Activation of the Unliganded Estrogen Receptor by Prolactin in Breast Cancer Cells

Home , Prolactin receptor

Oncogene (2009) 28, 1298–1308 & 2009 Macmillan Publishers Limited All rights reserved 0950-9232/09 $32.00 www.nature.com/onc ORIGINAL ARTICLE Activation of the unliganded estrogen receptor by prolactin in breast cancer cells

L Gonza´ lez1, A Zambrano1, I Lazaro-Trueba, E Lope´ z, JJA Gonza´ lez, J Martı´ n-Pe´ rez and A Aranda

Instituto de Investigaciones Biome´dicas, Consejo Superior de Investigaciones Cientı´ﬁcas and Universidad Auto´noma de Madrid, Madrid, Spain

Both prolactin (PRL) and estrogen (E2) are involved in receptor are expressed in mammary tumors and in the pathogenesis and progression of mammary neoplasia, breast cancer cell lines, thereby creating autocrine/ but the mechanisms by which these hormones interact to paracrine modes of action for PRL (Clevenger et al., exert their effects in breast cancer cells are not well 1995; Ginsburg and Vonderhaar, 1995; Touraine et al., understood. We show here that PRL is able to activate the 1998; Schroeder et al., 2002). Different studies have unliganded estrogen receptor (ER). In breast cancer cells, correlated PRL levels with breast cancer incidence, PRL activates a reporter plasmid containing estrogen showing that prolactin receptor (PRLR) levels are response elements (EREs) and induces the ER target gene generally higher in breast tumors than in normal pS2. These actions are blocked by the antagonist ICI mammary tissue (Clevenger et al., 2003; Tworoger and 182,780, showing that ER is required for the PRL- Hankinson, 2006). Furthermore, disruption of PRL mediated effect. Moreover, PRL leads to phosphorylation signaling in breast cancer cells causes cell growth of ERa in serine-118 (P-ERa), a modification related to inhibition apoptosis induction (Fuh and Wells, 1995). the potentiation of ligand-independent transcriptional Different signaling pathways that are activated on activation. In addition, PRL mimics the effect of E2 on PRL binding to its receptor have been demonstrated to target gene expression by inducing cyclical recruitment of lead to breast cancer cell proliferation. In breast cancer ERa and P-ERa to ERE-containing promoters, resulting cell lines PRL stimulates both the JAK1 and JAK2 in recruitment of co-activators and acetylation of histone pathways (Campbell et al., 1994; Liu et al., 1997; H3. Finally, PRL induces expression of c-Myc and Cyclin Neilson et al., 2007), and we have previously demon- D1 and leads to increased cell proliferation, which is strated that PRL stimulates Src kinases, which then specifically antagonized by ICI 182,780 or ERa depletion. independently activate Fak/ERK1/2 and the PI3K- These results show that ligand-independent ERa activa- dependent p70S6K and Akt kinases (Acosta et al., tion appears to be an important component of the 2003). Activated ERK1/2 and AKT lead among others proliferative and transcriptional actions of PRL in breast to increased AP-1 complexes, and induce the expression cancer cells. of cell-cycle progression genes such as cyclin D1 or Oncogene (2009) 28, 1298–1308; doi:10.1038/onc.2008.473; c-myc (Brockman et al., 2002; Acosta et al., 2003; published online 26 January 2009 Gutzman et al., 2005). Estrogen is also involved in normal breast develop- Keywords: prolactin; estrogen receptor; transcriptional ment, as well as in growth and progression of breast activation; breast cancer cells cancer. The biological actions of estrogens are mediated by binding to nuclear estrogen receptors (ERa and ERb). In breast cancer cells expressing ERs, estradiol (E2) has potent proliferative effects and treatment with ER antagonists is the current hormone therapy of choice Introduction for the treatment of ERa-positive breast cancers (Yager and Davidson, 2006). A key event for the antiprolifera- Prolactin (PRL) has a key function in mammary gland tive effects of antiestrogens appears to be the down- development and PRL involvement in breast cancer has regulation of Cyclin D1 and c-Myc (Musgrove et al., now been clearly established. Both the hormone and its 1993; Carroll et al., 2002). Normally, ERs act as ligand-dependent transcription Correspondence: Dr J Martı´ n-Pe´ rez or Dr A Aranda, Instituto de factors by binding as homodimers to estrogen response Investigaciones Biome´ dicas, Consejo Superior de Investigaciones elements (EREs) in target genes (Aranda and Pascual, Cientı´ ficas and Universidad Auto´ noma de Madrid, Arturo Duperier 2001). However, there is increasing evidence that the 4, 28029Madrid, Spain. presence of estrogen is not an absolute requirement for E-mails: [email protected] or [email protected] 1These authors contributed equally to this work. receptor activation, because growth factors, such as Received 23 July 2008; revised 27 November 2008; accepted 12 IGF-1 or EGF, and intracellular protein kinases can December 2008; published online 26 January 2009 induce an estrogen-independent ERa activation in Prolactin activates the estrogen receptor L Gonza´lez et al 1299 different model systems (Butt et al., 2005). In breast to ER activation, cells were incubated either with PRL cancer cells phosphorylation of the Ser-118 residue in or with E2 for 6 and 40 h in the absence and presence of the human ERa A/B domain by growth factors ICI. As shown in Figure 1b, the ER antagonist blocked stimulated mitogen-activated protein kinase (Kato not only the response to E2 but also to PRL. et al., 1995; Bunone et al., 1996; Chen et al., 2000; To determine the effect of PRL on endogenous ER- Medunjanin et al., 2005; Park et al., 2005; Weitsman dependent gene expression, we analysed transcripts for et al., 2006) results in the potentiation of the ER ligand- the well-known estrogen target gene pS2, containing independent transcriptional activation function (AF-1). EREs in its regulatory region (Nunez et al., 1989). PRL Interestingly, it has been proposed that phosphorylation significantly induced pS2 mRNA levels in T47D cells in Ser-118 may be associated with an increase in (Figure 2a), without increasing ERa mRNA levels estrogen agonism, progression of breast cancer, resis- (Figure 2b). PRL-mediated transcription was abolished tance to tamoxifen therapy and estrogen-independent when cells were treated with the ER antagonist, further growth of MCF-7 cells (Likhite et al., 2006; Murphy demonstrating that the lactotrophic hormone can et al., 2006). Recently, it has been shown that the pure stimulate the unliganded ER. ER antagonist ICI 182,780 (ICI) can also induce ER phosphorylation in Ser-118 (Lipfert et al., 2006; De los PRL induces Ser-118-ERa phosphorylation Santos et al., 2007). Post-translational modifications of ERa are emerging as In the normal mammary gland PRL and estrogen act important regulatory elements of cross talk between synergistically to favor mammary gland growth and different signaling pathways. In particular, phosphory- development (Hennighausen and Robinson, 2005). ERa lation at Ser-118 has been implicated in the ligand- and PRLR are co-expressed in many breast tumors dependent and -independent effects of ERa and in (Murphy et al., 1984; Ormandy et al., 1997). However, tamoxifen resistance of breast tumors (Lonard et al., the mechanisms by which these hormones interact to 2000; Wijayaratne and McDonnell, 2001; Murphy et al., affect breast cancer cell functions are not completely 2004). Thus, we analysed the effect of incubation with understood. One of the mechanisms of PRL and PRL on ERa phosphorylation in Ser-118 (P-ERa)by estrogen interaction is the cross-regulation of their western blot. As shown in Figure 3a, incubation with receptors (Ormandy et al., 1997; Dong et al., 2006). It PRL caused a rapid and sustained increase of P-ERa has been described that PRL increases ERa levels and levels, whereas total ERa levels remained unchanged E2 responsiveness in breast cancer cells (Shafie and through the period in which increased phosphorylation Brooks, 1977; Gutzman et al., 2004a). In addition, both was observed. Similar results were obtained in MCF-7 PRL and E2 increase ERK1/2 phosphorylation and cells (Figure 3b), demonstrating that the effect of PRL is cooperatively activate the transcription factor AP-1 not restricted to the T47D cell line. We also assessed the (Gutzman et al., 2004b, 2005). However, the possibility effect of ICI on PRL-induced ERa phosphorylation that PRL could also mediate some of its actions by (Figure 3c). Short-term incubation with ICI increased inducing estrogen-independent ERa activation has not the effect of PRL on P-ERa levels, but reduced been yet explored. phosphorylation was observed after 6 h of incubation, In this study we show that PRL increases ER activity due to the expected strong downregulation of ERa expr- in an estrogen-independent manner in T47D and ession caused by ICI (Lonard et al., 2000; Wijayaratne MCF-7 cells. We found that PRL induces Ser-118-ER and McDonnell, 2001). This reduction was not observed phosphorylation, increases the activity of an ERE- when PRL was combined with the ER antagonist containing reporter plasmid, upregulates the expression tamoxifen that does not induce receptor downregula- of the ER-dependent gene pS2 and induces the recruit- tion, and under these conditions a strong phosphoryla- ment of ERa and co-activators to ER target promoters. tion was found after 24 h of incubation in the presence Moreover, ICI abolishes the effect of PRL on ERE of both compounds (Figure 3d). Finally, induction of stimulation and pS2 gene expression and blocks PRL- Ser-118 phosphorylation by PRL was blocked in the dependent proliferation, demonstrating that ligand- presence of the inhibitors PD184352, LY294002 and independent ER activation by PRL is an important PP2 (Figure 3e), showing that ERK and PI3K activation mediator of the mitogenic and transcriptional effects of that we have previously shown to be secondary to c-Src this hormone in breast cancer cells. stimulation (Acosta et al., 2003) are required to induce receptor modification. Results PRL induces ERa recruitment to estrogen target genes PRL-dependent activation of unliganded ER Estrogen-dependent transcriptional activation has been To explore the interaction between PRL and ERs, we shown to involve occupancy of the target promoter by analysed the effect of this hormone on the transcription ER (Shang et al., 2000; Metivier et al., 2003) and P-Ser- of an estrogen-responsive reporter plasmid. As shown in 118-ERa is also recruited to the promoters of estrogen- Figure 1a, treatment of T47D cells with PRL, in the regulated genes (Weitsman et al., 2006). To analyse absence of E2, caused a significant increase in luciferase whether treatment of breast cancer cells with PRL could activity at the different time points examined. To prove also induce the recruitment of ERa and P-Ser-118-ERa that the effect of PRL on the reporter plasmid was due to the target promoter, we performed chromatin

Oncogene Prolactin activates the estrogen receptor LGonza´lez et al 1300

5 4

4 3

3 2 2

1 1 Luciferase activity (fold induction) activity (fold Luciferase Luciferase activity (fold induction) activity (fold Luciferase 0 0 0 1020304050 (h) E2 - - + + - - - - + + - - PRL PRL - - - - + + - - - - + + ICI - + - + - + - + - + - + 6 h 40 h Figure 1 Prolactin (PRL) induces estrogen response element (ERE)-dependent transactivation. (a) Luciferase activity was determined in T47D cells transfected with a reporter plasmid containing three copies of a consensus ERE and incubated with PRL for the indicated time periods. Data are shown as the mean±s.d. of values obtained in the corresponding untreated cells at each point. (b) Reporter activity in cells treated with PRL or E2 in the presence and absence of the ER antagonist ICI 182,780 (ICI). Data are expressed relative to the values obtained in the corresponding control cells.

5 1.5

1.0 3

2 0.5 mRNA (fold induction) mRNA (fold α pS2 mRNA (fold induction) pS2 mRNA (fold

1 ER

0 0.0 PRL - - + + - - + + ICI - + - + - + - + Figure 2 Prolactin (PRL) induces transcription of the estrogen receptor (ER) target gene pS2.(a) pS2 mRNA levels were measured by quantitative real-time PCR in cells preincubated with ICI 182,780 (ICI) for 1 h and then incubated for 24 h with PRL. Data are expressed relative to the values obtained in control untreated cells. (b)ERa mRNA levels determined in the same samples.

immunoprecipitation (ChIP) assays in a-amanitin-syn- 20 min of incubation with either E2 or PRL, decreasing chronized populations of cells to compare pS2 promoter at 60 min and increasing again at 80 min. These changes occupancy by the receptor in T47D cells incubated with were specific for the ERE-containing region because PRL or E2. As shown in Figure 4, E2 recruited ERa to no ER recruitment was observed when an irrelevant the pS2 promoter region containing the ERE and upstream region of the promoter was amplified. The known to be required for E2 regulation of transcription effect of PRL applies to other ER target promoters, as (Nunez et al., 1989). This binding increased at 40 min demonstrated by a significant recruitment of ERa and and then decreased, in agreement with data obtained P-Ser-118-ERa to the ERE-containing GREB1C pro- in other breast cancer cell line (MCF-7), where ERa moter (Bourdeau et al., 2004) in response to PRL. In recruitment to the promoter is cyclical (Shang et al., addition, PRL as well as E2 caused pS2 promoter 2000; Metivier et al., 2003). Interestingly, PRL induced occupancy by ERa and P-Ser-118-ERa in MCF-7 cells ER binding to the pS2 promoter with similar kinetics. (Supplementary Figure 1), showing that this finding can Moreover, we observed that the phosphorylated recep- be extended to other breast cancer cell lines. tor was also recruited in response to both hormones. Nuclear receptors stimulate transcription by recruit- P-Ser-118-ERa was essentially absent from the promoter ment of co-activators that lead to local alteration of in the untreated cells, but was already associated after chromatin structure generated by post-translational

Oncogene Prolactin activates the estrogen receptor L Gonza´lez et al 1301 PRL 0′ 5′ 10′ 15′ 30′ 2h 4h 6h 16h 24h 48h E2 PRL 0′ 2′ 5′ 10′ 152h 6h 16h 24h

pERα pERα

ERα * ERα *

15′ 30′ 6h 16h 24h 15′ 2h 6h 24h ICI - - + + ---+++ - + TMX - ----++++ PRL - + - +++++++++ PRL - ++ ++ ++ + +

pERα pERα

ERα ERα * *

PRL PRL + PD PRL + LY PRL + PP2

pERα

ERα 0′ 15′ 30′ 1h 3h 0′ 15′ 30′ 1h 3h 0′ 15′ 30′ 1h 3h 0′ 15′ 30′ 1h 3h Figure 3 Prolactin (PRL) induces Ser-118-ERa phosphorylation. (a) Levels of ERa phosphorylated in Ser-118 (P-ERa) and total ERa levels were determined by western blot in T47D cells incubated with PRL for increasing time periods or with E2 for 30 min. (b) P-ERa and ERa levels determined in MCF-7 cells incubated with PRL for various time periods. (c) T47D cells were preincubated with ICI 182,780 (ICI) for 2 h and then stimulated with PRL for kinetic analysis. In (d) cells were preincubated with tamoxifen and then treated with PRL. (e) P-ERa and ERa levels in T47D cells preincubated for 2 h with the inhibitors PD184352 (1 mM), LY29402 (10 mM) or PP2 (1 mM) and then stimulated with PRL for the times indicated. In a and b detection of ERa by western blot was carried out with an antibody, which detects a doublet where the specific band is labeled with an asterisk. In d, the anti-ERa used detects a single band. modifications of histones such as acetylation (Aranda concentrations) caused a marked reduction of [3H]thy- and Pascual, 2001). Therefore, we examined, by ChIP, midine uptake and abolished the mitogenic effect of the effect of PRL on the binding of the p160 co- PRL. ICI also blocked the increase in T47D cell number activator SRC-1 and acetylated histone H3 to the triggered not only by E2 but also by PRL (Figure 6b). pS2 promoter. As illustrated in Figure 5, PRL- and To further examine the effect of PRL and ICI on breast E2-induced ER binding correlated with the recruitment cancer cell proliferation, flow cytometry analysis was of the co-activator and increased promoter acetylation. performed in T47D and MCF-7 cells. Most cells were in These data are consistent with the occupancy of G0/G1 as corresponding with serum starvation, but PRL estrogen-regulated promoters by ERa and P-Ser-118- treatment caused a modest but consistent reduction in ERa in response to PRL treatment, strongly suggesting the number of cells in this phase of the cell cycle and that ligand-independent receptor recruitment may re- induced a concomitant increase in the percentage of cells present a mechanism for transcriptional regulation of in S phase. Significantly, incubation with ICI abolished ER-target genes by PRL. the effect of PRL in both T47D (Figure 6c) and MCF-7 cells (Supplementary Figure 2). These results suggest again that ER activity is required for the proliferative ICI blocks PRL-dependent breast cancer cell proliferation response of breast cancer cells to the lactotrophic Previous studies have demonstrated that PRL induces hormone. To prove this point, ERa was knocked down moderate T47D cell growth (Acosta et al., 2003). The by means of short-interfering RNA (siRNA) in T47D ability of ICI to antagonize the effect of PRL on cells (Figure 6d), and the effect of PRL on the cell cycle transcription suggested that this compound could also was analysed by flow cytometry (Figure 6e). Stimulation suppress PRL-dependent cell proliferation. As shown in of cell-cycle progression by PRL was abolished in ERa- Figure 6a, incubation of T47D cells with PRL induced depleted cells, indicating that this receptor is required an increase of [3H]thymidine incorporation. The mod- for PRL-induced proliferation. As expected, the mito- erate activation of cell proliferation above basal levels genic effect of E2 was also abolished in cells transfected induced by PRL may be due to the fact that these cells with siERa. produce PRL, which by an autocrine/paracrine Prolactin-induced proliferation of breast cancer cells feedback loop can affect breast cancer cells even appears to require ERK1/2 and AKT activation. under serum-free media conditions (Clevenger et al., Therefore, it was possible that ICI could inhibit the 1995; Reynolds et al., 1997; Schroeder et al., 2002). effect of the polypeptide hormone on proliferation In addition, ICI (from nanomolar to millimolar by blocking activation of these kinases. However,

Oncogene Prolactin activates the estrogen receptor LGonza´lez et al 1302 E2 PRL pS2

0′ 20′ 40′ 60′ 80′ 20′ 40′ 60′ 80′ min. Promoter Irrelevant region

INPUTS pS2 INPUT

INPUTS IRR. pS2 IgGs

INPUTS GREB1C ER

pS2 Irrelevant GREB1C SRC-1

α α α α α α H3-Ac IgG ER P-ER IgG ER P-ER IgG ER P-ER

0′ 0′ 60′ 60′ 100′ 140′ 0′ 60′ 60′ 100′ 140′ E2 PRL E2 PRL E2-20′ Figure 5 Histone H3 acetylation and co-activator recruitment at the pS2 promoter. Binding of ERa, the p160 co-activator SRC-1 E2-40′ and acetylated histone H3 (H3-Ac) to the proximal pS2 promoter and to an irrelevant region was determined by chromatin immunoprecipitation (ChIP) assay after treatment of T47D cells with E2 ′ E2-60 and prolactin (PRL) as indicated. Normal immunoglobulin G (IgG) was used as a negative control. E2-80′

PRL-20′ At short incubation times ICI was unable to repress c-Myc or Cyclin D1 levels. However, at incubations longer than 6 h, and concomitant with ERa depletion, PRL-40′ ICI caused a marked reduction of these proteins, reversing PRL induction and lowering expression to PRL-60′ levels below those of untreated T47D cells. Similar results were obtained in MCF-7 cells in which ICI PRL-80′ also caused a time-dependent reduction of Cyclin D1 expression (Supplementary Figure 3B). Figure 4 Prolactin (PRL) induces recruitment of P-ERa and ERa Recently, it has been reported that Cyclin D1 to estrogen response element (ERE)-containing promoters. Binding induction by E2 involves ERa recruitment to an of ERa and P-Ser-118-ERa to the pS2 and GREB1C promoter fragments containing the ERE, and to an irrelevant fragment of the enhancer located downstream from the coding region pS2 gene. Binding was determined by chromatin immunoprecipita- (Eeckhoute et al., 2006). We therefore analysed by ChIP tion (ChIP) assays in T47D cells at different time points after E2 binding of ERa and P-Ser-118-ERa to this enhancer and PRL treatment. The upper panels show the inputs for each (enh 2), as well as to an upstream promoter region ChIP and the lower panels the results obtained with the ERa and P-ERa antibodies and with normal immunoglobulin G (IgG) used (enh 1) also shown to bind ERa (Eeckhoute et al., 2006). as a negative control. Incubation not only with E2 but also with PRL caused association of total and phosphorylated ERa with these regions both in T47D (Figure 8a) and MCF7 cells (Figure 8b), thus providing a link between PRL and incubation with PRL for different time periods was breast cancer cell proliferation. equally effective in stimulating ERK1/2 and PI3K activity in the presence and absence of ICI in T47D (Figure 7a) and MCF-7 cells (Supplementary Figure 3A). In addition, lowering ERa by means of siRNA did Discussion not block phosphorylation (Figure 7b). Prolactin-stimulated mitogenic signaling cascades ERa normally acts as a transcription factor in response ultimately induce the expression of c-Myc and Cyclin to binding of its cognate ligand. However, there is D1 (Acosta et al., 2003), which are critical molecules for increasing evidence that the presence of estrogen is not G1/S cell-cycle progression. We therefore evaluated an absolute requirement for activation of this receptor, whether ICI could block the effect of PRL on expression because growth factors such as IGF-1 or EGF and of these proteins. As illustrated in Figure 7c, and in intracellular protein kinases can induce an ERa activa- agreement with its effect on proliferation, PRL induced tion independently of ligand binding (Butt et al., 2005). a detectable increase of cell-cycle proteins in T47D cells. In this work we show that PRL can also activate ER in

Oncogene Prolactin activates the estrogen receptor L Gonza´lez et al 1303 1995). These actions are observed in serum-free medium 300 2.0 lacking phenol red and therefore most likely represent 250

-4 ligand-independent receptor activation, but we cannot 1.5 200 ignore the possibility that PRL could synergize with residual levels of estrogen. The effects of PRL are 1.0 150 abolished by the pure antagonist ICI, demonstrating the 100 0.5 involvement of ER activation. Moreover, we show that Cell number x 10 Cell number 50 PRL mimics the effect of E2 promoting the recruitment

3H-Thymidine incorporation 3H-Thymidine 0.0 0 of ERa to endogenous target promoters, independently PRL - + - + - + - + E2 - - + + - - PRL - - - - + - of E2. PRL-induced binding of ERa to the promoter ICI 0 1 100 1000 (nM) ICI - + - + - + determines co-activator recruitment and histone

90 12 acetylation, both landmarks of receptor-mediated transcriptional activation. 10 Increased ERa levels are common in breast cancer 85 8 (Fabris et al., 1987; Holst et al., 2007) and have been 6 associated with aberrant promoter occupancy, increased

80 % S phase 4 gene expression and cell proliferation in the absence of

% G0/G1 phase hormonal stimulation (Fowler et al., 2004). Because it 2 has been reported that PRL increases ERa levels in 75 0 PRL - - + + - + + - + + PRL - - + + - + + - + + breast cancer cells (Shaﬁe and Brooks, 1977; Gutzman ICI - + - + + - + + - + ICI - + - + + - + + - + et al., 2004a), the stimulatory effect of PRL on the ER 6h 12h 24h 6h 12h 24h target genes could be due to ERa induction. However, we did not observe increased receptor expression over siControl siERα the period in which PRL stimulates ER-dependent gene ERα transcription.

Erk2 It has been proposed that phosphorylation in the AF-1 domain is associated with estrogen-independent ER activation. Ser-118 is a well-studied phosphorylation siControl siControl 90 siERα 17.5 α site in ERa and both ER ligands and growth factors can siER induce this modiﬁcation (Kato et al., 1995; Bunone 15 et al., 1996; Chen et al., 2000, 2002). Our results show

85 that PRL signaling pathways cause a rapid and 12.5 sustained Ser-118-ERa phosphorylation. Chemical inhibition of Src, ERK or AKT abolishes receptor 10 80 % S phase phosphorylation by PRL. This is surprising because % G0/G1 phase the inhibitors used are accepted to be specific and a 7.5 partial reduction would have been predicted if multiple pathways are involved in phosphorylation. Interestingly, 75 5 we have reported that each of these inhibitors also ControlE2 PRL Control E2 PRL blocks PRL-induced breast cancer cell proliferation Figure 6 ERa is required for prolactin (PRL)-dependent T47D (Acosta et al., 2003), and these pathways are also cell proliferation. (a)[3H]thymidine incorporation determined in cells treated for 72 h with PRL in the presence of increasing required for the mitogenic effects of estrogen (Migliaccio concentrations of ICI 182,780 (ICI). Results are expressed relative et al., 1996, 1998). to those obtained in the untreated cells. (b) Cell number was Our results also show that PRL causes recruitment counted in cells treated with PRL and E2 in the absence and of the phosphorylated receptor to target promoters. presence of ICI. (c) Flow cytometry analysis of cells treated with Therefore, increased receptor modification could also PRL and/or ICI. The left panel represents the percentage of cells in contribute to the activation of ER-dependent gene G0/G1 and the right panel the percentage of cells in S phase. (d) Cells were transfected with a control siRNA or with an siRNA transcription secondary to PRL treatment in breast targeting ERa. After 24 h, cells were shifted to serum- and phenol- cancer cells. However, it should be noted that this red-free medium and ERa levels were determined 48 h later. modification cannot be univocally linked to transcrip- (e) Cell-cycle analysis in cells transfected with control or ERa siRNAs and treated with E2 or PRL for the last 24 h. tional activation, because ER antagonists also cause Ser-118-ERa phosphorylation (Lipfert et al., 2006). In fact, we have previously shown that ICI is at least as strong as E2 to induce a sustained increase of Ser-118- breast cancer cells leading estrogen-independent induc- ERa phosphorylation in MCF-7 cells (De los Santos tion of ER target gene expression. We have demon- et al., 2007), and in this work we observe that ER strated that PRL increases ERE activity in transient antagonists cooperate with PRL to increase P-Ser-118- transfection assays and also induces transcription of the ERa levels. In addition to ERa phosphorylation, the endogenous ER target gene pS2, considered a marker of activation of ERK1/2 and AKT by PRL may also lead breast cancer progression and as a resistance predictor to phosphorylation of receptor co-regulators (co-acti- of breast tumors to ER antagonists (Johnston et al., vators and co-repressors). This modification alters their

Oncogene Prolactin activates the estrogen receptor LGonza´lez et al 1304 0 30′ 6h 24h siControl siERα PRL --+++ + + + ICI -+ -+ - + - + PRL - + - +

P-Erk1/2 P-Erk1/2

Erk2 P-Akt

ERα P-Akt

Erk2 Akt

0′ 30′ 2h 6h 12h 24h 0′ 30′ 2h 6h 12h 24h PRL -++++++++ + + ------ICI -+-+-+-+- + - +++++ +

c-Myc

Cyclin D1

ERα **

Tubulin

Figure 7 Role of estrogen receptor (ER) in prolactin (PRL) signaling in T47D cells. (a) Kinetics of ERK and AKT phosphorylation was determined by western blot in T47D cells pretreated with ICI 182,780 (ICI) for 2 h and then with PRL for the indicated time periods. Total levels of ERK and AKT were used as loading controls. (b) ERK and AKT phosphorylation as well as ERa levels were determined in cells transfected with siControl or siERa and incubated with PRL for 30 min. ERk2 was used as a loading control. (c) c-Myc, Cyclin D1 and ERa levels in cells stimulated with PRL for the indicated time periods. ICI was added 2 h before PRL. Tubulin was used as a loading control. The speciﬁc ERa band is labeled with an asterisk.

activity, their interaction with transcription factors and PRL is modest has been attributed to the production of their cellular redistribution (Jonas and Privalsky, 2004; PRL within the mammary cells themselves, because in Wu et al., 2005) and could therefore contribute to PRL- MCF-7 cells that do not express endogenous PRL dependent transcriptional stimulation. Furthermore, it treatment with the hormone has strong effects on the has been demonstrated that induction of progesterone levels of cell-cycle regulators and on cell proliferation target genes in breast cancer cells involves activation of (Schroeder et al., 2002). the ERK1/2 cascade and phosphorylation of histone The fact that repression of cell-cycle proteins by ICI is H3, as a prerequisite for recruitment of co-activators concomitant with ERa downregulation reinforces the and transcriptional activation (Vicent et al., 2006). It is idea that ER activation is involved in the effect of the plausible that activation of this signaling pathway by lactotrophic hormone on breast cancer cell prolifera- PRL could also induce histone H3 phosphorylation, tion. There was the possibility that the ability of ICI to promoting the transcription of ER target genes. repress cell proliferation in the presence of PRL could Both E2 and PRL induce breast cancer cell growth by simply indicate that the ER antagonist targets processes modifying the expression of key regulatory components essential for cell-cycle progression rather than inhibiting of cell-cycle progression such as Cyclin D1 or c-Myc processes activated by PRL. However, we found that (Musgrove et al., 1993, 1994; Carroll et al., 2002; depletion of ERa with siRNA was able to block the Schroeder et al., 2002; Acosta et al., 2003). These mitogenic effect of PRL, demonstrating a key function proteins are frequently overexpressed in human breast for ERa in this hormonal action. In contrast, other PRL cancers and have been implicated in the development of actions such as ERK activation appear to be ERa mammary hyperplasia and carcinogenesis (McNeil independent, because they are not blocked by treatment et al., 2006; Wang et al., 2007). We have confirmed that with ICI or by ERa downregulation. PRL induces moderate Cyclin D1 and c-Myc expres- In summary, our results indicate a novel layer of sion, and that the anti-estrogen counteracts their complexity in the interaction between ER and PRL induction, as well as PRL-dependent breast cancer cell signaling in breast cancer. The finding that the growth. The finding that induction of these proteins by polypeptide hormone can activate ERa in a ligand-

Oncogene Prolactin activates the estrogen receptor L Gonza´lez et al 1305 INPUTS INPUTS E2 PRL E2 PRL

C 20′ 40′ 60′ 80′ 20′ 40′ 60′ 80′ C 20′ 40′ 60′ 80′ 20′ 40′ 60′ 80′ Cyclin D1 enh 2

Cyclin D1 enh 1

enh2enh1 enh2 enh1

α α α α α α α α IgG ER P-ER IgG ER P-ER IgG ER P-ER IgG ER P-ER

Control Control

E2-20′ E2-20′

E2-40′ E2-40′

E2-60′ E2-60′

E2-80′ E2-80′

PRL-20′ PRL-20′

PRL-40′ PRL-40′

PRL-60′ PRL-60′

PRL-80′ PRL-80′

T-47D MCF-7 Figure 8 Prolactin (PRL) induces recruitment of P-ERa and ERa to the cyclin D1 gene sites responsible for E2 induction. Association of ERa and P-Ser-118-ERa to an enhancer located downstream the cyclin D1 coding region (enh 2), and to an upstream enhancer located at À2000 (enh 1). Binding was determined by chromatin immunoprecipitation (ChIP) assays in T47D (a) and MCF-7 cells (b) at the indicated time points of incubation with E2 and PRL. The upper panels show the inputs for each ChIP and the lower panels the results obtained with the ERa and P-ERa antibodies and with normal immunoglobulin G (IgG) used as a negative control. independent manner evidences the significance of this glutamine, 100 IU/ml penicillin and 100 mg/ml streptomycin, mechanism in the development of breast cancer and and 72 h before transfection were shifted to serum-free medium suggests that anti-estrogen therapy acts not only by lacking phenol red. Cells were transiently transfected with 5 mg inhibiting E2 actions but also by antagonizing PRL of a luciferase reporter plasmid that contains three copies of a effects on breast cancer cells. consensus ERE by incubation with a mixture of cationic liposomes (1.5 ml/mg DNA) for 6 h (De los Santos et al., 2007). Cells were then treated with 100 nM PRL, ICI or E2 for the indicated times, and luciferase activity was determined. Materials and methods Experiments were performed with triplicate cultures and each experiment was repeated at least three times. Data are Materials represented as means±standard deviations. siRNA for ERa Ovine PRL (NIDDK-oPRL-20; 31 IU/mg) was kindly pro- and non-target control were purchased from Dharmacon vided by the National Hormone and Pituitary Program of the (catalogue nos. L-003401-00 and D-001210-01-05). siRNA National Institute of Diabetes and Digestive and Kidney transfections were performed using 33 nM of each siRNA and Diseases (Bethesda, MD, USA). ICI was obtained from Tocris Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA), as Cookson (Ballwin, MO, USA) and estradiol (E2) and recommended by the manufacturer. The efficiency of knock- tamoxifen (TMX) were obtained from Sigma (San Luis, down was determined by western blot. MO, USA). PD184352, LY29402 and PP2 were obtained from Calbiochem (San Diego, CA, USA). [3H]thymidine was Cell proliferation obtained from Amersham Pharmacia Biotech (Buckingham- Exponentially growing T47D cells were inoculated in 24-well shire, UK). plates, and 24 h later were washed and shifted to serum- and phenol-free medium. Cells were kept in this medium for 48 h Cell culture and transfection and then treated for 72 h with PRL and/or ICI, and for the last T47D and MCF-7 cells were grown in Dulbecco’s modified 4 h with 0.5 mCi per well of [3H]thymidine (48 Ci/mmol). Cells Eagle’s medium supplemented with 10% fetal calf serum, 2 mM were disrupted and incorporated radioactivity was determined

Oncogene Prolactin activates the estrogen receptor LGonza´lez et al 1306 in a liquid scintillation Beta Wallac Counter. Alternatively, RNA following specifications of SuperScript First-Strand cells were inoculated in 60 mm Petri dishes and the number of Synthesis System (Invitrogen Life Technologies). PCR reac- cells was counted in Neubauer chambers after treatment with tions were performed in a Mx3005P thermocycler (Stratagene) PRL or E2 alone or in combination with ICI. and detected with SYBR Green using the following primers: pS2 50-tcccctggtgcttctatcctaa-30 (forward) and 50-AGTGTCTA Flow cytometry AAATTCACACTCCTCTTCT-30 (reverse), and ERa 50-CCA Duplicate cultures of T47D and MCF-7 cells grown in 60 mm CCAACCAGTGCACCATT-30 (forward) and 50-GGTCTT Petri dishes were transferred to the serum- and phenol-red- TTCGTATCCCACCTTTC-30 (reverse). Values obtained were depleted medium and after 72 h incubated with PRL and/or corrected by glyceraldehyde-3-phosphate dehydrogenase ex- ICI for different time periods. Cells were collected and stained pression determined with primers 50-ACAGTCCATGCCATC with propidium iodide for sorting as previously described ACTGCC-30 (forward) and 50-ctagctgacctccttgacctg-30 (re- (De los Santos et al., 2007). verse). Results were analysed by the CT comparative method (DDCT). Western blot Cells previously incubated for 48 h in serum- and phenol-red- ChIP assays depleted medium were treated with PRL (100 ng/ml) alone or Cells growing in p150 dishes were maintained in depleted in combination with ICI or tamoxifen (100 nM) for the times medium for 72 h, washed twice in serum-free medium and indicated. Antagonists were added 2 h before PRL. Cells were treated for 2.5 h with 2.5 mM a-amanitin (Sigma). As previously washed twice in ice-cold Tris-buffered saline (20 mM Tris-HCl, described this treatment is required for a preliminary silencing pH 7.6; 140 mM NaCl) with 0.1 mM Na3VO4 and lysed at 4 1C of the pS2 promoter (Metivier et al., 2003). Cells were then in 1 ml of lysis buffer (10 mM Tris-HCl, pH 7.6; 50 mM NaCl; washed and treated with PRL or E2. At the indicated time 30 mM sodium pyrophosphate; 5 mM EDTA; 0.5% Nonidet points, cells were fixed with 1% formaldehyde for 15 min at 1 P-40; 1% Triton X-100; 50 mM NaF; 0.1 mM Na3VO4;1mM 37 C. The Chromatin Immunoprecipitation Assay kit from phenylmethylsulfonylfluoride; 1 mM benzamidine; 1 mM iodo- Upstate (catalogue no. 17–295) was used. Sonication was acetamide and 1 mM phenantroline). Cell lysates were performed using a Bioruptor UCD-200TM (Diagenode) obtained by centrifugation at 17 000 g for 15 min at 4 1C, following manufacturer’s directions. For each immunopreci- protein concentration in the supernatant was determined by pitation 2.5–3.0 Â 106 cells and 3 mg of the following anti- the BCA protein assay (Pierce Chemical Co, Rockford, IL, bodies: anti-acetylated histone 3 (06-599; Upstate, Lake Placid, USA), and lysates were adjusted to equivalent concentrations NY, USA), anti-SRC-1 (sc-8995), anti-ER (sc-542), anti-pSer- with lysis buffer. Proteins from cell lysates were separated in 118-ERa (sc-12915-R) and normal rabbit serum immunoglo- SDS–polyacrylamide gel electrophoresis and transferred to bulins (sc-2027) were used. DNAs were subjected to 35 cycles polyvinylidene difluoride membranes (Immobilon; Millipore, of PCR with primers: forward 50-GCCATCTCTCACTAT Billerica, MA, USA) that were blocked for 1 h at room GAATC-30 and reverse 50-GGATTTGCTGATAGACAGAG- temperature with 4% bovine serum albumin. Incubation with 30 to amplify the ERE-containing pS2 promoter region antibodies was performed in blocking solution for 1 h at room (À392/À199 bp); forward 50-CAGTCTGGCAAATCATTCC temperature. Blots were visualized with enhanced chemilumi- CAAAC-30 and reverse 50-CACATCTGAGAGGTAAGAG nescence (ECL; Amersham Pharmacia Biotech). Rabbit GAGGTG-30 to amplify an irrelevant pS2 region (Weitsman polyclonal antibodies to Erk2 (C14), Akt (H-136), c-Myc et al., 2006); forward 50-TTGTTGTAGCTCTGGGAGCA-30 (9E10) and Cyclin D1 (H-295) were obtained from Santa Cruz and reverse 50-CAACCAGCCAAGAGGCTAAG-30 to Biotechnology Inc. (Santa Cruz, CA, USA). The ERa amplify the proximal GREB1C promoter region that contains antibodies used were either from Dako (1D5) (Carpinteria, the ERE; forward 50-CAGTTTGTCTTCCCGGGTTA-30 and CA, USA) or a kind gift of S Ramos. The later antibody reverse 50-TCATCCAGAGCAAACAGCAG-30 to amplify the detects a doublet where the upper band is the nonspecific. downstream enhancer (enh 2) of the cyclin D1 gene; and Antiphospho antibodies to pErk1/2, pAkt and pSer-118-ERa forward 50-GCTCTTTACGCTCGCTAACC-30 and reverse were from New England Biolabs (Beverly, MA, USA). 50-GGGCAGATCTCGACTAGGAA-30 to amplify the up- Secondary horseradish-peroxidase-conjugated antibodies were stream ER binding region (enh 1) of this gene (Eeckhoute purchased from BioSource International (Camarillo, CA, et al., 2006). USA), and the ECL kit was from Amersham Pharmacia Biotech. Primary antibodies were used at a 1:1000 dilution. Acknowledgements

Real-time PCR This work was supported by grants SAF2006-00371 and Cells were cultured for 48 h in depleted medium before BFU2007-62402 from the Ministerio de Educacio´ n y Ciencia, incubation with PRL and/or ICI. Total RNA was extracted RD06/0020/0036 and PIO40682 from the Fondo de Investi- using Tri Reagent (Sigma) and mRNA levels were analysed by gaciones Sanitarias, by a grant from the Fundacio´ n MMA and quantitative real-time PCR. RT was performed with 2 mgof by the EU Project CRESCENDO (FP6-018652).

References

Acosta JJ, Munoz RM, Gonzalez L, Subtil-Rodriguez A, Dominguez- Bourdeau V, Deschenes J, Metivier R, Nagai Y, Nguyen D, Caceres MA, Garcia-Martinez JM et al. (2003). Src mediates Bretschneider N et al. (2004). Genome-wide identiﬁcation of high- prolactin-dependent proliferation of T47D and MCF7 cells via the afﬁnity estrogen response elements in human and mouse. Mol activation of focal adhesion kinase/Erk1/2 and phosphatidylinositol Endocrinol 18: 1411–1427. 3-kinase pathways. Mol Endocrinol 17: 2268–2282. Brockman JL, Schroeder MD, Schuler LA. (2002). PRL activates the Aranda A, Pascual A. (2001). Nuclear hormone receptors and gene cyclin D1 promoter via the Jak2/Stat pathway. Mol Endocrinol 16: expression. Physiol Rev 81: 1269–1304. 774–784.

Oncogene Prolactin activates the estrogen receptor L Gonza´lez et al 1307 Bunone G, Briand PA, Miksicek RJ, Picard D. (1996). Activation Johnston SR, Saccani-Jotti G, Smith IE, Salter J, Newby J, Coppen M of the unliganded estrogen receptor by EGF involves the et al. (1995). Changes in estrogen receptor, progesterone receptor, MAP kinase pathway and direct phosphorylation. EMBO J 15: and pS2 expression in tamoxifen-resistant human breast cancer. 2174–2183. Cancer Res 55: 3331–3338. Butt AJ, McNeil CM, Musgrove EA, Sutherland RL. (2005). Jonas BA, Privalsky ML. (2004). SMRT and N-CoR corepressors are Downstream targets of growth factor and oestrogen signalling and regulated by distinct kinase signaling pathways. J Biol Chem 279: endocrine resistance: the potential roles of c-Myc, cyclin D1 and 54676–54686. cyclin E. Endocr Relat Cancer 12(Suppl 1): S47–S59. Kato S, Endoh H, Masuhiro Y, Kitamoto T, Uchiyama S, Sasaki H Campbell GS, Argetsinger LS, Ihle JN, Kelly PA, Rillema JA, et al. (1995). Activation of the estrogen receptor through phosphory- Carter-Su C. (1994). Activation of JAK2 tyrosine kinase by lation by mitogen-activated protein kinase. Science 270: 1491–1494. prolactin receptors in Nb2 cells and mouse mammary gland Likhite VS, Stossi F, Kim K, Katzenellenbogen BS, Katzenellenbogen explants. Proc Natl Acad Sci USA 91: 5232–5236. JA. (2006). Kinase-specific phosphorylation of the estrogen receptor Carroll JS, Swarbrick A, Musgrove EA, Sutherland RL. (2002). changes receptor interactions with ligand, deoxyribonucleic acid, Mechanisms of growth arrest by c-myc antisense oligonucleotides in and coregulators associated with alterations in estrogen and MCF-7 breast cancer cells: implications for the antiproliferative tamoxifen activity. Mol Endocrinol 20: 3120–3132. effects of antiestrogens. Cancer Res 62: 3126–3131. Lipfert L, Fisher JE, Wei N, Scafonas A, Su Q, Yudkovitz J et al. Clevenger CV, Chang WP, Ngo W, Pasha TL, Montone KT, (2006). Antagonist-induced, activation function-2-independent es- Tomaszewski JE. (1995). Expression of prolactin and prolactin trogen receptor alpha phosphorylation. Mol Endocrinol 20: 516–533. receptor in human breast carcinoma. Evidence for an autocrine/ Liu X, Robinson GW, Wagner KU, Garrett L, Wynshaw-Boris A, paracrine loop. Am J Pathol 146: 695–705. Hennighausen L. (1997). Stat5a is mandatory for adult mammary Clevenger CV, Furth PA, Hankinson SE, Schuler LA. (2003). The role gland development and lactogenesis. Genes Dev 11: 179–186. of prolactin in mammary carcinoma. Endocr Rev 24: 1–27. Lonard DM, Nawaz Z, Smith CL, O’Malley BW. (2000). The 26S Chen D, Riedl T, Washbrook E, Pace PE, Coombes RC, Egly JM et al. proteasome is required for estrogen receptor-alpha and coactivator (2000). Activation of estrogen receptor alpha by S118 phosphoryla- turnover and for efficient estrogen receptor-alpha transactivation. tion involves a ligand-dependent interaction with TFIIH and Mol Cell 5: 939–948. participation of CDK7. Mol Cell 6: 127–137. McNeil CM, Sergio CM, Anderson LR, Inman CK, Eggleton SA, Chen D, Washbrook E, Sarwar N, Bates GJ, Pace PE, Thirunuvakkarasu Murphy NC et al. (2006). c-Myc overexpression and endocrine V et al. (2002). Phosphorylation of human estrogen receptor alpha at resistance in breast cancer. J Steroid Biochem Mol Biol 102: 147–155. serine 118 by two distinct signal transduction pathways revealed by Medunjanin S, Hermani A, De Servi B, Grisouard J, Rincke G, phosphorylation-specific antisera. Oncogene 21: 4921–4931. Mayer D. (2005). Glycogen synthase kinase-3 interacts with and De los Santos M, Martinez-Iglesias O, Aranda A. (2007). Anti- phosphorylates estrogen receptor alpha and is involved in the estrogenic actions of histone deacetylase inhibitors in MCF-7 breast regulation of receptor activity. J Biol Chem 280: 33006–33014. cancer cells. Endocr Relat Cancer 14: 1021–1028. Metivier R, Penot G, Hubner MR, Reid G, Brand H, Kos M et al. Dong J, Tsai-Morris CH, Dufau ML. (2006). A novel estradiol/ (2003). Estrogen receptor-alpha directs ordered, cyclical, and estrogen receptor alpha-dependent transcriptional mechanism combinatorial recruitment of cofactors on a natural target controls expression of the human prolactin receptor. J Biol Chem promoter. Cell 115: 751–763. 281: 18825–18836. Migliaccio A, Di Domenico M, Castoria G, de Falco A, Bontempo P, Eeckhoute J, Carroll JS, Geistlinger TR, Torres-Arzayus MI, Brown Nola E et al. (1996). Tyrosine kinase/p21ras/MAP-kinase pathway M. (2006). A cell-type-specific transcriptional network required for activation by estradiol-receptor complex in MCF-7 cells. EMBO J estrogen regulation of cyclin D1 and cell cycle progression in breast 15: 1292–1300. cancer. Genes Dev 20: 2513–2526. Migliaccio A, Piccolo D, Castoria G, Di Domenico M, Bilancio A, Fabris G, Marchetti E, Marzola A, Bagni A, Querzoli P, Nenci I. Lombardi M et al. (1998). Activation of the Src/p21(ras)/Erk (1987). Pathophysiology of estrogen receptors in mammary tissue by pathway by progesterone receptor via cross-talk with estrogen monoclonal antibodies. J Steroid Biochem 27: 171–176. receptor. EMBO J 17: 2008–2018. Fowler AM, Solodin N, Preisler-Mashek MT, Zhang P, Lee AV, Murphy L, Cherlet T, Adeyinka A, Niu Y, Snell L, Watson P. (2004). Alarid ET. (2004). Increases in estrogen receptor-alpha concentra- Phospho-serine-118 estrogen receptor-alpha detection in human tion in breast cancer cells promote serine 118/104/106-independent breast tumors in vivo. Clin Cancer Res 10: 1354–1359. AF-1 transactivation and growth in the absence of estrogen. FASEB Murphy LC, Weitsman GE, Skliris GP, Teh EM, Li L, Peng B et al. J 18: 81–93. (2006). Potential role of estrogen receptor alpha (ERalpha) Fuh G, Wells JA. (1995). Prolactin receptor antagonists that inhibit phosphorylated at Serine118 in human breast cancer in vivo. the growth of breast cancer cell lines. J Biol Chem 270: 13133–13137. J Steroid Biochem Mol Biol 102: 139–146. Ginsburg E, Vonderhaar BK. (1995). Prolactin synthesis and secretion Murphy LJ, Murphy LC, Vrhovsek E, Sutherland RL, Lazarus L. by human breast cancer cells. Cancer Res 55: 2591–2595. (1984). Correlation of lactogenic receptor concentration in human Gutzman JH, Miller KK, Schuler LA. (2004a). Endogenous human breast cancer with estrogen receptor concentration. Cancer Res 44: prolactin and not exogenous human prolactin induces estrogen 1963–1968. receptor alpha and prolactin receptor expression and increases Musgrove EA, Hamilton JA, Lee CS, Sweeney KJ, Watts CK, estrogen responsiveness in breast cancer cells. J Steroid Biochem Mol Sutherland RL. (1993). Growth factor, steroid, and steroid Biol 88: 69–77. antagonist regulation of cyclin gene expression associated with Gutzman JH, Nikolai SE, Rugowski DE, Watters JJ, Schuler LA. changes in T-47D human breast cancer cell cycle progression. Mol (2005). Prolactin and estrogen enhance the activity of activating Cell Biol 13: 3577–3587. protein 1 in breast cancer cells: role of extracellularly regulated Musgrove EA, Lee CS, Buckley MF, Sutherland RL. (1994). Cyclin kinase 1/2-mediated signals to c-fos. Mol Endocrinol 19: 1765–1778. D1 induction in breast cancer cells shortens G1 and is sufficient for Gutzman JH, Rugowski DE, Schroeder MD, Watters JJ, Schuler LA. cells arrested in G1 to complete the cell cycle. Proc Natl Acad Sci (2004b). Multiple kinase cascades mediate prolactin signals USA 91: 8022–8026. to activating protein-1 in breast cancer cells. Mol Endocrinol 18: 3064–3075. Neilson LM, Zhu J, Xie J, Malabarba MG, Sakamoto K, Wagner KU Hennighausen L, Robinson GW. (2005). Information networks in the et al. (2007). Coactivation of janus tyrosine kinase (Jak)1 positively mammary gland. Nat Rev Mol Cell Biol 6: 715–725. modulates prolactin–Jak2 signaling in breast cancer: recruitment of Holst F, Stahl PR, Ruiz C, Hellwinkel O, Jehan Z, Wendland M et al. ERK and signal transducer and activator of transcription (Stat)3 (2007). Estrogen receptor alpha (ESR1) gene amplification is and enhancement of Akt and Stat5a/b pathways. Mol Endocrinol 21: frequent in breast cancer. Nat Genet 39: 655–660. 2218–2232.

Oncogene Prolactin activates the estrogen receptor LGonza´lez et al 1308 Nunez AM, Berry M, Imler JL, Chambon P. (1989). The 50 flanking by quantitative polymerase chain reaction in human breast region of the pS2 gene contains a complex enhancer region responsive tumors versus normal breast tissues. J Clin Endocrinol Metab 83: to oestrogens, epidermal growth factor, a tumour promoter (TPA), the 667–674. c-Ha-ras oncoprotein and the c-jun protein. EMBO J 8: 823–829. Tworoger SS, Hankinson SE. (2006). Prolactin and breast cancer risk. Ormandy CJ, Hall RE, Manning DL, Robertson JF, Blamey RW, Cancer Lett 243: 160–169. Kelly PA et al. (1997). Coexpression and cross-regulation of the Vicent GP, Ballare C, Nacht AS, Clausell J, Subtil-Rodriguez A, prolactin receptor and sex steroid hormone receptors in breast Quiles I et al. (2006). Induction of progesterone target genes requires cancer. J Clin Endocrinol Metab 82: 3692–3699. activation of Erk and Msk kinases and phosphorylation of histone Park KJ, Krishnan V, O’Malley BW, Yamamoto Y, Gaynor RB. H3. Mol Cell 24: 367–381. (2005). Formation of an IKKalpha-dependent transcription com- Wang Y, Thakur A, Sun Y, Wu J, Biliran H, Bollig A et al. (2007). plex is required for estrogen receptor-mediated gene activation. Synergistic effect of cyclin D1 and c-Myc leads to more aggressive Mol Cell 18: 71–82. and invasive mammary tumors in severe combined immunodeficient Reynolds C, Montone KT, Powell CM, Tomaszewski JE, Clevenger mice. Cancer Res 67: 3698–3707. CV. (1997). Expression of prolactin and its receptor in human breast Weitsman GE, Li L, Skliris GP, Davie JR, Ung K, Niu Y et al. (2006). carcinoma. Endocrinology 138: 5555–5560. Estrogen receptor-alpha phosphorylated at Ser118 is present at the Schroeder MD, Symowicz J, Schuler LA. (2002). PRL modulates cell promoters of estrogen-regulated genes and is not altered due to cycle regulators in mammary tumor epithelial cells. Mol Endocrinol HER-2 overexpression. Cancer Res 66: 10162–10170. 16: 45–57. Wijayaratne AL, McDonnell DP. (2001). The human estrogen Shafie S, Brooks SC. (1977). Effect of prolactin on growth and the receptor-alpha is a ubiquitinated protein whose stability is affected estrogen receptor level of human breast cancer cells (MCF-7). differentially by agonists, antagonists, and selective estrogen Cancer Res 37: 792–799. receptor modulators. J Biol Chem 276: 35684–35692. Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M. (2000). Cofactor Wu RC, Smith CL, O’Malley BW. (2005). Transcriptional regulation dynamics and sufficiency in estrogen receptor-regulated transcrip- by steroid receptor coactivator phosphorylation. Endocr Rev 26: tion. Cell 103: 843–852. 393–399. Touraine P, Martini JF, Zafrani B, Durand JC, Labaille F, Malet C Yager JD, Davidson NE. (2006). Estrogen carcinogenesis in breast et al. (1998). Increased expression of prolactin receptor gene assessed cancer. N Engl J Med 354: 270–282.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Oncogene