Transcriptional activation by the oestrogen receptor a is ... - Nature

Oncogene (2002) 21, 5773 – 5782 ª 2002 Nature Publishing Group All rights reserved 0950 – 9232/02 $25.00 www.nature.com/onc

Transcriptional activation by the oestrogen receptor a is modulated through inhibition of cyclin-dependent kinases Ge´rard Redeuilh1, Annette Attia1, Jan Mester1 and Miche`le Sabbah*,1 1

Institut National de la Sante´ et de la Recherche Me´dicale U 482, Hoˆpital Saint-Antoine, 184 Rue du Faubourg Saint-Antoine, 75571 Paris Cedex 12, France

We have investigated the interaction between the expression of p21WAF1/CIP1/SDI1, a stoichiometric inhibitor of Cdk, and the transcriptional activity of the oestrogen receptor a (ERa). Transient transfection experiments demonstrated that the expression of p21WAF1/CIP1/SDI1 amplified the transcriptional activation by ERa. A dominant negative mutant of Cdk2 also enhanced the ERa transcriptional activity, indicating that the underlying mechanism relies on the inhibition of Cdk2 activity and cell cycle arrest. In agreement with this conclusion, experiments with p21WAF1/CIP1/SDI1 mutants demonstrated that the domain involved in the binding of p21WAF1/CIP1/SDI1 to Cdks was indispensable for the modulation of ERa activity. In addition, we show that expression of p21WAF1/CIP1/SDI1 alleviates the block on CBP function mediated by Cdk2 and in turn stimulates transcriptional activation by ERa in a CBPhistone acetyltransferase (HAT)-dependent manner. These results suggest a novel mechanism by which p21WAF1/CIP1/SDI1 functions as an enhancer of ERa activity through the modulation of CBP function. Oncogene (2002) 21, 5773 – 5782. doi:10.1038/sj.onc. 1205753 Keywords: oestrogen receptor; p21waf1; cell cycle; coactivators; cyclin-dependent kinases

Introduction The oestrogen receptor (ER) belongs to a superfamily of structurally conserved transcription factors, called nuclear receptors (NR), which, upon hormone binding, enhance transcription of specific genes (Mangelsdorf et al., 1995). Like other members of the superfamily, ERa (NR3A1; Nuclear Receptors Nomenclature Committee, 1999) exhibits a modular structure composed of three functional regions: the amino-terminal region containing the transactivation function AF-1, the DNA binding central domain (DBD), and the carboxyl-

*Correspondence: M Sabbah, INSERM U482, Hoˆpital SaintAntoine, 184 rue du Faubourg Saint-Antoine, 75571 Paris Cedex 12, France; E-mail: [email protected] Received 29 January 2002; revised 16 May 2002; accepted 7 June 2002

terminal region that encompasses the ligand binding domain, a dimerization surface, and a ligand-dependent transactivation function AF-2 (Beato et al., 1995). Upon activation by hormone binding, the receptor interacts specifically with a cis-acting DNA sequence called the oestrogen response element (ERE), which is usually located upstream of the promoter and displays enhancer properties (Beato et al., 1995). For regulation of gene expression, the oestrogen receptor a has to interact with basal transcription factors, such as TFIIB (Ing et al., 1992; Sabbah et al., 1998), TBP (Cavailles et al., 1995), and TAFII 30 (Jacq et al., 1994), which help to stabilize the preinitiation complex at the promoter. These associations are an intrinsic property of ERa, independent of ligand, and can account for the constitutive enhancement of transcription observed in vitro. In addition, the binding of oestrogens to the receptor induces a conformational change in the hormone-binding domain (Brzozowski et al., 1997) which then binds to transcriptional co-activators, such as p160/SRC family, the co-integrators p300, CBP and p300/CBP associated factor p/CAF (Rosenfeld and Glass, 2001). These co-activators can form complexes with the receptor as well as with one another. They promote transcription, at least in part, by acetylating histones and by overcoming the repressive effects of chromatin structure on transcription (Rosenfeld and Glass, 2001). Additionally, p300/CBP can also interact with the RNA polymerase II complex and thus provide a link between the receptor and transcription initiation (Nakajima et al., 1997a,b). Recent research has shown that co-activators can bind to and stimulate the activity of a variety of transcription factors involved in multiple and sometimes opposite cellular activities (Glass and Rosenfeld, 2000). The same co-activator proteins target transcription factors involved in both cell proliferation such as E2F (Trouche et al., 1996) and AP-1 (Arias et al., 1994), and differentiation such as MyoD (Puri et al., 1997) and retinoic acid receptor (Kurokawa et al., 1998). The association of co-activators with either type of transcription factors can be regulated by diverse signals from mitogens or differentiation-inducing stimuli. For instance, the phosphorylation of the CREB transcription factor by protein kinase A regulates its interaction with p300/CBP and stimulates the transcriptional activity of CRE-dependent genes. However, regulated phosphorylation of p300/CBP on

p21waf1 activates ER G Redeuilh et al

5774

serine and threonine residues by cyclin-dependent kinase (Ait-Si-Ali et al., 1998; Yaciuk et al., 1991) suggests their possible role as adapter proteins that coordinate cell proliferation and differentiation. In this respect, it has been proposed that the inhibition of AP1 activity by nuclear receptors is a result of competition for the limiting concentration of p300/ CBP in cells (Kamei et al., 1996). Recently, it has been shown that the steroid receptor co-activators (SRC-1 and AIB1) are targets of the mitogen-activated protein kinase pathway, activated early in the G1 phase in growth factor stimulated cells (Font de Mora and Brown, 2000; Rowan et al., 2000a,b). ERa is required for cell proliferation or differentiation in oestrogen-dependent tissues, including the breast (Korach, 1994). It is well established that cyclin-dependent kinases (Cdks) control the transitions between successive phases of the cell cycle in all eukaryotic cells (Sherr, 1994). The activities of cyclinCdk complexes are regulated, in part, through the interaction with two classes of Cdk inhibitory proteins (CdkIs) (Harper and Elledge, 1996). The first class of CdkIs consists of p16INK4A, p15INK4B, p18INK4C, and p19INK4D. These CdkIs target Cdk4 and Cdk6 and inhibit their binding to cyclins. The second class of CdkIs consists of p21WAF1/CIP1/SDI1, p27KIP1 and p57KIP2 which bind to and inhibit all known cyclinCdk complexes. Apart from cyclin-dependent kinases, p21WAF1/CIP1/SDI1 (hereafter referred to as p21) can also directly prevent DNA synthesis by inhibiting DNA polymerase g and e (Chen et al., 1995; Zhang et al., 1993). Conversely, the inhibition of p21 expression in G0-arrested cells can induce DNA synthesis and cell cycle proliferation (Nakanishi et al., 1995). It has been proposed that p21 plays a positive role in the commitment to differentiate, as this protein is upregulated in the early stage of differentiation (Halevy et al., 1995; Liu et al., 1996). In this regard, the expression of p21 induced by MyoD has been demonstrated to promote myogenic differentiation. Both p300/CPB and p/CAF functions are required in this mechanism (Halevy et al., 1995; Puri et al., 1997; Sartorelli et al., 1999). A similar mechanism has been proposed for retinoic acid-induced F9-cell differentiation (Kawasaki et al., 1998). In addition, p21 inhibits the cell cycle and stimulates cell differentiation by increasing the transcriptional activity of NF-kB (Perkins et al., 1997). Previous studies indicated that Cdk-inhibitory activity in the human oestrogen-dependent MCF-7 breast cancer cell line is predominantly attributable to p21 (Prall et al., 1997). Furthermore, a significant correlation between p21 immunoreactivity and well differentiated histological grade in ER-positive breast carcinoma has been reported (Oh et al., 2001). As both ER and p21 have growth-regulatory effects in normal and tumour cells, we have investigated whether p21 can act as a modulator of actions of oestrogens. Our experiments demonstrate that p21 amplifies the transcriptional activation by ERa. Finally, we show that p21 enhances the transcriptional activation by ERa in Oncogene

a CBP histone acetyl transferase (HAT)-dependent manner. Results p21 functions as a positive regulator of hormonedependent oestrogen receptor transcriptional activity To determine whether ER function was influenced by cell cycle regulatory proteins, MELN cells derived from ER-positive MCF-7 breast cancer cells by stable transfection of the luciferase cDNA placed downstream of the oestrogen inducible element were transfected together with increasing amounts of a p21 expression vector or the empty expression vector cassette, and the cells were placed in medium containing oestradiol. We found that cotransfecting p21 led to a dose-dependent increase in ER activity (Figure 1a). It should be kept in mind that only a fraction of MELN cells express the transfected p21 while all cells respond to oestradiol by transcribing the integrated luciferase gene, so that the overall results underestimate the true enhancement of ER activity by p21. Similar results were observed when MCF-7 cells were transfected with an oestrogen responsive luciferase reporter construct (ERE-tk-luc) and increasing amounts of a p21 expression vector (Figure 1b). Expression of p21 did not affect either the basal activity of the ER or the activity of the receptor bound to the anti-oestrogen ICI 182 780, which blocks both activation functions of ER (AF-1 and AF-2) (Figure 1b). In the presence of tamoxifen, a mixed agonist/antagonist known to block only the AF-2 activation function of ER in MCF-7 cells, ERE-tkluc activity was slightly increased by expression of p21 (Figure 1b). Next, we have extended our analysis by using the HCT116 p21-deleted cells HCT116 p217/7. When these cells were co-transfected with an ERa expression vector together with the ERE-tk-luc construct, the level of induction by oestradiol (E2) was approximately twofold. A clear additional increase of luciferase activity was observed when p21 was cotransfected (Figure 1c). p21 did not potentiate the promoter activity in the absence of ER (data not shown). We also tested the ability of p21 to affect transactivation of the ERb isoform of the oestrogen receptor. The results indicate that p21 also increased the transcriptional activity of ERb in a dose-dependent manner (Figure 1c). These results indicate that p21 acts as a positive regulator of the ER activity. Subsequently, we investigated which domains of the ERa are involved in the mechanism by which p21 increases the transcriptional activity of ERa. Expression vectors encoding mutants HE19 (CDEF regions), HE15 (ABC regions) and HE11 (full length ERa deleted of the C region forming the core of the DBD) were co-transfected with p21 and ERE-tk-luc plasmids into HCT116 p217/7 cells (Figure 2a). The cells were incubated with oestradiol and their luciferase activity was determined. Previous work (Kumar et al., 1987) has established that each of these ERa variants is expressed at comparable levels from these vectors. In


5775

the presence of the wild type ERa (HEG0), p21 enhanced reporter gene activity 3.1-fold. HE19 transcriptional activity was induced twofold by expression of p21. HE15, which contains the constitutively active transactivating domain AF-1 and the DBD but lacks the ligand binding domain displayed no oestrogeninducible activity, with or without co-expression of p21. As expected, the receptor DBD was necessary as deletion of this domain (HE11) abolished the induction of transcription (Figure 2a). These experiments demonstrate that the DBD is required and that AF1 and AF2 have to cooperate in order to give an optimal p21-stimulation of ERa activity. We addressed the possibility that p21 could affect the DNA binding activity of ERa. We used COS-7 cells in order to obtain a high transfection efficiency (450%). Cells were transfected with vectors expressing HEGO together with a plasmid expressing p21. After stimulation with hormone for 1 h, nuclear extracts were prepared and tested by EMSA (Figure 2b). Under these experimental conditions, p21 did not modify ER DNA binding activity, although p21 was clearly expressed (Figure 2b). Importantly, over-expression of p21 had no effect on the expression of ERa (Figure 2b). Cdk2 kinase activity is implicated in the regulation of ERa transactivation mediated by p21

Figure 1 p21 enhances the ER transcriptional activity. (a) MELN cells were transiently transfected with increasing concentrations of p21 expression vector. After transfection, cells were incubated with 1078 M 17b-oestradiol for 24 h, lysed and assayed for luciferase activity. (b) MCF-7 cells were transiently transfected with 0.5 mg of ERE-tk-luc together with increasing concentrations of p21 expression vector. After transfection, cells were incubated with vehicle, 1078 M 17b-oestradiol, 1077 M tamoxifen or 1077 M ICI 182 780 as indicated. After 24 h, cells were lysed and assayed for luciferase activity. (c) HCT116(p217/7) cells were transiently transfected with 0.5 mg of ERE-tk-luc together with 0.1 mg of ERa or ERb expression vector and increasing concentrations of p21 expression vector. After transfection, cells were incubated with 1078 M 17b-oestradiol for 24 h and assayed for luciferase ac-

To further characterize the molecular mechanisms underlying p21-stimulated ERa activity, we used mutants p21cdk7 impaired in interaction with Cdks and p21PCNA7 impaired in interaction with the proliferating cell nuclear antigen (PCNA), an auxiliary factor for DNA polymerases d and e. Since in most cells p21 arrests cell proliferation in G1 phase, thereby preventing the induction of cyclin A expression (Zindy et al., 1992), we used the cyclin A promoter-based reporter system to monitor the p21-induced growth inhibition in transient transfection assays (Feng et al., 1995). The ability of the p21 mutants to arrest cell proliferation was examined in the cyclin A-luciferase assay using HCT116 p217/7 cells. Likewise wild type p21, the expression of the p21PCNA7 mutant downregulated the luciferase activity, contrary to the p21cdk7 mutant (Figure 3a). The expression levels of p21 mutant proteins were comparable to those of the wild-type protein as revealed by immunoblotting using an anti-p21 antibody (data not shown). These results confirm that p21 and p21PCNA7 have antiproliferative effects in HCT116 p217/7 cells. Next we compared the effects of these mutants on ERa activity. The mutant p21cdk7 was inefficient at stimulating ERa activity whereas the mutant p21PCNA7 was *90% as

tivity. The results shown represent the average of a minimum of three independent experiments assayed in duplicate. The activity derived from the ER in the absence of oestradiol was normalized to 1 and the activity of other effectors is expressed relative to this Oncogene


5776

Figure 2 (a) Effect of p21 on transcriptional activation of various ERa mutants. HCT116(p217/7) cells were transiently transfected with 0.5 mg of ERE-tk-luc and 0.1 mg of plasmids expressing different regions of ERa, with or without 1.2 mg of p21 expression vector. After transfection, cells were incubated with 1078 M 17b-oestradiol for 24 h and assayed for luciferase activity. The results shown represent the average of a minimum of three independent experiments assayed in duplicate. (b) Effect of p21 on ERa DNA binding activity. COS-7 cells were transfected with plasmids expressing ERa together with empty vector or with increasing concentrations of vectors expressing p21 (1, 2, 3, 4 and 5 mg). After transfection, cells were incubated with 1078 M 17boestradiol for 1 h. Nuclear extracts were analysed by EMSA or by Western blot

efficient as the wild type p21 (Figure 3b). These results suggest that p21-induced ERa activity is dependent of the binding of p21 to Cdks. To verify the role of Cdkdependent protein phosphorylation in ERa activity, we used an unrelated CdkI, p16INK4A, specific for Cdk4 and Cdk6 and Cdk2-dn, a catalytically inactive mutant dominant acting as a dominant negative form of Cdk2, that causes cell cycle arrest at G1/S when overexpressed (van den Heuvel and Harlow, 1993). As seen in Figure 3a,b, expression of p16 or of Cdk2-dn inhibited the cyclin A promoter activity and enhanced ERa activity. These observations indicate that ERa transcriptional activity is stimulated by proteins that inhibit phosphorylation by Cdks. Oncogene

Figure 3 Cdk2 kinase activity is involved in ERa transactivation mediated by p21. (a) HCT116(p217/7) were transfected with expression vectors (1 mg) of wild type p21, mutated p21, p16 or Cdk2-dn together with 0.5 mg of cyclin A-luc. Cyclin A-luciferase data are presented as the per cent inhibition of luciferase activity relative to control cells transfected with empty vector. In cells transfected with p21Cdk7 the cyclin A activity was higher than control, resulting in a negative inhibition. (b) HCT116(p217/7) were transfected with expression vectors (1 mg) of wild type p21, mutated p21, p16 or Cdk2-dn together with 0.1 mg of plasmids expressing the ERa and 0.5 mg of ERE-tk-luc. After transfection, cells were treated with 1078 M 17b-oestradiol for 24 h and assayed for luciferase activity. The results shown represent the average of a minimum of three independent experiments assayed in duplicate. The activity derived from the ERa in the presence of oestradiol was normalized to 1 and the activity of other effectors is expressed relative to this


5777

p21 mediates the effect of CBP on ERa activity Previous reports have shown that the cyclin E-Cdk2 complex associates with p300/CBP (Felzien et al., 1999; Perkins et al., 1997). In addition, p300/CBP has been shown to interact with ERa and to stimulate its transcriptional activity (Chakravarti et al., 1996; Hanstein et al., 1996; Smith et al., 1996). To investigate whether CBP is involved in the p21induced ERa activity, we transfected HCT116 and HCT116 p217/7 cells with the ERa expression vector and with increasing amounts of a CBP expression vector together with ERE-tk-luc. Remarkably, CBP did not enhance the ERa activity in HCT116 p217/7 cells even at the highest concentration used whereas in the parental cell line HCT116 we observed an increase in the ERa activity at all concentrations of CBP (Figure 4a). The additional co-transfection of p21 in HCT116 p217/ 7 cells resulted in enhancement of ERa activity by CBP similar to that observed in HCT116 cells (Figure 4a), in support of the hypothesis that p21 contributes to ERa-directed transcriptional activation by CBP. CBP promotes transcription in part through its ability to acetylate histones or non-histone proteins (Kouzarides, 2000). To determine whether or not this catalytic function is required to mediate the effects of p21 in stimulating ERa activity, we performed similar experiments as described above except that a dominant negative mutant of CBP (CBPDHAT) lacking acetylase activity was transfected instead of wild type CBP. The expression level of the CBP mutant protein was comparable to that of the wild type protein as revealed by immunoblotting using an anti-CBP antibody (data not shown). Consistent with previous results (Chen et al., 1999), the DHAT mutant was severely deficient in potentiation of ERa activity in HCT116 and HCT116 p217/7 cells (compare Figure 4b with Figure 4a). Additional cotransfection of p21 in HCT116 p217/7 cells did not result in enhancement of ERa activity by CBPDHAT. These results indicate that acetyltransferase activity of CBP is required for p21 to increase transcriptional activity of ERa. As p/CAF and SRC-1 as well as CBP can be recruited to the liganded ERa and enhance oestrogen induced ERa activity when over-expressed (Blanco et al., 1998; Smith et al., 1996), we decided to investigate the role of these co-activators in the p21-stimulated ERa activity. We performed similar experiments as with CBP (see above). As expected, we found that p/CAF and SRC-1 potentiated the transcriptional activity of ERa (Figure 5a,b). In contrast with CBP, these co-activators enhanced the transcriptional activity of ERa to the same extent in HCT116 or HCT116 p217/7 cells (Figure 5a,b). However, additional co-transfection of p21 in HCT116 p217/7 cells resulted in a further enhancement of ERa activity by p/CAF and SRC-1 (data not shown).

p21 modulates the phosphorylation status and the HAT activity of CBP Since CBP is a phosphoprotein and a potential substrate of Cdk (Yaciuk et al., 1991), we investigated

Figure 4 p21 mediates the effect of CBP on ERa activity. (a) HCT116 and HCT116(p217/7) cells were transfected with increasing concentrations of expression vector for CBP (0.1, 0.5 and 1 mg) together with 0.1 mg of plasmid expressing ERa and 0.5 mg of ERE-tk-luc. The same experiment was performed with HCT116(p217/7) transfected with expression vector for p21 (1 mg). (b) CBP HAT domain is required for ERa transcriptional activity. HCT116 and HCT116(p217/7) cells were transfected and treated as in A except that CBPDHAT was used instead of CBP wild type. In all these experiments, cells were treated with 1078 M 17b-oestradiol for 24 h and assayed for luciferase activity. The results shown represent the average of a minimum of three independent experiments assayed in duplicate. The activity derived from the ERa in the presence of oestradiol was normalized to 1 and the activity of other effectors is expressed relative to this Oncogene


5778

[32P]orthophosphate. Cell lysates were then analysed by immunoprecipitation with CBP-specific polyclonal antibodies and electrophoresis. As shown in Figure 6a, expression of p21 decreased the phosphorylation of endogenous CBP, whereas the level of CBP was not significantly affected. These results are consistent with in vitro and in vivo phosphorylation data suggesting that Cdk2 is involved in CBP phosphorylation at the G1/S boundary (Ait-Si-Ali et al., 1998; Perkins et al., 1997). These data underestimate the true decrease of CBP phosphorylation resulting from the expression of p21: CBP can be phosphorylated also by kinases other than cdk2 (Goodman and Smolik, 2000) and the fraction of transfected cells does not exceed 20 – 30%. Next, we tested the effect of p21 on CBP HAT activity. HAT assays were performed with HA-CBP immunoprecipitated from HCT116 p217/7 cells transfected or not with p21. We found that p21 increased the HAT activity of CBP whereas the level of HA-CBP was not significantly affected (Figure 6b). Discussion

Figure 5 Effects of p/CAF and SRC-1 on ERa activity in HCT116 and HCT116(p217/7) cells. (a) HCT116 and HCT116(p217/7) cells were transfected with increasing concentrations of expression vector for p/CAF (0.5, 1 and 1.5 mg) together with 0.1 mg of plasmid expressing ERa and 0.5 mg of ERE-tk-luc. (b) HCT116 and HCT116(p217/7) cells were transfected and treated as in A except that SRC-1 was used instead of p/CAF. In all these experiments, cells were treated with 1078 M 17b-oestradiol for 24 h and assayed for luciferase activity. The results shown represent the average of a minimum of three independent experiments assayed in duplicate. The activity derived from ERa in the presence of oestradiol was normalized to 1 and the activity of other effectors is expressed relative to this

the effect of p21 on its state of phosphorylation. HCT116 p217/7 cells were transfected with p21 expression vector and cells were labelled with Oncogene

In this report, we demonstrate that the inhibition of Cdk2 activity enhances transcriptional activation by the oestrogen receptor a. The inhibition of Cdk2 activity was obtained by the transfection of expression vectors encoding CdkI proteins (p21, p16INK4) or of a dominant negative variant of Cdk2. The expression of any of these proteins led to the arrest of the cell cycle as verified by the inhibition of the activity of the cyclin A promoter, confirming the efficient inhibition of the Cdk cascade. p16INK4A is a selective inhibitor of Cdk4/ Cdk6 activity, and indirectly of the activation of Cdk2 which relies on the expression of cyclins E and A following the phosphorylation of Rb by Cdk4/Cdk6 (Jiang et al., 1998; McConnell et al., 1999; Mitra et al., 1999). p21 inhibits Cdk2 directly and Cdk2-dn inhibits the phosphorylation of Cdk2 substrates by competition with the endogenous enzyme. Cdk2 activation is a critical event in the G1/S transition. Inhibition of Cdk2 leads to an arrest of cell proliferation and is often associated with induction of differentiation. In this context, it has been reported that overexpression of p21 in an ER-negative breast cancer cell line confers a differentiated phenotype reversed by antioestrogens, possibly through an ER-related protein (Chen et al., 2000). We have analysed in more detail the mechanisms of the interaction between the transcriptional activity of ERa and the expression of p21. The enhancement of ER activity by inhibition of Cdk2 was detected in oestrogen-dependent MCF-7 breast cancer cells transiently or stably transfected with an oestrogen responsive luciferase reporter construct as well as in HCT116 colon cancer cells with inactivated p21 gene. In the absence of agonist ligand, inhibition of Cdk2 did not induce the ER activity. Analysis of ERa deletion constructs revealed that the DBD is required and that AF1 and AF2 cooperate in order to mediate the p21-stimulated ERa activity. In addition,


5779

Figure 6 (a) p21 reduces the phosphorylation of CBP. HCT116(p217/7) cells were transfected with empty vector or with p21 expression vector. After 24 h, cells were metabolically labelled with [32P]orthophosphate for 4 h. Cell lysates were prepared and tested in immunoprecipitation reactions with CBPspecific polyclonal antibodies. Precipitated proteins were separated by SDS – PAGE and detected by autoradiography. The total CBP protein contents were evaluated by Western blot using a polyclonal anti-CBP antibodies. (b) p21 increases HAT activity of CBP. HCT116(p217/7) cells were transfected with HA-CBP together with empty vector or expression vector of p21. After 24 h, cell lysates were prepared, immunoprecipitated with antiHA antibody and incubated with calf thymus histones and [3H]acetyl-CoA for 1 h at 308C. [3H]acetyl-CoA incorporation into histones was determined by liquid scintillation counting. The total HA-CBP protein contents were evaluated by Western blot using a monoclonal anti-HA

we show that the expression of p21 stimulates transcriptional activation by ERa through a CBPdependent mechanism. Consistent with previous results (Chen et al., 1999), our data suggest that the stimulation of ERa activity by CBP is dependent on its HAT function. For the understanding of the role of CBP, it is important to note that CBP-stimulated ERa activity is dependent on the presence of p21 in the cells, although the expression of p21 does not augment the

intracellular levels of CBP or ERa nor does it increase the sequence-specific DNA-binding of ERa. It has been reported that CBP HAT activity is regulated in a cell cycle-dependent manner, through phosphorylation by the cyclin E-Cdk2 complex or by an unknown related kinase (Ait-Si-Ali et al., 1998). In contrast with these reports we found that the introduction of p21 into p21-negative cells increased the HAT activity of CBP although CBP was in hypophosphorylated state. These results are consistent with previous reports demonstrating that p21 strongly enhances transactivation by the RelA (p65) NF-kB subunit (Perkins et al., 1997). This was correlated with the inhibition of CBP/p300-associated cyclin E-Cdk2 activity. Furthermore, HAT activity of CBP potentiates different NF-kB-driven gene promoters (Vanden Berghe et al., 1999). More recently it has been shown that p300 and CBP contain a transcriptional repression domain (CRD1) and that transactivation by coactivators is stimulated through inactivation of this domain by p21 (Snowden et al., 2000). Although the precise mechanism has not been elucidated, it has been suggested that regulation of the CRD1 domain by p21 is indirect and arises from cell cycle arrest. We can hypothesize that inactivation of this repression domain can alleviate the block of the HAT activity of CBP. It is interesting to note that CBP/p300 HAT activity is critical for myogenic terminal differentiation downstream of the expression of p21 (Polesskaya et al., 2001). In summary, this work reveals a new mechanistic link between gene activation by ERa and signalling that controls cell cycle progression. In addition to the effects of oestradiol on cyclin D1 expression (Altucci et al., 1996; Planas-Silva et al., 2001; Sabbah et al., 1999), it has been recently demonstrated that cyclin D1, which is frequently over-expressed in ER-positive breast cancer, can act as a bridging factor between ERa, SRCs and p/CAF to form a transcriptionally active complex (McMahon et al., 1999; Zwijsen et al., 1998). The role of cyclin D1 in the control of proliferation of the mammary gland tissue has been also verified in mice with deleted cyclin D1 gene: such mice are defective in pregnancy-associated tissue (Sicinski et al., 1995). As expected, we find that SRC-1 and p/CAF increase the transcriptional activity of ERa. However, the lack of p21 does not significantly affect p/CAF nor SRC-1-stimulated ERa activity. Therefore, p21 and cyclin D1 can discriminate between the different coactivator families. These observations suggest that the ability of p21 and cyclin D1 to stimulate ERa may contribute to the transcriptional activation of specific genes involved in respectively differentiation or proliferation. The regulation of oestrogen-inducible differentiation-related genes by p21 is not an exclusive mechanism as female mice lacking p21 were found to develop normally and did not show defects in reproductive functions or lactation (Brugarolas et al., 1995; Deng et al., 1995). Thus, in vivo, the differentiation function of the ER remains intact in p217/7 Oncogene


5780

mice, and although p21 appears to play an important role in regulating cell cycle progression and differentiation, its activity may be functionally redundant in normal cells. As p300/CBP, SRC-1 and p/CAF are part of many transcription factor complexes, the role of these coactivators in the control of gene expression during proliferation or differentiation may be more complex than initially thought. The precise mechanisms through which p21 modulates the functions of these coactivators need to be further investigated. Materials and methods

and CMV-p16 were obtained from Drs E Harlow and RM Michalides, respectively. Western blot analysis Nuclear extracts from transfected COS-7 cells were prepared as described in Andrews and Faller (1991). Equal amounts of nuclear protein extracts were separated by sodium dodecyl sulphate (SDS)-polyacrylamide gel electrophoresis (PAGE) and blotted to a cellulose nitrate membrane. Immunoblot analysis was performed with an anti-oestrogen receptor (H222, Abott Laboratory) or anti-p21 (F5, Santa Cruz Biotechnology) monoclonal antibodies, and immunodetection was performed with an enhanced chemiluminescence system (Amersham Pharmacia Biotech).

Cell culture COS-7, an African green monkey kidney cell line and MCF7, a human breast cancer cell line were obtained from the American Type Culture Collection. MELN cells are MCF-7 cells stably transfected with the ERE-luciferase construct (Gagne et al., 1994). All these cell lines were grown in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% foetal bovine serum (FBS). HCT116, a human colorectal cancer cell line and a p21-null clone HCT116p217/7 derived by homologous recombination (Waldman et al., 1995) were grown in McCoy’s 5A medium supplemented with 10% FBS. All cell culture media were supplemented with 200 U/ml penicillin and 200 mg/ml streptomycin and all cells were grown at 378C in a humidified 5% CO2 atmosphere. Transient transfection Transfections were performed by the Lipofectamine method (Gibco – BRL) following manufacturer’s protocols. Prior to experiments, cells were placed for 24 h in medium without phenol red containing 10% charcoal-stripped serum. For assays of ER activity, 26105 cells were transfected with 0.5 mg of the ERE-tk-luc construct (Klock et al., 1987) and various expression vectors. For each transfection, the total quantity of transfected plasmid DNA was kept constant by the addition of pcDNA3 plasmid (Invitrogen). After 16 h, fresh medium containing vehicle, 1078 M 17b-oestradiol, 1077 M tamoxifen or 1077 M ICI 182 780 was added. After a further 24 h, cells were harvested and lysed in Reporter Lysis Buffer (Promega) prior to analysis of reporter activity using a luciferase assay system (Promega). Luciferase was measured using a Lumat LB 9507 luminometer (Berthold). All luciferase values were normalized by protein concentrations and expressed relative to the basal or oestradiol-induced promoter activity as fold induction. For cyclin A-luciferase assays, HCT116 (p217/7) were transfected with vector encoding wild-type p21, mutated p21, p16 or Cdk2-dn and the cyclin A-luciferase reporter (Feng et al., 1995). The luciferase activity was measured 48 h after transfection. Expression plasmids The expression plasmids for full length hERa and its fragments (HEG0, HE19, HE15, HE11) were described in (Sabbah et al., 1998). The expression plasmids pcDNA3-p21, pcDNA3-p21cdk7 and pcDNA3-p21PCNA7 were gifts from Dr B Ducommun. The expression plasmids pRc/RSV-CBP, pCMV-CBPDHAT, pCMXFlag-PCAF and pcDNA3-SRC-1 were gifts from Drs RH Goodman, D Trouche, Y Nakatani and B O’Malley respectively. The vectors pCMV-Cdk2-dn Oncogene

Phosphorylation assays HCT116 (p217/7) cells were transfected with empty vector or with a p21 expression vector. After 24 h, cells were metabolically labelled with [32P]orthophosphate (500 mCi/ml; Amersham) in phosphate-free DMEM for 4 h. The cells were lysed by incubation at 48C for 30 min in lysis buffer (20 mM tris-HCl, pH 8.0, 100 mM NaCl, 1 mM EDTA, 0.5% NP-40) supplemented with 10 mM NaF, 0.1 mM Na3VO4, 1 mM phenylmethylsulphonylfluoride, 1 mg/ml leupeptin, 0.5 mg/ml pepstatin and 1 mg/ml aprotinin. The lysates were centrifuged 30 min at 30 000 g and the supernatants were stored at 7808C. Proteins were immunoprecipitated with anti-CBP polyclonal antibody (A22, Santa Cruz Biotechnology) or rabbit IgGs as a control at 48C for 2 h, and collected on protein G-Sepharose beads (Amersham Pharmacia Biotech) at 48C for 1 h. The beads were washed five times with 1 ml of lysis buffer and proteins were analysed by SDS – PAGE. The gel was dried and exposed to X-ray film at 7708C. The level of proteins expression and the autoradiographs were quantified by densitometric scanning. The total CBP protein contents were evaluated by Western blot using a polyclonal anti-CBP. The immune complexes were detected with an enhanced chemiluminescence detection kit (Amersham Pharmacia Biotech). Immunoprecipitation histone acetyl transferase assays HCT116 (p217/7) cells were transfected with HA-CBP together with empty vector or with a p21 expression vector. After 24 h, cell lysates were immunoprecipitated with antiHA monoclonal antibody and the HAT assay was performed as described in (Fu et al., 2000) with 5 mg of calf thymus histones (Sigma, type II-A) and 90 pmoles of [3H]acetyl-CoA at 308C for 1 h. [3H]acetyl-CoA incorporation into the substrates was determined by liquid scintillation counting (Brownell and Allis, 1995). Electrophoretic mobility shift assays (EMSA) For DNA binding analysis, Cos-7 cells were transfected with plasmids expressing ERa together with empty vector or with increasing concentrations of vectors expressing p21. After transfection, cells were incubated with 1078 M 17b-oestradiol for 2 h. Nuclear extracts were prepared as described in Andrews and Faller (1991). A 40-bp double strand oligonucleotide containing the ERE element was end-labelled with [a-32P]dCTP and the Klenow fragment of the DNA polymerase I. Binding assays were performed as described (Sabbah et al., 1996). The products of the reactions were separated on 6% polyacrylamide, 0.25X TBE gels.


5781

Acknowledgments We are very grateful to Drs B Ducommun, D Gagne, R Goodman, E Harlow, Y Nakatani, B O’Malley, R Michalides, D Trouche, L Tora and B Vogelstein for generously providing various expression vectors and cell

lines. We thank D Catala for technical assistance. This work was supported by the Centre National de la Recherche Scientifique (CNRS), the Association pour la Recherche sur le Cancer and by the Ligue Nationale contre le Cancer, Comite´ de Paris.

References Ait-Si-Ali S, Ramirez S, Barre FX, Dkhissi F, MagnaghiJaulin L, Girault JA, Robin P, Knibiehler M, Pritchard LL, Ducommun B, Trouche D and Harel-Bellan A. (1998). Nature, 396, 184 – 186. Altucci L, Addeo R, Cicatiello L, Dauvois S, Parker MG, Truss M, Beato M, Sica V, Bresciani F and Weisz A. (1996). Oncogene, 12, 2315 – 2324. Andrews NC and Faller DV. (1991). Nucleic Acids Res., 19, 2499. Arias J, Alberts AS, Brindle P, Claret FX, Smeal T, Karin M, Feramisco J and Montminy M. (1994). Nature, 370, 226 – 229. Beato M, Herrlich P and Schutz G. (1995). Cell, 83, 851 – 857. Blanco JC, Minucci S, Lu J, Yang XJ, Walker KK, Chen H, Evans RM, Nakatani Y and Ozato K. (1998). Genes Dev., 12, 1638 – 1651. Brownell JE and Allis CD. (1995). Proc. Natl. Acad. Sci. USA, 92, 6364 – 6368. Brugarolas J, Chandrasekaran C, Gordon JI, Beach D, Jacks T and Hannon GJ. (1995). Nature, 377, 552 – 557. Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA and Carlquist M. (1997). Nature, 389, 753 – 758. Cavailles V, Dauvois S, L’Horset F, Lopez G, Hoare S, Kushner PJ and Parker MG. (1995). EMBO J., 14, 3741 – 3751. Chakravarti D, LaMorte VJ, Nelson MC, Nakajima T, Schulman IG, Juguilon H, Montminy M and Evans RM. (1996). Nature, 383, 99 – 103. Chen H, Lin RJ, Xie W, Wilpitz D and Evans RM. (1999). Cell, 98, 675 – 686. Chen J, Jackson PK, Kirschner MW and Dutta A. (1995). Nature, 374, 386 – 388. Chen X, Danes C, Lowe M, Herliczek TW and Keyomarsi K. (2000). J. Natl. Cancer Inst., 92, 1403 – 1413. Deng C, Zhang P, Harper JW, Elledge SJ and Leder P. (1995). Cell, 82, 675 – 684. Felzien LK, Farrell S, Betts JC, Mosavin R and Nabel GJ. (1999). Mol. Cell. Biol., 19, 4241 – 4246. Feng XH, Filvaroff EH and Derynck R. (1995). J. Biol. Chem., 270, 24237 – 24245. Font de Mora J and Brown M. (2000). Mol. Cell. Biol., 20, 5041 – 5047. Fu M, Wang C, Reutens AT, Wang J, Angeletti RH, Siconolfi-Baez L, Ogryzko V, Avantaggiati ML and Pestell RG. (2000). J. Biol. Chem., 275, 20853 – 20860. Gagne D, Balaguer P, Demirpence E, Chabret C, Trousse F, Nicolas JC and Pons M. (1994). J. Biolumin. Chemilumin., 9, 201 – 209. Glass CK and Rosenfeld MG. (2000). Genes Dev., 14, 121 – 141. Goodman RH and Smolik S. (2000). Genes Dev., 14, 1553 – 1577. Halevy O, Novitch BG, Spicer DB, Skapek SX, Rhee J, Hannon GJ, Beach D and Lassar AB. (1995). Science, 267, 1018 – 1021.

Hanstein B, Eckner R, DiRenzo J, Halachmi S, Liu H, Searcy B, Kurokawa R and Brown M. (1996). Proc. Natl. Acad. Sci. USA, 93, 11540 – 11545. Harper JW and Elledge SJ. (1996). Curr. Opin. Genet. Dev., 6, 56 – 64. Ing NH, Beekman JM, Tsai SY, Tsai MJ and O’Malley BW. (1992). J. Biol. Chem., 267, 17617 – 17623. Jacq X, Brou C, Lutz Y, Davidson I, Chambon P and Tora L. (1994). Cell, 79, 107 – 117. Jiang H, Chou HS and Zhu L. (1998). Mol. Cell. Biol., 18, 5284 – 5290. Kamei Y, Xu L, Heinzel T, Torchia J, Kurokawa R, Gloss B, Lin SC, Heyman RA, Rose DW, Glass CK and Rosenfeld MG. (1996). Cell, 85, 403 – 414. Kawasaki H, Eckner R, Yao TP, Taira K, Chiu R, Livingston DM and Yokoyama KK. (1998). Nature, 393, 284 – 289. Klock G, Strahle U and Schutz G. (1987). Nature, 329, 734 – 736. Korach KS. (1994). Science, 266, 1524 – 1527. Kouzarides T. (2000). EMBO J., 19, 1176 – 1179. Kumar V, Green S, Stack G, Berry M, Jin JR and Chambon P. (1987). Cell, 51, 941 – 951. Kurokawa R, Kalafus D, Ogliastro MH, Kioussi C, Xu L, Torchia J, Rosenfeld MG and Glass CK. (1998). Science, 279, 700 – 703. Liu M, Lee MH, Cohen M, Bommakanti M and Freedman LP. (1996). Genes Dev., 10, 142 – 153. Mangelsdorf DJ, Thummel C, Beato M, Herrlich P, Schutz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P and Evans RM. (1995). Cell, 83, 835 – 839. McConnell BB, Gregory FJ, Stott FJ, Hara E and Peters G. (1999). Mol. Cell. Biol., 19, 1981 – 1989. McMahon C, Suthiphongchai T, DiRenzo J and Ewen ME. (1999). Proc. Natl. Acad. Sci. USA, 96, 5382 – 5387. Mitra J, Dai CY, Somasundaram K, El-Deiry WS, Satyamoorthy K, Herlyn M and Enders GH. (1999). Mol. Cell. Biol., 19, 3916 – 3928. Nakajima T, Uchida C, Anderson SF, Lee CG, Hurwitz J, Parvin JD and Montminy M. (1997a). Cell, 90, 1107 – 1112. Nakajima T, Uchida C, Anderson SF, Parvin JD and Montminy M. (1997b). Genes Dev., 11, 738 – 747. Nakanishi M, Adami GR, Robetorye RS, Noda A, Venable SF, Dimitrov D, Pereira-Smith OM and Smith JR. (1995). Proc. Natl. Acad. Sci. USA, 92, 4352 – 4356. Nuclear Receptors Nomenclature Committee. (1999). Cell, 97, 161 – 163. Oh YL, Choi JS, Song SY, Ko YH, Han BK, Nam SJ and Yang YH. (2001). Pathol. Int., 51, 94 – 99. Perkins ND, Felzien LK, Betts JC, Leung K, Beach DH and Nabel GJ. (1997). Science, 275, 523 – 527. Planas-Silva MD, Shang Y, Donaher JL, Brown M and Weinberg RA. (2001). Cancer Res., 61, 3858 – 3862. Polesskaya A, Naguibneva I, Fritsch L, Duquet A, Ait-Si-Ali S, Robin P, Vervisch A, Pritchard LL, Cole P and HarelBellan A. (2001). EMBO J., 20, 6816 – 6825.

Oncogene


5782

Prall OW, Sarcevic B, Musgrove EA, Watts CK and Sutherland RL. (1997). J. Biol. Chem., 272, 10882 – 10894. Puri PL, Sartorelli V, Yang XJ, Hamamori Y, Ogryzko VV, Howard BH, Kedes L, Wang JY, Graessmann A, Nakatani Y and Levrero M. (1997). Mol. Cell, 1, 35 – 45. Rosenfeld MG and Glass CK. (2001). J. Biol. Chem., 276, 36865 – 36868. Rowan BG, Garrison N, Weigel NL and O’Malley BW. (2000a). Mol. Cell. Biol., 20, 8720 – 8730. Rowan BG, Weigel NL and O’Malley BW. (2000b). J. Biol. Chem., 275, 4475 – 4483. Sabbah M, Courilleau D, Mester J and Redeuilh G. (1999). Proc. Natl. Acad. Sci. USA, 96, 11217 – 11222. Sabbah M, Kang KI, Tora L and Redeuilh G. (1998). Biochem. J., 336, 639 – 646. Sabbah M, Radanyi C, Redeuilh G and Baulieu EE. (1996). Biochem. J., 314, 205 – 213. Sartorelli V, Puri PL, Hamamori Y, Ogryzko V, Chung G, Nakatani Y, Wang JY and Kedes L. (1999). Mol. Cell, 4, 725 – 734. Sherr CJ. (1994). Cell, 79, 551 – 555. Sicinski P, Donaher JL, Parker SB, Li T, Fazeli A, Gardner H, Haslam SZ, Bronson RT, Elledge SJ and Weinberg RA. (1995). Cell, 82, 621 – 630.

Oncogene

Smith CL, Onate SA, Tsai MJ and O’Malley BW. (1996). Proc. Natl. Acad. Sci. USA, 93, 8884 – 8888. Snowden AW, Anderson LA, Webster GA and Perkins ND. (2000). Mol. Cell. Biol., 20, 2676 – 2686. Trouche D, Cook A and Kouzarides T. (1996). Nucleic Acids Res., 24, 4139 – 4145. van den Heuvel S and Harlow E. (1993). Science, 262, 2050 – 2054. Vanden Berghe W, De Bosscher K, Boone E, Plaisance S and Haegeman G. (1999). J. Biol. Chem., 274, 32091 – 32098. Waldman T, Kinzler KW and Vogelstein B. (1995). Cancer Res., 55, 5187 – 5190. Yaciuk P, Carter MC, Pipas JM and Moran E. (1991). Mol. Cell. Biol., 11, 2116 – 2124. Zhang H, Xiong Y and Beach D. (1993). Mol. Biol. Cell, 4, 897 – 906. Zindy F, Lamas E, Chenivesse X, Sobczak J, Wang J, Fesquet D, Henglein B and Brechot C. (1992). Biochem. Biophys. Res. Commun., 182, 1144 – 1154. Zwijsen RM, Buckle RS, Hijmans EM, Loomans CJ and Bernards R. (1998). Genes Dev., 12, 3488 – 3498.