17?-Estradiol-Dependent Activation of Signal Trans
http://www.100md.com
内分泌学杂志 2005年第1期
Departments of Orthopedics (A.M.K., K.L.S., M.Z., R.T.T., A.M.) and Biochemistry and Molecular Biology (R.T.T., T.C.S.), Mayo Foundation, Rochester, Minnesota 55905
Address all correspondence and requests for reprints to: Dr. A. Maran, 3-69 Medical Sciences Building, Department of Orthopedics, Mayo Clinic College of Medicine, Rochester, Minnesota 55905. E-mail: maran@mayo.edu.
Abstract
Estrogen is essential for normal growth and remodeling of bone. Although the mechanism of estrogen action on bone cells has been widely investigated, the full spectrum of signal transduction pathways activated by estrogen is unknown. In this report, we investigate the effects of the gonadal hormone 17?-estradiol on the regulation of signal transducer and activator of transcription-1 (Stat1) protein in cultured human fetal osteoblast cells, devoid of the classical estrogen receptors (ERs). 17?-Estradiol (10 nM) led to rapid (within 15 min) activation of Stat1 protein as indicated by increases in tyrosine phosphorylation and DNA binding activity. Also, 17?-estradiol increased -activated sequence-dependent transcription in transient transfection assays, suggesting an increase in Stat protein-dependent transcription. Estrogen-dependent Stat1 activation was blocked in cells that transiently express dominant-negative Stat1 mutant protein. Activation of Stat1 by 17?-estradiol was not inhibited by ER antagonist ICI 182,780, providing further evidence that it is not dependent on classical ERs. 17?-Estradiol induced rapid (within 15 min) Stat1 phosphorylation and stimulated activated sequence-dependent transcription in ER-negative breast cancer cells, indicating that these results are not unique to bone cells. The rapid estrogenic effect involving the phosphorylation and activation of Stat1 was blocked in the presence of Src family kinase inhibitor PP2; activated Stat1 was associated with Src protein in estrogen-treated cells. These findings indicate the requirement for Src kinase pathways in estrogen-mediated Stat1 activation. Thus, the ER-independent activation of Stat1 in 17?-estradiol-treated osteoblast and breast cancer cells may partially mediate the actions of estrogen on target cells.
Introduction
THE MOST POTENT natural estrogen, 17?-estradiol is produced in the ovaries and acts as a hormone on peripheral target tissues. Estrogens, which are found in yeast, fish, reptiles, birds, and all mammals (1), orchestrate intricate regulatory pathways in bone, liver, the cardiovascular system, and reproductive tissues (2, 3, 4). The well-characterized genomic actions of estrogens are mainly transduced through two nuclear estrogen receptor (ER) isoforms, ER and ER? (5, 6, 7). Additionally, an orphan receptor, ERR-1, with homology to ER, is expressed in bone cells (8, 9). Multiple transactivation sites on the ER molecule provide the potential for differential ligand-dependent genomic regulation (10). The discovery of ER-mediated pathways for gene regulation that are not dependent on receptor binding to DNA (11, 12) but require the presence of other transcription factors suggests that there are additional, nongenomic mechanisms of estrogen action (13, 14).
Stats were first reported in the study of interferon signaling (15). Subsequently new members of this family have been identified that mediate signals from a variety of cytokines and growth factors (16, 17, 18, 19). Currently seven different Stat proteins are known to be encoded in distinct genes in mammals, designated as Stat 1, 2, 3, 4, 5a, 5b, and 6 (19, 20). The Stat signaling pathways involve ligand-induced receptor dimerization, tyrosine phosphorylation, homo- or heterodimerization of phosphorylated Stat proteins, translocation of Stat dimers to the nucleus, and binding to specific DNA sequences to regulate transcription of specific genes (15).
Stat proteins participate in IL-6-mediated regulation of osteoblast differentiation (21) and fibroblast growth factor-mediated inhibition of chondrocyte proliferation and bone development (22). In this study, we demonstrate that estrogen activates Stat1 protein in human fetal osteoblasts (hFOBs) and malondialdehyde (MDA) MB-231 breast cancer cells, both of which are devoid of ER, providing evidence for a previously unrecognized nongenomic estrogen action.
Materials and Methods
Estrogen, antiestrogen, kinase inhibitors, and plasmids
17?-Estradiol (E2) was purchased from Sigma Chemical Co. (St. Louis, MO), and the ER antagonist ICI 182,780 was kindly provided by Zeneca Pharmaceuticals (Macclesfield, Cheshire, UK). The kinase inhibitors PP2 (4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo[3,4-d] pyrimidine) and PP3 (4-amino-7-phenylpyrazol[3,4-d]pyrimidine) were purchased from Calbiochem (San Diego, CA). Luciferase constructs containing -activated sequence (GAS) elements were provided by Dr. Aseem Kumar (Laurentian University, Sudbury, Ontario, Canada). Luciferase constructs containing interferon-stimulated response element (ISRE) were purchased from Stratagene (La Jolla, CA). The dominant-negative Stat1 cDNA encoding the protein that lacks major phosphorylation site (due to mutation from tyrosine to phenylalanine at position 701) was kindly provided by Dr. Jaharul Haque (Cleveland Clinic Foundation, Cleveland, OH).
Construction of GAS mutant promoter-driven luciferase reporter plasmid
The 7 x GAS-mutant promoter-driven luciferase (GASM-luciferase) vector was generated by PCR and ligation as described (23). PGL3 Basic vector (Promega, Madison, WI) was linearized by digesting at SmaI site, and then the entire plasmid was amplified using forward primer that has seven repeats of mutant GAS elements at 5'-end (ATA GCG TAA ATA GCG TAA ATA GCG TAA ATA GCG TAA ATA GCG TAA ATA GCG TAA GGG CTC GAC ATC TGC GAT CTA AGT) and the reverse primer sequence (GGG CTA GCA CGC GTA AGA GCT CGG). Underlined sequences represent mutant GAS elements. Nonunderlined sequences represent vector. The amplified fragments were self-ligated and transformed, and large-scale plasmid purification was carried out as described (23). The cloning of GAS mutant sequences in the vector was confirmed by DNA sequencing at the Molecular Core Facility, Mayo Clinic.
Cell culture and estrogen treatment
The hFOB cell line containing the temperature sensitive T antigen expression vector with neomycin resistance gene (24) was maintained at 34 C in phenol red-free DMEM/F12 containing 10% charcoal-stripped fetal bovine serum supplemented with geneticin (300 μg/ml) as described (24, 25). MDA-MB231 (ER-negative) breast cancer cells were maintained at 37 C in phenol red-free DMEM/F12 medium containing 10% charcoal-stripped fetal bovine serum as described (26). The cells were plated 24 h before estrogen treatment. Cells were grown in T-75 flasks at 1 x 106 cells per flask for all experiments except for transient transfection assays in which cells were grown in 6-well plates at 1.5 x 105 cells/well. The cells were treated with 10 nM (protein studies) or 20–100 nM (transient transfections) E2 for various periods of time (5 min to 48 h). Cells were washed with PBS and harvested at the end of estrogen treatment.
Cell lysis and preparation of cytoplasmic and nuclear extracts
Estrogen-treated cell pellets were lysed by suspending in Nonidet P-40 lysis buffer containing 0.5% Nonidet P-40; 90 mM KCl; 1 mM magnesium acetate; 2 mM 2-mercaptoethanol; 10 μg/ml leupeptin; 1x phosphatase inhibitor cocktail 2 (a mixture of sodium orthovanadate, sodium molybdate, and sodium tartrate) (Sigma); and 10 mM HEPES (pH 7.6). After centrifugation at 10,000 x g for 10 min, the resultant supernatant containing cytoplasmic extract was used for immunoprecipitation, and the pellets containing nuclei were saved for preparing nuclear extract (27).
Immunoprecipitation analyses
Cytoplasmic cell extracts (60 μg protein) were used for immunoprecipitation with 1 μg of the anti-phosphoStat or anti-Stat1 antibody (Santa Cruz Biotechnology, San Diego, CA). Bound proteins were purified using protein A Sepharose. Immunoprecipitated proteins were separated by SDS-PAGE and analyzed by Western blotting. For coimmunoprecipitation studies, cell extracts containing equal amounts of protein were immunoprecipitated with anti-Src (Cell Signaling Technology, Beverly, MA), anti-Stat1 antibody, or nonimmune IgG (Dako Cytomation, Carpinteria, CA). The immunoprecipitates were incubated with protein A Sepharose for 1 h at 4 C and then washed and separated on SDS-PAGE gel and analyzed by Western blotting using anti-Src and anti-Stat1 antibodies.
EMSAs
Nuclear extracts (15 μg protein) prepared from hFOB cells were used in EMSAs as described in the manufacturer’s protocol (GENEKA, Montréal, Canada). Where indicated, the reactions were incubated with equal concentrations of cold and hot mutant and wild-type double-stranded DNA sis-inducible element (SIE) oligonucleotides provided in the kit as described (GENEKA). The following SIE oligonucleotides were used in this experiment (sequence of the sense strand is shown): wild-type SIE oligonucleotide, 5'-GTC GAC ATT TCC CGT AAA TCG TCG A-3'; and mutant SIE oligonucleotide, GTC GAC ATA TAG CGT AAA TCG TCG A-3'. The binding reactions were subjected to electrophoresis on a 5% polyacrylamide gel in 0.5x Tris-borate EDTA running buffer. Gels were analyzed on a PhosphorImager analyzer (Molecular Dynamics Inc., Sunnyvale, CA).
Luciferase assay
Luciferase constructs containing GAS elements (28) and luciferase constructs containing ISRE elements (Stratagene) were used in this study. Cells plated in 6-well plates were transfected at 60% confluence with pGAS-luciferase using the transfection agent lipofectamine as described in the manufacturer’s protocol (Invitrogen, Carlsbad, CA). The plasmids pISRE-luciferase, luciferase driven by pGASM-luciferase and dominant-mutant Stat1 cDNA (DN-Stat1) were used in various transfection experiments as indicated. All transfections performed had an internal transfection control plasmid containing Renilla luciferase (pRL-CMV-luciferase) (Promega), which also allowed normalization of luciferase units. Twenty-four hours post transfection, the cells were treated with 20 nM E2, 100 nM ER antagonist ICI 182,780, or 25 μM Src-kinase family inhibitor pyrazolopyrimidine compound (PP2). The cells were harvested after 48 h of treatment using 300 μl of passive lysis buffer provided in a luciferase assay kit (Promega). Luciferase assays were performed using the dual luciferase reporter assay system according to the manufacturer’s protocol (Promega) and read on a TD-20E luminometer (Turner, Sunnyvale, CA).
Statistical analysis
All values are expressed as mean ± SE. In each experiment, three to eight replicates of each treatment were measured. The data are representative of three independent experiments. Fisher’s protected least significant difference with a one-way ANOVA was used to compare the groups. P < 0.05 was considered statistically significant. Due to the small sample size in some of the groups, we supplemented the ANOVA with nonparametric Kruskal-Wallis tests.
Results
Effect of estrogen on Stat1 protein phosphorylation
E2 treatment resulted in an increased tyrosine phosphorylation of Stat1 within 15 min of treatment as shown by Western blot hybridization using an anti-phosphoStat1 antibody [p-Stat1 (Tyr 701)] directed against the tyrosine phosphorylated residue of Stat1 (Fig. 1A). The increase in phosphorylation appears to be transient and decreases with time. In contrast, there was no significant change in the total amount of Stat1 protein with E2 treatment as detected by anti-Stat1 p91 antibody (Fig. 1B).
FIG. 1. Time-course effects of E2 on Stat1 phosphorylation shows rapid transient increase in phosphorylation of Stat1 with no change in Stat1 protein levels. Cytoplasmic extracts prepared from hFOB cells 15, 30, 60, 120, and 240 min after E2 (10 nM) treatment were subjected to immunoprecipitation with anti-Stat1 p91 antibody (Santa Cruz Biotechnology) followed by Western blotting using p-Stat1 (Tyr 701) (Santa Cruz Biotechnology) directed against its phosphotyrosine residue (A). The filter was stripped and reprobed with the anti-Stat1 antibody (B). veh, Vehicle control.
Estrogen treatment leads to activation and formation of Stat1-DNA complex
Nuclear extracts prepared from hFOB cells treated with E2 were analyzed by EMSA (Fig. 2). Stat1 binding to radiolabeled wild-type oligonucleotide probe increased in hFOB cells in the presence of E2 (lane 7), compared with the vehicle control (lane 3). The specificity of Stat1 binding was confirmed by decreased binding in the presence of nonradiolabeled (cold) wild-type oligonucleotide (lane 5) and absence of binding in the presence of radiolabeled mutant oligonucleotide (lane 4). Stat1 binding to radiolabeled wild-type probe was not decreased in the presence of nonradioactive (cold) mutant oligonucleotides (lane 6), another indication of specificity. The specificity of Stat1-protein and DNA interaction has been further confirmed by the super shift in the presence of anti-Stat1 p91 (lane 8) and p-Stat1 (Tyr 701) antibody (lane 9).
FIG. 2. EMSA showing that estrogen treatment increases Stat1 binding to DNA. Nuclear extracts prepared from hFOB cells 16 h after E2 (10 nM) treatment were analyzed by EMSA by incubating with specific SIE oligonucleotide probes (GENEKA). Lower arrow indicates the migration of the Stat1-DNA complex. Upper arrow indicates the supershift and the migration of the Stat1-Ab-DNA complex. NA, No addition control; veh, vehicle control.
Estrogen treatment increases Stat1-dependent transcriptional activity in osteoblast cells
To determine whether estrogen treatment induces Stat1-stimulated transcription and whether Stat1-dependent promoter activities are involved in estrogen-mediated activation, the effects of E2 was tested on both the ISRE- and GAS-driven luciferase activities by transient transfection assays in hFOB cells. Relative to the activity in untreated cells, estrogen stimulated the GAS-luciferase reporter gene activity approximately 8-fold (vehicle 0.8 ± 0.3; estrogen 7 ± 0.2) (Fig. 3A). As expected, interferon- treatment led to even greater stimulation of the reporter activity (>100-fold) (vehicle 0.8 ± 0.3; interferon 212 ± 2.5) (Fig. 3A). A time-dependent increase in GAS-luciferase reporter gene activity occurred after E2 treatment (vehicle 0.9 ± 0.2; estrogen treatment 8 h, 2. ± 0.1; 12 h, 2.5 ± 0.8; 24 h, 3.7 ± 0.3; 48 h, 7.6 ± 0.1) (Fig. 3B). GAS promoter with mutations in the sequences was not activated in response to estrogen treatment, and no increase in luciferase activity was observed in cells transiently transfected with luciferase constructs driven by GAS mutant promoter (vehicle 6.6 ± 0.3; estrogen 8 ± 0.6) (Fig. 3C). Estrogen treatment led to a small change on ISRE-driven luciferase activity (vehicle 1.8 ± 0.1, estrogen 2.2 ± 0.1) (Fig. 3D), whereas interferon-? treatment stimulated ISRE-luciferase promoter functions (vehicle 6.6 ± 0.3, estrogen 217 ± 10.5) (Fig. 3D).
FIG. 3. Effect of E2 on Stat1-dependent transcription in hFOB cells. A, Effect on GAS-luciferase activity. Cells transfected with expression plasmid GAS-luciferase were treated for 48 h with vehicle control (veh), E2 (20 nM), or interferon- (IFN) (1000 U), and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 8 replicate cultures). #, P < 0.007 vs. veh; *, P < 0.0001 vs. veh; ##, P < 0.0001 vs. E2. B, Time course. Cells transfected with expression plasmid GAS-luciferase were treated with vehicle control (veh) or E2 (20 nM) for 8, 12, 24, and 48 h, and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). #, P < 0.009 vs. veh; ##, P < 0.002 vs. veh; *, P < 0.0001 vs. veh; C, Effect on GASM-luciferase activity: Cells transfected with expression plasmid GASM-luciferase were treated with vehicle control (veh) or E2 (20 nM) for 48 h, and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). D, Effect on ISRE-luciferase activity: Cells transfected with expression plasmid ISRE-luciferase were treated with veh, E2 (20 nM), or interferon-? (IFN-?) (1000 U) for 48 h, and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). *, P < 0.0001 vs. veh; #, P < 0.0001 vs. E2.
Activation of Stat1-dependent GAS promoter activity by estrogen is supported by experiments in cells cotransfected with dominant-negative Stat1 mutant cDNA, in which the E2-mediated activation of GAS luciferase activity was decreased (Fig. 4). Relative to the activity in untreated cells, estrogen stimulated GAS-luciferase reporter gene activity in the absence or presence of DN-Stat1 by 6.3- and 3.2-fold, respectively (vehicle 1.5 ± 0.1; estrogen 8.6 ± 0.6; estrogen+DN-Stat1 4.4 ± 0.4; DN-Stat1 2. 8 ± 0.2) (Fig. 4). However, ER-dependent activation of estrogen response element (ERE) and ERE-controlled expression of luciferase activity were not blocked (Fig. 4). Relative to the activity in untreated cells, estrogen stimulated ERE-luciferase reporter gene activity in the absence or presence of DN-Stat1 by 4.6- and 3.9-fold, respectively (vehicle 2.8 ± 0.1; estrogen 12.4 ± 0.6; estrogen+DN-Stat1 11 ± 0.5, DN-Stat1 2.5 ± 0.2) (Fig. 4).
FIG. 4. DN-Stat1 expression blocks GAS-luciferase transcription but not ERE-luciferase transcription in hFOB cells. Cells were cotransfected with the reporter plasmid GAS-luciferase or ERE-luciferase and expression vector containing DN-Stat1 and treated with vehicle control (veh) and E2 (20 nM) for 48 h. The luciferase reporter activity was analyzed. Values are the mean ± SE (n = 5 replicate cultures). *, P < 0.0001 vs. veh; #, P < 0.0001 vs. E2.
Estrogen-mediated activation of Stat1 is ER independent
ER antagonist ICI 182,780 blocked estrogen-stimulated ERE-luciferase activity in hFOB cells (data not shown). In contrast, ICI 182,780 did not block E2-stimulated GAS-driven luciferase activity (vehicle 1.2 ± 0.1; estrogen 9 ± 0.1; estrogen+ICI 182,780 8.6 ± 0.7) (Fig. 5).
FIG. 5. Estrogen-mediated activation of Stat1-dependent reporter gene expression is not ER dependent. hFOB cells transfected with the expression plasmid GAS-luciferase were treated for 48 h with vehicle control (veh), E2 (20 nM), or E2 plus ER antagonist ICI 182,780 (100 nM) (E2+ICI), and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). *, P < 0.0001 vs. veh.
Estrogen treatment induces Stat1 activation in ER-negative breast cancer cells
To determine target cell specificity, Stat1 phosphorylation was investigated in E2-treated breast cancer cells. Estrogen treatment resulted in an increased tyrosine phosphorylation of Stat1 within 15 min in ER-negative (MDA-MB231) breast cancer cells as shown by Western blot hybridization with p-Stat1(Tyr 701) antibody (Fig. 6A). As was the case in bone cells, there was no significant change in the total amount of Stat1 protein with estrogen treatment in the breast cancer cells (Fig. 6B).
FIG. 6. Estrogen treatment increases Stat1 phosphorylation in ER-negative breast cancer cells. Cytoplasmic extracts were prepared from MDA-MB 231 breast cancer cells 15 min after treatment with E2 (10 nM) and subjected to immunoprecipitation with anti-Stat1 p91 antibody (Santa Cruz Biotechnology) followed by Western blotting using p-Stat1 (Tyr 701) (Santa Cruz Biotechnology) directed against its phosphotyrosine residue (A). The filter was stripped and reprobed with the anti-Stat1 antibody (B).
The effects of E2 on GAS-driven luciferase activities were examined by transient transfection assays in MDA-MB231 breast cancer cells to determine whether E2 treatment induces Stat1-stimulated transcription and whether Stat1-dependent promoter activity is involved. Relative to the activity in untreated cells, E2 at 10, 20, and 100 nM stimulated the GAS-luciferase reporter gene activity by 1.2-, 4.9-, and 9-fold, respectively (vehicle 2.8 ± 0.2, estrogen 10 nM, 4.9 ± 0.5; 20 nM, 15.7 ± 0.4; 100 nM, 27.8 ± 0.4) (Fig. 7).
FIG. 7. Effect of E2 (20 nM) on Stat1-dependent transcription in MDA-MB 231 cells. Cells transfected with expression plasmid GAS-luciferase, were treated for 48 h with vehicle control (veh) or E2 (20 nM), and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). Data are from three independent experiments. #, P 0.005 vs. veh; *, P < 0.0001 vs. veh.
Estrogen-mediated activation of Stat1 requires Src kinase activity
Estrogen-stimulated GAS-luciferase expression was blocked in the presence of Src kinase inhibitor, PP2 (vehicle 2.1 ± 0.3; estrogen 14 ± 0; estrogen+PP2 4 ± 0.4) (Fig. 8), whereas the epidermal growth factor receptor kinase inhibitor PP3, which serves as a negative control for PP2 and does not inhibit Src kinase activity, had no effect (estrogen+PP3 13.4 ± 0.6) (Fig. 8). Neither PP2 nor PP3 showed any significant independent effect on GAS luciferase activity (PP2 4.8 ± 1; PP3 4.5 ± 1). In cells that transiently express dominant-negative Src protein, estrogen-dependent GAS-luciferase activity was inhibited (data not shown).
FIG. 8. Estrogen-mediated activation of Stat-dependent reporter gene activation is Src kinase-dependent. hFOB cells transfected with the expression plasmid GAS-luciferase were treated for 48 h with vehicle control (veh); E2 (20 nM); Src-kinase family inhibitor, PP2 (25 μM); E2+PP2; epithelial growth factor receptor kinase inhibitor, PP3 (25 μM); or E2+PP3. The luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). Data are from three independent experiments. *, P < 0.0001 vs. veh; #, P < 0.0001 vs. E2.
Src is associated with Stat1 in estrogen-treated osteoblasts
Coimmunoprecipitation studies in estrogen-treated hFOB cells analyzed by Western blot hybridization demonstrated c-Src protein coimmunoprecipitated with Stat1 (Fig. 9A). Both Stat1 and activated Stat1 proteins coimmunoprecipitated with c-Src as shown by anti-Stat1 and pStat1 antibodies (Fig. 9B). No signals were observed with nonimmune IgG controls (Fig. 9C).
FIG. 9. Stat1 is associated with Src in estrogen-treated bone cells. Cytoplasmic extracts prepared from hFOB cells treated for 15 min with vehicle (V) and E2 (10 nM) were subjected to immunoprecipitation (IP) with anti-Stat1 p91 antibody (Santa Cruz Biotechnology) (A), anti-c-Src antibody (Cell Signaling Technology) (B), or nonimmune IgG (Dako Cytomation) (C).
The luciferase activity results for all statistical analysis based on nonparametric Kruskal-Wallis tests were similar to those based on ANOVA (data not shown).
Discussion
Estrogen treatment leads to tyrosine phosphorylation and rapid (within 15 min) activation of Stat1 protein in ER-negative human osteoblast cells. This immediate response is followed by an increase in Stat1-specific DNA binding and transcription, which does not require classical ERs. In addition, the activation of Stat1 protein and the stimulation of Stat1-dependent transcription by estrogen in ER-negative breast cancer cells suggest that this could be a common signaling mechanism in estrogen target cells.
Many of the post-2 h actions of estrogen involve the genomic mechanism, which involves the binding of estrogen to ER, translocation of ligand bound receptors to the nucleus, followed by homo- or heterodimerization of ER and binding of ER dimers to estrogen-responsive elements on the DNA. The rapid (15 min) effects of estrogen described in this study occur too quickly for genomic-mediated mechanisms (13, 14, 29). Several other reports on the nongenomic actions of estrogens have been shown to occur much more rapidly, and some report the involvement of MAPK, Src, or phosphatidylinositol 3-kinase pathways (14, 29, 30, 31).
The estrogen-mediated activation of Stat1 described here is not inhibited by cycloheximide (Kennedy, A., A. Maran, and R. T. Turner, unpublished data). Unlike previously described nongenomic effects in bone cells, Stat1 activation by estrogen does not require classical ER, as shown by the failure of the ER antagonist ICI 182,780 to block the Stat1-dependent transcription. This finding may be analogous to the nongenomic activation of ERK MAPKs by estrogen in the explants of cerebral cortex (32). Estrogen increases type 1 collagen levels and alkaline phosphatase activity in hFOB cells (Maran, A., and R. T. Turner, unpublished data), indicating that activation of non-ER-mediated pathways by the hormone results in metabolic changes in bone cells.
Stat1 protein-dependent transcriptions involve the binding of Stat1 to two specific promoter elements, namely ISRE and GAS in cytokines and growth factor-treated cells (15, 33, 34). The Stat family members show specificity in DNA binding: Stat1 homodimer binds to GAS, whereas the Stat1/Stat2 heterodimer interacts with ISRE. E2 activation of Stat1 protein results in increased transcription of GAS element-containing genes. Estrogen’s preference for the GAS over the ISRE suggests that estrogen elicits responses similar to interferon- or ligands like IL-6 and platelet-derived growth factor (33, 35, 36). Stat1 binding GAS elements are present in the regulatory regions of genes expressed by osteoblasts, such as type 1 collagen, osteocalcin, and osteonectin. Further work is necessary to determine the effects of estrogen-mediated activation of Stat1 activity on osteoblast metabolism.
The involvement of Stat1 protein in estrogen-mediated actions of osteoblasts differs from a recent report showing estrogen-mediated, nongenomic activation in endothelial cells involving other Stat family members (Stat 3 and 5) in an ER-dependent manner (31). One report (14) has shown that an alternative form of ER could mediate some of the nongenomic actions of estrogens. Preliminary findings suggest that a subset of the classical ER is associated with the cell membrane (14, 32) and such putative membrane ERs have been found to be insensitive to antiestrogen ICI 182,780 treatment (32). The present findings do not address whether such membrane ERs are present in hFOB cells and could play a role in estrogen-mediated activation of Stat1 signaling.
Several known mechanisms lead to tyrosine phosphorylation and activation of Stat proteins in cells, including: 1) cytokine receptors that activate Janus kinases, which in turn phosphorylate Stat proteins (e.g. interferon and IL-6 receptors); 2) phosphorylation by growth factor receptors that have intrinsic tyrosine kinase activities (e.g. platelet-derived growth factor, epithelial growth factor receptors); or 3) direct phosphorylation by nonreceptor tyrosine kinases (e.g. Src, Fes, Abl). Stat1-dependent GAS promoter activity was inhibited in the presence of a chemical inhibitor for c-Src kinase, and both Stat1 and activated Stat1 are associated with c-Src kinase protein. The association of other Stat family members with Src protein has been demonstrated in cancer cells and cytokine-dependent signaling (37, 38, 39, 40). This suggests that estrogen-mediated activation of Stat1 protein in hFOB cells is dependent on the Src kinase pathway. Further work is necessary to determine whether Src kinase plays a role in Stat1 phosphorylation in estrogen-treated bone cells. These results are consistent with the rapid activation of Src kinase shown previously in osteoblasts and osteocytes after estrogen exposure (14). Thus, Stat1 association with activated Src kinase in estrogen-treated cells supports a potentially vital role for Src kinase pathway in Stat1 activation by estrogen in target cells.
In conclusion, ER-independent activation of Stat1 in E2-treated bone and breast cancer cells presents a novel nongenomic, non-ER-mediated mechanism for estrogen action and reinforces the concept that estrogen involves multiple signaling pathways in target cells.
Acknowledgments
The authors thank Ralf Janknecht (Mayo Clinic) and Dhan Kuppuswamy (Medical University of South Carolina) for helpful suggestions and comments on the manuscript. The authors thank the Center for Patient Oriented Research at Mayo Clinic for the statistical analysis, Gobinda Sarkar for helping with the PCRs, and Peggy Backup for editorial assistance.
References
Griffin C, Flouriot G, Sonntag-Buck V, Nestor P, Gannon F 1998 Identification of novel chicken estrogen receptor- messenger ribonucleic acid isoforms generated by alternative splicing and promoter usage. Endocrinology 139:4614–4625
Auchus RJ, Faqua SAWW 1994 Hormones-nuclear receptor in health disease. The oestrogen receptor. Baillieres Clin Endocrinol Metab 8:433–449
Sarrel PM, Lufkin EG, Oursler MJ, Keefe D 1994 Estrogen actions in arteries, bone and brain. Sci Am Sci Med 1:44–53
Turner RT, Riggs BL, Spelsberg TC 1994 Skeletal effects of estrogen. Endocr Rev 15:275–300
Kuiper GG, Gustafsson JA 1997 The novel estrogen receptor-? subtype: potential role in the cell- and promoter-specific actions of estrogens and anti-estrogens. FEBS Lett 410:87–90
Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA 1996 Cloning of a novel receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925–5930
Koike S, Sakai M, Muramatsu M 1987 Molecular cloning and characterization of rat estrogen receptor cDNA. Nucleic Acids Res 15:2499–2513
Bonnelye E, Merdad L, Kung V, Aubin JE 2001 The orphan nuclear estrogen receptor-related receptor (ERR) is expressed throughout osteoblast differentiation and regulates bone formation in vitro. J Cell Biol 153:971–984
Bonnelye E, Vanacker JM, Dittmar T, Begue A, Desbiens X, Denhardt DT, Aubin JE, Laudet V, Fournier B 1997 The ERR-1 orphan receptor is a transcriptional activator expressed during bone development. Mol Endocrinol 11:905–916
Tzukerman MT, Esty A, Santiso-Mere D, Danielian P, Parker MG, Stein RB, Pike JW, McDonnell DP 1994 Human estrogen receptor transactivational capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrinol 8:21–30
Yang NN, Venugopalan M, Hardikar S, Glasebrook A 1996 Identification of an estrogen response element activated by metabolites of 17?-estradiol and raloxifene. [Erratum (1997) 275:1249]. Science 273:1222–1225
El-Tanani MK, Green CD 1997 Two separate mechanisms for ligand-independent activation of the estrogen receptor. Mol Endocrinol 11:928–937
Monroe DG, Spelsberg TC 2003 A case for estrogen receptors on cell membranes and nongenomic actions of estrogen. Calcif Tissue Int 72:183–184
Kousteni S, Bellido T, Plotkin LI, O’Brien CA, Bodenner DL, Han L, Han K, DiGregorio GB, Katzenellenbogen JA, Katzenellenbogen BS, Roberson PK, Weinstein RS, Jilka RL, Manolagas SC 2001 Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell 104:719–730
Darnell Jr JE, Kerr IM, Stark GR 1994 Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264:1415–1421
Leaman DW, Leung S, Li X, Stark GR 1996 Regulation of STAT-dependent pathways by growth factors and cytokines. FASEB J 10:1578–1588
Kaplan MH, Schindler U, Smiley ST, Grusby MJ 1996 Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity 4:313–319
Ihle JN, Kerr IM 1995 Jaks and Stats in signaling by the cytokine receptor superfamily. Trends Genet 11:69–74
Stark GR, Kerr IM, Williams BR, Silverman RH, Schreiber RD 1998 How cells respond to interferons. Annu Rev Biochem 67:227–264
Bromberg J, Darnell Jr JE 2000 The role of STATs in transcriptional control and their impact on cellular function. Oncogene 19:2468–2473
Bellido T, Borba VZ, Roberson P, Manolagas SC 1997 Activation of the Janus kinase/STAT (signal transducer and activator of transcription) signal transduction pathway by interleukin-6-type cytokines promotes osteoblast differentiation. Endocrinology 138:3666–3676
Sahni M, Ambrosetti DC, Mansukhani A, Gertner R, Levy D, Basilico C 1999 FGF signaling inhibits chondrocyte proliferation and regulates bone development through the STAT-1 pathway. Genes Dev 13:1361–1366
Nishiyama T, Hatano H, Kurosaka M, Bolander ME, Sarkar G 2003 Cis-acting intronic elements that regulate cartilage-specific alternative splicing of the type II collagen (Col2) pre-mRNA lie at or near splice site junction sequences flanking exon 2 of the gene. J Bone Miner Res 18:1716–1722
Harris SA, Enger RJ, Riggs BL, Spelsberg TC 1995 Development and characterization of a conditionally immortalized human fetal osteoblastic cell line. J Bone Miner Res 10:178–186
Maran A, Zhang M, Spelsberg TC, Turner RT 2001 The dose-response effects of ethanol on the human fetal osteoblastic cell line. J Bone Miner Res 16:270–276
Reinholz GG, Getz B, Sanders ES, Karpeisky MY, Padyukova NSh, Mikhailov SN, Ingle JN, Spelsberg TC 2002 Distinct mechanisms of bisphosphonate action between osteoblasts and breast cancer cells: identity of a potent new bisphosphonate analogue. Breast Cancer Res Treat 71:257–268
Maran A, Maitra RK, Kumar A, Dong B, Xiao W, Li G, Williams BR, Torrence PF, Silverman RH 1994 Blockage of NF-B signaling by selective ablation of an mRNA target by 2–5A antisense chimeras. Science 265:789–792
Ramana CV, Grammatikakis N, Chernov M, Nguyen H, Goh KC, Williams BR, Stark GR 2000 Regulation of c-myc expression by IFN- through Stat1-dependent and -independent pathways. EMBO J 19:263–272
Pedram A, Razandi M, Aitkenhead M, Hughes CC, Levin ER 2002 Integration of the non-genomic and genomic actions of estrogen. Membrane-initiated signaling by steroid to transcription and cell biology. J Biol Chem 277:50768–50775
Wong C, McNally CEN, Komm BS, Cheskis BJ 2002 Estrogen receptor-interacting protein that modulates its nongenomic activity-cross-talk with Src/Erk phosphorylation cascade. Proc Natl Acad Sci USA 99:14783–14788
Bjornstrom L, Sjoberg M 2002 Signal transducers and activators of transcription as downstream targets of nongenomic estrogen receptor actions. Mol Endocrinol 16:2202–2214
Decker T, Lew DJ, Mirkovitch J, Darnell Jr JE 1991 Cytoplasmic activation of GAF, an IFN--regulated DNA-binding factor. EMBO J 10:927–932
Decker T, Kovarik P, Meinke A 1997 GAS elements: a few nucleotides with a major impact on cytokine-induced gene expression. J Interferon Cytokine Res 17:121–134
Shuai K, Horvath CM, Huang LH, Qureshi SA, Cowburn D, Darnell Jr JE 1994 Interferon activation of the transcription factor Stat91 involves dimerization through SH2-phosphotyrosyl peptide interactions. Cell 76:821–828
Sehgal PB 2000 STAT-signalling through the cytoplasmic compartment: consideration of a new paradigm. Cell Signal 12:525–535
Singh M, Setalo Jr G, Guan X, Warren M, Toran-Allerand CD 1999 Estrogen-induced activation of mitogen-activated protein kinase in cerebral cortical explants: convergence of estrogen and neurotrophin signaling pathways. J Neurosci 19:1179–1188
Xi S, Zhang Q, Dyer KF, Lerner EC, Smithgall TE, Gooding WE, Kamens J, Grandis JR 2003 Src kinases mediate STAT growth pathways in squamous cell carcinoma of the head and neck. J Biol Chem 278:31574–31583
Cao X, Tay A, Guy GR, Tan YH 1996 Activation and association of Stat3 with Src in v-Src-transformed cell lines. Mol Cell Biol 16:1595–1603
Chaturvedi P, Reddy MV, Reddy EP 1998 Src kinases and not JAKs activate STATs during IL-3 induced myeloid cell proliferation. Oncogene 16:1749–1758
Reddy EP, Korapati A, Chaturvedi P, Rane S 2000 IL-3 signaling and the role of Src kinases, JAKs and STATs: a covert liaison unveiled. Oncogene 19:2532–2547(Angela M. Kennedy, Kriste)
Address all correspondence and requests for reprints to: Dr. A. Maran, 3-69 Medical Sciences Building, Department of Orthopedics, Mayo Clinic College of Medicine, Rochester, Minnesota 55905. E-mail: maran@mayo.edu.
Abstract
Estrogen is essential for normal growth and remodeling of bone. Although the mechanism of estrogen action on bone cells has been widely investigated, the full spectrum of signal transduction pathways activated by estrogen is unknown. In this report, we investigate the effects of the gonadal hormone 17?-estradiol on the regulation of signal transducer and activator of transcription-1 (Stat1) protein in cultured human fetal osteoblast cells, devoid of the classical estrogen receptors (ERs). 17?-Estradiol (10 nM) led to rapid (within 15 min) activation of Stat1 protein as indicated by increases in tyrosine phosphorylation and DNA binding activity. Also, 17?-estradiol increased -activated sequence-dependent transcription in transient transfection assays, suggesting an increase in Stat protein-dependent transcription. Estrogen-dependent Stat1 activation was blocked in cells that transiently express dominant-negative Stat1 mutant protein. Activation of Stat1 by 17?-estradiol was not inhibited by ER antagonist ICI 182,780, providing further evidence that it is not dependent on classical ERs. 17?-Estradiol induced rapid (within 15 min) Stat1 phosphorylation and stimulated activated sequence-dependent transcription in ER-negative breast cancer cells, indicating that these results are not unique to bone cells. The rapid estrogenic effect involving the phosphorylation and activation of Stat1 was blocked in the presence of Src family kinase inhibitor PP2; activated Stat1 was associated with Src protein in estrogen-treated cells. These findings indicate the requirement for Src kinase pathways in estrogen-mediated Stat1 activation. Thus, the ER-independent activation of Stat1 in 17?-estradiol-treated osteoblast and breast cancer cells may partially mediate the actions of estrogen on target cells.
Introduction
THE MOST POTENT natural estrogen, 17?-estradiol is produced in the ovaries and acts as a hormone on peripheral target tissues. Estrogens, which are found in yeast, fish, reptiles, birds, and all mammals (1), orchestrate intricate regulatory pathways in bone, liver, the cardiovascular system, and reproductive tissues (2, 3, 4). The well-characterized genomic actions of estrogens are mainly transduced through two nuclear estrogen receptor (ER) isoforms, ER and ER? (5, 6, 7). Additionally, an orphan receptor, ERR-1, with homology to ER, is expressed in bone cells (8, 9). Multiple transactivation sites on the ER molecule provide the potential for differential ligand-dependent genomic regulation (10). The discovery of ER-mediated pathways for gene regulation that are not dependent on receptor binding to DNA (11, 12) but require the presence of other transcription factors suggests that there are additional, nongenomic mechanisms of estrogen action (13, 14).
Stats were first reported in the study of interferon signaling (15). Subsequently new members of this family have been identified that mediate signals from a variety of cytokines and growth factors (16, 17, 18, 19). Currently seven different Stat proteins are known to be encoded in distinct genes in mammals, designated as Stat 1, 2, 3, 4, 5a, 5b, and 6 (19, 20). The Stat signaling pathways involve ligand-induced receptor dimerization, tyrosine phosphorylation, homo- or heterodimerization of phosphorylated Stat proteins, translocation of Stat dimers to the nucleus, and binding to specific DNA sequences to regulate transcription of specific genes (15).
Stat proteins participate in IL-6-mediated regulation of osteoblast differentiation (21) and fibroblast growth factor-mediated inhibition of chondrocyte proliferation and bone development (22). In this study, we demonstrate that estrogen activates Stat1 protein in human fetal osteoblasts (hFOBs) and malondialdehyde (MDA) MB-231 breast cancer cells, both of which are devoid of ER, providing evidence for a previously unrecognized nongenomic estrogen action.
Materials and Methods
Estrogen, antiestrogen, kinase inhibitors, and plasmids
17?-Estradiol (E2) was purchased from Sigma Chemical Co. (St. Louis, MO), and the ER antagonist ICI 182,780 was kindly provided by Zeneca Pharmaceuticals (Macclesfield, Cheshire, UK). The kinase inhibitors PP2 (4-amino-5-(4-chlorophenyl)-7-(t-butyl)pyrazolo[3,4-d] pyrimidine) and PP3 (4-amino-7-phenylpyrazol[3,4-d]pyrimidine) were purchased from Calbiochem (San Diego, CA). Luciferase constructs containing -activated sequence (GAS) elements were provided by Dr. Aseem Kumar (Laurentian University, Sudbury, Ontario, Canada). Luciferase constructs containing interferon-stimulated response element (ISRE) were purchased from Stratagene (La Jolla, CA). The dominant-negative Stat1 cDNA encoding the protein that lacks major phosphorylation site (due to mutation from tyrosine to phenylalanine at position 701) was kindly provided by Dr. Jaharul Haque (Cleveland Clinic Foundation, Cleveland, OH).
Construction of GAS mutant promoter-driven luciferase reporter plasmid
The 7 x GAS-mutant promoter-driven luciferase (GASM-luciferase) vector was generated by PCR and ligation as described (23). PGL3 Basic vector (Promega, Madison, WI) was linearized by digesting at SmaI site, and then the entire plasmid was amplified using forward primer that has seven repeats of mutant GAS elements at 5'-end (ATA GCG TAA ATA GCG TAA ATA GCG TAA ATA GCG TAA ATA GCG TAA ATA GCG TAA GGG CTC GAC ATC TGC GAT CTA AGT) and the reverse primer sequence (GGG CTA GCA CGC GTA AGA GCT CGG). Underlined sequences represent mutant GAS elements. Nonunderlined sequences represent vector. The amplified fragments were self-ligated and transformed, and large-scale plasmid purification was carried out as described (23). The cloning of GAS mutant sequences in the vector was confirmed by DNA sequencing at the Molecular Core Facility, Mayo Clinic.
Cell culture and estrogen treatment
The hFOB cell line containing the temperature sensitive T antigen expression vector with neomycin resistance gene (24) was maintained at 34 C in phenol red-free DMEM/F12 containing 10% charcoal-stripped fetal bovine serum supplemented with geneticin (300 μg/ml) as described (24, 25). MDA-MB231 (ER-negative) breast cancer cells were maintained at 37 C in phenol red-free DMEM/F12 medium containing 10% charcoal-stripped fetal bovine serum as described (26). The cells were plated 24 h before estrogen treatment. Cells were grown in T-75 flasks at 1 x 106 cells per flask for all experiments except for transient transfection assays in which cells were grown in 6-well plates at 1.5 x 105 cells/well. The cells were treated with 10 nM (protein studies) or 20–100 nM (transient transfections) E2 for various periods of time (5 min to 48 h). Cells were washed with PBS and harvested at the end of estrogen treatment.
Cell lysis and preparation of cytoplasmic and nuclear extracts
Estrogen-treated cell pellets were lysed by suspending in Nonidet P-40 lysis buffer containing 0.5% Nonidet P-40; 90 mM KCl; 1 mM magnesium acetate; 2 mM 2-mercaptoethanol; 10 μg/ml leupeptin; 1x phosphatase inhibitor cocktail 2 (a mixture of sodium orthovanadate, sodium molybdate, and sodium tartrate) (Sigma); and 10 mM HEPES (pH 7.6). After centrifugation at 10,000 x g for 10 min, the resultant supernatant containing cytoplasmic extract was used for immunoprecipitation, and the pellets containing nuclei were saved for preparing nuclear extract (27).
Immunoprecipitation analyses
Cytoplasmic cell extracts (60 μg protein) were used for immunoprecipitation with 1 μg of the anti-phosphoStat or anti-Stat1 antibody (Santa Cruz Biotechnology, San Diego, CA). Bound proteins were purified using protein A Sepharose. Immunoprecipitated proteins were separated by SDS-PAGE and analyzed by Western blotting. For coimmunoprecipitation studies, cell extracts containing equal amounts of protein were immunoprecipitated with anti-Src (Cell Signaling Technology, Beverly, MA), anti-Stat1 antibody, or nonimmune IgG (Dako Cytomation, Carpinteria, CA). The immunoprecipitates were incubated with protein A Sepharose for 1 h at 4 C and then washed and separated on SDS-PAGE gel and analyzed by Western blotting using anti-Src and anti-Stat1 antibodies.
EMSAs
Nuclear extracts (15 μg protein) prepared from hFOB cells were used in EMSAs as described in the manufacturer’s protocol (GENEKA, Montréal, Canada). Where indicated, the reactions were incubated with equal concentrations of cold and hot mutant and wild-type double-stranded DNA sis-inducible element (SIE) oligonucleotides provided in the kit as described (GENEKA). The following SIE oligonucleotides were used in this experiment (sequence of the sense strand is shown): wild-type SIE oligonucleotide, 5'-GTC GAC ATT TCC CGT AAA TCG TCG A-3'; and mutant SIE oligonucleotide, GTC GAC ATA TAG CGT AAA TCG TCG A-3'. The binding reactions were subjected to electrophoresis on a 5% polyacrylamide gel in 0.5x Tris-borate EDTA running buffer. Gels were analyzed on a PhosphorImager analyzer (Molecular Dynamics Inc., Sunnyvale, CA).
Luciferase assay
Luciferase constructs containing GAS elements (28) and luciferase constructs containing ISRE elements (Stratagene) were used in this study. Cells plated in 6-well plates were transfected at 60% confluence with pGAS-luciferase using the transfection agent lipofectamine as described in the manufacturer’s protocol (Invitrogen, Carlsbad, CA). The plasmids pISRE-luciferase, luciferase driven by pGASM-luciferase and dominant-mutant Stat1 cDNA (DN-Stat1) were used in various transfection experiments as indicated. All transfections performed had an internal transfection control plasmid containing Renilla luciferase (pRL-CMV-luciferase) (Promega), which also allowed normalization of luciferase units. Twenty-four hours post transfection, the cells were treated with 20 nM E2, 100 nM ER antagonist ICI 182,780, or 25 μM Src-kinase family inhibitor pyrazolopyrimidine compound (PP2). The cells were harvested after 48 h of treatment using 300 μl of passive lysis buffer provided in a luciferase assay kit (Promega). Luciferase assays were performed using the dual luciferase reporter assay system according to the manufacturer’s protocol (Promega) and read on a TD-20E luminometer (Turner, Sunnyvale, CA).
Statistical analysis
All values are expressed as mean ± SE. In each experiment, three to eight replicates of each treatment were measured. The data are representative of three independent experiments. Fisher’s protected least significant difference with a one-way ANOVA was used to compare the groups. P < 0.05 was considered statistically significant. Due to the small sample size in some of the groups, we supplemented the ANOVA with nonparametric Kruskal-Wallis tests.
Results
Effect of estrogen on Stat1 protein phosphorylation
E2 treatment resulted in an increased tyrosine phosphorylation of Stat1 within 15 min of treatment as shown by Western blot hybridization using an anti-phosphoStat1 antibody [p-Stat1 (Tyr 701)] directed against the tyrosine phosphorylated residue of Stat1 (Fig. 1A). The increase in phosphorylation appears to be transient and decreases with time. In contrast, there was no significant change in the total amount of Stat1 protein with E2 treatment as detected by anti-Stat1 p91 antibody (Fig. 1B).
FIG. 1. Time-course effects of E2 on Stat1 phosphorylation shows rapid transient increase in phosphorylation of Stat1 with no change in Stat1 protein levels. Cytoplasmic extracts prepared from hFOB cells 15, 30, 60, 120, and 240 min after E2 (10 nM) treatment were subjected to immunoprecipitation with anti-Stat1 p91 antibody (Santa Cruz Biotechnology) followed by Western blotting using p-Stat1 (Tyr 701) (Santa Cruz Biotechnology) directed against its phosphotyrosine residue (A). The filter was stripped and reprobed with the anti-Stat1 antibody (B). veh, Vehicle control.
Estrogen treatment leads to activation and formation of Stat1-DNA complex
Nuclear extracts prepared from hFOB cells treated with E2 were analyzed by EMSA (Fig. 2). Stat1 binding to radiolabeled wild-type oligonucleotide probe increased in hFOB cells in the presence of E2 (lane 7), compared with the vehicle control (lane 3). The specificity of Stat1 binding was confirmed by decreased binding in the presence of nonradiolabeled (cold) wild-type oligonucleotide (lane 5) and absence of binding in the presence of radiolabeled mutant oligonucleotide (lane 4). Stat1 binding to radiolabeled wild-type probe was not decreased in the presence of nonradioactive (cold) mutant oligonucleotides (lane 6), another indication of specificity. The specificity of Stat1-protein and DNA interaction has been further confirmed by the super shift in the presence of anti-Stat1 p91 (lane 8) and p-Stat1 (Tyr 701) antibody (lane 9).
FIG. 2. EMSA showing that estrogen treatment increases Stat1 binding to DNA. Nuclear extracts prepared from hFOB cells 16 h after E2 (10 nM) treatment were analyzed by EMSA by incubating with specific SIE oligonucleotide probes (GENEKA). Lower arrow indicates the migration of the Stat1-DNA complex. Upper arrow indicates the supershift and the migration of the Stat1-Ab-DNA complex. NA, No addition control; veh, vehicle control.
Estrogen treatment increases Stat1-dependent transcriptional activity in osteoblast cells
To determine whether estrogen treatment induces Stat1-stimulated transcription and whether Stat1-dependent promoter activities are involved in estrogen-mediated activation, the effects of E2 was tested on both the ISRE- and GAS-driven luciferase activities by transient transfection assays in hFOB cells. Relative to the activity in untreated cells, estrogen stimulated the GAS-luciferase reporter gene activity approximately 8-fold (vehicle 0.8 ± 0.3; estrogen 7 ± 0.2) (Fig. 3A). As expected, interferon- treatment led to even greater stimulation of the reporter activity (>100-fold) (vehicle 0.8 ± 0.3; interferon 212 ± 2.5) (Fig. 3A). A time-dependent increase in GAS-luciferase reporter gene activity occurred after E2 treatment (vehicle 0.9 ± 0.2; estrogen treatment 8 h, 2. ± 0.1; 12 h, 2.5 ± 0.8; 24 h, 3.7 ± 0.3; 48 h, 7.6 ± 0.1) (Fig. 3B). GAS promoter with mutations in the sequences was not activated in response to estrogen treatment, and no increase in luciferase activity was observed in cells transiently transfected with luciferase constructs driven by GAS mutant promoter (vehicle 6.6 ± 0.3; estrogen 8 ± 0.6) (Fig. 3C). Estrogen treatment led to a small change on ISRE-driven luciferase activity (vehicle 1.8 ± 0.1, estrogen 2.2 ± 0.1) (Fig. 3D), whereas interferon-? treatment stimulated ISRE-luciferase promoter functions (vehicle 6.6 ± 0.3, estrogen 217 ± 10.5) (Fig. 3D).
FIG. 3. Effect of E2 on Stat1-dependent transcription in hFOB cells. A, Effect on GAS-luciferase activity. Cells transfected with expression plasmid GAS-luciferase were treated for 48 h with vehicle control (veh), E2 (20 nM), or interferon- (IFN) (1000 U), and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 8 replicate cultures). #, P < 0.007 vs. veh; *, P < 0.0001 vs. veh; ##, P < 0.0001 vs. E2. B, Time course. Cells transfected with expression plasmid GAS-luciferase were treated with vehicle control (veh) or E2 (20 nM) for 8, 12, 24, and 48 h, and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). #, P < 0.009 vs. veh; ##, P < 0.002 vs. veh; *, P < 0.0001 vs. veh; C, Effect on GASM-luciferase activity: Cells transfected with expression plasmid GASM-luciferase were treated with vehicle control (veh) or E2 (20 nM) for 48 h, and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). D, Effect on ISRE-luciferase activity: Cells transfected with expression plasmid ISRE-luciferase were treated with veh, E2 (20 nM), or interferon-? (IFN-?) (1000 U) for 48 h, and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). *, P < 0.0001 vs. veh; #, P < 0.0001 vs. E2.
Activation of Stat1-dependent GAS promoter activity by estrogen is supported by experiments in cells cotransfected with dominant-negative Stat1 mutant cDNA, in which the E2-mediated activation of GAS luciferase activity was decreased (Fig. 4). Relative to the activity in untreated cells, estrogen stimulated GAS-luciferase reporter gene activity in the absence or presence of DN-Stat1 by 6.3- and 3.2-fold, respectively (vehicle 1.5 ± 0.1; estrogen 8.6 ± 0.6; estrogen+DN-Stat1 4.4 ± 0.4; DN-Stat1 2. 8 ± 0.2) (Fig. 4). However, ER-dependent activation of estrogen response element (ERE) and ERE-controlled expression of luciferase activity were not blocked (Fig. 4). Relative to the activity in untreated cells, estrogen stimulated ERE-luciferase reporter gene activity in the absence or presence of DN-Stat1 by 4.6- and 3.9-fold, respectively (vehicle 2.8 ± 0.1; estrogen 12.4 ± 0.6; estrogen+DN-Stat1 11 ± 0.5, DN-Stat1 2.5 ± 0.2) (Fig. 4).
FIG. 4. DN-Stat1 expression blocks GAS-luciferase transcription but not ERE-luciferase transcription in hFOB cells. Cells were cotransfected with the reporter plasmid GAS-luciferase or ERE-luciferase and expression vector containing DN-Stat1 and treated with vehicle control (veh) and E2 (20 nM) for 48 h. The luciferase reporter activity was analyzed. Values are the mean ± SE (n = 5 replicate cultures). *, P < 0.0001 vs. veh; #, P < 0.0001 vs. E2.
Estrogen-mediated activation of Stat1 is ER independent
ER antagonist ICI 182,780 blocked estrogen-stimulated ERE-luciferase activity in hFOB cells (data not shown). In contrast, ICI 182,780 did not block E2-stimulated GAS-driven luciferase activity (vehicle 1.2 ± 0.1; estrogen 9 ± 0.1; estrogen+ICI 182,780 8.6 ± 0.7) (Fig. 5).
FIG. 5. Estrogen-mediated activation of Stat1-dependent reporter gene expression is not ER dependent. hFOB cells transfected with the expression plasmid GAS-luciferase were treated for 48 h with vehicle control (veh), E2 (20 nM), or E2 plus ER antagonist ICI 182,780 (100 nM) (E2+ICI), and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). *, P < 0.0001 vs. veh.
Estrogen treatment induces Stat1 activation in ER-negative breast cancer cells
To determine target cell specificity, Stat1 phosphorylation was investigated in E2-treated breast cancer cells. Estrogen treatment resulted in an increased tyrosine phosphorylation of Stat1 within 15 min in ER-negative (MDA-MB231) breast cancer cells as shown by Western blot hybridization with p-Stat1(Tyr 701) antibody (Fig. 6A). As was the case in bone cells, there was no significant change in the total amount of Stat1 protein with estrogen treatment in the breast cancer cells (Fig. 6B).
FIG. 6. Estrogen treatment increases Stat1 phosphorylation in ER-negative breast cancer cells. Cytoplasmic extracts were prepared from MDA-MB 231 breast cancer cells 15 min after treatment with E2 (10 nM) and subjected to immunoprecipitation with anti-Stat1 p91 antibody (Santa Cruz Biotechnology) followed by Western blotting using p-Stat1 (Tyr 701) (Santa Cruz Biotechnology) directed against its phosphotyrosine residue (A). The filter was stripped and reprobed with the anti-Stat1 antibody (B).
The effects of E2 on GAS-driven luciferase activities were examined by transient transfection assays in MDA-MB231 breast cancer cells to determine whether E2 treatment induces Stat1-stimulated transcription and whether Stat1-dependent promoter activity is involved. Relative to the activity in untreated cells, E2 at 10, 20, and 100 nM stimulated the GAS-luciferase reporter gene activity by 1.2-, 4.9-, and 9-fold, respectively (vehicle 2.8 ± 0.2, estrogen 10 nM, 4.9 ± 0.5; 20 nM, 15.7 ± 0.4; 100 nM, 27.8 ± 0.4) (Fig. 7).
FIG. 7. Effect of E2 (20 nM) on Stat1-dependent transcription in MDA-MB 231 cells. Cells transfected with expression plasmid GAS-luciferase, were treated for 48 h with vehicle control (veh) or E2 (20 nM), and the luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). Data are from three independent experiments. #, P 0.005 vs. veh; *, P < 0.0001 vs. veh.
Estrogen-mediated activation of Stat1 requires Src kinase activity
Estrogen-stimulated GAS-luciferase expression was blocked in the presence of Src kinase inhibitor, PP2 (vehicle 2.1 ± 0.3; estrogen 14 ± 0; estrogen+PP2 4 ± 0.4) (Fig. 8), whereas the epidermal growth factor receptor kinase inhibitor PP3, which serves as a negative control for PP2 and does not inhibit Src kinase activity, had no effect (estrogen+PP3 13.4 ± 0.6) (Fig. 8). Neither PP2 nor PP3 showed any significant independent effect on GAS luciferase activity (PP2 4.8 ± 1; PP3 4.5 ± 1). In cells that transiently express dominant-negative Src protein, estrogen-dependent GAS-luciferase activity was inhibited (data not shown).
FIG. 8. Estrogen-mediated activation of Stat-dependent reporter gene activation is Src kinase-dependent. hFOB cells transfected with the expression plasmid GAS-luciferase were treated for 48 h with vehicle control (veh); E2 (20 nM); Src-kinase family inhibitor, PP2 (25 μM); E2+PP2; epithelial growth factor receptor kinase inhibitor, PP3 (25 μM); or E2+PP3. The luciferase reporter activity was analyzed. Values are the mean ± SE (n = 3 replicate cultures). Data are from three independent experiments. *, P < 0.0001 vs. veh; #, P < 0.0001 vs. E2.
Src is associated with Stat1 in estrogen-treated osteoblasts
Coimmunoprecipitation studies in estrogen-treated hFOB cells analyzed by Western blot hybridization demonstrated c-Src protein coimmunoprecipitated with Stat1 (Fig. 9A). Both Stat1 and activated Stat1 proteins coimmunoprecipitated with c-Src as shown by anti-Stat1 and pStat1 antibodies (Fig. 9B). No signals were observed with nonimmune IgG controls (Fig. 9C).
FIG. 9. Stat1 is associated with Src in estrogen-treated bone cells. Cytoplasmic extracts prepared from hFOB cells treated for 15 min with vehicle (V) and E2 (10 nM) were subjected to immunoprecipitation (IP) with anti-Stat1 p91 antibody (Santa Cruz Biotechnology) (A), anti-c-Src antibody (Cell Signaling Technology) (B), or nonimmune IgG (Dako Cytomation) (C).
The luciferase activity results for all statistical analysis based on nonparametric Kruskal-Wallis tests were similar to those based on ANOVA (data not shown).
Discussion
Estrogen treatment leads to tyrosine phosphorylation and rapid (within 15 min) activation of Stat1 protein in ER-negative human osteoblast cells. This immediate response is followed by an increase in Stat1-specific DNA binding and transcription, which does not require classical ERs. In addition, the activation of Stat1 protein and the stimulation of Stat1-dependent transcription by estrogen in ER-negative breast cancer cells suggest that this could be a common signaling mechanism in estrogen target cells.
Many of the post-2 h actions of estrogen involve the genomic mechanism, which involves the binding of estrogen to ER, translocation of ligand bound receptors to the nucleus, followed by homo- or heterodimerization of ER and binding of ER dimers to estrogen-responsive elements on the DNA. The rapid (15 min) effects of estrogen described in this study occur too quickly for genomic-mediated mechanisms (13, 14, 29). Several other reports on the nongenomic actions of estrogens have been shown to occur much more rapidly, and some report the involvement of MAPK, Src, or phosphatidylinositol 3-kinase pathways (14, 29, 30, 31).
The estrogen-mediated activation of Stat1 described here is not inhibited by cycloheximide (Kennedy, A., A. Maran, and R. T. Turner, unpublished data). Unlike previously described nongenomic effects in bone cells, Stat1 activation by estrogen does not require classical ER, as shown by the failure of the ER antagonist ICI 182,780 to block the Stat1-dependent transcription. This finding may be analogous to the nongenomic activation of ERK MAPKs by estrogen in the explants of cerebral cortex (32). Estrogen increases type 1 collagen levels and alkaline phosphatase activity in hFOB cells (Maran, A., and R. T. Turner, unpublished data), indicating that activation of non-ER-mediated pathways by the hormone results in metabolic changes in bone cells.
Stat1 protein-dependent transcriptions involve the binding of Stat1 to two specific promoter elements, namely ISRE and GAS in cytokines and growth factor-treated cells (15, 33, 34). The Stat family members show specificity in DNA binding: Stat1 homodimer binds to GAS, whereas the Stat1/Stat2 heterodimer interacts with ISRE. E2 activation of Stat1 protein results in increased transcription of GAS element-containing genes. Estrogen’s preference for the GAS over the ISRE suggests that estrogen elicits responses similar to interferon- or ligands like IL-6 and platelet-derived growth factor (33, 35, 36). Stat1 binding GAS elements are present in the regulatory regions of genes expressed by osteoblasts, such as type 1 collagen, osteocalcin, and osteonectin. Further work is necessary to determine the effects of estrogen-mediated activation of Stat1 activity on osteoblast metabolism.
The involvement of Stat1 protein in estrogen-mediated actions of osteoblasts differs from a recent report showing estrogen-mediated, nongenomic activation in endothelial cells involving other Stat family members (Stat 3 and 5) in an ER-dependent manner (31). One report (14) has shown that an alternative form of ER could mediate some of the nongenomic actions of estrogens. Preliminary findings suggest that a subset of the classical ER is associated with the cell membrane (14, 32) and such putative membrane ERs have been found to be insensitive to antiestrogen ICI 182,780 treatment (32). The present findings do not address whether such membrane ERs are present in hFOB cells and could play a role in estrogen-mediated activation of Stat1 signaling.
Several known mechanisms lead to tyrosine phosphorylation and activation of Stat proteins in cells, including: 1) cytokine receptors that activate Janus kinases, which in turn phosphorylate Stat proteins (e.g. interferon and IL-6 receptors); 2) phosphorylation by growth factor receptors that have intrinsic tyrosine kinase activities (e.g. platelet-derived growth factor, epithelial growth factor receptors); or 3) direct phosphorylation by nonreceptor tyrosine kinases (e.g. Src, Fes, Abl). Stat1-dependent GAS promoter activity was inhibited in the presence of a chemical inhibitor for c-Src kinase, and both Stat1 and activated Stat1 are associated with c-Src kinase protein. The association of other Stat family members with Src protein has been demonstrated in cancer cells and cytokine-dependent signaling (37, 38, 39, 40). This suggests that estrogen-mediated activation of Stat1 protein in hFOB cells is dependent on the Src kinase pathway. Further work is necessary to determine whether Src kinase plays a role in Stat1 phosphorylation in estrogen-treated bone cells. These results are consistent with the rapid activation of Src kinase shown previously in osteoblasts and osteocytes after estrogen exposure (14). Thus, Stat1 association with activated Src kinase in estrogen-treated cells supports a potentially vital role for Src kinase pathway in Stat1 activation by estrogen in target cells.
In conclusion, ER-independent activation of Stat1 in E2-treated bone and breast cancer cells presents a novel nongenomic, non-ER-mediated mechanism for estrogen action and reinforces the concept that estrogen involves multiple signaling pathways in target cells.
Acknowledgments
The authors thank Ralf Janknecht (Mayo Clinic) and Dhan Kuppuswamy (Medical University of South Carolina) for helpful suggestions and comments on the manuscript. The authors thank the Center for Patient Oriented Research at Mayo Clinic for the statistical analysis, Gobinda Sarkar for helping with the PCRs, and Peggy Backup for editorial assistance.
References
Griffin C, Flouriot G, Sonntag-Buck V, Nestor P, Gannon F 1998 Identification of novel chicken estrogen receptor- messenger ribonucleic acid isoforms generated by alternative splicing and promoter usage. Endocrinology 139:4614–4625
Auchus RJ, Faqua SAWW 1994 Hormones-nuclear receptor in health disease. The oestrogen receptor. Baillieres Clin Endocrinol Metab 8:433–449
Sarrel PM, Lufkin EG, Oursler MJ, Keefe D 1994 Estrogen actions in arteries, bone and brain. Sci Am Sci Med 1:44–53
Turner RT, Riggs BL, Spelsberg TC 1994 Skeletal effects of estrogen. Endocr Rev 15:275–300
Kuiper GG, Gustafsson JA 1997 The novel estrogen receptor-? subtype: potential role in the cell- and promoter-specific actions of estrogens and anti-estrogens. FEBS Lett 410:87–90
Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA 1996 Cloning of a novel receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925–5930
Koike S, Sakai M, Muramatsu M 1987 Molecular cloning and characterization of rat estrogen receptor cDNA. Nucleic Acids Res 15:2499–2513
Bonnelye E, Merdad L, Kung V, Aubin JE 2001 The orphan nuclear estrogen receptor-related receptor (ERR) is expressed throughout osteoblast differentiation and regulates bone formation in vitro. J Cell Biol 153:971–984
Bonnelye E, Vanacker JM, Dittmar T, Begue A, Desbiens X, Denhardt DT, Aubin JE, Laudet V, Fournier B 1997 The ERR-1 orphan receptor is a transcriptional activator expressed during bone development. Mol Endocrinol 11:905–916
Tzukerman MT, Esty A, Santiso-Mere D, Danielian P, Parker MG, Stein RB, Pike JW, McDonnell DP 1994 Human estrogen receptor transactivational capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrinol 8:21–30
Yang NN, Venugopalan M, Hardikar S, Glasebrook A 1996 Identification of an estrogen response element activated by metabolites of 17?-estradiol and raloxifene. [Erratum (1997) 275:1249]. Science 273:1222–1225
El-Tanani MK, Green CD 1997 Two separate mechanisms for ligand-independent activation of the estrogen receptor. Mol Endocrinol 11:928–937
Monroe DG, Spelsberg TC 2003 A case for estrogen receptors on cell membranes and nongenomic actions of estrogen. Calcif Tissue Int 72:183–184
Kousteni S, Bellido T, Plotkin LI, O’Brien CA, Bodenner DL, Han L, Han K, DiGregorio GB, Katzenellenbogen JA, Katzenellenbogen BS, Roberson PK, Weinstein RS, Jilka RL, Manolagas SC 2001 Nongenotropic, sex-nonspecific signaling through the estrogen or androgen receptors: dissociation from transcriptional activity. Cell 104:719–730
Darnell Jr JE, Kerr IM, Stark GR 1994 Jak-STAT pathways and transcriptional activation in response to IFNs and other extracellular signaling proteins. Science 264:1415–1421
Leaman DW, Leung S, Li X, Stark GR 1996 Regulation of STAT-dependent pathways by growth factors and cytokines. FASEB J 10:1578–1588
Kaplan MH, Schindler U, Smiley ST, Grusby MJ 1996 Stat6 is required for mediating responses to IL-4 and for development of Th2 cells. Immunity 4:313–319
Ihle JN, Kerr IM 1995 Jaks and Stats in signaling by the cytokine receptor superfamily. Trends Genet 11:69–74
Stark GR, Kerr IM, Williams BR, Silverman RH, Schreiber RD 1998 How cells respond to interferons. Annu Rev Biochem 67:227–264
Bromberg J, Darnell Jr JE 2000 The role of STATs in transcriptional control and their impact on cellular function. Oncogene 19:2468–2473
Bellido T, Borba VZ, Roberson P, Manolagas SC 1997 Activation of the Janus kinase/STAT (signal transducer and activator of transcription) signal transduction pathway by interleukin-6-type cytokines promotes osteoblast differentiation. Endocrinology 138:3666–3676
Sahni M, Ambrosetti DC, Mansukhani A, Gertner R, Levy D, Basilico C 1999 FGF signaling inhibits chondrocyte proliferation and regulates bone development through the STAT-1 pathway. Genes Dev 13:1361–1366
Nishiyama T, Hatano H, Kurosaka M, Bolander ME, Sarkar G 2003 Cis-acting intronic elements that regulate cartilage-specific alternative splicing of the type II collagen (Col2) pre-mRNA lie at or near splice site junction sequences flanking exon 2 of the gene. J Bone Miner Res 18:1716–1722
Harris SA, Enger RJ, Riggs BL, Spelsberg TC 1995 Development and characterization of a conditionally immortalized human fetal osteoblastic cell line. J Bone Miner Res 10:178–186
Maran A, Zhang M, Spelsberg TC, Turner RT 2001 The dose-response effects of ethanol on the human fetal osteoblastic cell line. J Bone Miner Res 16:270–276
Reinholz GG, Getz B, Sanders ES, Karpeisky MY, Padyukova NSh, Mikhailov SN, Ingle JN, Spelsberg TC 2002 Distinct mechanisms of bisphosphonate action between osteoblasts and breast cancer cells: identity of a potent new bisphosphonate analogue. Breast Cancer Res Treat 71:257–268
Maran A, Maitra RK, Kumar A, Dong B, Xiao W, Li G, Williams BR, Torrence PF, Silverman RH 1994 Blockage of NF-B signaling by selective ablation of an mRNA target by 2–5A antisense chimeras. Science 265:789–792
Ramana CV, Grammatikakis N, Chernov M, Nguyen H, Goh KC, Williams BR, Stark GR 2000 Regulation of c-myc expression by IFN- through Stat1-dependent and -independent pathways. EMBO J 19:263–272
Pedram A, Razandi M, Aitkenhead M, Hughes CC, Levin ER 2002 Integration of the non-genomic and genomic actions of estrogen. Membrane-initiated signaling by steroid to transcription and cell biology. J Biol Chem 277:50768–50775
Wong C, McNally CEN, Komm BS, Cheskis BJ 2002 Estrogen receptor-interacting protein that modulates its nongenomic activity-cross-talk with Src/Erk phosphorylation cascade. Proc Natl Acad Sci USA 99:14783–14788
Bjornstrom L, Sjoberg M 2002 Signal transducers and activators of transcription as downstream targets of nongenomic estrogen receptor actions. Mol Endocrinol 16:2202–2214
Decker T, Lew DJ, Mirkovitch J, Darnell Jr JE 1991 Cytoplasmic activation of GAF, an IFN--regulated DNA-binding factor. EMBO J 10:927–932
Decker T, Kovarik P, Meinke A 1997 GAS elements: a few nucleotides with a major impact on cytokine-induced gene expression. J Interferon Cytokine Res 17:121–134
Shuai K, Horvath CM, Huang LH, Qureshi SA, Cowburn D, Darnell Jr JE 1994 Interferon activation of the transcription factor Stat91 involves dimerization through SH2-phosphotyrosyl peptide interactions. Cell 76:821–828
Sehgal PB 2000 STAT-signalling through the cytoplasmic compartment: consideration of a new paradigm. Cell Signal 12:525–535
Singh M, Setalo Jr G, Guan X, Warren M, Toran-Allerand CD 1999 Estrogen-induced activation of mitogen-activated protein kinase in cerebral cortical explants: convergence of estrogen and neurotrophin signaling pathways. J Neurosci 19:1179–1188
Xi S, Zhang Q, Dyer KF, Lerner EC, Smithgall TE, Gooding WE, Kamens J, Grandis JR 2003 Src kinases mediate STAT growth pathways in squamous cell carcinoma of the head and neck. J Biol Chem 278:31574–31583
Cao X, Tay A, Guy GR, Tan YH 1996 Activation and association of Stat3 with Src in v-Src-transformed cell lines. Mol Cell Biol 16:1595–1603
Chaturvedi P, Reddy MV, Reddy EP 1998 Src kinases and not JAKs activate STATs during IL-3 induced myeloid cell proliferation. Oncogene 16:1749–1758
Reddy EP, Korapati A, Chaturvedi P, Rane S 2000 IL-3 signaling and the role of Src kinases, JAKs and STATs: a covert liaison unveiled. Oncogene 19:2532–2547(Angela M. Kennedy, Kriste)