MCF-7and ZR-75 cell lines with plasmids containing a series of5deleted segments of the human ER beta promoter.Schematic representation of these constructs is shown inFigure 3B. In transfection experiments performed usingp-1568/+315 (A), p1256/+315 (B) and p-400/+315(C) plasmids, the responsiveness to mibolerone wasstill maintained (Figure 3B), suggesting that the regionbetween 400 to +315 might be involved in the trans-activation mechanisms exerted by mibolerone. Thus, wefocused our attention on the latter construct, p-400/+315(C), and we identified, upstream to the initiation transcrip-tion site, one ARE site, which is the putative effector ofAR signaling. We observed that in MCF-7 cells transientlytransfected with the ER beta promoter plasmid bearingthe ARE-mutated site or with a deleted construct of ERbeta promoter without the ARE site (p-355/+315, D),mibolerone was no longer able to induce ER beta pro-moter activity (Figure 3C). Similar results were obtainedin ZR75 breast cancer cells (data not shown). Taken to-gether, our findings demonstrated that the up-regulatory

ERGAPDH

Mib 10 nMOH-Fl 1 M

+

+

++

0

50

Figure 2 Hydroxyflutamide (OH-Fl) reverses mibolerones effects on Elevels in cells treated as indicated. Each sample was normalized to GAPDHseparate RNA samples expressed as the percentage of control assumed toin cells treated with vehicle () or Mib 10 nM in the presence or absence opanel, the histograms represent the mean S.D. of three independent expdensity arbitrary units and expressed as the percentage of the control assumecompared to Mib treated cells. ER, estrogen receptor.100

150

200

250

50

100

150

200

250

ZR-75

ZR-75

0Mib 10 nM

OH-Fl 1 M

+

+

++

ptica

l den

sity

bitra

ry u

nits

%)

ER

/GAPD

H

tra

nscr

ipt l

evels

(% of

contr

ol )

ER

/GAPD

H m

RNA effects exerted by mibolerone on ER promoter geneexpression require an ARE sequence motif.

The AR protein is recruited at an ARE site to ER betapromoter regionThe specific role of the ARE motif in mediating thestimulatory role of mibolerone on ER beta gene expressionwas then investigated using EMSAs and chromatin immu-noprecipitation (ChIP) assays. Using synthetic radiola-beled oligonucleotides bearing the ARE site present in theER beta promoter region (Figure 4A, lane 1), we observedthe formation of a protein complex in nuclear extractsfrom MCF-7 cells, which was abrogated by incubationwith 100-fold molar excess of unlabeled probe (Figure 4A,lane 2), demonstrating the specificity of the DNA-bindingcomplex. This inhibition was no longer observed when amutated oligodeoxyribonucleotide probe was used asa competitor (Figure 4A, lane 3). Interestingly, treat-ment with mibolerone strongly increased the DNA-binding protein complex compared with control samples(Figure 4A, lane 4). The inclusion of the anti-AR antibody

0

50

ERGAPDH

Mib 10 nMOH-Fl 1 M

+

+

++

(O ar

R beta expression. (A) Real-Time RT-PCR for analyzing ER beta mRNARNA content. Data represent the mean S.D. of values from threebe 100%. (B) Bottom panel, Western blot analysis of ER beta expressionf OH-Fl 1 M for 48 hours. GAPDH was used as loading control. Uppereriments in which band intensities were evaluated in terms of opticald to be 100%. *, P

OC

Rizza et al. Breast Cancer Research 2014, 16:R21 Page 6 of 13http://breast-cancer-research.com/content/16/1/R21-1568

AP-1

MCF-7

A

EREin the reaction supershifted the specific band, confirmingthe presence of this protein in the complex (Figure 4A,lanes 5). Non specific IgG did not affect AR complexformation (Figure 4A, lane 6).Moreover, to better evaluate the functional importance

of the ARE site at the ER beta promoter level, ChIPassays were performed. Protein-chromatin complexeswere immunoprecipitated from MCF-7 cells treated withmibolerone using specific antibodies against AR or RNA-

0 0.5 1.0 1.5 2.0 2.5 3.0

pGL3

-1283

-1568 +315

+315

+1

+1

-400

ARE= TGTTCT

-355

-400

ARE mut= TACCCT

-400 +315+1

00.51.01.5

2.02.5

AP-1

OCT-1

GATA

Luciferase activity (Fold Change)

Luci

fera

seac

tivity

(Fold

Chan

ge)

AREAP-1 OCT-1 GATA

Mib 10 nM

OH-Fl 1 M

MCF-7B

C

A

B

C

CM ARE mut

D

ERE

(-)MIB 10 nM

Figure 3 ER beta promoter activity is up-regulated by mibolerone thrconstruct of the ER beta gene promoter used in this study. Bottom panel,transfected in MCF-7 and ZR-75 cells treated with vehicle () or Mib 10 nMvehicle-treated cells; *, P

7Rizza et al. Breast Cancer Research 2014, 16:R21 Page 7 of 13http://breast-cancer-research.com/content/16/1/R21A1 2 3 4 5 6AR activation enhances ER beta expression in MDA-MB-231breast cancer cellsTo extend the results obtained, we also evaluated theeffects of mibolerone on ER beta expression in the ERalpha-negative, ER beta-positive MDA-MB-231 breastcancer cell line. As previously shown for MCF-7 andZR-75 cells, treatment with mibolerone for 48 hoursenhanced ER beta mRNA and protein levels that werecompletely reversed in the presence of the androgenantagonist OH-Fl (Figure 5A and B). Again, treatmentwith mibolerone significantly increased ER beta pro-moter activity and this induction was abrogated bothin cells treated with OH-Fl and in cells transfected withthe ER beta promoter plasmid bearing the ARE mutatedsite (M ARE mut) (Figure 5C). Therefore, AR activationresulted in an up-regulation of ER beta expression andactivity in different cellular backgrounds.

MUT

Nuclear extractMib

Competitor (100 fold excess)Ab-AR

IgG

Figure 4 Mibolerone recruits AR protein to an ARE site in the ER betavehicle () or Mib 10 nM for 16 hours were incubated with a double-strandto electrophoresis in a 6% polyacrylamide gel (lanes 1 and 4). Competitionmolar excess of unlabeled probe (lane 2 ) or a 100-fold molar excess of unNuclear extracts from MCF-7 cells treated with Mib 10 nM for 16 hours werpresence of the probe. Lane 7, probe alone. (B,C), MCF-7 cells treated with veand lysed. The pre-cleared chromatin was immune-precipitated with specific(IgG) as a negative control. For each sample and input, a 5 l volume was anaSection, to amplify the ER beta promoter sequence containing the ARE sitexperiments. *, P

Rizza et al. Breast Cancer Research 2014, 16:R21 Page 8 of 13http://breast-cancer-research.com/content/16/1/R21

0

50

100

150

200

250

AMDA-MB- 231

Mib 10 nM

tra

nscr

ipt l

evels

(% of

contr

ol )

ER

/GAPD

H m

RNA at both the mRNA and protein levels. No changes wereobserved after transfection of cells with a scrambledsiRNA control (Figure 6A and B). Accordingly, ER betagene silencing partially reversed the inhibition medi-ated by mibolerone on MCF-7 and ZR-75 cell growth(Figure 6C and E), suggesting how the anti-proliferativeeffects exerted by mibolerone may also be related to aninduction of ER beta levels.

DiscussionIn this study, we show, for the first time, that androgensincrease ER beta gene expression in ER-positive breastcancer cells. This occurs through an enhanced recruit-ment of AR to the ARE site located at 383 to 377 bp

OH-Fl 1 M

+

+

++

0

0.5

1.0

1.5

2.0

2.5

Luci

fera

se a

ctiv

ity (F

old C

hang

e)

Mib 10 nM

OH-Fl 1 M

+

+

MDA

C

CFigure 5 Mibolerone increases ER beta expression in MDA-MB-231 brcells treated with vehicle () or mibolerone (Mib) 10 nM in the presence oGAPDH mRNA content. Data represent the mean S.D. of values from threassumed to be 100%. (B) Bottom panel, Western blot analysis of ER beta inin the presence or absence of OH-Fl for 48 hours. GAPDH was used as loadthree independent experiments in which band intensities were evaluated in tof the control assumed to be 100%. (C) ER beta luciferase promoter activity inmeans S.D. of three different experiments each performed in triplicate. *, P