Expression of estrogen receptor β increases integrin α1 and integrin β1 levels and enhances...

ORIGINAL ARTICLE 156J o u r n a l o fJ o u r n a l o f

CellularPhysiologyCellularPhysiology
Expression of Estrogen
Receptor b Increases Integrina1 and Integrin b1 Levels andEnhances Adhesion of BreastCancer Cells
KAROLINA LINDBERG,1* ANDERS STROM,2 JOHN G. LOCK,1 JAN-AKE GUSTAFSSON,1,2
LARS-ARNE HALDOSEN,1 AND LUISA A. HELGUERO1

1Department of Biosciences and Nutrition, Novum, Karolinska Institutet, Huddinge, Stockholm, Sweden2Department of Biology and Biochemistry, University of Houston, Houston, Texas

Estrogen effects on mammary gland development and differentiation are mediated by two receptors (ERa and ERb). Estrogen-bound ERainduces proliferation of mammary epithelial and cancer cells, while ERb is important for maintenance of the differentiated epithelium andinhibits proliferation in different cell systems. In addition, the normal breast contains higher ERb levels compared to the early stage breastcancers, suggesting that loss of ERb could be important in cancer development. Analysis of ERb�/� mice has consistently revealedreduced expression of cell adhesion proteins. As such, ERb is a candidate modulator of epithelial homeostasis and metastasis.Consequently, the aim of this study was to analyze estrogenic effects on adhesion of breast cancer cells expressing ERa and ERb. As ERb iswidely found in breast cancer but not in cell lines, we used ERa positive T47-D and MCF-7 human breast cancer cells to generate cells withinducible ERb expression. Furthermore, the colon cancer cell lines SW480 and HT-29 were also used. Integrin a1 mRNA and proteinlevels increased following ERb expression. Integrin b1—the unique partner for integrin a1—increased only at the protein level. ERbexpression enhanced the formation of vinculin containing focal complexes and actin filaments, indicating a more adhesive potential. Thiswas confirmed by adhesion assays where ERb increased adhesion to different extracellular matrix proteins, mostly laminin. In addition,ERb expression was associated to less cell migration. These results indicate that ERb affects integrin expression and clustering andconsequently modulates adhesion and migration of breast cancer cells.

J. Cell. Physiol. 222: 156–167, 2010. � 2009 Wiley-Liss, Inc.

Jan-Ake Gustafsson is a shareholder and consultant of KaroBio AB.

Additional Supporting Information may be found in the onlineversion of this article.

Contract grant sponsor: The Swedish Cancer Fund.Contract grant sponsor: KaroBio AB.Contract grant sponsor: Lars-Hiertas Minne Foundation.

*Correspondence to: Karolina Lindberg, Unit for NuclearReceptor Biology, Department of Biosciences and Nutrition,Novum, Karolinska Institutet, Halsovagen 7, 141 57 Stockholm,Sweden. E-mail: [email protected]

Received 28 June 2009; Accepted 11 August 2009

Published online in Wiley InterScience(www.interscience.wiley.com.), 24 September 2009.DOI: 10.1002/jcp.21932

Breast cancer is the most common form of cancer in women(Parkin et al., 2005; Ferlay et al., 2007) and the ovarian hormoneestrogen is associated with its promotion and growth.Two-thirds of breast cancers express estrogen receptors (ERs)and to date this is the most important predictive criterion ofhormone responsiveness (Osborne, 1998). ERs mediateestrogen actions by enhancing transcription of genes eitherdirectly through binding to classical estrogen responseelements (EREs) in the DNA, or through alternative pathwaysinvolving regulatory interactions with other transcriptionfactors, such as AP-1 and SP-1 (reviewed by Gustafsson, 2000).

Two ERs, alpha (ERa) and beta (ERb) (Nilsson et al., 2004)are found throughout the normal breast. ERa and ERb sharehigh structural homologies and bind the endogenous estrogen17b-estradiol (E2) with similar affinity but differently to othernatural or synthetic ligands, and in many cases ERa and ERb canantagonize each other’s actions (reviewed by Koehler et al.,2005). Proliferation and progression of breast cancer is drivenby ERa and most human breast cancers, at least initially, expressthis receptor. On the contrary, ERb is constitutively expressedin normal mammary glands but in human breast cancer, ERbexpression is lower than in the normal breast (Leygue et al.,1998; Iwao et al., 2000; Roger et al., 2001). Furthermore, ERþhuman breast cancers express ERa alone, ERa and ERb(reviewed by Speirs et al., 2004) or ERb alone, (Novelli et al.,2008), where ERb expression is associated with less invasiveand proliferating tumors (Jarvinen et al., 2000) and bettersurvival in patients with ERa�/progesterone receptor (PR�)and ERa�/PR�/EGFR2� tumors (Honma et al., 2008). ERbexpression inhibits E2-induced proliferation of breast cancer

� 2 0 0 9 W I L E Y - L I S S , I N C .

cells in culture and in vivo through modulation of cell cycleproteins (Lazennec et al., 2001; Strom et al., 2004; Hartmanet al., 2006). Loss of ERb can also affect the normaldevelopment of the mammary gland. ERb�/� mice showreduced expression of the adhesion proteins, E-cadherin,occludin, connexin 32, and integrin a2 and an increase inproliferating cells in the lactating stage of the mammary gland(Forster et al., 2002). Furthermore, knockdown of ERb inHC11 mouse mammary epithelial cells is associated withincreased proliferation, loss of contact growth inhibition(Helguero et al., 2005) and E-cadherin downregulation(Helguero et al., 2008). Overall, these findings indicate that

R O L E O F E S T R O G E N R E C E P T O R b I N C E L L A D H E S I O N 157

there is a role for ERb in the regulation of cell adhesion andpoint out ERb as a modulator of invasion and proliferation. Still,more studies are needed in order to understand these effects.

Invasive cells undergo dramatic alterations in levels ofintegrin expression and integrin affinity for extracellular matrix(ECM) substrates (Pignatelli and Stamp, 1995). Integrins occuras heterodimeric transmembrane receptors, consisting of an aand ab subunit, which are involved in attachment, proliferation,migration, and differentiation of cells (Hynes, 1987, 1992).Upon ligand binding, integrins cluster into focal contacts. Fromfocal contacts, cells connect the ECM to the cytoskeletonthrough a variety of actin binding proteins resulting in formationof actin filaments and recruitment of signaling molecules to thefocal adhesions (Aplin et al., 1998; Hynes, 2002). Changes inexpression and signaling of integrinsb1,b4,a2,a3, anda6 havebeen observed in cancer cell lines and in tumor tissue sections,and have also been shown to be associated with tissuedisorganization, loss of polarity, increased tumoraggressiveness, and metastasis (Natali et al., 1992; Rossen et al.,1994; Gui et al., 1995; Zutter et al., 1995; Chao et al., 1996;Lipscomb et al., 2005; Guo et al., 2006; Park et al., 2006).Furthermore, overexpression of the actin binding proteinvinculin reduces cell migration, while its downregulation has anopposite effect, possibly by reducing the number and size offocal adhesions and adhesion of cells to the substrate (Xu et al.,1998; Subauste et al., 2004).

Nearly 90% of all breast cancer deaths arise from metastasis,and overexpression of ERb has been shown to inhibit invasionof ER-negative MDA-MB-231 invasive breast cancer cells andDU-145 prostate cancer cells (Lazennec et al., 2001; Chenget al., 2004). In this work, the aim was to investigate estrogeneffects on integrin expression and the modulation of adhesiveand migratory properties of cancer cells in a context whereboth ERa and ERb are expressed. We focused our studies onthe regulation of integrin expression and found that ERbupregulates integrins a1 and b1. This results in formation ofvinculin containing focal complexes and actin assembly andcorrelates to increased cell adhesion to ECM proteins anddecreased cell migration. Overall, findings from these studiesindicate that ERb is a potential target in the treatment of breastcancer.

Materials and MethodsCell cultures

T47-D cells with tetracycline-regulated expression of ERb485(T47-DERb) and control cells, T47-DPBI (mock) (Strom et al.,2004), were grown in RPMI 1640 (Invitrogen, Carlsbad, CA)supplemented with L-glutamine (Invitrogen), 5% heat-inactivatedfetal bovine serum (FBS; Invitrogen), 1% penicillin–streptomycin(Invitrogen) and 10 ng/ml doxycycline (Sigma, St. Louis, MO). Threeto 24 h or 1 week prior to analysis, the medium was replacedby phenol red-free RPMI 1640 medium plus L-glutamine, 2%heat-inactivated fetal bovine serum and treated with 10 or0.01 ng/ml doxycycline, from now onwards: �ERb or þERb,respectively. Cells were treated with vehicle alone (EtOH) or10 nM 2,3-bis(4-hydroxy-phenyl)-propionitrile (DPN, Tocris,Bristol, UK), 10 nM 17b-estradiol (E2, Sigma) or 100 nM ICI 182780 (ICI, Tocris). MCF-7 breast cancer cells and HT29 coloncancer cells with tetracycline-regulated expression of ERb485(MCF-7ERb, HT29ERb) were treated in a similar manner and usedto confirm results obtained in T47-DERb cells. ERa-negativeSW480 colon cancer cells with expression of lentivirus transducedERb (SW480ERb) were also used.

Transient Transfections

Cells were treated for 24 h or 1 week and transiently transfectedwith 2mg of 3� ERE-TATA-LUC reporter plasmid and 60 ng of

JOURNAL OF CELLULAR PHYSIOLOGY

b-galactosidase gene expressing plasmid using LipofectamineTM

2000 (Invitrogen) according to the manufacturer’s protocol. ERbexpression was induced for 48 h with or without DPN, E2, or ICI.Reporter gene activity was normalized to b-galactosidase enzymeactivity.

FACS Analysis

Following 1 week of treatment, cells were dissociated with anenzyme free cell dissociation buffer (Chemicon, Temecula, CA) and0.5� 106 cells were incubated with anti-integrin b1 (MAB1951Z,Chemicon), anti-integrin a1 (FB12, Chemicon), or anti-integrina2 (ab24697, Abcam, Cambridge, UK) antibodies for 1 h at roomtemperature. After washing with Dulbecco’s PBS (D-PBS;Invitrogen)/1% BSA, cells were incubated with goat anti-mouse IgGPE (R-phycoerythrin; Jackson Immunolabs, West Grove, PA) andanalyzed with a FACSCalibur flow cytometer (Becton Dickinson,Franklin Lakes, NJ).

Quantitative real-time PCR

Cells were grown in 6-well tissue culture plates for 3–24 h or 1week, lysed in Trizol reagent (Invitrogen), RNA extracted, andcDNA was synthesized as described earlier (Muller et al., 2002).Real-time PCR was performed with SYBR-Green PCR MasterMix in an ABI PRISM 7500 (Applied Biosystems, Foster City, CA).Primers used are listed in Supplemental Table 1. Quantification wascarried out following the supplier’s protocol and using the standardcurve method.

Whole-cell extracts

For ER and PR detection, cell pellets were freeze–thawed andwhole cell extracts were prepared with buffer C (10 mM Hepes-KOH, pH 7.9, 0.4 M NaCl, 0.1 mM EDTA, 5% glycerol, 1 mM DTT).For integrin b1 and vinculin, PBS-TDS buffer (PBS with 1% TritonX-100, 0.5% sodium deoxycholate, 0.1% SDS, 1 mM EDTA, 1 mMPMSF) was used. Protein quantification was carried out usingBradford reagent (BioRad, Hercules, CA) or BCA protein assay kit(Pierce, Rockford, IL).

Western blotting

Fifty micrograms total cellular protein was separated on a 7.5%SDS–PAGE, and electrotransferred to a nitrocellulose membrane(HybondTM-C Amersham, Buckinghamshire, UK). The followingantibodies were used: anti-ERb Ab14021 (Abcam), anti-ERaHC-20 (Santa Cruz Biotechnology, Santa Cruz, CA), anti-integrinb1 MAB 1951Z (Chemicon), anti-vinculin hVIN-1 (Sigma), anti-PRC-20 (Santa Cruz Biotechnology), anti-tubulin 11H10 (CellSignaling, Danvers, MA), and anti-b actin (Sigma). The secondaryantibodies were HRP-conjugated (Sigma). Visualization was carriedout using ECL plus (Amersham) or West Pico (Pierce) kits. Bandintensity was measured with Quantity one software (BioRad) andnormalized to beta actin or tubulin content. The intensity ratiobetween PR isoforms A/B was calculated by dividing the intensity ofthe bands in each sample. For comparison between experimentsthe values in the control cells were arbitrarily set to one.

Immunofluorescence

Glass coverslips were coated with mouse collagen type IV(BD Biosciences, Franklin Lakes, NJ) and cells treated for 1 week asdescribed above, were allowed to adhere to the coverslips for 1 h.The cells were fixed in 3% paraformaldehyde, 5% sucrose in D-PBS,and permeabilized in PBS containing 0.1% Tween-20 (PBS-T).Blocking of unspecific binding was done with 5% goat serum(Jackson Immunolabs), 5 mg/ml BSA (Sigma) in PBS-T. Thefollowing primary antibodies were diluted in D-PBS containing 0.1%Tween-20 (D-PBS-T), anti-vinculin (hVIN-1, Sigma), or anti-integrina1 (sc-10728, Santa Cruz Biotechnology). As secondary antibody,goat anti-mouse or anti-rabbit IgG coupled to AlexaFluor568 dye (Invitrogen) was used. Actin filaments were stained with

158 L I N D B E R G E T A L .

phalloidin-AlexaFluor 647 (Invitrogen); Hoechst 33342 (MolecularProbes, Eugene, OR) was used to stain the nucleus. The sampleswere mounted with ProLong Gold (Invitrogen).

Fluorescence imaging and quantitative image analysis

Pictures of the fluorescence stainings were captured with aninverted fluorescence microscope (Olympus IX71) with thefollowing lenses: PlanApo 60�/1.40 oil and UplanApo100�/1.35 oil Iris. Images were acquired with ORCA-ERHamamatsu camera using Wasabi imaging program under the samesettings, and if edited with Adobe Photoshop 6.0, the sameadjustments were applied to all images. Images were imported intocustom-designed image analysis software (Digital Cell ImagingLaboratories, Keerbergen, Belgium) for quantification. Cellboundaries were automatically detected and manually correctedwhere necessary to delimit subsequent analyses to individual cells.Hoechst labeling of nuclei assisted in the determination of cellnumber and position in densely clustered cell populations.Objects (cell–matrix adhesions) within delimited cell areas wereautomatically detected en masse using a second tunablealgorithm. Algorithm performance was supervised by visualverification of object segmentation. Segmented objects were thenquantitatively characterized by extracting mean and maximumintensity values (8 bit) per object, as well as object area and totalobject number per cell. These data were exported as text files,converted to excel files (Microsoft) and then visualized andanalyzed for trends and variations using Tableau software(Seattle, WA).

Cell adhesion assay

T47-DERb, MCF-7ERb, and HT29ERb cells treated for 1 week, asdescribed above, were dissociated with an enzyme-free celldissociation solution, and assayed on cytomatrix cell adhesionstrips (Chemicon) coated with collagen I, collagen IV,fibronectin, laminin, or vitronectin, following the manufacturer’ssuggestion.

Wound assay

Following 1 week treatment, a confluent monolayer of cells wasdisrupted by scratching a line through the layer using a pipette tip.Cells were treated with a stimulator of migration, 50 ng/ml EGF(Frey et al., 2004) (Sigma), or vehicle alone. Brightfield images of thecells were captured on a Zeiss Axiovert S100 microscope, usingthe Axiovision AC software, at 0 and 24 h. The area of the woundwas measured and analyzed with Image J. Three areas werequantified in duplicate treatments and experiments repeated threetimes.

Transwell migration assay

Cells were treated for 1 week, thereafter cells were labeled withDiIC12(3) Fluorescent Dye (BD Biosciences) and placed in BDFalcon FluoroBlok Inserts (8.0mm pore size) in the presence ofchemoattractant (10% fetal bovine serumþ 10% matrigel) in thelower chamber. Chemotaxis was analyzed after 48 h by measuringthe fluorescence of cells migrating through the pores to the lowerchamber using a fluorescence microplate reader. NIH3T3 cellswere used as a positive control for migration. Random migrationwas measured by incubating both, the upper and lower chamberwith 2% FBS.

Statistical analysis

The analysis was carried out using GraphPad Prism 5. Differencesbetween two groups were analyzed with student’s ‘‘t’’ test andbetween more than two groups with one-way ANOVA andTukey’s multiple post-test. Differences between the ratios of PR


isoforms A/B were analyzed with one-sample t-test for significantdifferences from a hypothetical value of one.

ResultsTet-regulated expression of ERb in T47-D humanbreast cancer cells

In T47-DERb cells (Strom et al., 2004), ERbmRNA and proteinexpression was induced upon withdrawal of doxycycline (S.1).In order to avoid unspecific effects such as squelching, aconcentration of 0.01 ng/ml doxycycline was chosen for use inall experiments. Under these experimental conditions 200-foldinduction of ERbmRNA levels at 24 h and more than 1000-foldinduction at 1 week were achieved. ERb protein levelsincreased but were higher after 24 h than 1 week of induction,which, given the growth inhibitory effects of ERb on these cells(Strom et al., 2004), might indicate an adaptation response bythe cells. ERb mRNA and protein expression persisted after1 week which, taking into consideration the long half-life ofintegrins, is the incubation time needed to measure differencesin integrin levels. The functionality of ERb was evaluatedmeasuring transcriptional activation of an ERE-luciferasereporter following 24 h or 1 week induction of ERb expressionand treatment with E2 or the ERb-specific agonist DPN(Fig. 1A). An ERb-dependent increase in transactivation wasfound, which was repressed by co-incubation with the anti-estrogen ICI 182 780 (ICI). Note that E2- and DPN-inducedtransactivation was lower after 1 week of induction, mostprobably due to lower levels of ERb in the system compared tothose at 24 h (S.1 and Fig. 1C). Expression of PR isoforms PRAand PRB, which are endogenous ER downstream targets, wasanalyzed. Following ERb expression and independently of theaddition of ligand, no significant differences were observed after24 h incubation while PRA was slightly decreased following1 week of ERb expression (Fig. 1B).

Interestingly, ERa protein and mRNA levels were reduced incells expressing ERb (Fig. 1C, lower part; mRNA not shown).Although this result may have implications on E2 signaling, themechanism remains to be studied.

Interestingly and as expected, ICI-induced ERadownregulation (Dauvois et al., 1992; Pink and Jordan, 1996),although it had no significant effect on ERb protein levels(Fig. 1C). The differential effects of ICI on ERa and ERb turnover have been reported in ERb transfected MCF-7 cells(Peekhaus et al., 2004) and were corroborated in MCF-7ERbcells (S.2). However, given the artificial nature of the expressionconstructs used in our study and the ones referred to above, itis premature to state that ICI does not affect the turn over ofendogenously expressed ERb.

In summary, since in T47-DERb cells, ERb is functional andcan be transcriptionally blocked by anti-estrogens, weproceeded to use this system to study effects of ERb oncell–ECM adhesion.

ERb increases cell surface expression of integrina1 and integrin b1

As a preliminary experiment, we performed an ELISA-basedintegrin screening, where expression of seven integrin alphasubunits (a1, a2, a3, a4, a5, aV, and aVb3) and seven integrinbeta subunits (b1, b2, b3, b4, b6, aVb5, and a5b1) wasanalyzed. The results indicated a possible upregulation ofintegrin a1 and integrin b1 by ERb, whereas levels of otherintegrins were not significantly changed (data not shown).These data were confirmed by flow cytometry, which showedthat expression of ERb induced an increase in surfaceexpression of integrin a1 and integrin b1 which was furtherenhanced by treatment with the ERb agonist DPN (Fig. 2A). Nosignificant effect was observed in surface levels of integrin a2(Fig. 2A), confirming the data from the ELISA-based integrin

Fig. 1. Expressionandfunctionalityof inducibleERb inT47-DERbcells.A:EREreportertransactivation.Cellsweretreatedfor24 hor1weekwith10 ng/ml (�ERb) or 0.01 ng/ml doxycycline (RERb), transiently transfected with a 3T ERE-LUC reporter plasmid and stimulated with 10 nM DPNor 10 nM E2, in the presence of 100 nM ICI or vehicle alone. All data were normalized to b-galactosidase enzyme activity. Experiments arerepresentative of three carried out in triplicates. Mean andSEM are shown (MP < 0.05 vs. EtOH aP < 0.05 vs. same treatment�ICI; bP < 0.05vs. sametreatment�ICI cP < 0.05 vs. same treatment�ICI). B: Progesterone receptor expression. Immunoblot analysis of cells treated for 24 h or 1 weekwith 10 ng/ml (�ERb) or 0.01 ng/ml doxycycline (RERb), in the presence or absence of 10 nM DPN alone or with 100 nM ICI. Results arerepresentative of three experiments. C: ERb and ERa protein expression. Immunoblot analysis of whole cell extract from cells treated for24 h or 1 week with 10 ng/ml (�ERb) or 0.01 ng/ml (RERb) doxycycline, in the presence or absence of 10 nM DPN or vehicle alone or with 100 nMICI. ERa protein expression (lower part), ERb protein expression (upper part). In all cases results are representative of three independentexperiments.


screening and indicating that the effects observed following ERbexpression are specific for some but not all integrins. Additionof the antagonist ICI completely repressed ERb effects onintegrina1 and partially reversed those observed on integrinb1(Fig. 2B). No significant changes were seen in the control cellline (T47-DPBI) under the same conditions (data not shown).As integrin a1 only exists as a heterodimer with integrin b1,these results indicate that expression of ERb in T47-DERb cellsresults in a specific increase of the cell surface ECM receptorintegrin a1b1.


ERb expression is associated to increased mRNAexpression of integrin a1 but not integrin b1

We next investigated if the increase of integrin a1b1 at the cellsurface could be through ERb effects on transcription.Quantitative real-time PCR was used to analyze integrina1 andintegrin b1 mRNA levels following up to 1 week of ERbexpression in T47-DERb cells. Induction of ERb expressionresulted in a time-dependent increase of integrin a1 mRNA(Fig. 3A), which was evident from 12 h and onwards and after

Fig. 2. ERb expression increases cell surface integrinsa1 andb1. Cell surface integrin expression analyzed by flow cytometry. Histograms showdistribution of fluorescence intensity of 2 T 104 T47-DERb cells after 1 week treatment with 10 ng/ml (�ERb), or 0.01 ng/ml (RERb)doxycycline inthepresenceorabsenceof10 nMDPN(A),andwithorwithout100 nMICI(B).Cellswere immunostainedwithantibodiesagainsta1integrin chain,b1 integrin chain, ora2 integrin chain. As a negative control secondary antibody alone (PE) was used. The bar graphs represent themean fold change and the data shown is representative of three experiments.


the induction of ERb expression (Fig. 3C). No change in integrinb1 mRNA level was observed (Fig. 3B). These results wereconfirmed in MCF-7ERb cells (S.3A), albeit to a lower extentwith regard to integrin a1. We speculate that the differenceobserved between the magnitude of integrin a1 induction inT47-DERb and MCF-7 ERb cells, respectively, is due to thelower ERb expression (S.3B) and higher ERa levels in MCF-7cells. The results were corroborated in a third cell line—SW480ERb—where integrin a1 mRNA also increased in cellsexpressing ERb compared to the mock control (not shown).Furthermore, integrin a1 mRNA increase in SW480ERb cellswas also corroborated by microarray analysis (Edvardsson,unpublished results).

In order to test that the induction of integrina1 transcriptionwas an ERb-mediated effect, the cells were co-incubated withICI for 24 h or 1 week, which partially reversed integrin a1mRNA upregulation (Fig. 3D,E, left part, respectively), while noeffect was detected regarding levels of integrin b1 mRNA(Fig. 3E, right part).

Since no significant effect on integrin b1 mRNA level wasobserved after DPN treatment and the potency of DPN onpromoter transactivation by ERb is lower than that of E2(Harrington et al., 2003), we measured integrin a1 andintegrin b1 mRNA levels in cells treated with the endogenousligand E2, obtaining similar results to those observed with DPN(Fig. 3F).


In conclusion, in T47-D breast cancer cells, the increase ofcell surface integrin b1 observed following ERb expression isnot due to a direct transcriptional event. On the other hand, theincrease in cell surface integrin a1 is most likely the result ofactivation of gene transcription.

Integrin a1 and integrin b1 protein levels increasefollowing long-term ERb expression

To further corroborate ERb induction of integrin a1b1,total protein levels were analyzed by immunoblot.Following 24 h, integrin b1 protein was expressed at lowlevels independently of ERb presence (Fig. 4A), whereasafter 1 week, integrin b1 was increased in cells expressingERb, but neither DPN nor E2 addition enhanced this effect(Fig. 4A,B, respectively). A time curve spanning from 24 h to1 week showed that integrin b1 total protein levels increasedsignificantly from 3 days and onwards (not shown).Furthermore, ERb induction of integrin b1 was partlyrepressed by ICI (Fig. 4C).

Several antibodies against integrin a1 were tested for use inWestern blot, but unfortunately, none was useful. Instead,integrin a1 was detected using immunofluorescence staining ofpermeabilized cells (Fig. 4D) and the digital data quantified usingan automated image analysis software. Adhesion complexeswere detected and counted on a per cell basis (Fig. 4E, top part),

Fig. 3. Integrina1butnotintegrinb1mRNAincreasesuponERbexpression.Quantitativereal-timePCRwascarriedoutonT47-DERbcells.A,B:Cells treatedfor3–24 hwith10 ng/ml(�ERb)or0.01 ng/mldoxycycline(RERb), inthepresenceorabsenceof10 nMDPN.Thecolorcodeappliestoparts A–E. C: ERb expression in T47-D ERb cells. D: Integrin a1 mRNA expression after 24 h treatment with 10 ng/ml (�ERb) or 0.01 ng/mldoxycycline (RERb), in the presence or absence of 10 nM DPN, with or without 100 nM ICI. E: Integrina1 and integrinb1 mRNA expression after1 week treatment with 10 or 0.01 ng/ml doxycycline, in the presence or absence of 10 nM DPN and with or without 100 nM ICI. F: Integrin a1 andintegrinb1mRNAexpressionafter1weektreatmentwith10or0.01 ng/mldoxycycline, inthepresenceorabsenceof10 nME2,andwithorwithout100 nM ICI. Data shown is representative of three experiments, carried out in triplicates. Mean and standard deviation are shown(MP < 0.05 vs. �ERb; aP < 0.05 vs. RERb; bP < 0.05 vs. RERb RDPN or RERb RE2).


where ERb expression caused a four fold increase in thenumber of adhesions detected per cell. ERb also significantlyenhanced integrin-clustering density within adhesions, asmeasured by mean fluorescence intensity per adhesion


(Fig. 4E, bottom part). Thus, it is reasonable to assume thatERb increases total cellular levels of integrin a1 as its mRNA(Fig. 3A,D–F) and membrane protein levels (Fig. 2A) increasedin the presence of ERb.

Fig. 4. Integrin a1b1 protein levels increase following ERb expression. A,B: Immunoblotting analysis of integrin b1 levels in whole cell extractfromT47-DERbcells.Cellsweretreatedfor24 hor1weekwith10 ng/ml(�ERb)or0.01 ng/ml(RERb)doxycycline, inthepresenceofvehiclealone,10 nMDPN(A)or10 nME2(B).C:Cellsweretreatedfor1weekwith10or0.01 ng/mldoxycycline, inthepresenceorabsenceof10 nMDPN,andwithorwithout100 nMICI.Resultsarerepresentativeof threeexperiments.D: Integrina1staining.Cellsweretreated for1weekwith10 ng/ml(�ERb)or0.01 ng/mldoxycycline(RERb), inthepresenceorabsenceof10 nMDPN.Thereafter,cellsweredetachedandallowedtoadheretocollagentypeIV coated coverslips. Following fixation, cells were permeabilized and immunostained for integrin a1. Scale bar, 10mm. E: Quantification ofintegrina1staining. Digital data was quantified usingan automated image analysis software. Adhesion complexeswere detected andcountedon aper cell basis; n U cell number (upper part), or integrin-clustering density within adhesions, as measured by mean fluorescence intensity peradhesion; n U adhesion number (bottom part). Mean and standard deviation are shown (MP < 0.05 vs. �ERb).


ERb induces formation of vinculin-containing focalcomplexes and actin filaments

To investigate if the increase in integrin a1b1 had any effect onthe formation of focal contacts, ERb was expressed inT47-DERb cells for 1 week with or without DPN andthereafter, cells were plated on collagen type IV coatedcoverslips and allowed to adhere for 1 h. An increase information of vinculin-containing focal complexes was observedupon ERb expression; addition of the ERb agonist DPN furtherincreased their size and number (Fig. 5A, upper row). In orderto study a possible regulation of vinculin expression by ERb,vinculin mRNA and protein were analyzed (Fig. 5B left and rightpart, respectively). No significant changes were observed. Thus,it is likely that pre-existing vinculin is recruited to ERb-inducedfocal complexes. Integrin activation also induces formation ofactin filaments, which are important for cells to adhere and


spread (Hynes, 2002). In the absence of ERb, the cells weresmaller and did not spread as much as compared to cellsexpressing ERb. Immunofluorescence staining of actin showedthat in ERb expressing cells, more actin filaments were formed(Fig. 5A, middle row), indicating that the cells had acquired amore adhesive potential. Whether ERb influences theexpression or activation of other proteins involved in theformation of focal complexes and focal adhesions remains to bestudied.

ERb enhances adhesion to extracellularmatrix proteins

We next investigated if ERb regulation of integrin a1b1 couldaffect adhesion of T47-DERb cells to different matrix proteins:collagen type I, collagen type IV, and laminin, which aresubstrates for integrin a1b1, and fibronectin and vitronectin

Fig. 5. ERb expression enhances formation of focal contacts and assembly of the actin cytoskeleton. Cells were treated for 1 week with 10 ng/ml(�ERb) or 0.01 ng/ml doxycycline (RERb), in the presence or absence of 10 nM DPN. A: immunostaining of cells attached to collagentype IV: vinculin (red), actin (green), and DNA (blue). Scale bar, 10mm. B: Vinculin expression in T47-DERb cells. Cells were treated for 24 h or1 week with 10 ng/ml (�ERb) or 0.01 ng/ml doxycycline (RERb), in the presence or absence of 10 nM DPN. Left part: real-time PCR analysis ofvinculin mRNA expression; right part: Western blot analysis of vinculin total protein levels.


which are substrates for integrin b1 partnered with other asubunits (Humphries et al., 2006). ERb expression increasedadhesion to all matrix proteins, with the most pronouncedeffect seen on laminin and collagen type IV. Addition of DPN orE2 (Fig. 6A, B, respectively) in the presence of ERb, slightly, butsignificantly, increased the adhesion to all matrix proteins,except vitronectin. In the absence of ERb, a significant decreasein adhesion to collagen type I, fibronectin, and vitronectin wasobserved following DPN treatment (Fig. 6A). This is mostprobably due to the fact that T47-D cells express high ERalevels and that DPN can also bind to ERa, albeit with muchlower affinity (Meyers et al., 2001). This was corroborated byincubation of the cells with E2, which in the absence of ERb also


decreased binding to these matrices (Fig. 6B). The effectsobserved were not due to doxycycline itself, as adhesion of theT47-DPBI (mock) cells was not affected (S.4).

In addition to integrin expression, adhesion to ECMsubstrates depends on expression and activation of manyproteins which can be dependent on the cell type and context.Thus, estrogen effects on adhesion of MCF-7ERb andHT29ERb cells were studied (S.5). Following ERb expression,adhesion of MCF-7ERb cells to fibronectin and lamininincreased, an effect further enhanced by E2. However, bindingto collagen type I and collagen type IV was increased by E2independently of ERb expression, indicating that in MCF-7 cellsregulation of adhesion to collagens by ERa is different from that

Fig. 6. Effects of ERbon cell adhesion to extracellular matrix proteins and migration. A, B: Cells were treated for 1 week with 10 ng/ml (�ERb)or0.01 ng/mldoxycycline(RERb), inthepresenceorabsenceof10 nMDPN(A)or10 nME2(B).Foldchangeisshown,whereadhesionofcellsculturedin the presence of 10 ng/ml doxycycline (�ERb) was set to 1. Adhesion to BSA was used as a control for unspecific binding. Experiments arerepresentative of three. Mean and standard deviation are shown (MP < 0.05 vs. �ERb; aP < 0.05 vs. RERb; bP < 0.05 vs. �ERb).


observed in T47-DERb cells. Expression of ERb in HT29ERbcells enhanced adhesion to all ECM proteins in a similar way as inT47-DERb cells. However, in collagen type I, collagen type IV,and fibronectin, E2 partially reversed this effect.

From these experiments, we conclude that effects of ERaand ERb on adhesion are specific for each ECM protein and arecell type specific. However, adhesion to laminin was repeatedlyincreased following ERb expression in the three cancer celllines studied.

ERb expression reduces cell migration

Due to the ERb-induced changes in expression of integrina1b1, formation of vinculin-containing focal complexes andeffects on adhesion to ECM substrates, we examined influenceof ERb on migration of T47-D cells. ERb was expressed for1 week with or without DPN or E2 and a confluent layer of cellswas disrupted by scratching a line with a pipette tip. The woundwas measured at 0 and 24 h. Expression of ERb and its activationby DPN treatment significantly reduced migration. Similarresults were seen following E2 incubation (Fig. 7A, left part).Interestingly, when cells were treated with the scatterfactor EGF or FGF, expression of ERb alone decreasedmigration, an effect enhanced by addition of ligand (Fig. 7B, rightpart; FGF not shown). Representative pictures of one scratchquantified for the bar graph are shown (Fig. 7B). The effect ofERb expression on cell migration was corroborated with achemotactic assay (Fig. 7C), showing reduced cellmigration when ERb was expressed both in the absence orpresence of ligand. These results indicate a correlation betweenERb-induced integrin a1b1 expression and inhibition ofmigration.


Discussion

Earlier studies indicate that estrogens can affect integrinexpression. Two putative ERE half sites and several AP-1 siteshave been identified in the promoter region of human integrina2 (Zutter et al., 1994) and an increase in integrin a5, integrina6, and integrin b1 mRNA was observed in endothelial cellsafter estrogen treatment (Cid et al., 1999). Few studies haveaddressed estrogen effects on integrin expression in breastcancer cells; however, these have not taken into account theopposing effects displayed by ERa and ERb. In the presentstudy, we show that ERb has an important role as modulator ofcell adhesion. Specifically, expression of ERb in T47-DERbbreast cancer cells increased cell surface levels of integrina1b1.In addition, we show that ERb expression enhanced adhesion toECM proteins—mainly laminin—and reduced migration ofcells.

The regulatory mechanisms behind the increase in integrina1b1 were specific for each integrin subunit. Integrina1 mRNAwas upregulated 6 h after the initial appearance of ERb mRNAand could be partly reversed by the antagonist ICI, indicatingthat ERb regulates its transcription through the nuclearreceptor pathway. Screening of the integrin a1 promoterregion spanning 10 kb, showed one ERE, one SP1, and six AP1sites, which could be possible sites subjected to ER regulation(data not shown). Whether ERb itself affects any of theseregulatory elements or whether it inhibits a possible repressiveERa function at some of these sites—as previously reportedfor AP1 sites (Kushner et al., 2000)—remains to be studied.However, the observation that ICI treatment (Fig. 3D)—whichdownregulates ERa protein—decreased rather than increasedintegrin a1 mRNA levels, suggests that upregulation of integrina1 mRNA expression is likely to be achieved by ERb. On the

Fig. 7. Effects of ERb on cell migration. A: Wound assay, (% wound closure) was measured on cells treated for 1 week with 10 ng/ml (�ERb) or0.01 ng/mldoxycycline(RERb), inthepresenceorabsenceof10 nMDPNor10 nME2,andwithorwithout50 ng/mlEGF.Resultsarerepresentativeof three independent experiments. Mean and SEM is shown (aP < 0.05 vs. �ERb; MP < 0.05 vs. �ERbR DPN). Note that during the 24 h incubationperiod used to measure wound filling no differences were observed in T47-DERb cell number (data not shown). B: Representative figure of thewound measured at 0 and 24 h. C: Transwell migration assay. ERbwas induced for 1 week and cells grown with or without 10 nM DPN, whereafterthe assay was carried out as described in Materials and Methods Section.


other hand, ERb expression did not affect integrin b1 mRNAlevels even though integrin b1 protein levels increased. Sinceupregulation of integrin b1 protein occurred after 3 days ofinduction of ERb and since integrin a1 (which is upregulatedafter 12 h) only heterodimerizes with integrin b1, it is possiblethat integrin b1 protein becomes stabilized uponheterodimerization with integrin a1 and accumulates. It is alsopossible that ERbmodulates integrinb1 expression by affectingproteins that bind to its intracellular domain, thereby activatingand stabilizing it. In both cases integrin b1 would be less proneto degradation.

Studies in mice have shown that integrin b1 is critical for thealveolar morphogenesis and functional differentiation of themammary epithelium (Faraldo et al., 2000; Naylor et al., 2005).Moreover, re-expression of the a2b1 integrin dimer intoMm5MT mouse mammary cancer cells restored their ability todifferentiate into gland-like structures in three-dimensionalmatrices and markedly reduced the in vivo tumorigenicity ofthese cells (Zutter et al., 1995). However, in MCF-7 humanbreast cancer cells grown on reconstituted basementmembrane, inhibition of integrin b1 resulted in a decrease inproliferation and increase in apoptosis (Park et al., 2006).Integrin a1 is expressed in the myoepithelial cells of the normalbreast (Gould et al., 1990) and is also detected in ducts of thepregnant breast (Suzuki et al., 2002). Integrin a1 is expressed ata low level in human breast cancer cell lines (Shaw, 1999) and isdownregulated in breast and ovarian cancer (Su et al., 2001)suggesting that its loss might be associated to malignancy.Furthermore, integrin a1b1 can negatively regulate EGFR


signaling in PC3 prostate cancer cells, HeLa cervical cancer cells(Mattila et al., 2005), and mesangial cells (Chen et al., 2007). Inthis context, it is worth mentioning that in the present work, weshowed that ERb expression—which upregulates integrina1b1— inhibited EGF-induced migration.

ERb expression in T47-DERb cells resulted in increased sizeand number of vinculin containing focal complexes andformation of actin fibers indicating that the cells had acquired amore adhesive potential, which was more markedly observedon collagen type IV and laminin, an effect also seen in endothelialcells treated with estrogen (Cid et al., 1999). However, onlyincreased adhesion to laminin was consistently observed in allcell lines tested in our study.

In this work, we have focused our studies on ERb regulationof integrin a1 and integrin b1, but the contribution of otherintegrins and associated proteins to increase adhesion cannotbe excluded. Particularly, increased adhesion to fibronectin andvitronectin, which are not substrates for integrin a1b1heterodimers, indicates that ERb could have an effect on earlysignaling events involved in cell adhesion, such as activation ofERK, FAK, and RhoGTPases, which is the focus of our currentstudies.

The rate of cell migration is determined by the level ofadhesiveness of the cells to their substrate (DiMilla et al., 1991,1993). ERb decreased the migratory potential, which is inagreement with our previous observations in mouse mammaryepithelial cells (Helguero et al., 2008).

It has been shown that in T47-D cells, an increase of PRA/PRBratios (due to increased expression of PRA) to levels higher


than 4:1, results in decreased adhesion and focal contactformation, but for this to occur, it is necessary to treat the cellswith progestins for a period of at least 48 h (Graham et al.,2005). In line with this, upon ERb expression, we have observeda decrease of PRA/PRB ratios of 25% and therefore, somelong-term effects of ERb expression such as increased integrinb1 protein levels or adhesive potential of T47-DERb cells, maywell in part be the result of these changes in PR isoform ratio.Nevertheless, it is worth mentioning that in our experimentalsetup, the cells were not incubated with progestins as was donein the study referred to above. One interesting correlation isthat breast cancer patients with node-positive disease and aprimary tumor expressing higher PRA than PRB, are likely torespond more poorly to endocrine therapy than those with ahigher expression of PRB (Hopp et al., 2004), and that ERb,which negatively regulates PRA in T47-D cells, is a goodprognostic factor for the response to endocrine treatment(Honma et al., 2008).

Some effects of ERb reported in this study were ligandindependent, while others were enhanced by ligand addition,that is, mixed effects (Table 1). Ligand-independent effects wereobserved at the gene expression and total protein levels, but thecell surface amounts of integrin a1b1 and vinculin focal contactformation were enhanced by ligand addition. Furthermore,adhesion to different ECM proteins was clearly regulated byboth E2 and DPN.

Ligand-independent effects of ERb were initially reported byHall and McDonnell (1999) who showed that the ERb-AF-2domain functions independently within the receptor and thatERb interacts with target gene promoters in a ligand-independent manner. Since then, this type of effect at differentcellular levels have been observed by us and others (Lazennecet al., 2001; Strom et al., 2004). In this particular work, onepossibility could be that squelching may play a role, but the factthat ICI blocked the ligand-independent effects on integrin a1transcription, while ERbwas still highly expressed in the systemsuggests that this is not the case. The mechanisms behind theligand-independent effects of ERb need further studies.Alternatively, other growth factor activated pathways might beimportant for maintaining membrane stability of adhesionproteins. Recently, we found that in mammary epithelial cells,ERb affects E-cadherin membrane stability and subcellularlocalization in a ligand-dependent and- independent manner(Helguero et al., 2008). Thus, we do not rule out the possibilitythat a similar mechanism could be regulating other membraneproteins.

The finding that ERb expression negatively influences ERalevels is in agreement with the idea that the relative ratio ofthese receptors dictates the response to estrogens andindicates that some effects such as changes in PRA/PRB ratio

TABLE 1. Summary of various effects observed following ERb expression

as to ligand dependency

Effect Independent

Mixed (partiallyligand

dependent)

PRA downregulation XERa downregulation XIncreased integrin a1 mRNA XIncreased integrin a1 total cellular protein XIncreased integrin b1 total cellular protein XIncreased surface integrin a1 XIncreased surface integrin b1 XIncreased formation of FC XModulation of adhesion to ECM substrates XMigrationa X

FC, focal contact; ECM, extracellular matrix.aMeasured by wound assay, the effects were only ligand dependent, on the other hand,transwell assays showed ligand-independent effects.


and cell adhesion to ECM proteins may be influenced byERa/ERb levels.

Finally, in this work we have shown that ERb upregulatesintegrina1b1; consequently it enhances adhesion to some ECMproteins and inhibits migration of breast cancer cells. Theseresults indicate that ERb is a prospective prognostic marker inbreast cancer with potential as a drug target in future therapieswhere the expression and/or activation of ERb could be analternative for preventing invasion of breast cancer cells.

Acknowledgments

We are grateful to Dr. Johan Hartman for help withimmunofluorescence experiments, Ms. Karin Edvardsson forintegrin a1 measurement in SW480ERb cells and Dr. StaffanStromblad for good advice throughout the development of thismanuscript. This study was supported by grants from TheSwedish Cancer Fund, KaroBio AB, David and Astrid HagelensPostdoctoral fellowship to LAH and Lars-Hiertas MinneFoundation.

Literature Cited

Aplin AE, Howe A, Alahari SK, Juliano RL. 1998. Signal transduction and signal modulation bycell adhesion receptors: The role of integrins, cadherins, immunoglobulin-cell adhesionmolecules, and selectins. Pharmacol Rev 50:197–263.

Chao C, Lotz MM, Clarke AC, Mercurio AM. 1996. A function for the integrin alpha6beta4 inthe invasive properties of colorectal carcinoma cells. Cancer Res 56:4811–4819.

Chen X, Abair TD, Ibanez MR, Su Y, Frey MR, Dise RS, Polk DB, Singh AB, Harris RC, Zent R,Pozzi A. 2007. Integrin alpha1beta1 controls reactive oxygen species synthesis bynegatively regulating epidermal growth factor receptor-mediated Rac activation. Mol CellBiol 27:3313–3326.

Cheng J, Lee EJ, Madison LD, Lazennec G. 2004. Expression of estrogen receptor beta inprostate carcinoma cells inhibits invasion and proliferation and triggers apoptosis. FEBSLett 566:169–172.

Cid MC, Esparza J, Schnaper HW, Juan M, Yague J, Grant DS, Urbano-Marquez A, HoffmanGS, Kleinman HK. 1999. Estradiol enhances endothelial cell interactions with extracellularmatrix proteins via an increase in integrin expression and function. Angiogenesis 3:271–280.

Dauvois S, Danielian PS, White R, Parker MG. 1992. Antiestrogen ICI 164,384 reducescellular estrogen receptor content by increasing its turnover. Proc Natl Acad Sci USA89:4037–4041.

DiMilla PA, Barbee K, Lauffenburger DA. 1991. Mathematical model for the effects ofadhesion and mechanics on cell migration speed. Biophys J 60:15–37.

DiMilla PA, Stone JA, Quinn JA, Albelda SM, Lauffenburger DA. 1993. Maximal migration ofhuman smooth muscle cells on fibronectin and type IV collagen occurs at an intermediateattachment strength. J Cell Biol 122:729–737.

Faraldo MM, Deugnier MA, Thiery JP, Glukhova MA. 2000. Development of mammary glandrequires normal beta 1-integrin function. Adv Exp Med Biol 480:169–174.

Ferlay J, Autier P, Boniol M, Heanue M, Colombet M, Boyle P. 2007. Estimates of the cancerincidence and mortality in Europe in2006. Ann Oncol 18:581–592.

Forster C, Makela S, Warri A, Kietz S, Becker D, Hultenby K, Warner M, Gustafsson JA. 2002.Involvement of estrogen receptor beta in terminal differentiation of mammary glandepithelium. Proc Natl Acad Sci USA 99:15578–15583.

Frey MR, Golovin A, Polk DB. 2004. Epidermal growth factor-stimulated intestinal epithelialcell migration requires Src family kinase-dependent p38 MAPK signaling. J Biol Chem279:44513–44521.

Gould VE, Koukoulis GK, Virtanen I. 1990. Extracellular matrix proteins and their receptorsin the normal, hyperplastic and neoplastic breast. Cell Differ Dev 32:409–416.

Graham JD, Yager ML, Hill HD, Byth K, O’Neill GM, Clarke CL. 2005. Altered progesteronereceptor isoform expression remodels progestin responsiveness of breast cancer cells.Mol Endocrinol 19:2713–2735.

Gui GP, Wells CA, Browne PD, Yeomans P, Jordan S, Puddefoot JR, Vinson GP, Carpenter R.1995. Integrin expression in primary breast cancer and its relation to axillary nodal status.Surgery 117:102–108.

Guo W, Pylayeva Y, Pepe A, Yoshioka T, Muller WJ, Inghirami G, Giancotti FG. 2006. Beta 4integrin amplifies ErbB2 signaling to promote mammary tumorigenesis. Cell 126:489–502.

Gustafsson JA. 2000. An update on estrogen receptors. Semin Perinatol 24:66–69.Hall JM, McDonnell DP. 1999. The estrogen receptor beta-isoform (ERbeta) of the human

estrogen receptor modulates ERalpha transcriptional activity and is a key regulator of thecellular response to estrogens and antiestrogens. Endocrinology 140:5566–5578.

Harrington WR, Sheng S, Barnett DH, Petz LN, Katzenellenbogen JA, Katzenellenbogen BS.2003. Activities of estrogen receptor alpha- and beta-selective ligands at diverse estrogenresponsive gene sites mediating transactivation or transrepression. Mol Cell Endocrinol206:13–22.

Hartman J, Lindberg K, Morani A, Inzunza J, Strom A, Gustafsson JA. 2006. Estrogen receptorbeta inhibits angiogenesis and growth of T47D breast cancer xenografts. Cancer Res66:11207–11213.

Helguero LA, Faulds MH, Gustafsson JA, Haldosen LA. 2005. Estrogen receptors alfa(ERalpha) and beta (ERbeta) differentially regulate proliferation and apoptosis of thenormal murine mammary epithelial cell line HC11. Oncogene 24:6605–6616.

Helguero LA, Lindberg K, Gardmo C, Schwend T, Gustafsson JA, Haldosen LA. 2008.Different roles of estrogen receptors alpha and beta in the regulation of E-cadherin proteinlevels in a mouse mammary epithelial cell line. Cancer Res 68:8695–8704.

Honma N, Horii R, Iwase T, Saji S, Younes M, Takubo K, Matsuura M, Ito Y, Akiyama F,Sakamoto G. 2008. Clinical importance of estrogen receptor-beta evaluation in breastcancer patients treated with adjuvant tamoxifen therapy. J Clin Oncol 26:3727–3734.


Hopp TA, Weiss HL, Hilsenbeck SG, Cui Y, Allred DC, Horwitz KB, Fuqua SA. 2004. Breastcancer patients with progesterone receptor PR-A-rich tumors have poorer disease-freesurvival rates. Clin Cancer Res 10:2751–2760.

Humphries JD, Byron A, Humphries MJ. 2006. Integrin ligands at a glance.J Cell Sci 119:3901–3903.

Hynes RO. 1987. Integrins: A family of cell surface receptors. Cell 48:549–554.Hynes RO. 1992. Integrins: Versatility, modulation, and signaling in cell adhesion.

Cell 69:11–25.Hynes RO. 2002. Integrins: Bidirectional, allosteric signaling machines. Cell 110:673–687.Iwao K, Miyoshi Y, Egawa C, Ikeda N, Noguchi S. 2000. Quantitative analysis of estrogen

receptor-beta mRNA and its variants in human breast cancers. Int J Cancer 88:733–736.Jarvinen TA, Pelto-Huikko M, Holli K, Isola J. 2000. Estrogen receptor beta is coexpressed

with ERalpha and PR and associated with nodal status, grade, and proliferation rate inbreast cancer. Am J Pathol 156:29–35.

Koehler KF, Helguero LA, Haldosen LA, Warner M, Gustafsson JA. 2005. Reflections on thediscovery and significance of estrogen receptor beta. Endocr Rev 26:465–478.

Kushner PJ, Agard DA, Greene GL, Scanlan TS, Shiau AK, Uht RM, Webb P. 2000. Estrogenreceptor pathways to AP-1. J Steroid Biochem Mol Biol 74:311–317.

Lazennec G, Bresson D, Lucas A, Chauveau C, Vignon F. 2001. ER beta inhibits proliferationand invasion of breast cancer cells. Endocrinology 142:4120–4130.

Leygue E, Dotzlaw H, Watson PH, Murphy LC. 1998. Altered estrogen receptor alpha andbeta messenger RNA expression during human breast tumorigenesis. Cancer Res58:3197–3201.

Lipscomb EA, Simpson KJ, Lyle SR, Ring JE, Dugan AS, Mercurio AM. 2005. The alpha6beta4integrin maintains the survival of human breast carcinoma cells in vivo. Cancer Res65:10970–10976.

Mattila E, Pellinen T, Nevo J, Vuoriluoto K, Arjonen A, Ivaska J. 2005. Negative regulation ofEGFR signalling through integrin-alpha1beta1-mediated activation of protein tyrosinephosphatase TCPTP. Nat Cell Biol 7:78–85.

Meyers MJ, Sun J, Carlson KE, Marriner GA, Katzenellenbogen BS, Katzenellenbogen JA.2001. Estrogen receptor-beta potency-selective ligands: Structure-activity relationshipstudies of diarylpropionitriles and their acetylene and polar analogues. J Med Chem44:4230–4251.

Muller P, Kietz S, Gustafsson JA, Strom A. 2002. The anti-estrogenic effect of all-trans-retinoic acid on the breast cancer cell line MCF-7 is dependent on HES-1 expression. J BiolChem 277:28376–28379.

Natali PG, Nicotra MR, Botti C, Mottolese M, Bigotti A, Segatto O. 1992. Changes inexpression of alpha 6/beta 4 integrin heterodimer in primary and metastatic breast cancer.Br J Cancer 66:318–322.

Naylor MJ, Li N, Cheung J, Lowe ET, Lambert E, Marlow R, Wang P, Schatzmann F,Wintermantel T, Schuetz G, Clarke AR, Mueller U, Hynes NE, Streuli CH. 2005. Ablationof beta1 integrin in mammary epithelium reveals a key role for integrin in glandularmorphogenesis and differentiation. J Cell Biol 171:717–728.

Nilsson M, Dahlman-Wright K, Gustafsson JA. 2004. Nuclear receptors in disease: Theoestrogen receptors. Essays Biochem 40:157–167.

Novelli F, Milella M, Melucci E, Di Benedetto A, Sperduti I, Perrone-Donnorso R, PerracchioL, Venturo I, Nistico C, Fabi A, Buglioni S, Natali PG, Mottolese M. 2008. A divergent rolefor estrogen receptor-beta in node-positive and node-negative breast cancer classified


according to molecular subtypes: An observational prospective study. Breast Cancer Res10:R74.

Osborne CK. 1998. Steroid hormone receptors in breast cancer management. Breast CancerRes Treat 51:227–238.

Park CC, Zhang H, Pallavicini M, Gray JW, Baehner F, Park CJ, Bissell MJ. 2006. Beta1 integrininhibitory antibody induces apoptosis of breast cancer cells, inhibits growth, anddistinguishes malignant from normal phenotype in three dimensional cultures and in vivo.Cancer Res 66:1526–1535.

Parkin DM, Bray F, Ferlay J, Pisani P. 2005. Global cancer statistics, 2002. CA Cancer J Clin55:74–108.

Peekhaus NT, Chang T, Hayes EC, Wilkinson HA, Mitra SW, Schaeffer JM, Rohrer SP. 2004.Distinct effects of the antiestrogen Faslodex on the stability of estrogen receptors-alphaand -beta in the breast cancer cell line MCF-7. J Mol Endocrinol 32:987–995.

Pignatelli M, Stamp G. 1995. Integrins in tumour development and spread. Cancer Surv24:113–127.

Pink JJ, Jordan VC. 1996. Models of estrogen receptor regulation by estrogens andantiestrogens in breast cancer cell lines. Cancer Res 56:2321–2330.

Roger P, Sahla ME, Makela S, Gustafsson JA, Baldet P, Rochefort H. 2001. Decreasedexpression of estrogen receptor beta protein in proliferative preinvasive mammarytumors. Cancer Res 61:2537–2541.

Rossen K, Dahlstrom KK, Mercurio AM, Wewer UM. 1994. Expression of the alpha 6 beta 4integrin by squamous cell carcinomas and basal cell carcinomas: Possible relation toinvasive potential? Acta Derm Venereol 74:101–105.

Shaw LM. 1999. Integrin function in breast carcinoma progression. J Mammary Gland BiolNeoplasia 4:367–376.

Speirs V, Carder PJ, Lane S, Dodwell D, Lansdown MR, Hanby AM. 2004. Oestrogen receptorbeta: What it means for patients with breast cancer. Lancet Oncol 5:174–181.

Strom A, Hartman J, Foster JS, Kietz S, Wimalasena J, Gustafsson JA. 2004. Estrogen receptorbeta inhibits 17beta-estradiol-stimulated proliferation of the breast cancer cell line T47D.Proc Natl Acad Sci USA 101:1566–1571.

Su AI, Welsh JB, Sapinoso LM, Kern SG, Dimitrov P, Lapp H, Schultz PG, Powell SM, MoskalukCA, Frierson HF Jr, Hampton GM. 2001. Molecular classification of human carcinomas byuse of gene expression signatures. Cancer Res 61:7388–7393.

Subauste MC, Pertz O, Adamson ED, Turner CE, Junger S, Hahn KM. 2004. Vinculinmodulation of paxillin-FAK interactions regulates ERK to control survival and motility.J Cell Biol 165:371–381.

Suzuki RN, Entwistle A, Atherton AJ, Clarke C, Lakhani SR, O’Hare MJ. 2002. The expressionpatterns of integrin subunits on human breast tissues obtained during pregnancy. Cell BiolInt 26:593–598.

Xu W, Baribault H, Adamson ED. 1998. Vinculin knockout results in heart and brain defectsduring embryonic development. Development 125:327–337.

Zutter MM, Santoro SA, Painter AS, Tsung YL, Gafford A. 1994. The human alpha 2 integringene promoter. Identification of positive and negative regulatory elements important forcell-type and developmentally restricted gene expression. J Biol Chem269:463–469.

Zutter MM, Santoro SA, Staatz WD, Tsung YL. 1995. Re-expression of the alpha 2 beta 1integrin abrogates the malignant phenotype of breast carcinoma cells. Proc Natl Acad SciUSA 92:7411–7415.

Expression of estrogen receptor β increases integrin α1 and integrin β1 levels and enhances...

Documents

Transcript of Expression of estrogen receptor β increases integrin α1 and integrin β1 levels and enhances...