Abstract
Estrogen treatment of MCF-7 cells grown in serum-free medium induced a modification of the intracellular distribution of p53 protein. Western blot analysis and immunofluorescence staining showed that p53 was localized in the nucleus of untreated cell and that after 48 h of hormone treatment, it was mostly localized in the cytoplasm. This effect was blocked by the antiestrogen ICI182,780. Intracellular redistribution of p53 was correlated to a reduced expression of the WAF1/CIP1 gene product and to the presence of degradation fragments of p53 in the cytosol. Estradiol treatment prevented the growth inhibition induced by oligonucleotide transfection, simulating DNA damage. This observation indicated that the wild-type p53 gene product present in the MCF-7 cell could be inactivated by estradiol through nuclear exclusion to permit the cyclin-dependent phosphorylation events leading to the G1-S transition. In addition, the estradiol-induced inactivation of p53 could be involved in the tumorigenesis of estrogen-dependent neoplasm.
Introduction
Different types of stress (e.g., hypoxia or DNA damage) induce a cellular response leading to growth arrest or to apoptosis through a functional activation of p53 gene product (1) . DNA-dependent protein kinase (2) or the gene product mutated in ataxia telangiectasia, ATM (3 , 4) , are involved in alternative pathways leading to p53 activation. Activated p53 protein acts as a transcription factor, increasing the expression of genes such as p21 (5) , GADD45 (6) , or bax (7) , whose products inhibit cell cycle progression or induce apoptosis. Activation of p53 protein is limited by a short feedback loop involving the MDM2 gene product (8) . p53 stimulates transcription of the MDM2 gene (9) , and MDM2 protein binds to activated p53 protein (10) . This interaction inhibits p53 transcriptional activity and promotes its export to the cytoplasm for proteasome-mediated degradation (11) . In many human and animal tumors, the p53 gene is functionally inactivated by deletion or point mutations, participating, by this mechanism, in the process of neoplastic transformation as an antioncogene (12) .
The MCF-7 breast cancer cell line responds to stimulation by a physiological concentration of estradiol with an increase in proliferation rate (13) . This cell line contains a wild-type p53 gene (14) whose product is mostly localized in the nucleus during the G1 phase and moves to the cytoplasm after the G1-S transition (15) . In this report, we present evidence that, in MCF-7 cells, estradiol was able to induce functional inactivation of p53 protein by intracellular redistribution.
Materials and Methods
Cell Culture.
MCF-7 breast cancer cells were grown in 75-cm2 flasks in DMEM, supplemented with 5% FCS, 100 units/ml penicillin G, 100 units/ml streptomycin, 50μ g/ml gentamicin, and 2 mm l-glutamine in a 5-% CO2 atmosphere. For induction experiments, 30% confluent cells were grown in phenol red- and serum-free medium for 2 days (cells were transferred to fresh medium twice a day) and then 10 μg/ml insulin and 10 nm estradiol were added in the absence or the presence of 1 μm ICI182,780. All tissue culture media, sera, and reagents were from Life Technologies, INC., Grand Island, NY.
For oligonucleotide transfection, cells were grown in 6-cm plates. Transfection was performed with 8 μg/dish of 17-mer double-stranded oligonucleotides with a 5′-end overhang (GATC) premixed with 30 μl of DOTAP 3 liposomal transfection reagent (1 mg/ml; Boehringer Mannheim, Mannheim, Germany), according to manufacturer’s instructions. In mock transfections, only DOTAP liposomal transfection reagent was added. A rough estimate of the cell number per well was obtained by staining the cells with crystal violet and measuring the absorbance on a densitometer (16) .
Colony Formation Assay in Methylcellulose.
MCF-7 cells were cultured in 50% DMEM and 50% RPMI, supplemented with 20% FCS, and treated with dextran-coated charcoal and 0.8% methylcellulose (MethoCult H4100; Terry Fox laboratories, Vancouver BC, Canada) in 6-well plates (5 × 103 cells/well). Colonies were analyzed after 7 days. Photographs were taken at ×100 magnification.
Cytosol and Nuclear Extract Preparation.
Cells were rinsed with cold PBS containing 1 mm EDTA, and harvested with 2 mm EDTA. The cell pellet was washed twice with ice-cold PBS, once with ice-cold buffer A[ 10 mm HEPES (pH 7.9), 10 mm KCl, 0.5 mm DTT, 1.5 mm MgCl2], resuspended in three volumes of buffer A, and homogenized in a Dounce homogenizer (10 strokes with pestle B). The homogenate was centrifuged at 3300 × g for 30 min at 4°C to obtain a nuclear pellet. The supernatant was collected and centrifuged for 1 h at 100,000 × g at 4°C; the supernatant referred to as cytosol. The nuclear pellet was resuspended with buffer B [20 mm HEPES (pH 7.9), 450 mm NaCl, 0.2 mm DTT, 1.5 mm MgCl2, 0.5 mm EDTA, 0.5 mm phenylmethylsulfonyl fluoride, 25% (v/v) glycerol], homogenized (two to five strokes in a Dounce homogenizer, pestle B), incubated for 30 min with gentle shaking, and centrifuged for 1 h at 25,000 × g at 4°C. The supernatant, referred to as nuclear extract, was clarified and aliquoted. Complete protease inhibitor cocktail TM was added to all buffers (1 tablet/50 ml; Boehringer Mannheim).
Electrophoresis and Western Blot Analysis.
SDS PAGE was performed in reducing conditions in 10 or 11% acrylamide gels (acrylamide/bisacrylamide ratio, 40/1, w/w). Twenty-five μg of protein form cytosol or nuclear extract were applied to each lane. For Western blot analysis, proteins were electrophoretically transferred to a 0.45 μm nitrocellulose sheet. Membranes were then blocked by 5% nonfat milk for 1 h in TBST buffer [20 mm Tris-HCl (pH 7.5), 135 mm NaCl, and 0.05% Tween 20]. After repeated washes with TBST buffer, the membranes were incubated with the primary antibodies for 1 h at room temperature in the same buffer. Proteins were detected using mouse monoclonal antibodies Ab-6 (1 μg/ml; Calbiochem-Oncogene Research Products, Cambridge, MA) or BP53-12 (1 μg/ml; Sigma Immunochemicals, St. Louis, MO) to p53, or mouse monoclonal antibody to WAF1 (Calbiochem-Oncogene Research Products). At the end of the incubation, the membrane was washed once for 15 min and three time for 5 min with TBST buffer. Antibody reactions were revealed by incubation for 1 h, at room temperature with enzyme-coupled antimouse IgG (1:10000 dilution; Amersham, Aylersbury, Bucks, United Kingdom), in TBST buffer containing a 0.5% solution of blocking reagent, followed by a washing cycle, and development with a chemiluminescent substrate (ECL; Amersham), according to manufacturer’s instructions.
EMSA.
Whole-cell extract was obtained from cells transfected with oligonucleotides (or mock-transfected) by freeze-thawing (three cycles) in 50 mm Tris-HCl (pH 8.0), 1 mm DTT, 0.2 mm EDTA, 0.02 mg/ml BSA, and 0.2 μg/ml poly(dI-dC·dI-dC). Ten pmol of the GADD45 p53-binding site (5′-GATCCTGCAGCAGAACATGTCTAAAGCATGCTGGGCTCGAG-3′) were incubated with 10 units of polynucleotide kinase and 50 μCi ofγ[ 32P]ATP at 37°C for 30 min. The forward and reverse strands were annealed, and double-stranded sequences were purified by 12% PAGE. Each reaction mixture contained 5 μg of protein from whole-cell extract, 50,000 cpm of labeled probe (∼3000 Ci/mmol), and where indicated, 20 pmol of double-stranded oligomers containing a 5′ overhang. EMSA was performed as described (17) .
Immunocytochemistry.
Cells were grown on slides for immunofluorescence (bioMerieux SA, Marcy l’Etoile, France), fixed with 4% (w/v) formaldehyde in PBS for 10 min and washed three times with PBS. Slides were treated at 4°C with methanol and 0.1% hydrogen peroxide for 10 min, permeabilized with 0.2% Triton X-100 in PBS for 10 min (where indicated), and blocked with PBS containing 1% goat serum (Life Technologies) and 5% BSA (Sigma) for 1 h. Cells were stained overnight at 4°C with antibodies to p53 (clone BP53; Sigma) diluted 1:3000, washed twice with PBS, and incubated with FITC-conjugated goat antimouse IgG antibody (1:40 dilution; Ortho Diagnostic Systems Inc., Raritan, NJ) for 45 min. After two washes with PBS, slides were mounted in Fluo permounting medium (bioMerieux SA).
Results
Estradiol produced an increase in the cell proliferation rate of MCF-7 cells, either serum-starved or grown in the presence of charcoal-stripped serum. We tested our cell line in a soft-agar colony assay (Fig. 1 ⇓ ). A 7-day treatment of cells grown in serum-free medium with a physiological concentration of estradiol induced a 2–3-fold increase in colony number and size. Simultaneous treatment with the pure antagonist ICI182,780 prevented this effect, thus confirming that the mitogenic effect of the steroid was mediated by estrogen receptor.
Estradiol effect on in a colony formation assay. MCF-7 cells cultured in hormone-free medium were plated on methylcellulose and incubated for 7 days in the presence of vehicle alone (A); 10 nm estradiol (B), and 10 nm estradiol and 1 μm ICI182,780 (C). The bar corresponds to 50 μm.
To investigate whether p53 protein was affected by hormone action, Western blot analysis of cytosol or nuclear extract was performed on those cells. Analysis with antibodies to p53 protein showed that the protein was present in both compartments of hormone-deprived cells and that a 48-h estradiol treatment in the absence of serum produced a dramatic decrease in the p53 signal in the nuclear extract. This effect was complemented by a signal increase in the cytosol. The simultaneous presence of an excess of ICI182,780 also produced a minimal increase of p53 signal in the cytosol, probably as a consequence of the incomplete antagonist effect at the concentration used. However, it preserved or somewhat increased the nuclear signal (Fig. 2A ⇓ ). This result suggested that estradiol induced a redistribution of p53 in the MCF-7 cells, and it was confirmed by immunofluorescence staining of the cells (Fig. 3 ⇓ ). In serum-starved cells, antibodies to p53 protein revealed a prevalent nuclear distribution, whereas the fluorescence was concentrated in the cytoplasm of cells treated with estradiol for 48 h (Fig. 3 ⇓ ). Simultaneous treatment with the antiestrogen ICI182,780 preserved the prevalent nuclear staining in most cells. This effect was specific for estradiol, because insulin, another mitogen for MCF-7 cell, was unable to induce it.
Western blot analysis of MCF-7 cell extracts. A, Western blot analysis with monoclonal antibody Ab-6 to p53 protein (53) in cytosol or nuclear extract, as indicated at the bottom of the panels (25 μg protein/lane), from MCF-7 cells treated with vehicle alone (Lane a), 10 nm estradiol (Lane b), or 10 nm estradiol and 1 μm ICI182,780 (Lane c) for 48 h, as indicated in “Materials and Methods.” SDS-PAGE was performed on a 11-% polyacrylamide gel. B, Western blot analysis with monoclonal antibody Ab-6 to p53 protein (53; Lanes a and b) or with mouse IgG (Lanes c and d) of cytosol (25 μg protein/lane) from MCF-7 cells treated with vehicle alone (Lanes a and c) or 10 nm estradiol (Lanes b and d) for 48 h, as indicated in “Materials and Methods.” SDS-PAGE was performed on a 10% polyacrylamide gel. C, Western blot analysis with monoclonal antibody Ab-1 to p21-WAF-1 protein (21) of nuclear extract (25 μg protein/lane), from MCF-7 cells treated with vehicle alone (Lane a); 10 nm estradiol (Lane b), or 10 nm estradiol and 1μ m ICI182,780 (Lane c) for 48 h, as indicated in “Materials and Methods.” SDS-PAGE was performed on an 11% polyacrylamide gel.
Immunofluorescence detection of p53 protein in MCF-7 cells. Cells were grown on coverslips in hormone-free medium for 2 days, then with vehicle alone (A), with 10 nm estradiol (B), with 10 nm estradiol and 1 μm ICI182,780 (C), or with 10μ m insulin (D) for an additional 48 h. Coverslips were then processed and mounted as indicated in“ Materials and Methods.” The bar corresponds to 10μ m.
Estradiol treatment of cells also modified the electrophoretic pattern of p53 immunoreactive peptides. As evident in Fig. 2B ⇓ , Lane b, in a number of experiments, the p53 immunoreactivity in the cytosol of estradiol-treated cells was shared by more than one band of variable mass, but smaller than 53 kDa. Although this result was not constant, it was repeated several times in conditions of strict proteolysis control during cell disruption and fractionation (see “Materials and Methods”). This evidence suggested that the observed increase in p53 protein in the cytosol after estradiol treatment was associated with its proteolytic degradation. To evaluate whether the reduced concentration of nuclear p53 was correlated to its functional inactivation, Western blot analysis was performed on p21 (WAF/CIP1), a protein whose expression is known to be induced by p53. In an untreated cell, the presence of p53 in the nucleus was associated with p21 expression, as expected in the presence of a functional p53 gene product. In the nuclear extract of cells treated with estradiol for 48 h, the amount of p21 was markedly reduced compared with untreated cells or cells treated in the presence of antihormone (Fig. 2C ⇓ ). This evidence indicated that the estradiol treatment specifically reduced the expression of this p53-controlled gene. There is no evidence to suppose that this was a direct effect. It could be consequent to the postulated p53 inactivation by estradiol. This inactivation could be mediated by the estradiol-induced expression of the mdm2 gene, observable by Western blot analysis of the mdm2 protein (data not presented) or Northern blot analysis of the mRNA (18) .
To determine whether the absence of p53 protein in the nucleus was correlated with a reduced response to DNA damage, cell survival was assessed after oligonucleotide transfection simulating DNA damage. Short single-stranded DNA are able to induce p53 binding to the GADD45 site in vitro (19) , and DNA strand breaks trigger p53 activation in the cell (20) . EMSA of a probe containing the GADD45 p53-binding site showed that the whole-cell extract from MCF-7 cells transfected with double-stranded oligomers containing a 5′ overhang was able to retard more probe than an extract from mock-transfected cells (Fig. 4A ⇓ ). In vitro addition of the oligonucleotides to the extract from mock-transfected cells was also able to increase the intensity of the retarded band. This evidence indicated that these oligonucleotides (either transfected into living cells or added to the cellular extract) were able to induce p53 activation in MCF-7 cells. The oligonucleotide transfection induced growth arrest of transfected cultures (Fig. 4B ⇓ ). Two days after transfection of oligomers with a 5′ overhang, only 45% of the cells cultured in the absence of hormone survived, whereas a 2-day treatment with a physiological concentration of hormone prevented the p53-mediated growth arrest. This evidence strongly supports the hypothesis that estradiol-induced p53 inactivation prevented the DNA damage-induced growth arrest mediated by p53 protein.
Estradiol effect on MCF-7 cell survival after simulated DNA damage. A, EMSA of the probe containing the GADD45 p53-binding site incubated with MCF-7 cell extracts (5 μg of protein). Cells were mock-transfected with DOTAP (Lanes a and b) or transfected with DOTAP premixed with oligonucleotides (Lanes c and d) and incubated for an additional 48 h in the same medium containing 10% charcoal-treated FCS. The probe was incubated with cell extract in the absence (Lanes a and c) or presence (Lanes b and d) of 20 pmol of double-stranded oligomers containing a 5′ overhang. The arrow indicates migration of uncomplexed probe. B, cells were grown in 6-well microtiter plates to 60% confluency and incubated for 36 h in hormone-free medium before the addition of 10 nm estradiol or vehicle. After 48 h, cells were mock-transfected with DOTAP (hatched columns) or transfected with DOTAP premixed with oligonucleotides (gray columns) and incubated for an additional 48 h in the same medium containing 10% charcoal-treated FCS in the absence or in the presence of hormone. Cells were stained with crystal violet, and the absorbance was measured at 595 nm. Bars, SD.
Discussion
MCF-7 is a breast cancer cell line extensively studied for the expression of a functional estrogen receptor (13) . Growth of these cells is stimulated by estradiol and by other factors such as insulin. The mechanism underlying the mitogenic effect of estradiol has not been elucidated definitively. The hypothesis that the mitogenic effect was mediated solely by hormone-regulated autocrine growth factor secretion has been contradicted by the evidence that estrogens can directly promote G1 progression through direct activation of cell cycle genes or through the ras-mitogen-activated protein kinase pathway. This evidence suggests that, as the result of hormone-receptor interaction, more than one intracellular pathway becomes activated or inactivated for the proliferating response. Inactivation of p53 negative control of the G1-S transition by estradiol could greatly contribute to the hormonal control of cell proliferation. Expression of the mdm2 gene correlates with the estrogen receptor status of many human breast carcinoma cell lines, and mdm2 mRNA is accumulated after estradiol treatment of MCF-7 cell line (18) . The mdm2 gene product inhibits p53 function by direct binding, hindering the transactivation domain and targeting p53 protein for degradation. mdm2 protein contains the nuclear export signal for the complex that is eventually degraded by the proteasome. It is conceivable that the mdm2 gene product mediated the observed effect, although p53 protein has a leucine-rich nuclear export signal necessary and sufficient for nuclear export (21) . The biochemical and morphological evidences presented in this report suggest that estradiol was able to induce a redistribution of p53 from the nucleus to the cytoplasm. This was correlated to a reduction of p53 activity, as evident from the reduced expression of p21 in the estrogen-treated cells and the presence of immunoreactive fragments of p53 protein in the cytosol. The estradiol effect was produced in cells grown in serum-free medium, and it was mediated by estrogen receptor, because the estrogen antagonist ICI182,780 prevented it.
Nuclear exclusion is one of the mechanisms of p53 inactivation in breast cancer, and cytoplasm staining for p53 of neoplastic cells strongly correlated with the presence of a wild-type, intact p53 protein by sequence analysis (22) . In addition, cytoplasm sequestration of p53 is visible in normal lactating breast tissue (22) . Our observations, therefore, strengthen the hypothesis that p53 inactivation by nuclear exclusion is a necessary step for estrogen-induced cell proliferation in the presence of an active, wild-type p53 gene. This p53 inactivation has a permissive role for the observed induction of cyclin-dependent kinase activity and the consequent phosphorylation of retinoblastoma protein induced by estradiol and inhibited by coadministration of ICI182,780 in the MCF-7 cell line (23) . Apparently contradictory evidence that in the T47D breast cancer cell line estradiol is able to increase the expression of p53 protein (24) further confirms our hypothesis that nuclear exclusion is a nonmutational mechanism for abrogating the inhibitory function of wild-type p53 protein. This cell line, in fact, expresses an inactive p53 protein harboring a point mutation at codon 194 (14) and responds to estradiol treatment with increased proliferation and retinoblastoma protein phosphorylation despite the induced expression of p53 protein (25) .
These results, indicating that estradiol treatment prevented the p53-mediated growth arrest induced by the presence of intracellular DNA fragments, have allowed further speculation. Through inactivation of the tumor suppressor function of p53, estradiol could play an additional role in the multistep process of tumorigenesis of target tissues in which the p53 gene is not altered by genetic mutations. Absence of an active p53 protein by nuclear exclusion could, in fact, contribute to the genetic instability of target cells in which estradiol is able to induce a proliferative response. A hormone-independent phenotype usually emerges during the process of tumor progression, both in human breast cancer and in mouse mammary tumor models, despite the presence of a hormonal environment conferring a growth advantage to the hormone-dependent cells.
Footnotes
-
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
-
↵1 This investigation was supported by grants from the Italian Ministry for University and Scientific and Technological Research, and from Regione Campania (L.R. 41/90, rep. 8397).
-
↵2 To whom correspondence should be addressed, at Istituto di Patologia Generale ed Oncologia, Facoltà di Medicina e Chirurgia, Seconda Università degli Studi di Napoli, Larghetto Sant’Aniello a Caponapoli, 2, I-80138 Naples, Italy. Phone: 39 081 5665686; fax: 39 081 5665695; E-mail: bruno.moncharmont{at}unina2.it
-
↵3 The abbreviations used are: DOTAP, N-[1-(2,3-dioleoyloxy)propyl]-N,N,N-trimethylammonium methylsulfate; TBST, Tris-buffered saline-Tween 20; EMSA, electrophoretic mobility shift assay.
- Received January 17, 2000.
- Accepted March 31, 2000.
- ©2000 American Association for Cancer Research.
References
Volume 60, Issue 10
