This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lopez, D. Articles by McLean, M. P. Search for Related Content PubMed PubMed Citation Articles by Lopez, D. Articles by McLean, M. P.

Estrogen Activates the High-Density Lipoprotein Receptor Gene via Binding to Estrogen Response Elements and Interaction with Sterol Regulatory Element Binding Protein-1A

Departments of Obstetrics & Gynecology (D.L., M.D.S., W.S.-E., M.P.M.) and Biochemistry & Molecular Biology (M.P.M.), College of Medicine, University of South Florida, Tampa, Florida 33606

Address all correspondence and requests for reprints to: Dr. Mark P. McLean, Department of Obstetrics & Gynecology, University of South Florida, 4 Columbia Drive, Room 529, Tampa, Florida 33606. E-mail: . mmclean{at}hsc.usf.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
The effects of E2 on the high-density lipoprotein receptor (HDL-R) scavenger receptor class B type I (SR-BI) gene were examined. Four putative estrogen response element half-site motifs (ERE) (-2176, -1726, -1622, and -1211, designated ERE-1, 2, 3, and 4, respectively) were identified in the HDL-R SR-BI promoter. Transfection studies and mutation analysis demonstrated that E2 significantly increased HDL-R SR-BI promoter activity and that mutating ERE-1, 2, and 4 resulted in a loss of E2 responsiveness. Both ER and ERß formed specific complexes with ERE-1, 2, and 4 but did not bind ERE-3 in vitro. Interestingly, ERE-3 was the motif shown not to be important for E2-activation of the HDL-R SR-BI promoter in the mutational analysis studies. The influence of SREBP-1a (sterol regulatory element binding protein-1a) on E2 regulation of the HDL-R SR-BI gene was also examined. SREBP-1a was able to bind directly to the ERE motifs and enhanced ER binding when both ER subtypes were present. ER and ß also bound to a sterol response element motif, but they did not enhance SREBP-1a binding. Cotransfection studies demonstrated that the presence of the three factors, ER, ERß, and SREBP-1a, enhanced the overall luciferase activity produced from the HDL-R SR-BI promoter construct in the presence of only one of the factors. Interaction of SREBP-1a with both ERs was demonstrated using a mammalian two-hybrid assay. The data confirmed that E2 through the ERs can positively regulate the HDL-R SR-BI through binding and activation of three ERE motifs and identified SREBP-1a as a potential coactivator of the E2-ER-dependent effects on the HDL-R SR-BI gene.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
CARDIOVASCULAR DISEASES ARE the leading cause of death and disability in the United States for both sexes (1, 2). Although women show less cardiovascular diseases than age-matched men in the premenopausal period (3), the risk of death from cardiovascular diseases increases with a woman’s age, with the sharpest increases in the postmenopausal years (4). These changes appear to correlate with decreases in estrogen levels after menopause (3, 5). In addition, epidemiological studies consistently report that endogenous and exogenous estrogens reduced cardiovascular morbidity and mortality in women (6, 7) and oophorectomized primates (8, 9). One putative mechanism by which estrogen may reduce the risk of atheroma formation is by increasing high-density lipoprotein (HDL) cholesterol levels (10, 11) and reverse cholesterol transport, and through this protective mechanism, decrease the risk of developing cardiovascular diseases (12).

The HDL receptor (HDL-R) scavenger receptor class B type I (SR-BI) has been shown to mediate cholesterol efflux to lipoproteins from atheromatous arteries (13) and HDL cholesterol uptake by the liver (reverse cholesterol transport) (14, 15) and steroidogenic tissues (15, 16, 17). Several studies have identified this receptor as the major determinant of plasma HDL concentrations (18, 19) and hence, a key element in cholesterol homeostasis. In addition to its role in lipid metabolism, the HDL-R SR-BI has also been shown to be critical in reproductive physiology as demonstrated by infertility of female HDL-R SR-BI knockout mice (17). Thus, elucidation of the regulatory mechanisms underlying HDL-R SR-BI gene expression, especially by factors known to alter HDL levels such as estrogen (10, 11), is extremely important. Landschulz et al. (15) reported that estrogen treatment increased HDL-R SR-BI expression in steroidogenic tissues but dramatically decreased hepatic HDL-R SR-BI gene expression demonstrating tissue specific regulation. The mechanisms by which tissue specific factors influence estrogen regulation of the HDL-R SR-BI gene have not been clearly defined.

Estrogen influences gene transcription through activation of its nuclear receptors (20). In the classical pathway, once activated by estrogen, the ERs can bind as a homo- or heterodimer to the EREs found in the promoter of estrogen-sensitive genes, to activate or repress transcription (21, 22, 23, 24, 25). Estrogen has also been shown to alter transcription of an estrogen-sensitive gene through nonclassical pathways (26, 27, 28, 29, 30, 31). In these cases, activated ER does not bind to EREs in the promoter of the target gene (26, 27, 28, 29, 30, 31). Instead, the ER interacts with other transcription factors bound to their response elements resulting in enhancement or repression of transcription (26, 27, 28, 29, 30, 31). Transcriptional factors that have been shown to interact with the ER include steroidogenic factor-1 (SF-1; 26), specific factor-1 (Sp1; 27, 28, 29), nuclear factor-Y (29), and activator protein-1 (30, 31).

Currently, two ER isoforms have been identified, and ß (32, 33). Both isoforms exhibit a high degree of homology within the DNA-binding domain and appear to bind to the same ERE motifs (25) but differ in their transactivation domains suggesting distinct roles in gene activation. In addition, their transcriptional activity has been shown to be influenced not only by ligands, but also by coregulatory proteins with which they associate (34, 35). It has been suggested that the ERß is the form responsible for the cardiovascular benefits of estrogen (36). However, one cannot exclude the possible mechanisms of HDL-R SR-BI gene regulation by ER.

In the current investigation, transfection studies followed by luciferase assays and EMSAs were used to examine the effects of E2 on the regulation of the HDL-R SR-BI gene. The influence of a known regulator of HDL-R SR-BI gene transcription, SREBP-1a (37), on estrogen-dependent regulation of this gene was also examined.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Materials
Integrated DNA Technologies, Inc. (Coralville, IA) synthesized all oligonucleotides and primers. The pGL3-basic luciferase vector, the renilla luciferase vector, and the Dual Luciferase Reporter Assay System were obtained from Promega Corp. (Madison, WI). The full-length rat ERß cDNA, which encodes a protein of 549 amino acids, cloned in the p-cytomegalovirus (CMV)-5 vector (ERß-pCMV5) was obtained from Dr. Jan-Åke Gustafsson (Department of Biosciences at Novum, Karolinska Institute, Huddinge, Sweden). The human breast carcinoma MCF-7 cells, the human bladder carcinoma HTB-9 cells, and the human ER cDNA in pBluescript were obtained from American Type Culture Collection (Manassas, VA). For the experiments, the full-length ER cDNA was cloned into the pCMV5 expression vector. The rat luteal GG-CL cell line was kindly provided by Dr. Geula Gibori (Department of Physiology and Biophysics, University of Illinois College of Medicine, Chicago, IL). ER subtype ( and ß) specific antibodies and recombinant proteins were purchased from PanVera (Madison, WI). The NH₂-terminal segment (active fragment) of SREBP-1a under the control of the CMV promoter (SREBP-1a-pCMV5) and SREBP-1a-polyhistidine-tagged in the pRSET B vector were kindly provided by Dr. Tim Osborne (Department of Molecular Biology and Biochemistry, University of California, Irvine, CA). The rabbit polyclonal SREBP-1 specific antibody was purchased from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). [³²P]-deoxy-CTP (3000 Ci/mmol) and the polydeoxyinosinic-deoxycytidylic acid were obtained from Amersham Pharmacia Biotech (Piscataway, NJ). The QuikChange Site-directed Mutagenesis Kit was purchased from Stratagene (La Jolla, CA). All restriction enzymes and the Fugene 6 Transfection reagent were obtained from Roche Biochemicals Inc. (Indianapolis, IN). DMEM: nutrient mixture F-12 (DMEM: F-12) and the Roswell Park Memorial Institute medium were obtained from Life Technologies, Inc./BRL (Grand Island, NY). FBS was purchased from Summit Biotechnology (Ft. Collins, CO). BioMax-MR films were obtained from Fisher Scientific (Norcross, GA). All other chemicals were purchased from Fisher Scientific or Sigma (St. Louis, MO).

Preparation of MCF-7 and GG-CL nuclear extracts
Cells (5 x 10⁷) were treated with 0.25% trypsin in 1 mM EDTA and then centrifuged. Nuclear extracts were prepared from the cell pellets as previously described (37). Protein samples were concentrated 4- to 6-fold using Centricon-10 concentrators (Millipore Corp., Bedford, MA), and concentrations were determined using the Bio-Rad Laboratories, Inc. (Hercules, CA) protein assay. Nuclear proteins were then used in Western blot analysis.

Western blot analysis
Nuclear proteins (50 µg) were denatured at 100 C in loading buffer for 10 min and subjected to electrophoresis on a 12% SDS-PAGE as previously described (37). After electrophoresis, samples were electroblotted onto nitrocellulose membranes (0.2-µm pore) in buffer containing 0.25 M Tris hydrochloride (Tris-HCl; pH 8.3) and 1.92 M glycine for 16 h at 4 C. To verify equal protein loading, nitrocellulose membranes were stained with 0.1% Ponceau S (in 5% acetic acid) and destained in water. Western blot analysis was carried out with a 1:1000 dilution of the rabbit polyclonal specific antiserum to ER or ERß in 3% nonfat dry milk. Immunoreactive proteins were then visualized using a 1:6000 dilution of horseradish peroxidase-conjugated goat antirabbit antisera (Santa Cruz Biotechnology, Inc.) in 3% nonfat dry milk and the SuperSignal ULTRA Chemiluminescent Substrate method (Pierce Chemical Co., Rockford, IL).

Cell transfections and luciferase assays
The p-2267 HDL-R SR-BI promoter-luciferase gene construct used in these experiments was prepared as previously described (37). MCF-7 cells were cultured in 24-well (2.5 x 10⁵ cells per well) plates DMEM: F-12 medium without phenol red + 5% charcoal-stripped FBS. GG-CL cells were cultured in 12-well (5 x 10⁵ cells per well) tissue culture plates in Roswell Park Memorial Institute medium without phenol red + 5% charcoal-stripped FBS at 33 C for 24 h and then transferred to 40 C for 24 h before transfection. Cells were transfected with 0.5 (for 24-well) or 1.0 (for 12-well) µg each of appropriate plasmid using Fugene 6 (ratio 1:2) and incubated for 48 h. Twenty-four hours before the end of the incubation period, cells were incubated with 17ß-E2 (0.001–100 nM; E2; Sigma), 4-hydroxytamoxifen (4-OHT; 100 nM), or an equal volume of ethanol (vehicle). Cells were washed twice with PBS and treated with a passive lysis buffer for 20 min. Lysates were transferred to a microcentrifuge tube and stored at -80 C until the determination of luciferase activity. Luciferase assays were performed as previously described (37). Cotransfection of a plasmid containing the renilla luciferase gene under control of the simian virus 40 early enhancer/promoter region (0.25 µg for 24-well or 0.5 µg for 12-well plates) was used as a control to correct for differences in transfection efficiencies. Luciferase data were expressed as mean fold induction ± SEM, where the value of luciferase activity for the promoter construct in the absence of E2 or any exogenous transcriptional factor was set to 1.0. Each luciferase assay experiment was performed in triplicate and repeated for the indicated number of times in the figure legends. Data from the individual parameters were compared by ANOVA followed by Student’s-Newman-Keuls multiple comparison test when applicable (38). A value of P < 0.05 was considered significant for all tests.

Site-directed mutagenesis
Site-directed mutants were obtained using the QuikChange Site- Directed Mutagenesis Kit (Stratagene) as previously described (39). Oligonucleotides used for site-directed mutagenesis to mutate the four ERE motifs were 5'-GACTCCTCCTGAATCACATAGAATG-3' (for ERE-1), 5'-GGTCTCACTAATCAGAGAAGTCTGG-3' (for ERE-2), 5'-GTGCACCATTTCTGAGGAGG-3' (for ERE-3), and 5'-CTGACTCTGAATCATGTTGCTGAAG-3' (for ERE-4), and their complements. The nucleotides that are underlined correspond to the mutated bases. All mutations were confirmed by sequencing using the T7 Sequenase DNA sequencing kit and [³⁵S]-deoxy-ATP.

Fusion protein production
Histidine-tagged SREBP-1a protein was overexpressed in Escherichia coli by induction of mid-logarithmic-phase cultures with 1 mM isopropyl-ß-D-thiogalactopyranoside. After incubating for 6 h at 27 C, cells were sedimented by centrifuging at 7700 x g for 10 min at 4 C. The cell pellet after the initial centrifugation was resuspended in guanidium lysis buffer provided in the XPRESS System (Invitrogen, Carlsbad, CA). Purification was performed using an immobilized metal affinity column according to the manufacturer’s recommendations. Recombinant protein samples were concentrated 4- to 6-fold using Centricon-10 concentrators (Millipore Corp.). Protein concentrations were determined using the Bio-Rad Laboratories, Inc. protein assay. The purified SREBP-1a protein was then used in mobility shift assays.

EMSA
Oligonucleotides corresponding to each putative ERE motif and to the distal SRE (dSRE) found in the promoter of the HDL-R SR-BI gene were synthesized and annealed in 10 mM Tris-HCl, pH 7.5; 1 mM EDTA; 25 mM NaCl; 10 mM MgCl₂; and 1 mM dithiothreitol. The oligonucleotides used were ERE-1 (5'-GACTCCTCCTGGGTCACATAGAATGGC-3'), ERE-2 (5'-CGGGTCTCACTGGTCAGAGAAGTCTGG-3'), ERE-3 (5'-TCTGTGCACCAGGTCAGAGGAGGGCAT-3'), ERE-4 (5'-TCCTGACTCTGGGTCATGTTGCTGAAG-3'), and dSRE (5'-CTGCCCCCCTCACACCCTCCTCTGTAG-3'). An oligonucleotide (5'-CCAGGTCAGAGTGACCTGAGCTAAAAT-3') that contains a consensus ERE motif was also synthesized and used as control in some experiments. Nucleotides in boldface underlined letters correspond to the putative ERE sites. The oligonucleotide probes were labeled using the Klenow fragment of DNA polymerase and [³²P] deoxy-CTP (3000 Ci/mmol). Unlabeled oligonucleotides were used as competitors in some experiments. Recombinant proteins (100–200 ng) were incubated either in the presence or absence of competitor for 15 min at 15 C in binding buffer (12 mM HEPES, pH 7.9; 12% glycerol; 60 mM KCl; 1 mM EDTA; 1 mM dithiothreitol; and 4 mM Tris-HCl, pH 8.0), 2 µg of polydeoxyinosinic-deoxycytidylic acid, and 1 µg of nonfat dry milk proteins. After incubation, 50,000–100,000 cpm of the radiolabeled probe was added and the mixture incubated for 15 min at 15 C. Where indicated, ER, ERß (PanVera), or SREBP-1 (Santa Cruz Biotechnology, Inc.) specific antibody was also added to the reaction for supershift analysis. The protein/DNA complexes were resolved on a 4% nondenaturing acrylamide gel in 0.5x TGE (0.25 M Tris, 1.9 M glycine, 10 mM EDTA final). Gels were vacuum-dried and developed using a Cyclone Storage Phosphor Screen (Packard Bioscience, Meriden, CT) or Kodak BioMax MR films (Fisher Scientific). Densitometric quantitation of the autoradiograms was performed using the Ultra-Violet Products imaging system and Labworks software (Ultra-Violet Products, Upland, CA).

Mammalian two-hybrid assay
The plasmids pBIND, pACT and pG5-luciferase used in these assays were provided in the Mammalian Two-hybrid System (Promega Corp.). The pBIND refers to a vector containing the yeast GAL4 DNA binding domain upstream of a multiple cloning region and the pACT refers to a vector containing the herpes simplex virus VP16 activation domain upstream of a multiple cloning region. The pG5-luciferase refers to a vector containing five GAL4 binding sites upstream of a minimal TATA box in front of the firefly luciferase gene. The plasmid SREBP-1a-pBIND was constructed using full-length human SREBP-1a cDNA inserted into the EcoRI and KpnI sites. The ER-pACT and ERß-pACT plasmids containing full-length ER and ERß, respectively, were kindly provided by Dr. Katarina Pettersson (Department of Medical Nutrition, Karolinska Institute, Sweden). For this, human bladder carcinoma HTB-9 cells plated in six-well plates (at a density of 3 x 10⁶ cells per well) were cotransfected with the pG5-luciferase vector either in the presence or absence of SREBP-1a-pBIND ± ER-pACT (or ERß-pACT). For these experiments, transfections were performed with 2 µg of each plasmid using the Fugene 6 transfection reagent according to the manufacturer’s instructions. After transfection, cells were incubated for 48 h at 37 C before harvesting. Luciferase assays and statistical analysis were performed as described in Cell transfections and luciferase assays. Transfection efficiencies were corrected using the renilla activity expressed by the pBIND plasmid.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Because estrogen is known to influence gene transcription through activation of the ERs (20), the differential expression of these receptors within distinct cell lines was examined. For this, nuclear extracts were prepared from human breast carcinoma MCF-7 and rat luteal GG-CL cells, and Western blot analysis using receptor subtype ( and ß) specific antibodies was performed as described in Materials and Methods. MCF-7 cells, the first candidate for these studies, were selected because these cells have been extensively used for investigating the effects of estrogen/ER on gene expression (27, 29, 40, 41). Although they have not been as widely used as the MCF-7 cells, the GG-CL cells were also chosen for these studies because they have been previously shown to express functional ERß (42, 43). Figure 1A demonstrates that ER (66.4 kDa) was mainly expressed in MCF-7 cells, whereas ERß (53.4 kDa) was mainly expressed in GG-CL cells. These results correlate with the previous reports showing the presence of ER and ERß in MCF-7 (40, 41) and GG-CL (42, 43) cells, respectively.

View larger version (48K): [in this window] [in a new window]   Figure 1. Effects of E2 on the HDL-R SR-BI promoter activity in human breast carcinoma MCF-7 (B) and rat luteal GG-CL (C) cells. A, Expression of ER and ß proteins in MCF-7 and GG-CL cells. Nuclear extracts were prepared from each cell type and Western blot analysis was performed as described in Materials and Methods using ER subtype ( and ß) specific antibodies. Typical Western blots are shown. B and C, Effects of adding increasing amounts of E2 on the HDL-R SR-BI promoter activity. MCF-7 (B) and GG-CL (C) cells were transfected with the p-2264 HDL-R SR-BI promoter construct as described in Materials and Methods. E2 (0.001–100.0 nM) was added 24 h before lysing the cells. 4-OHT (100.0 nM) was used as a negative control in these experiments and was also added 24 h before lysing the cells. The data are represented as mean fold-induction ± SEM, where the value of luciferase activity for the promoter construct without treatment was set to 1.0. Data are from a typical experiment performed in triplicate. These experiments were repeated two times. *, P < 0.01; **, P < 0.002; and ***, P < 0.001, were obtained by comparing to the HDL-R SR-BI promoter activity in the absence of any treatment.

To determine whether estrogen regulates the HDL-R SR-BI gene, transfection studies followed by luciferase assays were performed. For this, the p-2267 HDL-R SR-BI promoter construct, which contained all four putative ERE motifs, was transfected into MCF-7 (Fig. 1B) and GG-CL (Fig. 1C) cells. Some plates received increasing amounts of E2 (0.001 to 100 nM) 24 h before lysing the cells. The data were represented as fold-induction where the value of luciferase activity for the construct with no treatment was set to 1.0 (Fig. 1, B and C). Using these conditions, it was shown that E2 activated, in a dose-dependent manner, the p-2267 HDL-R SR-BI promoter in both MCF-7 and GG-CL cells (Fig. 1, B and C). Higher concentrations of E2 (0.1–2.5 µM), however, did not further induce the HDL-R SR-BI promoter-driven luciferase activity (data not shown). Maximum activation levels (7.5-fold in MCF-7 and 6.8-fold in GG-CL) of the HDL-R SR-BI promoter were obtained at the E2 dose of 100 nM (Fig. 1, B and C). Even though the GG-CL cells appeared to express more ERß than MCF-7 cells express ER (see Fig. 1A), the levels of luciferase activity produced from the HDL-R SR-BI promoter construct were comparable in both cell lines (Fig. 1, B and C). These effects appear to be specific for E2 because the addition of 100 nM of the antiestrogen 4-OHT was unable to increase the activity of the HDL-R SR-BI promoter (Fig. 1, B and C). To determine whether 4-OHT was able to block E2’s effects on the HDL-R SR-BI promoter, 1 nM of E2 was added to the cells in the presence or absence of increasing amounts of 4-OHT (1–100 nM). As shown, 1 nM of 4-OHT was able to prevent E2’s enhancement of the HDL-R SR-BI in both cell lines (Fig. 2, A and B).

View larger version (17K): [in this window] [in a new window]   Figure 2. Effects of 4-OHT on E2-dependent activation of the HDL-R SR-BI promoter in MCF-7 (A) and GG-CL (B) cells. Cells were transfected with the p-2264 HDL-R SR-BI promoter construct as described in Materials and Methods. E2 (1 nM) and 4-OHT (1–100.0 nM) were added 24 h before lysing the cells. The data are represented as mean fold-induction ± SEM, where the value of luciferase activity for the promoter construct without treatment was set to 1.0. Data are from a typical experiment performed in triplicate. *, P < 0.001, was obtained by comparing to the HDL-R SR-BI promoter activity in the presence of E2 but not 4-OHT.

View larger version (87K): [in this window] [in a new window]   Figure 3. EMSA depicting ER subtype ( or ß) binding to the ERE motifs found in the HDL-R SR-BI promoter. 32P- labeled double-stranded oligonucleotide probes (100,000 cpm per lane) containing each ERE motif found in the HDL-R SR-BI promoter were incubated with recombinant ER (A) or ß (B) proteins (100 ng) in the presence or absence of unlabeled oligonucleotide (Cold; 125x) or ER (A) or ERß (B) specific Ab (5 µg of IgG). Control refers to an oligonucleotide containing a consensus ERE (see Materials and Methods). The DNA-protein complexes (indicated by arrows) were resolved on a 4% nondenaturing acrylamide gel at 4 C in 0.5x Tris-glycine-EDTA buffer. Typical mobility shift assay autoradiograms are shown. This experiment was repeated two times.

View larger version (49K): [in this window] [in a new window]   Figure 4. Competition analysis of ER subtype ( or ß) binding to EREs found in the HDL-R SR-BI promoter. 32P-Labeled double-stranded oligonucleotide probes (100,000 cpm per lane) containing ERE-4 (A) or ERE-1 (B) were incubated with recombinant ER (A) or ERß (B) (100 ng) protein in the presence or absence of increasing competing amounts (12.5–125x) of unlabeled ERE-4 (A) or the ERE-1 (B). The DNA-protein complexes (indicated by arrows) were resolved as described in Fig. 3. Representative gel mobility shift assay autoradiographs are shown. C, Calculating the amount of competitor required to achieve 50% competition. Densitometric quantitation of the autoradiograms shown in A and B was performed using the Ultra-Violet Products imaging system and Labworks software (Ultra-Violet Products Laboratory Products). The percent remaining was calculated by comparing the density of the major protein/DNA band seen with ER subtype ( or ß) alone with the band formed in the presence of ER ( or ß) and competitor. The percent remaining was then plotted against the picomoles of competitor used in each lane. The amount of competitor required to achieve 50% competition was then obtained from the graph and the results shown in D. These experiments were repeated twice for each ERE motif.

View this table: [in this window] [in a new window]   Table 1. Mutation analysis of the HDL-R SR-BI ERE motifs

View larger version (33K): [in this window] [in a new window]   Figure 5. Mutation analysis of the ERE motifs in the HDL-R SR-BI promoter. Site-directed mutagenesis and transfections of the mutated constructs into the MCF-7 (A) and GG-CL (B) cells were carried out as described in Materials and Methods. E2 (1 nM) was added 24 h before lysing the cells. The data are represented as mean fold-induction ± SEM, where the value of luciferase activity for the wild-type (WT) HDL-R SR-BI promoter construct without treatment was set to 1.0. Numbers after the "M" (mutation) refer to the specific ERE motif (1, 2, 3, or 4) that has been mutated in that promoter construct. Data are from a typical experiment performed in triplicate. *, P < 0.001, was obtained by comparing to the activity of the WT HDL-R SR-BI promoter treated with E2.

View larger version (47K): [in this window] [in a new window]   Figure 6. Effects of both ERs ( and ß) on the HDL-R SR-BI promoter. A, Mobility shift assay depicting ER and ß binding the ERE-1 motif in the HDL-R SR-BI promoter. Binding reactions were performed using recombinant ER and ß proteins (100 ng each) in the presence or absence of unlabeled oligonucleotide (Cold; 125x) or ER subtype ( or ß) specific Ab (Ab and ßAb, respectively; 5 µg of IgG). A representative mobility shift assay autoradiograph is shown. This experiment was repeated three times. The DNA-protein and supershifted complexes are indicated. B, Effects of ER and ERß on the HDL-R SR-BI promoter activity in MCF-7 cells. Cells were transfected with the p-2267 HDL-R SR-BI promoter construct in the presence or absence of ER-pCMV5 ± ERß-pCMV5 expression plasmids as described in Materials and Methods. E2 (1 nM) was added 24 h before lysing the cells. The data are represented as mean fold-induction ± SEM, where the value of luciferase activity for the promoter construct without treatment was set to 1.0. Data are from a typical experiment performed in triplicate. These experiments were repeated four times. *, P < 0.001, was obtained by comparing to the HDL-R SR-BI promoter activity in the absence of any exogenous factor or treatment.

View larger version (34K): [in this window] [in a new window]   Figure 7. Effects of SREBP-1a on E2-dependent HDL-R SR-BI promoter activity in MCF-7 (A) and GG-CL (B) cells. Cells were transfected with the p-2267 HDL-R SR-BI promoter construct in the presence or absence of SREBP-1a-pCMV5 expression plasmid. E2 (1 nM) was added 24 h before lysing the cells. The data are represented as mean fold-induction ± SEM, where the value of luciferase activity for the HDL-R SR-BI promoter construct without treatment was set to 1.0. The data are from a typical experiment performed in triplicate. These experiments were repeated four times in the MCF-7 cells and two times in the GG-CL cells. *, P < 0.001, and **, P < 0.01, were obtained by comparing to the HDL-R SR-BI promoter activity in the absence of any factor.

View larger version (47K): [in this window] [in a new window]   Figure 8. Binding of ER, ERß, and SREBP-1a to the ERE-1 in the HDL-R SR-BI promoter. A, Binding of SREBP-1a to ERE-1 in the presence of the ERs ( and ß). A 32P-labeled double-stranded oligonucleotide probe (100,000 cpm per lane) corresponding to ERE-1 was incubated with recombinant ER and ERß (100 ng each) in the presence or absence of SREBP-1a protein (200 ng). Unlabeled ERE-1 was added to one lane as a competitor (Cold; 125x). ER subtype ( or ß) (Ab and ßAb, respectively) and SREBP-1 (1Ab) specific antibodies (5 µg of IgG each) were also added to some of the binding reactions. B, Binding of SREBP-1a to the ERE-1 in the absence of the ERs ( and ß). Binding reactions were performed with SREBP-1a protein in the presence or absence of unlabeled ERE-1 motif (Cold) or SREBP-1 (1Ab) specific antibody. Binding to a probe containing the dSRE of the HDL-R SR-BI was used as a positive control. Representative gel mobility shift assay autoradiographs are shown in both A and B. These experiments were repeated four times. The DNA protein and supershifted complexes are indicated.

View larger version (47K): [in this window] [in a new window]   Figure 9. Binding of SREBP-1a, ER and ERß to the dSRE in the rat HDL-R SR-BI promoter. A, Binding of ER and ERß to the dSRE in the absence of SREBP-1a. A 32P-labeled double-stranded oligonucleotide probe (100,000 cpm per lane) corresponding to the dSRE was incubated with recombinant ER or ERß (100 ng) in the presence or absence of unlabeled dSRE (Cold; 125x) or ER subtype ( or ß) (Ab and ßAb, respectively; 5 µg of IgG) specific antibodies. Binding to ERE-1 was used as positive control. B, Binding of ER and ERß to dSRE in the presence of SREBP-1a. Binding reactions were performed with SREBP-1a protein (200 ng) in the presence or absence of recombinant ER ± ERß (100 ng each). Unlabeled dSRE was added to some of the lanes as a competitor (Cold; 125x). ER subtype ( or ß) (Ab and ßAb, respectively) and SREBP-1 (1Ab) specific antibodies (5 µg of IgG) were also added to some of the binding reactions. Representative gel mobility shift assay autoradiographs are shown in both (A) and (B). These experiments were repeated two times. The DNA-protein and supershifted complexes are indicated.

View larger version (16K): [in this window] [in a new window]   Figure 10. Interaction between the ERs ( and ß) and SREBP-1a using the mammalian two-hybrid assay. Human bladder carcinoma HTB-9 cells (plated in six-well plates) were transfected with the pG5-luciferase vector in the presence of the indicated vectors (2 µg of each plasmid DNA per well). ER and ß were cloned into the pACT vector, whereas SREBP-1a was cloned into the pBIND vector. The data are represented as mean fold induction ± SEM, where the value of luciferase activity for the pG5-basic plasmid transfected in the presence of the pACT/pBIND negative controls was set to 1.0. The data are from a typical experiment performed in triplicate. These experiments were repeated two times. *, P < 0.001, was obtained by comparing to the pACT/pBIND negative control.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Another important finding of the current report is that similar to other genes regulated by estrogen (24, 25), the ERs were able to bind as homo- or heterodimers to the HDL-R SR-BI EREs as shown by the supershift analysis although the formation of homodimers was preferred over heterodimers. However, the possibility that the preferential formation of homodimers over heterodimers resulted from the use of recombinant protein in the experiment cannot be eliminated. In addition, it was demonstrated that the presence of both ERs in the transfection studies enhanced the overall luciferase activity produced in the presence of only one ER subtype. It is well known that in humans and rodents, both ER and ERß are found in most cell types within steroidogenic tissues (46, 47, 48, 49). Reports have shown that ER and ß are both expressed in ovarian follicles (56) and the corpus luteum of pregnancy (43). Thus, the presence of the ERs in these tissues simultaneously may contribute to maximize the response of HDL-R SR-BI gene to estrogen treatment. In the liver, however, although both ERs could be found, only ER is expressed in hepatocytes (46, 47), the main cell type within this tissue that mediates selective uptake of HDL-cholesterol esters (45, 57). Herein, it was shown that ER could also mediate activation of the HDL-R SR-BI gene by estrogen. These data would appear to conflict with previous findings using a rat model or human liver cells in culture that demonstrated that estrogen treatment dramatically decreased hepatic (15) and HepG2 cell (58) HDL-R SR-BI expression and HDL-cholesterol uptake (59). The human cell culture experiments (58) demonstrated that treating HepG2 cells with 10 pM of E2 for 12 h eliminated the expression of the HDL-R SR-BI gene and reduced selective-cholesterol ester uptake (58), confirming the rat liver studies (15). Interestingly, that same study (58) showed that the scavenger receptor class B type II (SR-BII), an alternatively spliced product of the HDL-R SR-BI, significantly increased after E2 treatment in both HepG2 cells and rat liver. Normally, SR-BII represents approximately 12% of the total immunodetectable SR-BI/SR-BII present in the liver (60). SR-BI and SR-BII bind HDL with similar affinity but SR-BII is 4-fold less efficient at selective HDL-uptake compared with SR-BI (60), which explains the decrease in overall cholesterol ester uptake (58) following E2 administration. It appears that E2 reduces hepatic HDL-R SR-BI gene expression by altering the posttranscriptional processing of the HDL-R SR-BI RNA. However, the finding that E2-treatment was able to increase SR-BII protein levels 3.2-fold more than the starting level of SR-BI (58) suggests that transcription of the HDL-R SR-BI gene may also increase in response to estrogen to compensate for the increase in the alternative form. Because the liver expresses only ER (46, 47), the increase in transcription may not be as high as in steroidogenic tissues that express both ERs but is adequate to enhance the increase in SR-BII. This would correlate with the findings from the current report, but additional study will be required to confirm transcriptional control of the shift in receptor subtype expression.

SREBP-1a was shown in a previous report to activate the HDL-R SR-BI gene (37). The experiments reported here now identify an additional function for SREBP-1a as a coactivator of the ERs for the HDL-R SR-BI gene. These studies demonstrate that SREBP-1a enhances both binding and activation of the HDL-R SR-BI promoter by E2/ERs. The mechanism for this regulatory process involves direct protein-protein interaction between SREBP-1a and the ERs as demonstrated using a mammalian two-hybrid assay. Interaction of the ERs with other factors have been previously demonstrated by GST-pull-down and two-hybrid assays (26, 29, 31, 61, 62, 63). These factors include SF-1 (26), c-Jun (31, 62), Sp1 (29, 62), and dosage-sensitive sex adrenal hypoplasia congenital critical region on the X chromosome gene-1 (63). However, this is the first study to demonstrate interaction between ERs and SREBP-1a. It was also demonstrated that SREBP-1a was able to bind directly to the ERE containing oligonucleotides in the absence of ERs with the same intensity, as it would bind to an SRE. Binding of SREBP-1a to non-SRE sequences has been previously demonstrated (64). SREBP-1 has been shown to bind E-box motifs (CANNTG) (64). None of the ERE oligonucleotides used in these studies contain E-box or SRE motifs. Thus, the possibility that SREBP-1a could be binding directly to the ERE core sequence cannot be discarded. The ERs also bound to the SRE but with lower intensity than it would bind to an ERE. Recently, Li et al. (28) demonstrated that mutation at the SRE-1 motif found in the low density lipoprotein receptor (LDL-R) promoter diminished ER activation of this gene. This suggests that SREBP-1a may be involved in the ER-dependent regulation of the LDL-R (28), or that ER may bind to the SRE-1 and that the mutation prevents the binding. In contrast to the HDL-R SR-BI, the LDL-R promoter does not contain a functional ERE site and consequently, estrogen regulation of this gene occurs through nonclassical pathways via the Sp1 binding sites that surround the SRE-1 (28). Thus, interaction of the ER with nonconsensus motifs such as an SRE could be crucial in this case. Whether the binding of the ERs to the SRE and/or the binding of SREBP-1a to an ERE occurs in vivo will required additional analysis.

Because SREBP-1a is not a tissue-specific factor (65), the question remains whether this transcription factor could, under certain conditions, enhance estrogen’s effects on the HDL-R SR-BI gene. Steroidogenic tissues use HDL-cholesterol mainly as substrate for steroid hormone production (16, 66, 67). E2 has been shown to enhance steroid hormone production in tissues such as the adrenal and the ovary (15, 68, 69). Thus, to satisfy the high demand for substrate after E2 treatment, cholesterol synthesis and transport as well as lipoprotein receptor content of these tissues, including the HDL-R SR-BI, are dramatically enhanced (15, 68, 69). Due to the increase in steroidogenesis, intracellular cholesterol levels may decrease rapidly resulting in the activation of SREBP-1a. In fact, depletion in ovarian cholesteryl ester content and increased cell-surface HDL binding in response to estrogen stimulation has been previously reported (69) suggesting that synergistic ER-SREBP-1a activation of the HDL-R SR-BI may occur in the ovary. While SREBP-1a is also expressed in the liver (65), it is well known that estrogen treatment enhances the expression of hepatic LDL-R (28, 70, 71, 72) and 3- hydroxy-3-methylglutaryl-coenzyme A reductase (71, 72) genes. Thus, in this tissue cholesterol accumulation may reduce SREBP-1a and consequently, the interaction between ER and SREBP-1a is prevented.

In summary, E2 through its nuclear receptors (ER and ERß) positively activated the HDL-R SR-BI promoter through the classical pathway by the binding and activation of three EREs. This activation of the HDL-R SR-BI by E2 also involved the formation of heterodimers between the two ER subtypes and interaction with the coactivator SREBP-1a. Although the current data suggest that SREBP modulation of the ER-dependent transactivation could occur via DNA- dependent and/or -independent (coactivation) pathways, further analysis will be required to determine the exact mechanism for this modulation. The results presented here are the first to report on a possible mechanism through which estrogen via its nuclear receptors activates the HDL-R SR-BI promoter in steroidogenic tissues.

	Acknowledgments

 
We thank Drs. Geula Gibori, Jan-Åke Gustafsson, Katarina Pettersson, and Tim Osborne for gifts of materials.

	Footnotes

 
This work was supported by grants from the American Heart Association Florida Affiliate Grant-in-Aid 0150973B and the National Institute of Health R01-HD-35163 (to M.P.M.). D.L. was supported by a National AHA Scientist Development Grant 0030172N.

¹ These individuals contributed equally to the research presented in this article.

Abbreviations: Ab, Antibody; CMV, cytomegalovirus; dSRE, distal SRE; ERE, estrogen response element; ERE, ERE half-site motifs; HDL, high-density lipoprotein; HDL-R, HDL receptor; 4-OHT, 4-hydroxytamoxifen; SF-1, steroidogenic factor-1; Sp1, specific factor-1; SRE, sterol response element; SR-BI, scavenger receptor class B type I; SREBP-1a, sterol regulatory element binding protein-1a.

Received November 2, 2001.

Accepted for publication February 22, 2002.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Fuster V 1999 Epidemic of cardiovascular disease and stroke: the three main challenges. Circulation 99:1132–1137[Free Full Text]
2. 1997 Heart and Stroke Statistical Update. American Heart Association, Dallas, TX
3. Heiss G, Tamir I, Davis CE, Tyroler HA, Rifkand BM, Schonfeld G, Jacobs D, Frantz Jr ID 1980 Lipoprotein-cholesterol distributions in selected North American populations: the lipid research clinics program prevalence study. Circulation 61:302–315[Free Full Text]
4. 1995 National Center for Health Statistics: women’s deaths from heart disease in 1994. NCHS, Vital Statistics Report, Hyattsville, MD
5. Kuller LH, Gutal JP, Meilahn EN, Matthews KA, Plantinga P 1990 Relationship of endogenous sex steroid hormones to lipids and apoproteins in postmenopausal women. Arteriosclerosis 10:1058–1066[Abstract/Free Full Text]
6. Henderson BE, Paganini-Hill A, Ross RK 1991 Decreased mortality in users of estrogen replacement therapy. Arch Intern Med 151:75–78[Abstract]
7. Stampfer MJ, Colditz GA, Willett BSWC, Manson JE, Rosner B, Speizer FE, Hennekens CH 1991 Postmenopausal estrogen therapy and cardiovascular disease. N Engl J Med 325:756–762[Abstract]
8. Kushwaha RS, Lewis DS, Carey KD, McGill Jr HC 1991 Effects of estrogen and progesterone on plasma lipoproteins and experimental atherosclerosis in the baboon (Papio sp.). Arterioscler Thromb 11:23–31[Abstract/Free Full Text]
9. Adams MR, Kaplan JR, Manuck SB, Koritnik DR, Parks JS, Wolfe MS, Clarkson TB 1990 Inhib