Opposite effects of estrogen receptors alpha and beta on MCF-7 ...

Oncogene (2005) 24, 4789–4798

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

Opposite effects of estrogen receptors alpha and beta on MCF-7 sensitivity to the cytotoxic action of TNF and p53 activity Sebastian A Lewandowski1,4, Jerome Thiery2, Abdelali Jalil2, Guy Leclercq3, Cezary Szczylik1 and Salem Chouaib*,2 1 Department of Oncology, Military Institute of Medicine, Szaserow 128 Street, 00-909 Warsaw, Poland; 2INSERM U487, ‘Cytokines et Immunologie des Tumeurs Humaines’ Institut Gustave Roussy, 94805 Villejuif Cedex, France; 3Laboratoire J-C Heuson de Cance´rologie Mammaire, Service de Me´decine Interne, Institut Jules Bordet, Universite´ Libre de Bruxelles, 1 rue He´ger-Bordet, B-1000, Brussels, Belgium; 4Postgraduate School of Molecular Medicine, 3 Pasteur Street, 02-093 Warsaw, Poland

We have investigated the effect of estrogen on p53 cellular location and its influence on tumor cell susceptibility to tumor necrosis factor (TNF)-mediated cytotoxic action. For this purpose, we have used the TNF-sensitive human breast adenocarcinoma MCF-7 and its derivative, the TNF-resistant 1001 clone. Our data indicate that although estrogen receptor (ER)a is present in both cell lines, estrogen treatment (1 108 M) has an influence only on the MCF-7 cells and protects these cells from the TNF cytotoxicity. This protective effect is associated with translocation of p53 from the nucleus to the cytoplasm in p53 wild-type MCF-7 and not in p53-mutated 1001 cells. The translocation of p53 in MCF-7 cells results in a decrease in its transcriptional activity, as revealed by diminished p21WAF1/CIP1 induction and an altered ratio of Bax and Bcl-2 proteins. The estrogen-induced effects are reversed by the selective estrogen inhibitor 182, 780 (1 106 M). Interestingly, transient transfection of MCF-7 cells with ERb but not ERa cDNA encoding plasmid results in retention of p53 in the nucleus, a subsequent potentiation of its transcriptional activity, and in an increased MCF-7 sensitivity to TNF. The estrogen effects on p53 location and transcriptional activity may involve the mdm2 protein since both events were reversed following MCF-7 transfection with plasmid encoding the ARF cDNA. These studies suggest that estrogen-induced MCF-7 cell survival in the presence of TNF requires a transcriptionally active p53 and, more importantly, indicate that introduction of ERb can attenuate the estrogen effects on the p53 protein location, its transcriptional activity and also results in a potentiation of cell sensitivity to TNF-mediated cell death. Oncogene (2005) 24, 4789–4798. doi:10.1038/sj.onc.1208595; published online 2 May 2005 Keywords: estrogen receptor; p53; cell death

*Correspondence: S Chouaib; E-mail: [email protected] Received 16 July 2004; revised 4 February 2005; accepted 8 February 2005; published online 2 May 2005

Introduction It is well established that a better understanding of the interplay between tumor-associated proapoptotic and antiapoptotic pathways may offer novel modalities of manipulating tumor cell growth and thereby improving the effectiveness of cancer treatment. Estrogen (17b-estradiol), apart from its role in reproduction, is also known to regulate the growth of hormoneresponsive tissues, eventually leading to carcinogenous progression (Sommer and Fuqua, 2001). Clinical and epidemiological studies correlate the prolonged estrogen treatment of post-menopausal women with an increase rate of breast cancer occurrences (Rossouw et al., 2002). Besides growth stimulation, estrogen can also exert protective effect on breast cancer cells undergoing stress stimuli such as UV, taxol (Razandi et al., 2000) oxidative stress (Sudoh et al., 2001; Kanda and Watanabe, 2003) or tumor necrosis factor (TNF) (Burow et al., 2001). There are two distinct types of estrogen receptor (ER), the ERa and ERb (Green et al., 1986; Mosselman et al., 1996; Nilsson et al., 2001). Both share similar domain design, estrogen-binding affinity (Kuiper et al., 1997) and recognize the same DNA sequence, the estrogen response element (ERE) (Klein-Hitpass et al., 1986) in promoters of estrogen responsive genes (Pace et al., 1997; Hyder et al., 1999; Hall and Korach, 2002). On the other hand, the receptors have diverse expression patterns and exert dissimilar transcriptional activities in response to estrogen (Hall and McDonnell, 1999; Kuiper et al., 1997). The knockout mouse models for ERa and ERb further support the view on diverse actions of both receptors. While the ductal growth of mammary gland is reduced in ERa knockout mice (Mueller et al., 2002), it remains unaffected in ERb knockouts (Krege et al., 1998). Both receptors have also been shown to have different roles in the proliferation of the normal mammary gland (Cheng et al., 2004). The precise roles of ERa and ERb on the proliferation and protection are still unknown. They have been so far shown to exhibit unique properties on the cell cycle progression, activation of the mitogen-activated protein kinase pathway and protein kinase C expression (Burow

Effects of ERa and ERb on p53 activity and TNF response SA Lewandowski et al

4790

et al., 2000; Wade et al., 2001; Liu et al., 2002; Paruthiyil et al., 2004; Strom et al., 2004). The p53 protein is one of the most extensively studied tumor suppressors that promotes DNA repair, cell cycle arrest and apoptosis (Vousden and Lu, 2002). There are two most prominent features of p53. Firstly, it arrests the cell cycle by the induction of p21 (WAF1/CIP1) gene. Secondly, it controls two types of the apoptotic pathways: the extrinsic, based on cell surface receptors (Owen-Schaub et al., 1995; Wu et al., 1997; Muller et al., 1998), and the intrinsic, which centers on the mitochondrial proteins from the Bcl-2 family (Miyashita et al., 1994b; Kroemer and Reed, 2000). Acquisition of somatic p53 mutation results in a decreased sensitivity to cytotoxic drugs and host immune responses. Recently, it has been reported that estrogen was able to alter p53 location and modulate the p53 pathway in ERpositive cells (Molinari et al., 2000; Kato et al., 2002), indicating that estrogen has an ability to modulate p53 action. However, the mechanisms by which ERs mediate the protective effect remains unknown, yet is of great importance for improving breast cancer cytotoxic treatment. These studies were designed to better understand the functional interaction between estrogen, the tumor suppressor p53 protein and the influence of such interaction on the cytotoxic action of TNF. The latter has a powerful direct tumor-killing capability (Chouaib et al., 1991) and has been shown to play a role in tumor regression mediated by killer cells. TNF is released by NK cells, cytotoxic T lymphocytes (CTLs), and significantly contributes to the local immune response to the tumor. Thus, when its secretion is confined to the area of tumor growth, TNF may fulfill its promise as an anticancer agent. Previously, we have provided evidence for the association between the resistance of human breast adenocarcinoma cell line MCF-7 to the cytotoxic actions of TNF and the loss of p53 function (Cai et al., 1997b). This insensitivity can be reversed by the reintroduction of wild-type, transcriptionally active p53 (Ameyar et al., 1998). Here we describe the protective effect of estrogen on ERa-positive MCF-7 cells on TNF-mediated cytotoxicity and its association with altered cellular localization and transcriptional activity of p53. Interestingly, those properties can be reversed not only with antiestrogen ICI 182, 780 treatment but also by the introduction of ERb cDNA.

Results The differential response to estrogen in the presence of TNF is not related to altered expression of ERa by MCF-7 and 1001 cells We have first investigated the effect of estrogen on the TNF cytotoxic action in the TNF-sensitive MCF-7 and TNF-resistant derivative 1001 cells. Data depicted in Figure 1a indicate that estrogen-deprived MCF-7 cells were sensitive to TNF, while estrogen treatment resulted in resistance of MCF-7 cells to the cytotoxic action of Oncogene

Figure 1 The differential response to estrogen in the presence of TNF is not related to altered expression of ERa. (a) Estrogendeprived MCF-7 (’) and 1001 (&) cells were treated with DMSO (0.01%), estrogen (E2) (1 108 M) and ICI 182, 780 (1 106 M) for 24 h and exposed to 100 ng/ml of TNF. Cell viability was measured using the viability assay as described in Materials and methods. Data presented are the mean7s.d. of triplicate determinations. The comparator treatment for both MCF-7 and 1001 cells was 0 ng/ml TNF without steroid treatment (DMSO). The level of significance was Po0.004 (*) for compared DMSO vs E2 and E2 vs E2 þ ICI groups in MCF-7 cells as determined by two-tailed Student’s t-test. (b). Expression of ERa in MCF-7 and 1001 cells. In all, 50 mg of cell lysate was analysed using Western blotting for ERa expression as described in Materials and methods. Similar results were obtained in three independent experiments

TNF, similar to the 1001 clone. The protective effect of estrogen was reversed by co-treatment with specific inhibitor ICI 182, 780. Results from cells treated only with dimethylsulfoxide (DMSO) were used as point of reference for each cell line. Estrogen stimulation of starved cells in the absence of TNF has significantly increased the viability of only MCF-7 cells (data not shown). The level of significance for the MCF-7 TNF sensitivity test of DMSO- vs E2- and E2- vs E2 þ ICItreated groups was Po0.004, as determined by twotailed Student’s t-test. We have next examined if the differential responsiveness to estrogen was due to an altered expression of ERa in MCF-7 and 1001 cells. As shown in Figure 1b, Western blotting analysis indicates a comparable ERa expression by both cell lines, suggesting that the lack of response to estrogen in TNF-treated 1001 cells is not associated with altered ERa expression. Exogenous estrogen stimulates the translocation of p53 protein in MCF-7 but not in 1001 cells, and results in a decrease in its transcriptional activity It has been previously reported that MCF-7 cells stimulated with estrogen displayed an export of the p53 protein from the nucleus to the cytoplasm. We have


4791

thus asked if this effect might occur in the TNF-resistant clone 1001 carrying a mutated p53. When cultured in FBS supplemented media, MCF-7 cells show p53 mostly positioned in the cytoplasm (data not shown). However, after estrogen deprivation p53 is mostly accumulated in the nucleus. Following estrogen stimulation, MCF-7 cells exhibit a substantial change by promoting p53 export from the nucleus to the cytoplasm. The later event was not observed in 1001 cells. Co-treatment with both estrogen and estrogen inhibitor ICI 182, 780 diminished the p53 export in MCF-7 and showed no effect in 1001 (Figure 2a). This indicates that estrogen has an effect on the wt p53 protein location and is unable to export it in the 1001 clone. To test whether the p53 translocation in MCF-7 was associated with a functional effect on its transcriptional activity, we have examined the influence of estrogen on three p53regulated proteins: p21, the regulator of the cell cycle and the ratio between Bax and Bcl-2 proteins, a common marker for mitochondria contribution to apoptosis. Data shown in Figure 2b indicate that estrogen induced a significant decrease in p21 protein expression following irradiation in MCF-7 but not in 1001 cells. Co-treatment with estrogen and ICI 182, 780 partially diminished estrogen-mediated effect. Since some of the Bcl-2 family members are directly regulated by p53, we have asked if the p53 shift from the nucleus to cytoplasm affects the expression of Bax and Bcl-2 proteins. To determine the expression level of Bax and Bcl-2, we have carried out Western blotting of estrogendeprived MCF-7 and 1001 cells treated with estrogen and ICI 182, 780 for 24 h. Data shown in Figure 2c indicate that estrogen treatment decreased the level of Bax and increased the level of Bcl-2, changing the overall ratio between the two proteins only in MCF-7 cells (Figure 2b). These data indicate that the estrogendependent cytoplasmic export of p53 results in a subsequent inhibition of its transcriptional activity and affects expression of p53-regulated proteins involved in the control of cell cycle and apoptotic process. Differential influence of ERa and ERb on p53 transcriptional activity Since the balance between ERa and ERb may control the estrogen effect, we have asked whether the introduction of ERb protein affects the observed estrogen response. Two expression plasmids pERa and pERb were introduced to MCF-7 cells. The transfection efficiency was determined by Western blotting (Figure 3a). The functional properties of introduced receptors were compared by examining their transcriptional activity on a construct with ERE and luciferase gene (Figure 3b). The endogenous ERa in MCF-7 cells induced transcription after estrogen treatment and this activity was reversed with specific inhibitor. With overexpression of ERa, the luciferase activity was increased as compared to endogenous level and also reversible by this inhibitor as shown in Figure 3b. The introduction of ERb resulted in a decrease in the ERE transcriptional activity level after estrogen stimulation.

To examine if the introduction of ERb has an effect on estrogen-mediated p53 transcriptional activity, we have analysed the p21 protein induction after gamma irradiation. The data depicted in Figure 3c indicate that overexpression of ERa induced a decrease in p21 expression while the introduction of ERb resulted in almost twofold increase of p21 as compared with basal induction level. These results further confirm the differential properties of the two ERs on p53 transcriptional activity. ERa and ERb have antagonistic effects on MCF-7 cell sensitivity to TNF To compare the effects of ERs on the sensitivity to TNF, we have transiently transfected the cells grown in regular FBS-supplemented media with pERa and pERb for 3 h. After an additional 21 h, the cells were exposed to various concentrations of TNF (Figure 4). Cells transfected with pERb showed an increased sensitivity to TNF at concentrations ranging from 5 to 100 ng/ml than those transfected with pERa (Po0.001) or pcDNA 3.1 (Po0.05) as determined by Student’s t-Test. These results demonstrate the differential influence of ERa and ERb on cell viability in the presence of TNF and suggest an antagonistic effect of ERb and ERa. Introduction of ERb prevents p53 export in estrogentreated MCF-7 cells To find out if ERb has an effect on estrogen-stimulated p53 translocation, we have examined estrogen-deprived MCF-7 cells transfected with pERa and pERb after estrogen stimulation. Confocal microscopy analysis using anti-p53 antibody (Figure 5) indicates that empty vector transfected cells with endogenous ERa showed estrogen-dependent, inhibitor-reversible translocation of the p53 protein similarly to those transfected with pERa. However, the pERb-transfected cells maintained p53 in the nucleus after estrogen treatment. These data suggest that the introduction of ERb can potently reduce estrogen-stimulated p53 protein translocation. Mdm2 accumulation is associated with estrogen-mediated effects on p53 location and transcriptional activity We have asked if estrogen effects on p53 involve the mdm2 protein. For this purpose, we have transiently transfected MCF-7 cells with a plasmid encoding the ARF protein known to induce rapid degradation of mdm2. Such transfection resulted in accumulation of p53 in the nucleus and a lack of p53 export to the cytoplasm after estrogen stimulation (Figure 6a) as determined by immunofluorescent staining and confocal microscopy. Additionally, MCF-7 cells transfected with pARF were resistant to estrogen-mediated decrease of gamma irradiation stimulated expression of p21 protein (Figure 6b). These results suggest that estrogen effects on p53 localization and activity are at least in part mediated by mdm2. In order to track the cellular localization of mdm2 after estrogen stimulation, we Oncogene


4792

Figure 2 Effect of exogenous estrogen on the translocation of p53 protein and its transcriptional activity in MCF-7 and 1001 cells. MCF-7 and 1001 cells were preincubated in hormone-free media for 6 days and then treated with DMSO (0.01%), estrogen (E2) (1 108 M) and ICI 182, 780 (1 106 M) for 24 h. (a) p53 protein localization in MCF-7 and 1001 cells determined by confocal microscope imaging and immunofluorescence. (b) Expression of p21 protein following estrogen treatment. MCF-7 and 1001 cells were exposed to 5 Gy dose of gamma irradiation (’). Nonirradiated cells (&) were used for comparison. p21WAF1/CIP1 expression was determined after subsequent 6 h by Western blotting. Basal expression of p21 in MCF-7 was used as control [C] for 1001 samples. Relative band intensity in respect to actin was calculated using Bio-Profil software and is shown as a graph. (c) Expression of Bax and Bcl-2 following estrogen treatment. In all, 50 mg of cell lysate was analysed by Western blotting for Bax and Bcl-2 expression as described in Materials and methods. Relative band intensity was calculated in respect to actin using Bio-Profil software and is shown as a graph presenting the ratio between expression level of Bax and Bcl-2

have transiently transfected estrogen-deprived MCF-7 cells with pERa and pERb plasmids followed by steroid treatment. Confocal microscopy analysis using antimdm2 antibody indicates that MCF-7 cells exposed to Oncogene

estrogen accumulate the mdm2 protein in the nucleus. The transfection with pERb plasmid significantly impaired such accumulating effects after estrogen treatment (Figure 6c).


4793

Figure 4 ERa and ERb have opposing effects on MCF-7 cells sensitivity to TNF. MCF-7 cells were cultured in FBS supplemented medium and transiently transfected with pERa (&), pERb (’) encoding plasmids and pcDNA 3.1 ( ) as control. At 24 h after transfection, the cells were exposed to a range of TNF concentrations. Cell viability was measured using the MTT assay as described in Materials and methods. Results from nontreated cells (0 ng/ml TNF) were used as point of reference within each transfection. Data presented are the means7s.d. of triplicate determinations. The level of significance was Po0.005 (*) for ERa vs ERb transfected cells and Pp0.01 (**) for ERb vs pcDNA 3.1, as determined by the Student’s t-test. Similar results were obtained in three independent experiments

Discussion

Figure 3 Differential influence of ERa and ERbon p53 transcriptional activity. (a) MCF-7 cells were transiently transfected with pERa and pERb plasmids and pcDNA 3.1 as control. After 24 h 50 mg of cell lysate was analysed by Western blotting for presence of ERa and ERb as described in Materials and methods. (b) Transcriptional activity of introduced receptors. MCF-7 cells were preincubated in hormone-free media for 6 days. Then cells were transfected with the respective ER plasmids together with two luciferase reporter plasmids 3 Vit-ERE-Luc and pRL as described in Materials and methods, followed by treatment with DMSO (’), estrogen (’) or estrogen and ICI 182, 780 ( ) containing media. After 24 h 1 104 cells were used for analysis of luciferase activity. Result is shown in relative luciferase units (3 Vit-ERE-Luc/pRL) of triplicate measurements. (c) Expression of p21 protein after transfection with pERa and pERb. MCF-7 cells cultured in FBS supplemented medium were transiently transfected with pERa, pERb encoding plasmids and pcDNA 3.1 as control. After 24 h MCF-7 cells were exposed to 5 Gy dose of gamma irradiation (’) and nonirradiated cells (&) were used for comparison. p21WAF1/CIP1 expression was determined after subsequent 6 h by Western blotting. Relative band intensity with respect to actin was calculated using Bio-Profil software and is shown as a graph. Similar results were obtained in two independent experiments

ER plays an important role in the development, progression and treatment of breast cancer (Sommer and Fuqua, 2001). It has been proven to be a valuable predictive and prognostic factor in the clinical management of this disease. Consequently, its inhibition has become one of the major strategies for the prevention and treatment of breast cancer. However, there is increasing evidence that this receptor interacts with coregulatory proteins and several signal transduction pathways other than the so-called ‘classical’ ligandmediated stimulation of genes containing EREs (Driggers and Segars, 2002; Segars and Driggers, 2002). Several reports have demonstrated that estrogen promotes cell survival in estrogen-responsive cells (Razandi et al., 2000; Sudoh et al., 2001; Kanda and Watanabe, 2003). We have first examined, in the course of these studies, the ability of estrogen to influence cell survival in response to TNF taking advantage of our cell model based on the sensitive MCF-7 breast carcinoma cell line and its resistant derivative 1001 clone, displaying a mutated p53, to the cytotoxic action of TNF. It is well established that, among the players involved in the regulation of cell survival and death, p53 has a major function in transducing stress to apoptotic machinery of the cell. In a previous report, we have shown that wildtype p53 is involved in the cytotoxic action of TNF and that p53 function contributes to resistance of tumor cells to TNF-induced killing (Cai et al., 1997b). This is consistent with the importance of p53 status as a determinant of cellular response to DNA-damaging Oncogene


4794

Figure 5 The export of p53 after estrogen treatment in MCF-7 cells is altered following the transfection with pERb. Confocal microscopy analysis of p53 protein localization. MCF-7 cells were preincubated in hormone-free media for 6 days and then transiently transfected with pERa, pERb encoding plasmids and pcDNA 3.1 as control. At 6 h after transfection, the cells were exposed to estrogen/inhibitor-containing media. After another 24 h, cells were fixed and analysed by immunofluorescence for p53 protein staining. Transfection efficiency was determined using cotransfection of respective plasmids together with pGFP plasmid in the ratio 1 : 1. Cells were analysed using FACS with an average of 68%75.8% positive fluorescent cells (data not shown). Similar results were obtained in two independent experiments

factors (Brown and Wouters, 1999) and its key role in the development or progression of many types of cancer. We have therefore attempted to elucidate the cross-talk between estrogen, p53 and TNF cytotoxic action. Here we demonstrated, based on the p53 wild-type parental MCF-7 cells and the p53-mutated 1001 clone, that although both cell types express ERa they respond differentially to the effect of estrogen under TNF stimulation. This suggests a possible relationship between the effect of estrogen and p53 function. Accumulating evidence has been provided indicating that the activity of p53, including post-translational modifications and the ability to interact with other proteins, can be further controlled by regulation of the subcellular localization of components of the p53-response pathways. Obviously, p53 is actively transported into and out of the nucleus and can be localized to distinct structures in both the nucleus and cytoplasm. Here we show that the estrogen effect on MCF-7 is associated with an export of the p53 protein from the nucleus to the cytoplasm and a subsequent decrease in its transcriptional activity. Interestingly, no such distribution was observed in 1001 cells displaying a mutated p53 that has lost its transcriptional activity (Cai et al., 1997b). The mutation was described in the DNA-binding domain and does not explain the inability to migrate between the Oncogene

nucleus and cytoplasm. Besides impaired transcription, p53 protein in the 1001 cells is likely unable to respond to the mechanisms of its turnover such as ubiquitination, phosphorylation or acetylation (Brooks and Gu, 2003). It is clearly established that the ERa mediates breast cancer-promoting effects of estrogen. However, the role of ERb in breast cancer remains elusive. Several studies show that ERb has a counteracting effect on the ERa activity (Cowley et al., 1997; Pettersson et al., 1997; Hall and McDonnell, 1999; Wade et al., 2001; Liu et al., 2002). In this regard, we have shown that ERb can antagonize the effect of ERa. In fact, the introduction of ERb- to ERa- positive MCF-7 cells results in three observed effects. (i) It allows retention of p53 in the nucleus after estrogen stimulation. (ii) It increases p53 transcriptional activity. (iii) It consequently leads to an increased sensitivity to TNF. These observations further underline that estrogen-induced effects on MCF-7 need a transcriptionally active p53. Hence, our model can reflect the mammary cancer progression, which usually advances from hormone-dependent, p53-wt to hormone-independent, p53-mutated phenotype (Moll et al., 1992; Berns et al., 2000; Caleffi et al., 1994). Our results confirm the previously observed effects of estrogen under TNF treatment in MCF-7 cells (Burow


4795

Figure 6 Mdm2 accumulation is associated with estrogen-mediated effects on p53 location and transcriptional activity. (a) Confocal microscopy analysis of p53 staining in MCF-7 cells following pARF transfection and estrogen stimulation. Cells were preincubated in hormone-free media for 6 days and then plated on glass coverslips in a 24-well plate. Following overnight attachments, cells were transiently transfected with pARF and pcDNA 3.1 plasmids for 12 h. The cells were assayed 24 h after estrogen stimulation. Cells showing accumulated p53 are indicated with arrows. (b) Expression of p21 protein after transfection with pARF and pcDNA 3.1 plasmids. MCF-7 cells were cultured in estrogen-deprived medium and transfected with the respective plasmids for 12 h and stimulated with estrogen for 24 h. Then, cells were exposed to 5 Gy dose of gamma irradiation and assayed by Western blotting. (c) Confocal microscopy analysis of mdm2 protein localization. MCF-7 cells were preincubated in hormone-free media for 6 days and then transiently transfected with pERa, pERb encoding plasmids and pcDNA 3.1 as control. After transfection the cells were exposed to estrogen/inhibitor-containing media. After another 24 h, cells were fixed and analysed by immunofluorescence for mdm2 protein staining as described in Materials and methods. Cells displaying accumulated mdm2 protein are indicated with arrows. Similar results were obtained in two independent experiments

et al., 2001) and also point to p53 as the target of estrogen action. Recently, it has been reported that the induced expression of ERb in the breast cancer cell line T47D reduced 17b-estradiol-stimulated proliferation when expression of ERb equals that of ERa and that such an effect involves a functional interaction with components of the cell cycle machinery in these cells (Strom et al., 2004). In addition, recently it has been shown that ERb inhibits proliferation by repressing c-myc, cyclin D and cyclin A gene transcription and increases the expression of p21 and p27, which leads to a G2 cell cycle arrest (Paruthiyil et al., 2004).

Our findings can also shed light on the paradoxical regulation of Bcl-2 family proteins by estrogen (Leung and Wang, 1999). As reported by other authors, estrogen induces the expression of Bcl-2 protein (Teixeira et al., 1995; Burow et al., 2001; Kanda and Watanabe, 2003), changing the overall ratio between Bcl-2 and Bax (Huang et al., 1997; Leung and Wang, 1999) – a marker for mitochondria contribution in apoptosis (Oltvai et al., 1993). Since p53 is shown to be a positive transcriptional regulator for Bax (Miyashita et al., 1994b; Miyashita and Reed, 1995) and a negative for Bcl-2 (Miyashita et al., 1994a, b), estrogen-mediated Oncogene


4796

decrease in p53 transcriptional activity would explain the contradictory regulation of these proteins by estrogen. Nevertheless, Bcl-2 may only represent one pathway by which estrogen and ER function to promote cell survival. Indeed, the effects of estrogen on survival signaling may also occur through activation of other signaling cascades. The constitutive high Bcl-2 levels in 1001 cells are also in line with what other teams have observed in drug-resistant derivatives of MCF-7 (Lilling et al., 2000). There are two plausible explanations on how ERs can mediate their effects on p53 protein. Firstly, ERa and p53 were shown to interact directly, which results in a decreased ERE-dependent transcription of ERa (Yu et al., 1997; Liu et al., 1999). However, this interaction occurs regardless of estrogen or inhibitor presence and results in a protective rather than deactivating effect on p53 (Liu et al., 2000). Secondly, estrogen stimulation of ERa-positive cells can increase the expression of mdm2, the main regulator of p53 localization and stability (Kato et al., 2002; Phelps et al., 2003). Nevertheless, the ERa and ERb expression status has not been clearly defined in these reports. Our findings suggest that mdm2 protein may be involved in the mediation of estrogen effects on p53. Firstly, the estrogen stimulation accumulates mdm2 protein in the nuclei of ERa-positive cells. Secondly, the exogenous expression of the ARF protein results in the accumulation of the p53 protein in the nucleus and in an increase in its transcriptional activity despite estrogen stimulation. ERa and ERb did not display similar effects on the mdm2 accumulation following estrogen treatment. The antagonistic roles of ERa and ERa on the mdm2 expression could be explained by previously described opposing activities of both receptors on the AP-1 sequence within several gene promoters (Paech et al., 1997; Weatherman and Scanlan, 2001). Such effect could also apply to the AP-1 site present in the mdm2 gene promoter. Experiments will be designed to address this issue in our experimental model. There are several implications resulting from estrogen-mediated suppression of p53 protein. First, it can allow efficient growth of mammary gland cells by suppressing the p53 activity and consequently the growth-inhibiting proteins under its control. The concept of constantly blocked wild-type p53 protein implies low pressure for p53 gene mutations (Wynford-Thomas and Blaydes, 1998) and could explain low (9%) mutation frequency in mammary gland (Caleffi et al., 1994) as compared with overall rate (50–55%) (Hollstein et al., 1996) in all human cancers. Secondly, prolonged exposure to estrogen can allow damaged cells to bypass the p53-regulated mechanisms of DNA repair and cell elimination by p53-dependent apoptosis (Vousden and Lu, 2002). This can ultimately result in acquisition of the mutation in p53 itself, a key step in tumor progression resulting in an increased cytotoxic and chemotherapeutic resistance (Cai et al., 1997b; Berns et al., 2000; Geisler et al., 2001; Thiery et al., 2003). In summary, our data show a differential effect of estrogen on TNF-sensitive cells and their resistant Oncogene

counterparts with compromised p53 transcriptional activity, and highlight the ERb dominant-negative action on ERa. Our observations also argue that the balance between ERa and ERb plays a role in the control of cell survival by a mechanism involving at least in part the status of p53.

Materials and methods Plasmids and reagents Plasmids pRST7 ERa, (pERa), pRST7 ERb (pERb) and 3 ERE vitellogenin luciferase (3 Vit-ERE-Luc) reporter construct were kind gifts of Dr Kenneth S Korach (NIEHS NIH NC). The pARF-Myc plasmid was a kind gift from Dr Yue Xiong (University of North Carolina Chapel Hill, NC, USA). The Renilla-Luciferase (pRL) reporter was purchased from Promega (Madison, WI, USA) and pcDNA3.1 plasmid was purchased from Invitrogen (Carlsbad, CA, USA). Estrogen (17b-estradiol) was purchased from Sigma (St Louis, MO, USA) and ICI 182, 780 was purchased from Tocris (Northpoint, UK). TNF-a (Beromun) was purchased from Boehringer Ingelheim (Ingelheim, Germany). FuGENE 6 transfection reagent was purchased from Roche Molecular Biochemicals (Mannheim, Germany). 3-[4,5-Dimethylthiazol-2-yl]-2,5diphenyl tetrazolium bromide (MTT) and DMSO were purchased from Sigma. Cell culture and hormone treatments MCF-7 TNF-resistant clone 1001 was established as described (Cai et al., 1997a). All cells were cultured in RPMI with glutamax (GIBCO, Carlsbad, CA, USA) supplemented with 5% fetal bovine serum (FBS) and 100 U/ml penicillin– streptomycin solution in a humidified atmosphere with 5% CO2. For 1001 cells, geneticin (GIBCO) was used as selective agent at a concentration of 250 mg/ml. Before estrogen treatments, the cells were cultured for 6 days in RPMI phenol red-free medium (GIBCO) supplemented with 5% charcoaldextran-treated serum (HyClone, Logan, UT, USA) and 100 U penicillin-streptomycin. For hormonal treatments, estrogen was used at concentration of 1 108 M, ICI 182, 780 was used at concentration of 1 106 M. DMSO was used as diluent for stock solutions of E2, ICI and also as mock control treatment at 0.01% v/v. For gamma irradiation, MCF-7 cells (5 105) were exposed to 5 Gy dose of radiation and cultured for an additional 6 h before protein harvest. Western blotting Cells were harvested in protein extraction buffer (25 mM Hepes, pH 7.5,. 300 mM NaCl, 1.5 mM MgCl2, 0.2 mM EDTA, 1% Triton -100, 0.5 mM DTT, 0.5% sodium deoxycholate, 10% glycerol). Complete Mini proteinase inhibitors cocktail from Roche was added as suggested by the manufacturer. The protein concentration was determined using BCA Protein Assay Kit from Pierce (Rockford, IL, USA). Protein extracts (50 mg) were separated on a 12% sodoum dodecyl sulfate (SDS)–polyacrylamide gel and electroblotted on nitrocellulose Hybondt membranes (Amersham Pharmacia Biotech, Aylesbury, UK). The transfer efficiency and protein equivalence were determined with Ponceau Red (Sigma) staining. After blocking, the membranes were blotted with anti-ERa antibody (Cell Signaling, Beverly, MA, USA), ERb antibody (Zymed, San Francisco, CA, USA), anti-p21/WAF1 (Oncogene, Merck-KgaA, Darmstadt, Germany), anti-Myc mouse


4797 monoclonal antibody developed in our laboratory and Bax, Bcl-2 or alpha-actin antibodies (Santa Cruz Biotechnology, Santa Cruz, CA, USA). Secondary antibodies against rabbit, mouse and donkey (Santa Cruz) conjugated with horseradish peroxidase (HRP) were used in the second blotting step. Complexes were detected using ECL chemiluminescence kit (Amersham Pharmacia Biotech) and autoradiography. Subsequent results were analysed using BioProfil Bio1D Software (Vilber Lourmat, Paris, France). Confocal scanning immunofluorescence microscopy 3 104 cells grown on sterile slides were subjected to hormone treatments as described above. Slides were washed once with PBS and fixed for 10 min in 4% paraformaldehyde (PFA) solution in PBS and washed three times with PBS. Membranes were permeabilized for 10 min with 0.1% SDS in PBS and washed three times with PBS. Nonspecific sites were blocked for 20 min with 10% FBS in PBS and washed once with PBS. The cells were incubated for 2 h with anti-p53 monoclonal antibody (DO-1) (Santa Cruz) or anti-mdm2 antibody (SMP14, Santa Cruz). Nuclei were counterstained using SYTOX green reagent (Molecular Probes, Eugene, OR, USA). After three washes with PBS, anti-mouse secondary antibody conjugated with Alexa 546 (Molecular Probes, Eugene, OR, USA) was used for 1 h. Coverslips were mounted with antifading Vectashield from (Vector, Burlingame, CA, USA). Confocal microscopy analysis was performed on Zeiss LSM 510 microscope. Cell viability MCF-7 cells (5 103) or 1001 (7.5 103) were plated in flatbottom 96-well plates. For TNF treatments the estrogendeprived cells were preincubated with estrogen, ICI 182, 780 or DMSO for 24 h and then the media was switched to mixed steroid and TNF solutions. After 72 h, MTT solution (2.5 mg/ ml in PBS) was added to a final concentration of 0.8 mg/ml. Cells were incubated for additional 2 h and lysed with MTT lysis buffer (50% N-N0 dimethyl formamide, 20%, SDS pH 4.7). The OD was measured at 550 nm with 630 nm wavelength

as reference. The percentage of cell viability was calculated as follows: percentage of viability ¼ (A1/A0) 100, where A1 and A0 represent absorbances obtained, respectively, for treated and untreated cells. The average value of a triplicate measurement was used for data analysis. Transient transfection and luciferase reporter assays In all, 3–4 104 cells/cm2 were transfected 24 h after plating with 0.5 mg/ml. of either pERa, pERb or pcDNA3.1 plasmid using the FuGENE 6 transfection reagent according to the manufacturer’s instructions. For viability assays TNF treatments were applied 24 h after transfection. For luciferase reporter assays, 1 104 MCF-7 cells were transfected for 6 hours with 0.15 mg of either pERa or pERb plasmid together with 0.15 mg of the 3 Vit-ERE-Luc and 0.01 mg of pRL using the FuGENE 6 reagent according to the manufacturer’s instructions. The total amount of DNA was brought to 0.5 mg with pcDNA3.1 plasmid. After 6 h, the transfection solution was switched to hormonal treatments and incubated for an additional 20 h. Cells were assayed with Dual-Luciferases Reporter Assay System kit from Promega (Madison, WI, USA) and data were collected using MicroLumat LB96P luminomiter counter (Berthold, Bad Wildbad, Germany).

Abbreviations ERs, estrogen receptors; TNF, tumor necrosis factor a; ERb, estrogen receptor beta; ERa, estrogen receptor alpha; ERE, estrogen response element; DMSO, dimethylsulfoxide; E2, 17b-estradiol; ICI, ICI 182, 780. Acknowledgements This work was supported by grants from INSERM, the Association pour la Recherche sur le Cancer (Grants 5253– 5129). SAL is a recipient of a fellowship from Marie Curie (5th PCRD, contract QLGA-1999-50406). JT is supported by a fellowship from the Ligue Nationale Francaise de Recherche Contre Le Cancer.

References Ameyar M, Atfi A, Cai Z, Stancou R, Shatrov V, Bettaieb A and Chouaib S. (1998). J. Biol. Chem, 273, 29002–29008. Berns EM, Foekens JA, Vossen R, Look MP, Devilee P, Henzen-Logmans SC, van Staveren IL, van Putten WL, Inganas M, Meijer-van Gelder ME, Cornelisse C, Claassen CJ, Portengen H, Bakker B and Klijn JG. (2000). Cancer Res., 60, 2155–2162. Brooks CL and Gu W. (2003). Curr. Opin. Cell. Biol., 15, 164–171. Brown JM and Wouters BG. (1999). Cancer Res., 59, 1391–1399. Burow ME, Weldon CB, Chiang TC, Tang Y, Collins-Burow BM, Rolfe K, Li S, McLachlan JA and Beckman BS. (2000). Int. J. Oncol., 16, 1179–1187. Burow ME, Weldon CB, Tang Y, McLachlan JA and Beckman BS. (2001). J. Steroid. Biochem. Mol. Biol., 78, 409–418. Cai Z, Bettaieb A, Mahdani NE, Legres LG, Stancou R, Masliah J and Chouaib S. (1997a). J. Biol. Chem., 272, 6918–6926.

Cai Z, Capoulade C, Moyret-Lalle C, Amor-Gueret M, Feunteun J, Larsen AK, Paillerets BB and Chouaib S. (1997b). Oncogene, 15, 2817–2826. Caleffi M, Teague MW, Jensen RA, Vnencak-Jones CL, Dupont WD and Parl FF. (1994). Cancer, 73, 2147–2156. Cheng G, Weihua Z, Warner M and Gustafsson JA. (2004). Proc. Natl. Acad. Sci. USA, 101, 3739–3746. Chouaib S, Branellec D and Buurman WA. (1991). Immunol. Today, 12, 141–142. Cowley SM, Hoare S, Mosselman S and Parker MG. (1997). J. Biol. Chem., 272, 19858–19862. Driggers PH and Segars JH. (2002). Trends Endocrinol. Metab., 13, 422–427. Geisler S, Lonning PE, Aas T, Johnsen H, Fluge O, Haugen DF, Lillehaug JR, Akslen LA and Borresen-Dale AL. (2001). Cancer Res., 61, 2505–2512. Green S, Walter P, Greene G, Krust A, Goffin C, Jensen E, Scrace G, Waterfield M and Chambon P. (1986). J. Steroid. Biochem., 24, 77–83. Hall JM and Korach KS. (2002). J. Biol. Chem., 277, 44455–44461. Oncogene


4798 Hall JM and McDonnell DP. (1999). Endocrinology, 140, 5566–5578. Hollstein M, Shomer B, Greenblatt M, Soussi T, Hovig E, Montesano R and Harris CC. (1996). Nucleic. Acids Res., 24, 141–146. Huang Y, Ray S, Reed JC, Ibrado AM, Tang C, Nawabi A and Bhalla K. (1997). Breast Cancer Res. Treat., 42, 73–81. Hyder SM, Chiappetta C and Stancel GM. (1999). Biochem. Pharmacol., 57, 597–601. Kanda N and Watanabe S. (2003). J. Invest. Dermatol., 121, 1500–1509. Kato K, Horiuchi S, Takahashi A, Ueoka Y, Arima T, Matsuda T, Kato H, Nishida Ji J, Nakabeppu Y and Wake N. (2002). J. Biol. Chem., 277, 11217–11224. Klein-Hitpass L, Schorpp M, Wagner U and Ryffel GU. (1986). Cell, 46, 1053–1061. Krege JH, Hodgin JB, Couse JF, Enmark E, Warner M, Mahler JF, Sar M, Korach KS, Gustafsson JA and Smithies O. (1998). Proc. Natl. Acad. Sci. USA, 95, 15677–15682. Kroemer G and Reed JC. (2000). Nat. Med., 6, 513–519. Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S and Gustafsson JA. (1997). Endocrinology, 138, 863–870. Leung LK and Wang TT. (1999). Br. J. Cancer, 81, 387–392. Lilling G, Hacohen H, Nordenberg J, Livnat T, Rotter V and Sidi Y. (2000). Cancer Lett., 161, 27–34. Liu G, Schwartz JA and Brooks SC. (1999). Biochem. Biophys. Res. Commun., 264, 359–364. Liu G, Schwartz JA and Brooks SC. (2000). Cancer Res., 60, 1810–1814. Liu MM, Albanese C, Anderson CM, Hilty K, Webb P, Uht RM, Price Jr RH, Pestell RG and Kushner PJ. (2002). J. Biol. Chem., 277, 24353–24360. Miyashita T, Harigai M, Hanada M and Reed JC. (1994a). Cancer Res., 54, 3131–3135. Miyashita T, Krajewski S, Krajewska M, Wang HG, Lin HK, Liebermann DA, Hoffman B and Reed JC. (1994b). Oncogene, 9, 1799–1805. Miyashita T and Reed JC. (1995). Cell, 80, 293–299. Molinari AM, Bontempo P, Schiavone EM, Tortora V, Verdicchio MA, Napolitano M, Nola E, Moncharmont B, Medici N, Nigro V, Armetta I, Abbondanza C and Puca GA. (2000). Cancer Res., 60, 2594–2597. Moll UM, Riou G and Levine AJ. (1992). Proc. Natl. Acad. Sci. USA, 89, 7262–7266. Mosselman S, Polman J and Dijkema R. (1996). FEBS Lett., 392, 49–53. Mueller SO, Clark JA, Myers PH and Korach KS. (2002). Endocrinology, 143, 2357–2365. Muller M, Wilder S, Bannasch D, Israeli D, Lehlbach K, Li-Weber M, Friedman SL, Galle PR, Stremmel W, Oren M and Krammer PH. (1998). J. Exp. Med., 188, 2033–2045.

Oncogene

Nilsson S, Makela S, Treuter E, Tujague M, Thomsen J, Andersson G, Enmark E, Pettersson K, Warner M and Gustafsson JA. (2001). Physiol. Rev., 81, 1535–1565. Oltvai ZN, Milliman CL and Korsmeyer SJ. (1993). Cell, 74, 609–619. Owen-Schaub LB, Zhang W, Cusack JC, Angelo LS, Santee SM, Fujiwara T, Roth JA, Deisseroth AB, Zhang WW, Kruzel E and Radinsky R. (1995). Mol. Cell. Biol., 15, 3032–3040. Pace P, Taylor J, Suntharalingam S, Coombes RC and Ali S. (1997). J. Biol. Chem., 272, 25832–25838. Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson J, Kushner PJ and Scanlan TS. (1997). Science, 277, 1508–1510. Paruthiyil S, Parmar H, Kerekatte V, Cunha GR, Firestone GL and Leitman DC. (2004). Cancer Res., 64, 423–428. Pettersson K, Grandien K, Kuiper GG and Gustafsson JA. (1997). Mol. Endocrinol., 11, 1486–1496. Phelps M, Darley M, Primrose JN and Blaydes JP. (2003). Cancer Res., 63, 2616–2623. Razandi M, Pedram A and Levin ER. (2000). Mol. Endocrinol., 14, 1434–1447. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, Jackson RD, Beresford SA, Howard BV, Johnson KC, Kotchen JM and Ockene J. (2002). JAMA, 288, 321–333. Segars JH and Driggers PH. (2002). Trends Endocrinol. Metab., 13, 349–354. Sommer S and Fuqua SA. (2001). Semin. Cancer Biol., 11, 339–352. Strom A, Hartman J, Foster JS, Kietz S, Wimalasena J and Gustafsson JA. (2004). Proc. Natl. Acad. Sci. USA, 101, 1566–1571. Sudoh N, Toba K, Akishita M, Ako J, Hashimoto M, Iijima K, Kim S, Liang YQ, Ohike Y, Watanabe T, Yamazaki I, Yoshizumi M, Eto M and Ouchi Y. (2001). Circulation, 103, 724–729. Teixeira C, Reed JC and Pratt MA. (1995). Cancer Res., 55, 3902–3907. Thiery J, Dorothee G, Haddada H, Echchakir H, Richon C, Stancou R, Vergnon I, Benard J, Mami-Chouaib F and Chouaib S. (2003). J. Immunol., 170, 5919–5926. Vousden KH and Lu X. (2002). Nat. Rev. Cancer, 2, 594–604. Wade CB, Robinson S, Shapiro RA and Dorsa DM. (2001). Endocrinology, 142, 2336–2342. Weatherman RV and Scanlan TS. (2001). J. Biol. Chem., 276, 3827–3832. Wu GS, Burns TF, McDonald III ER, Jiang W, Meng R, Krantz ID, Kao G, Gan DD, Zhou JY, Muschel R, Hamilton SR, Spinner NB, Markowitz S, Wu G and elDeiry WS. (1997). Nat. Genet., 17, 141–143. Wynford-Thomas D and Blaydes J. (1998). Carcinogenesis, 19, 29–36. Yu CL, Driggers P, Barrera-Hernandez G, Nunez SB, Segars JH and Cheng S. (1997). Biochem. Biophys. Res. Commun., 239, 617–620.