help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lethimonier, C.
Right arrow Articles by Ducouret, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lethimonier, C.
Right arrow Articles by Ducouret, B.
Endocrinology Vol. 143, No. 8 2961-2974
Copyright © 2002 by The Endocrine Society

ARTICLE

The Glucocorticoid Receptor Represses the Positive Autoregulation of the Trout Estrogen Receptor Gene by Preventing the Enhancer Effect of a C/EBPß-Like Protein

Christèle Lethimonier, Gilles Flouriot, Olivier Kah and Bernadette Ducouret

Equipe d’Endocrinologie Moléculaire de la Reproduction, Unité Mixte de Recherche Centre National de la Recherche Scientifique 6026, Université de Rennes 1, 35042 Rennes Cedex, France

Address all correspondence and requests for reprints to: B. Ducouret, Equipe d’Endocrinologie Moléculaire de la Reproduction, Unité Mixte de Recherche Centre National de la Recherche Scientifique 6026, Campus de Beaulieu, Bat 13, Université de Rennes 1, 35042 Rennes Cedex, France. E-mail: . bernadette.ducouret{at}univ-rennes1.fr


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Stress and cortisol are known to have negative effects on vitellogenesis in oviparous species. This provides a physiological context in which to explore in more detail the molecular mechanisms involved in transcriptional interferences between two steroids receptors, the estradiol receptor (ER) and the glucocorticoid receptor (GR). We have previously shown that the cortisol inhibitory effect on rainbow trout (rt) vitellogenesis is the result of a repression of the estradiol-induced ER-positive autoregulation by activated GR. In the present study, we demonstrate that the GR repression involves a proximal region of the rtER promoter that is unable to bind GR. This inhibition is counteracted in part by the orphan receptor COUP-TF1 that has been previously shown to cooperate with ERs on the same promoter. A detailed analysis allowed us to identify a C/EBPß-like protein that is implicated in both the maximal stimulatory effect of estradiol and the GR repression. Indeed, GR, through its DNA-binding domain, suppresses the binding of C/EBPß on the rtER promoter by protein-protein interactions and thereby prevents the enhancer effect of this transcription factor.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
ESTRADIOL (E2) PLAYS a key role in the control of reproductive functions by directing the establishment and maintenance of sex differentiation patterns and controlling the female reproductive cycle (1, 2). The physiological changes induced by E2 often result from modifications in the expression patterns of specific target genes and are mediated by intracellular receptors, the estrogen receptors (ERs)-{alpha} and -ß (3, 4). These two receptors belong to the superfamily of steroid/thyroid hormone/retinoic acid receptors (RARs) whose members act as ligand-inducible transcriptional factors (5). They regulate gene expression by interacting either in a protein/DNA manner with cognate DNA sequences called estrogen-responsive elements (EREs) (GGTCAnnnTGACC) (6) or in a protein/protein manner with other transcriptional factors such as stimulating protein 1, steroidogenic factor-1, activator protein-1(AP1), CCAAT/enhancer-binding protein (C/EBP), or nuclear factor {kappa}B (NF-{kappa}B) (7, 8, 9, 10, 11, 12). Structure/function analysis of these receptors reveals that they contain a DNA-binding domain (DBD), a ligand-binding domain (LBD), and two transactivation function (AF) domains, AF1 and AF2, located in the N- and C-terminal regions of ERs, respectively. These AF domains allow the receptor to promote gene transcription by interacting with components of the basal transcriptional machinery (13) or with coactivators, in a ligand-dependent manner (14). ER coactivators have been well described and include steroid receptor coactivator-1 (15), glucocorticoid receptor-interacting protein 1 (16), transcriptional intermediary factor 1ß (17), receptor interacting protein 140 (18), and cAMP response element-binding protein/p300 (19, 20, 21).

Factors involved in ER transactivation functions may also interact with other members of the steroid hormone receptor family (19, 22, 23, 24, 25, 26). Therefore, transcription interferences between ERs and other steroid hormone receptors, resulting from competition for a functionally limiting transcription factor or coactivator, are phenomena that may occur in cello. Such interferences between ER{alpha} and glucocorticoid receptors (GRs) have been demonstrated in vitro by transfection experiments (27, 28, 29). However, although ERs and GRs are coexpressed in several tissues (30, 31), there are no data reporting transcriptional interferences between these two receptors in vivo. Only antagonistic effects between estrogen and glucocorticoids on breast cell line, uterine growth, or osteoblast development have been described (32, 33, 34).

A physiological situation, recently reported in fish, gave us the opportunity to explore in more detail the molecular mechanisms involved in a transcriptional interference between ER{alpha} and GRs. In oviparous species, the synthesis of vitellogenin (Vg) takes place in the liver according to a strictly estrogen-dependent mechanism that first involves an up-regulation of ER{alpha} by its own ligand. Reports from the literature indicate that, in trout, stress or cortisol causes a reduction of cytosolic E2-binding sites in the liver and a decrease in plasma Vg levels (35, 36). We recently demonstrated that this physiological effect of glucocorticoids results from a transcriptional inhibition by GRs of the trout E2-induced (rt)ER{alpha} gene (37). The up-regulation of the rtER{alpha} gene by E2 is mediated by an ERE (38) and by a perfect ERE half site located in the proximal promoter of the gene (39) and can be enhanced by a synergistic effect between rtER{alpha} and the orphan receptor COUP-TF (40, 41). These ERE (FP1) and half ERE (FP2) sequences are protected by trout liver nuclear extracts in foot-printing experiments. A third region (FP3: -132/-104) was also protected (39), but its function remained unknown. In this study, we demonstrate that the rtER gene transcriptional inhibition by GRs involves both the ERE and FP3 regions. By performing a detailed analysis of the FP3-dependent repression, we show that the GR prevents the binding of FP3-specific transcription factors identified as C/EBPß-like proteins.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Chemicals and materials
Culture media, hormones, and antibiotics were obtained from Sigma (St. Louis, MO). Reduced glutathione, glutathione-agarose beads, and protease inhibitors were also from Sigma. Fetal calf serum and TRIzol reagent were obtained from Life Technologies, Inc.-BRL (Eragny, France). Ultroser-SF was obtained from Biosepra (Villeneuve La Garenne, France). Hepatocytes were aggregated on a Novotron gyratory shaker (INFORS AG, Massy, France). Klenow fragment of Escherichia coli DNA polymerase I, T4 DNA ligase, and collagenase A obtained from Clostridium histolyticum were obtained from Boehringer (Mannheim, Germany). T4 polynucleotide kinase, TNT-coupled reticulocyte lysate, and pGL2 basic vector was obtained from Promega Corp. (Madison, WI). {alpha}-32P dCTP, {gamma}-32P dATP, and 35S-methionine were obtained from ICN Pharmaceuticals, Inc. (Irvine, CA). The T7 DNA polymerase sequencing kit was obtained from USB Corp. (Cleveland, OH). The luciferase assay system was obtained from Promega Corp., and luciferase activity was measured on a luminometer (Dynatech Corp., Guynancourt, France). Dynabeads M-280 were obtained from DynAl (Oslo, Norway). The pcDNA-3 vector was obtained from Invitrogen PV/Novex (Groningen, The Netherlands) and the pGEX-3X vector from Amersham Pharmacia Biotech (Orsay, France). Nylon Biodine A membrane was from Pall (St. Germain en Laye, France). The rabbit polyclonal anti-C/EBP{alpha} (sc-61) and the rabbit polyclonal anti-C/EBPß (sc-150) were obtained from TEBU (Le Perray en Yvelines, France). The anti-Oct1 antiserum was a gift from Dr. W. Herr (Cold Spring Harbor Laboratory). Oligonucleotides were synthesized by GENSET SA (Paris, France), which were used as double-stranded DNA for the EMSA and as single-stranded DNA for the site-directed mutagenesis experiment (SDM) and for fusion protein construction by PCR. Only noncoding strands are listed: FP3: 5'-CCACTGGGAGTTGCATAATGTTGGCATCAGAAGGGT-3' (EMSA); GRE1: 5'-GGTACTGTACATTCTGTTCTAC-3' (EMSA); Oct 5': 5'-CGAGCTTCACCTAGTTGCATGGGCGATTG-3' (EMSA); Oct 3': 5'-CGAGCTTCACCTAATGTTGGCATGCGA-3' (EMSA); Oct 1'a: 5'-CGAGCTTCACCTAGTTGCATAGCGATTG-3' (EMSA); Oct 1'aa: 5'-CGAGCTTCACCTAGTTGCATAAGCGATTG-3' (EMSA); Oct: 5'-CGAGCTTCACCTATTTGCATAAGCGATTG-3' (EMSA); CEBP/FP3: 5'-CGAGC TTCACCTGTTGCATAAGCGATTG-3' (EMSA); -gCEBP/FP3: 5'-CGAGCTTCACCTTTGCATAAGCGATTG-3' (EMSA); CEBP cons: 5'-CGAGCTTCACCTATTGCATAATGTGCGATTG-3' (EMSA); CEBP comp: 5'-CCACTGGGACGAGCTTCACCTTGGCATCAGAAGGGT -3' (EMSA); FP3m: 5'(P)-TCCCCACTGGGAGGTGAATATTGTTGGCATCAG-3' (SDM); DR24: 5'-AGGGTGTTGACCTCACATACTGTTTGCTGTGTCATGTTGACCTG-3' (EMSA); rtGRLBD down: 5'-CGCGGATCCGGTCCATGCCCCAGCTGG-3' (PCR); rtGRLBD up: 5'-CGCGAATTCTTAAGGCATTGTGTATGGT-3' (PCR); rtGRA/B down: 5'-CGCGGATCCTAGATCCAGGTGGACTGAAA-3' (PCR); rtGRA/B up: 5'-CGCGAATTCTTATGGGTTGGACCGCTGGG-3' (PCR).

Double-stranded oligonucleotides used for EMSA were labeled either with {alpha}32P dCTP by Klenow fill-in or with {gamma}-32P dATP by polynucleotide kinase. After labeling, probes were purified on Sephadex G50 columns. For mutagenesis experiments, single-stranded oligonucleotides were phosphorylated with ATP and polynucleotide kinase.

Animals
Animals were treated in agreement with the European Union regulation concerning the protection of experimental animals. Investigations were conducted in accordance with the Guiding Principles for the Care and Use of Research Animals promulgated by the Society for the Study of Reproduction. Male rainbow trout (rt; 1 kg) (Oncorhynchus mykiss) were supplied by the INRA fish farm (Le Drennec, France) and kept in the laboratory in a recirculating water system at 12–15 C under an artificial light regimen mimicking the natural photoperiod (46° North). The animals were fed a trout diet ad libitum.

Hepatocyte isolation and culture
The trout liver was dissociated by collagenase A perfusion, as described by Seglen (42) and adapted to trout (43). The cell suspension was filtered through a 70-mesh sieve and the pellet was collected by centrifugation (50 g for 5 min) at 18 C. The nonparenchymal and damaged cells were removed by centrifugation. The hepatocyte pellet was resuspended in a serum-free medium/DMEM/Ham’s F-12 nutrient mixture (1:1 mixture, with L-glutamine and 15 mM HEPES, without phenol red), supplemented with 15 mM N-Tris [hydroxymethyl] methyl-2 amino-ethanesulfonic acid (TES), 12 mM NaHCO3, 1% (vol/vol) antibiotics (penicillin, streptomycin, and amphotericin B) and 2% (vol/vol) ultroser SF. The cells were plated in 60-mm untreated plastic Petri dishes (Falcon; 1–2 x 107 cells/5 ml medium per dish). Aggregates were obtained by constant gyratory shaking at 55 rpm at 18 C. The culture medium was changed every 2 d. Steroid treatments were performed 8 d after plating: E2 (100 nM) dexamethasone (Dex) (1000 nM) were dissolved in ethanol and added to the culture medium (1/1000 vol/vol). After hormonal treatment, cells were harvested, pelleted by centrifugation (at 50 g for 5 min), and stored at -70 C until use.

Dot blot hybridization
Total RNA was prepared from cultured hepatocytes using TRIzol reagent. Total RNA samples (5 µg) were spotted onto a nylon Biodyne A membrane, using a dot blot apparatus (Bio-Rad Laboratories, Inc., Hercules, CA), as described by Cheley and Anderson (44). The membrane was prehybridized [50% formamide, 5x saline sodium citrate (SSC), 5x Denhart’s solution, 5 mM NaH2PO4 (pH 6.5), 0.1 mg/ml Thymus calf DNA, and 0.1% SDS] at 42 C for 6 h and hybridized under stringent conditions [50% formamide, 5x SSC, 1x Denhart’s solution, 20 mM NaH2PO4 (pH 6.5), 0.05 mg/ml calf thymus DNA, 0.1% SDS, and 2 x 106 cpm/ml radiolabeled probe]. Rainbow trout (rt) Vg (rtVg) and actin cDNA were radiolabeled by random priming as previously described (45). The E2 receptor (rtER) single-strand probe was labeled using Dynabeads M-280. After 16 h of hybridization, blots were washed four times with 2x SSC, 0.1% SDS for 5 min at room temperature, and then three times for 15 min at 50 C with 0.1x SSC and 0.1% SDS. The washed blots were autoradiographed at -70 C. Radioactivity was quantified using an instant imager (Packard, Wellesley, MA).

Transient transfections
Chinese hamster ovary (CHO)-K1 cells were grown in 24-well plates at 37 C in phenol red-free DMEM-F12 containing 10% fetal calf serum. One hour before transfection, the medium was replaced with 10% charcoal-treated serum medium. Cells were transfected by using the calcium phosphate DNA precipitation method (46) with 875 ng of a reporter vector. In each case, both expression vectors cytomegalovirus (CMV)rtER (50 ng) and CMVrtGR (50 ng) were transfected together with or without CMV-human chicken ovalbumin upstream promoter transcription factor 1 (hCOUP-TF1) (25 ng). Each dish received 1 µg DNA, using Bluescript vector for complementation. After 18 h of incubation at 37 C with 2% CO2, cells were washed once with PBS, and 1 µM of the steroids were added. Cells were harvested 36 h later for luciferase and protein assays. Luciferase activities were normalized to total cellular proteins. The internal controls, CMV-ß-galactosidase or pCH110-ß-galactosidase, were not used in those experiments because the ß-galactosidase control activity was modified after Dex treatment.

Statistics
Statistics analysis was performed using Statview 4.02 software (SAS Inc., Cary, NC). Comparisons were made using ANOVA and Fisher test. Any P values less than 0.05 were considered significant.

Reporter vectors
"Basic" is the pGL2 basic vector with the luciferase reporter gene from Promega Corp. Deleted constructs of the rtER promoter gene were prepared by Lazennec et al. (40) in pGL2. These reporter vectors contain, respectively, 2.0 Basic (-2079 to +28), 0.2 Basic (-207 to +28), Kpn3 (-154 to +28), Kpn2 (-112 to +28), and Kpn1 (-87 to +28).

Expression vectors
CMV-rtER (containing the strong CMV promoter from pCMV5) was provided by Dr. B. Katzenellenbogen (University of Illinois, Urbana, IL). CMV-rtGR was also derived from pCMV5 (47). CMV-hCOUP-TF1 is derived from pcDNA3 and contains the human COUP-TF1.

Plasmids for in vitro translation. CMV-C/EBPß was a generous gift from Prof. A. Le Cam (Institut de Génétique Moléculaire, Institut de la Santé et de la Recherche Médicale, Montpellier, France).

SDM
SDM was performed as previously described (48) on the Kpn3 construct by using the FP3m oligonucleotide. To destroy the C/EBP-binding site, three mutations were introduced and led to a new reporter gene construct called Kpn3m. Briefly, a culture of CJ236 E. coli containing the plasmid of interest was infected by 1011 pfu/ml of phage helper VCSM13 to produce a single-stranded uracil DNA template. Single-stranded DNA (1 µg) was annealed with 10 pmol phosphorylated FP3m. The annealed primer was elongated with T4 DNA polymerase and ligated with T4 DNA ligase. The mutated double-stranded plasmid containing mutations was used to transform DH5{alpha} E. coli, and mutations were checked by sequencing.

Recombinant plasmids
Glutathione-S-transferase (GST)-rtGR-DBD (pGEX-3X) plasmid construct was already prepared by Tujague et al. (49). The constructs of GST fusion proteins with the A/B or LBD domains of rtGR were performed by subcloning of PCR fragments. The A/B domain and LBD domains were amplified by PCR on pUC19-rtGR cloned cDNA (47) as a matrix and with rtGRA/B oligonucleotides, or rtGRLBD oligonucleotides as primers, respectively. The down primers contained a BamH1 site and the up primers contained an EcoR1 site. These inserts constructed by PCR were subcloned into BamH1/EcoR1 sites of the pGEX-3X expression. For these two constructions, the reading frame at the junction with the GST peptide was checked by recombinant plasmid DNA sequencing.

Production of proteins
The coupled transcription-translation TNT reticulocyte lysate system was used according to the manufacturer’s protocol to produce C/EBPß for EMSA experiments. GST-rtGR fusion proteins were expressed and purified from E. coli. Briefly, bacteria expressing fusion proteins were grown at 37 C to a density of OD600 0.6–0.8. Protein expression was induced by adding 0.1 mM isopropyl ß-D-thiogalactopyranoside for 4 h at 37 C. Cells were centrifuged, and the bacterial pellet was resuspended in NETN buffer [0.5% Nonidet, 1 mM EDTA, 20 mM Tris (pH 8.0), 100 mM NaCl, 10 µg/ml protease inhibitors (leupeptin, pepstatin, aprotinin), 1 mM phenylmethylsulfonyl fluoride] and lysed by sonication three times for 15 sec. Aliquots of cleared bacterial lysates were loaded on SDS-PAGE to check the presence of the proteins of interest. To purify fusion proteins, cleared lysates were mixed with glutathione-agarose beads prepared in NETN buffer and incubated overnight on a rotary shaker at 4 C. After centrifugation, the bead pellets containing the bound protein were washed twice with NETN buffer. For EMSA, GST rtGR proteins were eluted from beads by adding reduced glutathione for 1 h at 4 C on a rotary shaker. After elution, fusion proteins were loaded on SDS-PAGE to determine protein concentrations.

Preparation of whole-cell extracts
CHO-K1 cells were harvested, washed in PBS, and resuspended in whole-cell extract (WCE) buffer [20 mM HEPES (pH 7.9), 0.4 M KCl, 2 mM dithiothreitol, and 20% glycerol]. Cells were frozen at -80 C and then transferred at 37 C for few seconds. Three cycles (-80 C/37 C) were performed to break cells before pelleting. The supernatant was kept at -80 C until use.

Preparation of trout liver nuclear extracts
Male rt were intraperitoneally injected for 48 h with a solution of 17ß-estradiol in ethanol at a concentration of 1.5 mg/kg. Nuclear extracts were prepared as previously described (50). Briefly, nuclei were purified on a sucrose gradient and lysed in a nuclear lysis buffer [10 mM HEPES (pH 7.6), 100 mM KCl, 0.1 mM EDTA, 3 mM MgCl2, and 10% glycerol containing 0.5 M dithiothreitol, 14 µg/ml aprotinin, 2 µM L-1-chloro-3-[4-tosylamido]-7-amino-2-heptanone, 0.1 µg/ml pepstatin and leupeptin, and 0.1 mM phenylmethylsulfonyl fluoride]. Nuclear extracts were homogenized in a Dounce homogenizer, DNA was precipitated with ammonium sulfate, and the pellet was then resuspended in the dialysis buffer (25 mM HEPES, pH 7.6; 40 mM KCl; 0.1 mM EDTA; 1 mM dithiothreitol; and 10% glycerol) and dialyzed twice for 2 h against 1000 volumes of the same buffer.

EMSAs
Either WCEs of CHO-K1 cells or nuclear extract from rt liver was used for these experiments. Six micrograms WCE or 1 µg nuclear extract were preincubated, or not, with recombinant proteins or antibodies at room temperature for 40 min. Then 1 µg poly(dIdC) was added in 20 µl binding buffer [10 mM Tris-HCl (pH 7.5), 1 mM dithiothreitol, 100 mM KCl, 0.05 mM phenylmethylsulfonyl fluoride, and 10% glycerol] either with or without competitors at room temperature for 20 min. The samples were then incubated with 32P-labeled probes (30,000 cpm) for 20 min at room temperature. Protein-DNA complexes were separated from the free probe by nondenaturing electrophoresis in 6% polyacrylamide gels in 0.5x TBE (45 mM Tris base, 45 mM boric acid, 1 mM EDTA) at 4 C. The same protocol was used for hC/EBPß produced by in vitro translation in reticulocyte lysate and for GST fusion proteins with rtGR fragments.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The inhibitory effect of Dex on vitellogenesis observed in hepatocyte cultures involves a 0.2-kb proximal region of the rtER promoter gene
The in vivo inhibitory effects of Dex on rtER and Vg expression were reproduced in vitro in hepatocyte cultures. After 8 d of culture, aggregated hepatocytes were treated with E2 and/or Dex for 24 h. Total RNAs were extracted and hybridized with rtER and rtVg cDNA probes. As expected, the E2-treated cells presented a strong increase of rtER mRNA levels (Fig. 1Go, A and B), and an induction was also seen on the rtVg mRNA levels because of the increase of rtER proteins available to stimulate rtVg gene expression. Dex treatment strongly decreased both basal and E2-stimulated levels of rtER and consequently rtVg mRNA.



View larger version (39K):
[in this window]
[in a new window]
  		Figure 1. Effect of Dex on rtER and rtVg mRNA levels in cultured hepatocytes. Hepatocytes from male trout liver were isolated by collagenase perfusion. Aggregated hepatocytes were treated 8 d after plating, with or without 100 nM E2, 1000 nM Dex, or with a combination of both steroids. After 24 h of treatment, rtER, rtVg, and actin mRNA were monitored by dot blot hybridization. A, Autoradiograph of dot blots from triplicated cultures. The dots were quantified and normalized with actin (B). Values are expressed as the mean ± SE. *, Significantly different between Dex-treated and Dex untreated cells, P < 0.01.

 
To detail the inhibitory transcriptional effect of rtGR on the rtER promoter, cotransfection experiments on hepatocyte primary cultures were tested. Unfortunately, these studies produced unsatisfactory results because of the low efficiency of the DNA transfer into fish hepatocytes (data not shown). Thus, the transcriptional activity was studied in CHO-K1 cells that have been used previously to investigate the rtER gene positive autoregulation (39, 40). CHO-K1 cells were cotransfected with different 5'-deleted fragments of the rtER promoter together with the two expression vectors CMVrtER and CMVrtGR. All these constructs derived from the 2.0 Basic construct, containing -2079 to +28 of the rtER gene with its own start site, in front of the luciferase coding sequence of the basic vector pGL2 (Fig. 2AGo). A first set of transfections were performed with either the entire promoter 2.0 Basic vector or with 200 bp of rtER upstream of the initiation start site (0.2 Basic; Fig. 2Go, A and B). The luciferase activity was stimulated by E2 in both constructs, and Dex treatment resulted in a similar repression of the E2-stimulated luciferase activity in both 2.0 Basic and 0.2 Basic rtER constructs. No significant Dex effect was detected on the basal activity of the two constructs. These results suggest that no particular sequence, located between -2079 and -207 bp of the rtER promoter, is involved in the glucocorticoid inhibitory effect.



View larger version (43K):
[in this window]
[in a new window]
  		Figure 2. Effect of rtGR on the transcriptional activity of two constructs of the rtER gene promoter in transfection experiments in the CHO-K1 cell line. Either full-length (2.0 Basic) or 5' deletion containing the fragment (-207; +28) (0.2 Basic) of the rtER promoter were inserted upstream of the luciferase gene of the pGL2 Basic (A). Each reporter vector (875 ng) was mixed with 50 ng CMV-rtGR and 50 ng CMV-rtER and cotransfected in CHO-K1 cells by the calcium phosphate precipitation method (46 ). Cells were washed with PBS 18 h after transfection and treated with steroids (1000 nM) E2 and/or Dex. Cells were harvested 36 h after steroid treatment for luciferase and protein assays. Luciferase activities were normalized to total cellular proteins and were expressed as fold induction of treated vs. untreated cells. Each value represents the mean of 12 culture dishes ± SEM.

 
GR repression involves both FP3 and FP1 promoter sequences
Three other constructs, further deleted at the 5' end (Fig. 3AGo) were tested under the same conditions. Cotransfection analyses demonstrated that the E2 stimulation was equally efficient on the two shorter constructs Kpn1 and Kpn2 (Fig. 3BGo), which contain an ERE and both the ERE and half ERE, respectively, but was much more efficient on the longer Kpn3 construct. On Kpn1 and Kpn2, the stimulation by rtER can be explained by the classical ERE-dependent response, including a coactivator effect, whereas the -154/-112 region of Kpn3 could contain a binding site for a transcription factor that would enhance E2 stimulation of ERE-dependent rtER promoter activity. When the cells were treated with Dex, the rtGR inhibition is clearly evident on the E2-stimulated activities but never on the basal activities. The Dex inhibitory effect is more efficient on the Kpn3 construct than on the Kpn2 and Kpn1 constructs.



View larger version (36K):
[in this window]
[in a new window]
  		Figure 3. Effect of rtGR on the transcriptional activity of 5'-deleted mutants of the rtER gene promoter. A, Schematic diagram of 5' promoter deletion mutants inserted upstream of the luciferase gene of the pGL2 basic vector. Each construct (875 ng) was transiently transfected in CHO-K1 cells together with the expression vectors: CMV-rtGR (50 ng), CMV-rtER (50 ng) in the absence (B) or in the presence (C) of CMVhCOUP-TF1 (25 ng). Transfected cells were treated as described in Fig. 2Go. Promoter activities were normalized to total cellular proteins and expressed as fold induction of treated vs. untreated cells. Values represent the mean of 15 culture dishes ± SEM of four independent experiments. D, The rtGR dose effect (0–200 ng) with 50 ng CMVrtER in presence or not of 25 ng CMVhCOUP-TF1. Means of three values of a representative experiment are presented as percent inhibition of cells treated with both steroids vs. E2 treatment alone. **, Significantly different from control without E2 (ethanol) (P < 0.01); ##, significantly different from the corresponding group without Dex (P < 0.01); #, significantly different from the corresponding group without Dex (P < 0.05).

 
Because the orphan receptor COUP-TF1 enhances the E2 stimulation of the rtER gene through the FP1 region (51), a third set of experiments (Fig. 3CGo) was performed with the further addition of a COUP-TF1 expression vector. The enhancing effect of COUP-TF1 on E2-stimulated levels was clearly demonstrated on Kpn1 and Kpn2 contructs (Fig. 3CGo), in agreement with previous results (40). After Dex addition, the inhibitory effect of rtGR disappeared on the two shorter constructs in the presence of COUP-TF1 but was still obvious on Kpn3. This last experiment confirmed the implication of the -154/-112 sequence in an enhancing effect of the E2 stimulation and also pointed to an implication of this region in the rtGR inhibitory effect on rtER expression.

The inhibition of rtER-stimulated transcriptional activity of Kpn3 by rtGR was tested at different doses, and the results are expressed as percentages of inhibition by Dex vs. controls (Fig. 3DGo). The data demonstrate an increased inhibition by Dex with increasing amounts of rtGR, reaching a plateau with the highest doses of rtGR. COUP-TF1 partially neutralized the Dex inhibitory effect for the lower doses of rtGR (25 and 50 ng) but had no effect at higher doses of rtGR. Therefore, it seems that the higher stability of the rtER complex, mediated by COUP-TF1 (Métivier, R., personal communication), is not affected by low doses of rtGR, but high concentrations of rtGR are able to counteract this effect.

These experiments demonstrate that the rtGR interference on the ERE-dependent activation, observed on the Kpn1 and Kpn2 constructs, is counteracted by COUP-TF1. On the other hand, the rtGR strongest repression, evident on the Kpn3, construct persists in the presence of the COUP-TF1 expression vector. Therefore, the -154/-112 promoter sequence, specific of the enhancer effect on E2 stimulation, seems to be a specific target for the Dex inhibitory effect. This fragment contains the FP3 region (-134/-106), already known to be protected by fish liver nuclear extracts in footprint experiments (40).

A C/EBPß-like transcription factor binds to the FP3 region
Our working hypothesis was to assume that the binding of unknown transcription factors on the FP3 sequence could enhance the rtER-stimulated activity and that the rtGR could block the enhancing activity of this transcription factor. To investigate this hypothesis, we first analyzed potential interactions between liver proteins and the FP3 DNA sequence.

Nuclear protein extracts were prepared from trout liver and tested using EMSA on the FP3 sequence oligonucleotide (Fig. 4AGo). Because CHO cells were used for transfection assays and allowed to identify the FP3 region, EMSAs were also performed with CHO WCE (Fig. 4BGo). As shown by the presence of DNA complexes with shifted mobility (lane 1) and by the complete competition of the complexes by a 20-fold excess of unlabeled FP3 (lane 2), these extracts contain proteins able to specifically bind the FP3 region. These shifted complexes were unaffected by competition with a 100-fold excess of unlabeled GRE (lane 3), indicating the absence of rtGR in these complexes. This last result was confirmed by the fact that rtGR antibodies had no effect on the complex mobility with liver nuclear extract (data not shown).



View larger version (45K):
[in this window]
[in a new window]
  		Figure 4. Binding of trout liver nuclear extract on FP3 oligonucleotide as shown by EMSA. Trout liver nuclear extract (NE; 1 µg) was used to study the binding of trout liver proteins on the 32P-labeled FP3 oligonucleotide (A). CHO WCE (6 µg) as used in the same conditions (B). A 20-fold excess of FP3 oligonucleotide (lane 2) or a 100-fold excess of unlabeled GRE oligonucleotide (lane 3) were used as competitors. The protein-DNA complexes were separated from the free probe on a 6% polyacrylamide gel.

 
To identify the components of the FP3-bound protein complex, a computer analysis of potential binding sites of transcription factors was performed by using the Citi2 software (Citi2, Paris, France). This indicated that several transcription factors can potentially bind the FP3 region (Fig. 5Go). Two of them, Oct and C/EBP, were already known to interfere with GR in different physiological situations (52, 53, 54, 55, 56) and prompted us to focus on their potential involvement.



View larger version (12K):
[in this window]
[in a new window]
  		Figure 5. The FP3 region contains multiple binding sites for transcription factors. Potential binding sites for transcription factors on the FP3 region indexed by computer analysis using the Citi2 software. The FP3 protected region (-134/-106), resulting from footprint experiment with trout liver nuclear extract is presented.

 
We performed competition with oligonucleotides containing potential binding sites for transcription factors, using gel shift experiments on the FP3-labeled probe. The sequences of the different oligonucleotides used are presented in Fig. 6AGo. All unlabeled oligonucleotides were used with a 20-fold excess. Both liver nuclear extracts and CHO WCE showed specific binding on the FP3 region (Fig. 6Go, B and C, lane 2), which was abolished by a 20-fold excess of unlabeled FP3 (lane 3). Only three oligonucleotides were able to compete with the labeled FP3 probe: Oct1'aa, C/EBP/FP3, and C/EBP consensus (lanes 8, 9, 11). Thus, the specific binding sequence of FP3 deduced from the EMSA competitions was GTTGCATAA. This sequence of nine nucleotides is a potential binding site for the C/EBP transcription factor and shows only one mutation on the first position (G), compared with the consensus sequence (A). The liver nuclear extract and the CHO WCE showed the same pattern of competition, indicating that the transcription factors involved in the rtER gene regulation in trout liver were also present in CHO cells with the same DNA sequence specificity, despite the different size of the complexes. This further validates the use of the CHO heterologous system for the study of the inhibition of the rtER promoter in transfection experiments.



View larger version (55K):
[in this window]
[in a new window]
  		Figure 6. Competition analysis of in vitro binding of liver nuclear extract or CHO WCE to the FP3 oligonucleotide. Competitor oligonucleotides used in EMSA are represented (A). The sequence in common with FP3 is underlined, and bold letters show mutated nucleotide vs. the consensus sequence of the related transcription factor. Trout liver nuclear extract (1.5 µg) (B) or CHO WCE (6 µg) (C) were incubated with the 32P-labeled FP3 oligonucleotide for 20 min at room temperature in 1x binding buffer. The protein-DNA complexes were separated from the free probe (lane 1) on polyacrylamide gels. For competition analysis, a 20-fold molar excess of each competitor was mixed with proteins for 20 min before the addition of the labeled probe.

 
To check whether the C/EBP-binding site is really involved in the regulation of the rtER gene expression, we used a mutated FP3 oligonucleotide (called FP3m) containing three mutations on the C/EBP-binding site (Fig. 7AGo). These mutations were chosen to destroy the C/EBP-binding site without creating a new binding site for any known transcription factor, and this was checked by using the MatInspector version 2.2 software (57). This mutant was tested in competition assays on the FP3 probe by gel shift experiments using CHO WCE (Fig. 7BGo), trout liver nuclear extract (Fig. 7CGo), and hC/EBPß translated in reticulocyte lysate (Fig. 7DGo). For each extract, specific complexes were checked with a 20-fold excess of unlabeled FP3 and C/EBP/FP3 DNA sequence. However, with all protein extracts, the mutant FP3 oligonucleotide failed to compete with the probe. These data suggest that C/EBP could be the transcriptional factor interacting with the FP3 region. To check this hypothesis, antibodies directed against two isoforms of the hC/EBP proteins (hC/EBP{alpha} and hC/EBPß) were used. We first tested the specificity of both antibodies on C/EBPß translated in the reticulocyte lysate system (Fig. 7DGo, lanes 5 and 6). We observed that the anti-C/EBPß antibody, but not the anti-C/EBP{alpha} antibody, caused a supershift, demonstrating that this last antibody does not cross-react with hC/EBPß. When CHO WCEs were incubated with the FP3 probe, only the C/EBPß antibody was able to supershift the resulting complex (Fig. 7BGo). However, these antibodies failed to cross-react with trout proteins in EMSA (data not shown). All of these results strongly suggest that a C/EBPß-like protein could interact with the FP3 region and may be involved in the regulation of the trout ER gene.



View larger version (45K):
[in this window]
[in a new window]
  		Figure 7. Competition analysis by EMSA with mutated FP3 sequence on the C/EBP-binding site. To study the involvement of C/EBP on the FP3 region, a mutated competitor oligonucleotide called FP3m was prepared. This oligonucleotide contains three mutations (stars) in the C/EBP-binding site (A). Competition analyses by EMSA were performed using 6 µg CHO WCE (B), 0.5 µg liver nuclear extract (C), or 2 µl hC/EBPß in vitro translated protein (D) incubated with 32P-labeled FP3 oligonucleotide. Proteins were incubated with a 20-fold excess of each unlabeled competitor: FP3 (lane 2), CEBP/FP3 (lane 3), or FP3m (lane 4). For CHO WCE and in vitro-translated C/EBPß, a supershift analysis was performed using 0.5 µg antibodies directed against C/EBPß (lane 5) or C/EBP{alpha} (lane 6). These antibodies were mixed to protein extracts for 40 min before the addition of unlabeled competitors.

 
To test the potential implication of Oct in the protein complex bound to the FP3 region, an antibody directed against the human Oct-1 protein was used (Fig. 8Go). We first checked the ability of the Oct antibody to cross-react with CHO proteins using gel shift experiments with an Oct- labeled probe (Fig. 8AGo). We observed that the major protein complex that specifically bound to the Oct probe was decreased by the addition of 1 and 5 µl anti-Oct antibody (Fig. 8AGo). When CHO WCEs were incubated with the FP3 probe (Fig. 8BGo), the Oct antibody had no effect on the complex mobility, showing that this complex did not contain the Oct protein. These results were confirmed by titration with unlabeled Oct 5' and CEBP/FP3 in EMSA experiments with CHO WCEs and the labeled FP3 probe. Whereas a 3-fold excess of unlabeled CEBP/FP3 was sufficient to completely displace the shifted complex, a 50-fold excess of unlabeled Oct 5' was unable to compete (data not shown).



View larger version (57K):
[in this window]
[in a new window]
  		Figure 8. Analysis of the potential binding of Oct1 transcription factor on the FP3 region. To study the potential implication of Oct1 transcription factor on the FP3 region, EMSAs were performed with an antibody directed against human Oct1. CHO WCEs (8 µg) were incubated with 1 µl (lane 4), 5 µl (lane 5) of the anti-Oct1 antibody, or with 5 µl of a preimmune serum for 40 min before adding of the Oct (A) or FP3 (B) 32P-labeled oligonucleotides. This Oct oligonucleotide contains the Oct consensus sequence ATTTGCAT. The specificity of protein binding was checked by competition analysis with a 20-fold excess of unlabeled Oct (A) (lane 3) or FP3 (B) (lane 3) oligonucleotides. The arrow indicates specific Oct-shifted complex.

 
GR represses E2-induced rtER activity by squelching of the C/EBPß-like protein
To check whether the C/EBP-binding site was involved in the rtER gene regulation in transfection assays, we first performed experiments with a reporter gene consisting of Kpn3 with a mutated C/EBP binding site (Kpn3m; same mutations as in Fig. 7Go) upstream of the luciferase gene. The Kpn3m activity was compared with the wild-type Kpn3 and to the Kpn2 reporter constructs in CHO cells. As shown in Fig. 9Go, we observed that the E2 stimulation was more efficient on the Kpn3 construct than on the Kpn2 reporter gene. Moreover, the Dex inhibitory effect was higher on Kpn3 in agreement with the previous results (Fig. 3Go), whereas the Kpn3m had the same activity as Kpn2. We concluded that when the C/EBP site is disrupted, the enhancing effect on E2-stimulation is lost together with the rtGR inhibitory effect on this enhancing activity.



View larger version (28K):
[in this window]
[in a new window]
  		Figure 9. Effect of C/EBP-binding site mutations on the transcriptional activity of rtER promoter. Kpn3m is the Kpn3 construct containing the three same point mutations as on the FP3m oligonucleotide (Fig. 7Go). This new reporter construct was obtained using site-directed mutagenesis. Kpn3, Kpn3m, or Kpn2 (875 ng) were transfected with 50 ng CMV-rtGR and 50 ng CMV-rtER in CHO-K1 cells by the calcium phosphate precipitation method. Transfected cells were then treated with 1000 nM E2 and/or Dex for 36 h. Luciferase activities were normalized to total cellular proteins and expressed as fold induction of treated vs. untreated cells. Values represent the mean of nine culture dishes ± SEM.

 
To investigate the rtGR influence on C/EBPß binding on its own site, we performed EMSA with different domains of the rtGR fused with GST: GST-A/B-rtGR, GST-DBD-rtGR, GST-LBD-rtGR (Fig. 10AGo). Three different protein extracts were used: C/EBPß translated in reticulocyte lysate system (Fig. 10BGo), trout liver nuclear extracts (Fig. 10Go, C and E), or CHO WCE (Fig. 10DGo). Each protein extract formed specific complexes with the FP3 probe, as demonstrated by competition with a 20-fold excess of C/EBP consensus oligonucleotide or by a supershift with the C/EBPß antibody. When the DBD-rtGR/GST fusion protein was added with protein extracts, the protein-DNA complex was abolished. The specificity of this effect was checked by using A/B domain (lane 4) and LBD domain (lane 2) of the rtGR fused with GST or GST alone (lane 5). GST LBD-rtGR was also tested after incubation with Dex (data not shown). None of these proteins had the same effect as the DBD-rtGR/GST. Furthermore, the GST/DBD-rtGR was unable to interfere with the binding of nuclear extract proteins to the DR24 probe containing both FP1 and FP2 regions of the rtER promoter (Fig. 10EGo). This last result confirmed that the DNA binding domain of rtGR specifically blocked the binding of C/EBP on the FP3 sequence.



View larger version (41K):
[in this window]
[in a new window]
  		Figure 10. Study of the rtGR effect on the binding of proteins on the FP3 probe by EMSA. GST fusion proteins were constructed by using different fragments of the rtGR protein containing either the A/B domain (GST-A/B-rtGR), the DNA binding domain (GST-DBD-rtGR) or the LBD (GST-LBD-rtGR) (A). The purified fusion proteins were checked on SDS-PAGE (A) and quantified by comparison to a standard bovine serum albumin scale. Then 150 ng of each fusion protein or GST alone were incubated with 1.5 µl in vitro-translated hC/EBPß (B), 1 µg trout liver nuclear extract (C), or 6 µg CHO WCE (D) for 40 min at room temperature. Antibodies directed against C/EBPß were also used, together with in vitro-translated C/EBPß (B) (lane 7) or CHO WCE (D) (lane 8). A 20-fold excess of unlabeled competitors (C/EBP consensus or FP3 oligonucleotides) was then added in 1x binding buffer for 20 min before the addition of the 32P-labeled FP3 probe. A control experiment (E) was performed by incubating 500 ng GST/DBD-rtGR protein with an oligonucleotide 32P-labeled DR24 probe containing the FP2 and FP1 regions of rtER promoter (DR24 probe) in presence or not of 20-fold excess of unlabeled DR24 oligonucleotide.

 
These results suggest that the rtGR can affect the enhancing effect of C/EBPß on the E2-stimulation by preventing the binding of C/EBPß on its DNA binding site. Different molecular mechanisms could explain the inhibitory effect of rtGR on the DNA binding of C/EBPß: rtGR can interact with C/EBPß by protein-protein interactions or bind to the FP3 DNA sequence. This last hypothesis is unlikely because as in the case of the rtGR of liver nuclear extract already shown in Fig. 4Go, the fusion protein GST/DBD-rtGR was unable to form a complex with an oligonucleotide containing the FP3 region (data not shown).


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this work, we bring new information on the molecular interactions involved in the positive autoregulation of the rtER gene by its own product, and we analyze potential mechanisms explaining the inhibitory effect of cortisol via its cognate receptor on the expression of the rtER. It is well established that the molecular mechanism of the positive autoregulation of rtER gene expression takes place at the transcriptional level (58) and involves an imperfect ERE and a half ERE in the 0.2-kb fragment located just upstream of the transcription start site (38, 39). Indeed, footprint analyses on the rtER promoter with trout liver nuclear extracts result in three protected regions. The first region (FP1) contains the imperfect ERE, which is necessary and sufficient to observe an E2-induction by rtER, whereas the FP2 region contains a half ERE that cooperates with the ERE to induce higher transcriptional activity in yeast (41). In addition, the orphan receptor COUP-TF1 has been demonstrated to mediate a cooperative effect with ER on the positive autoregulation of rtER gene expression in cell lines and yeast (40, 41). The present work is in agreement with these reports and further validates the functionality of our experimental system. Two potential molecular mechanisms for this ER/COUP-TF1 cooperation have emerged from previous studies. First, COUP-TF1 could modify the nucleosomal structure of the rtER promoter gene by binding on a DR24 (direct repeat) element, consisting of the half-ERE (FP2) and the 3' half-site of the ERE (FP1) (40) and subsequently facilitate ER binding on its site (51). The second potential mechanism is based on the interplay between ER and COUP-TF1, which would favor the recruitment of an ER-specific coactivator. Recent studies in our laboratory examining this latter hypothesis demonstrated that these two nuclear receptors are able to physically interact in an E2-independent manner (Métivier, R., personal communication). This interaction requires the DBD of COUP-TF and the AF-1 domain of rtER and mediates phosphorylation of the rtER protein, therefore increasing its transcriptional activity.

In the present report, we demonstrated by cotransfection of 5'-deleted rtER promoter constructs that the third protected region (FP3) mediates a potentiation of rtER transactivation because deletion or mutation of this sequence reduces the E2-stimulated promoter activity. FP3 implication has also been indicated by other previous experiments. Indeed, footprinting analysis performed with pituitary and liver nuclear extracts showed that the FP3 region was protected only in liver and not in pituitary (40), in agreement with the fact that the positive autoregulation exists only in the liver and not in the pituitary (59, 60). Taken together, these results suggest that the FP3 region plays a role in the positive autoregulation of the rtER gene, specifically in the liver.

The FP3 region appears to be recognized by a transcription factor able to bind on a C/EBP-binding site, present in both CHO cells and trout hepatocytes and thus validating the CHO cell line as a suitable transfection system to study the regulatory effect on the FP3 region. Antibody supershift experiments identified a C/EBPß-like protein as the C/EBP family member binding to the FP3 region. C/EBPß corresponds to a leucine zipper protein of the CCAAT/enhancer-binding protein family that currently encompasses six members, some of which have several isoforms (61). C/EBPs control the transcription of genes involved in a broad range of physiological processes, such as adipocyte differentiation, inflammatory functions, and regulation of multiple hepatocyte-specific genes (62). C/EBPß, a 32-kDa protein, is constitutively expressed in the intestine, lung, adipose tissue, and testis (61, 63) but is also present in high concentrations in the nucleus of liver cells (64). During development, C/EBPß is involved in the acquisition of glucocorticoid sensitivity (65) and in the up-regulation of the {alpha}-1 acid glycoprotein gene by GR in vitro and in vivo (55, 66). C/EBPß is also involved in the down-regulation of the albumin gene (67) or in the up-regulation of Vg gene expression (11). In this latter study, a cooperation among C/EBPß, HNF3, and ER was found to be necessary for maximal gene activation in response to the hormone. C/EBPß is able to physically interact with different transcription factors, notably members of the NF-{kappa}B family, resulting in agonistic or antagonistic effects, depending on the promoter architecture and cell specificity (68). Cross-talk mechanisms with nuclear receptors such as RAR, ER, or GR have also been demonstrated (12, 56, 66, 69).

Evidence for the potentialization effect of rtER gene autostimulation is strongly supported by cotransfection experiments using the Kpn3m construct, which contains three mutations in the C/EBP binding site. The luciferase activity of the Kpn3m was comparable with that of the Kpn2 construct. Moreover, failure of the FP3m oligonucleotide, containing the same three mutations, to compete with the FP3 probe for the C/EBPß-DNA complex was verified by EMSA. All these convergent data point to a C/EBPß-like protein as the liver transcriptional factor involved in enhancing the positive autoregulation of the rtER gene expression by binding on the FP3 region.

Our results also provide information on the molecular mechanisms mediating the Dex-induced inhibition of rtER expression. Different models of transcriptional repression by GRs have already been proposed. Some genes negatively regulated by GRs contain either negative GR elements (nGREs) or composite GREs (70, 71, 72). For example, upon ligand-induced dimerization, the GR binds to an nGRE to repress the proopiomelanocortin gene expression (70). On the proliferine promoter gene, the GR has to bind to DNA together with AP-1 transcription factor on a composite GRE (71). However, mutual interference between GRs and other transcription factors such as AP-1 and NF-{kappa}B, does not require GR binding to DNA but rather involves protein-protein interactions (13, 73, 74). In this last case, GR action does not involve dimerization. This is shown by experiments in which mutations were introduced in the GR dimerization interface that generated a receptor unable to bind DNA and transactivate GRE-dependent promoters in vitro. This mutated GR was still able to transrepress the collagenase promoter gene as effectively as the wild-type receptor (75). The physiological importance of these protein-protein interactions has been strongly established by the use of homologous transgenic mice. Indeed, mice carrying a defective GR gene (76) died shortly after birth with severe abnormalities, whereas mice carrying a dimerization-defective GR unable to bind on its responsive element were viable (77).

Given these different ways of repression used by the GR, we wanted to investigate the molecular mechanisms involved in our physiological model. To localize the transcriptional action site, we studied the effect of 5' deletions of the rtER promoter by cotransfection experiments, which indicated that the activated GR had no effect on basal levels but only on E2-stimulated levels of rtER-luciferase constructs. This result suggests that the rtGR inhibits rtER transcription by interference with the stimulatory mechanism and not by interfering with the basal transcriptional machinery, as described, for example, for the human osteocalcin promoter, which contains a GRE overlapping with the TATA box (78, 79).

Cotransfection analyses in the present study indicate two different pathways of GR inhibition, consistent with the different mechanisms of rtER stimulation demonstrated so far. One pathway, strictly ERE dependent, is clearly seen on the Kpn1 and Kpn2 constructs and is counteracted by the addition of the hCOUP-TF1 expression vector in cotransfection assays. This GR-mediated inhibition could be a competition between GR and ER for a common coactivator, as already described between GR and RAR (19). This hypothesis is fully consistent with the new enhancer effect of COUP-TF1 on rtER gene expression described by (Métivier, R., personal communication), which shows that COUP-TF1 interacts with rtER protein, mediates phosphorylation of rtER, and enhances the recruitment of a coactivator. Thus, the higher stability of the coactivator/rtER complex would prevent the squelching of the coactivator by the activated GR.

On the other hand, we found by cotransfection experiments with 5'-deleted rtER promoter constructs that the FP3 region, which mediates the rtER transactivation potentiation by binding of a C/EBPß-like protein, was also involved in the rtGR-induced inhibition. To dissect the molecular mechanism of this rtGR-mediated effect, we looked for GR DNA binding on the FP3 region using EMSA and found no binding of GR on this sequence. The other potential mechanism was presumed to be a direct protein-protein interaction of rtGR with C/EBPß, an interaction that has already been described between GR and a large number of transcription factors, such as AP-1 (13). This hypothesis was tested with EMSA experiments using fusion proteins containing three parts of the rtGR protein or GST alone and showed that the DBD of the rtGR was able to prevent the binding of C/EBPß protein translated in reticulocyte lysate system on the FP3 probe. The same effect was observed by using CHO WCE or liver nuclear extract. Previous studies based on GST pull-down experiments (56) have shown that rat GR and C/EBPß were able to physically interact. This interaction occurred through the DBD of GR, which suggests that GR prevents C/EBPß DNA binding by direct protein-protein interaction. Thus, it seems that in response to Dex, activated rtGR interacts with C/EBPß through its DNA-binding domain, thus preventing its binding on the rtER promoter (Fig. 11Go). This observation is consistent with other reports showing that, although the DBD of a steroid receptor is required for transrepression, specific binding to the DNA is not required. The inhibition, mediated by the interaction between ER and C/EBPß or NF-{kappa}B on the cytokine IL-6 gene, depends on the DBD of ER and two domains of NF-{kappa}B and C/EBP, the DNA binding and dimerization domains (12).



View larger version (14K):
[in this window]
[in a new window]
  		Figure 11. Schematic presentation of the inhibition by GR of C/EBPß-like enhancer effect. C/EBPß binds to its site on the FP3 region to enhance the E2 activation of rtER gene. The DNA binding domain of the liganded GR prevents the binding of C/EBPß to FP3 sequence and blocks the C/EBPß enhancer effect.

 
Meanwhile, such interactions were also described in a positive way because GR is able to mediate a synergistic effect on the C/EBPß activation of the rat {alpha}-1 acid glycoprotein gene by direct GR-C/EBPß interaction. This protein-protein interaction involves the leucine-zipper domain of C/EBPß and the DNA binding domain of GR (66). It seems that direct protein interaction is not the only model for nuclear receptor-C/EBPß cross-talk. Indeed, a positive functional cross-talk between GR and C/EBPß was described on the Herpes simplex virus-thymidine kinase promoter but was mediated by an indirect GR-C/EBPß interaction (56). Indeed, the potentiation occurs through the AF-2 domain of GR, although in vitro pull-down experiments have shown that the direct GR-C/EBPß interaction requires the DBD of GR. In the same way, the AF-2 transactivation domain of RAR is necessary for interference with C/EBPß function (69). This study suggests that this cross-talk, which cannot occur by direct protein-protein interactions, could involve a coactivator complex.

Therefore, our results report the first example of negative interactions between GR and C/EBPß. This negative interference that requires the DBD of rtGR could be assimilated to a direct protein-protein interaction. This provides further evidence that the molecular mechanism of nuclear receptor action is cell dependent because the same GR domain can mediate either an activation (66) or a repression by cross-talk with C/EBPß. In this work, GR inhibits the ER-mediated up-regulation of the rtER gene by preventing C/EBPß-ER potentiation through inhibition of C/EBPß DNA binding.

This study aimed at dissecting the mechanisms of transcriptional interference between two nuclear receptors, rtGR and rtER, in the regulation of rtER gene expression. Such an interference is likely to have a physiological significance in the context of reproduction in oviparous species. Indeed, the response to chronic and severe stress requires energy and forces the organism to make adaptive choices. Energy normally available for growth, immune response, and reproduction is channeled under stress conditions into the restoration of disturbed homeostasis. As a result, negative effects of stress on reproduction have been documented in many species and seem to affect different levels of the reproductive axis (80). In the case of oviparous species, vitellogenesis carries a high-energy cost, and many studies have already documented the deleterious effects of stress or cortisol on Vg levels (81) and egg quality (36). In this study, a strong inhibitory effect of Dex on basal and E2-stimulated rtER and Vg mRNA levels was clearly observed in cultured trout hepatocytes, and in a previous report a consistent decrease was also described in trout liver Vg and ER mRNA levels after in vivo glucocorticoid treatment of female trout (37). Such effects can be interpreted as adaptive mechanisms interrupting the process of vitellogenesis to direct energy toward the restoration of homeostasis.


   Acknowledgments
 
We are grateful to Dr. Guennadi Iline for his generous gift of C/EBP antibodies and to Dr. Winship Herr for his generous gift of Oct1 antiserum. We also thank Dr. Fréderique Gay for her technical contribution, Sarah and Nic Bury for manuscript revision.


   Footnotes
 
This work was supported by grants from the Ministère de l’Enseignement et de la Recherche, Centre National de la Recherche Scientifique, and Fondation Langlois.

Abbreviations: AF, Transactivation function; AP1, activator protein-1; C/EBP, CCAAT/enhancer-binding protein; CHO, Chinese hamster ovary; CMV, cytomegalovirus; COUP-TF1, chicken ovalbumin upstream promoter transcription factor 1; DBD, DNA-binding domain; Dex, dexamethasone; E2, estradiol; ER, estrogen receptor; ERE, estrogen-responsive element; GR, glucocorticoid receptor; GRE, GR element; GST, glutathione-S-transferase; LBD, ligand-binding domain; NF-{kappa}B, nuclear factor {kappa}B; nGRE, negative GR element; RAR, retinoic acid receptor; rt, rainbow trout; SDM, site-directed mutagenesis experiment; SSC, saline sodium citrate; Vg, vitellogenin; WCE, whole-cell extract.

Received January 9, 2002.

Accepted for publication April 19, 2002.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Georges FW, Wilson JD 1988 Sex determination and sex differentiation. In: Knobil E, Ewing LL, Greenwald GS, Market CL, Pfaff DW, eds. The physiology of reproduction. New York: Raven Press; 3–26
  2. Korach KS 1994 Insights from the study of animals lacking functional estrogen receptor. Science 266:1524–1527[Abstract/Free Full Text]
  3. Green S, Walter P, Kumar V, Krust A, Bornert JM, Argos P, Chambon P 1986 Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A. Nature 320:134–139[CrossRef][Medline]
  4. 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[Abstract/Free Full Text]
  5. Mangelsdorf D, Thummel C, Beato M, Herrlich P, Schütz G, Umesono K, Blumberg B, Kastner P, Mark M, Chambon P, Evans RM 1995 The nuclear receptor superfamily: the second decade. Cell 83:835–839[CrossRef][Medline]
  6. Dana SL, Hoener PA, Wheeler DA, Lawrence CB, McDonnell DP 1994 Novel estrogen response elements identified by genetic selection in yeast are differentially responsive to estrogens and antiestrogens in mammalian cells. Mol Endocrinol 8:1193–207[Abstract]
  7. Porter W, Saville B, Hoivik D, Safe S 1997 Functional synergy between the transcription factor Sp1 and the estrogen receptor. Mol Endocrinol 11:1569–1580[Abstract/Free Full Text]
  8. Le Dréan Y, Lui D, Wong AOL, Xiong F, Hew CL 1996 Steroidogenic factor 1 and estradiol receptor act in synergism to regulate the expression of the salmon gonadotropin II ß subunit gene. Mol Endocrinol 10:217–229[Abstract]
  9. Gaub MP, Bellard M, Scheuer I, Chambon P, Sassone-Corsi P 1990 Activation of ovabumine gene by the estrogen receptor involves the Fos-Jun complex. Cell 63:1267–1276[CrossRef][Medline]
  10. Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson J, Kushner PJ, Scanlan TS 1997 Differential ligand activation of estrogen receptors ER{alpha} and ERß at AP1 sites. Science 277:1508–1510[Abstract/Free Full Text]
  11. Marilley D, Robyr D, Schild-Poulter C, Wahli W 1998 Regulation of the vitellogenin gene B1 promoter after transfer into hepatocytes in primary cultures. Mol Cell Endocrinol 141:79–93[CrossRef][Medline]
  12. Stein B, Yang MX 1995 Repression of the interleukine-6 promoter by estrogen receptor is mediated by NF-{kappa}B and C/EBPß. Mol Cell Biol 15:4971–4979[Abstract]
  13. Beato M, Sanchez-Pacheco A 1996 Interaction of steroid hormone receptors with the transcription initiation complex. Endocr Rev 17:587–609[CrossRef][Medline]
  14. Robyr D, Wolffe AP, Wahli W 2000 Nuclear hormone receptor coregulators in action: diversity for shared tasks. Mol Endocrinol 14:329–347[Free Full Text]
  15. Oñate SA, Tsai SY, Tsai M-J, O’Malley BW 1995 Sequence and characterization of a coactivator for the steroid hormone receptor superfamily. Science 270:1354–1357[Abstract/Free Full Text]
  16. Hong H, Kohli K, Trivedi A, Johnson DL, Stallcup MR 1996 GRIP1, a novel mouse protein that serves as a transcriptional coactivator for the hormone binding domains of steroid receptors. Proc Natl Acad Sci USA 93:4948–4952[Abstract/Free Full Text]
  17. Le Douarin B, Zechel C, Garnier J-M, Lutz Y, Tora L, Pierrat B, Heery D, Gronemeyer H, Chambon P, Losson R 1995 The N-terminal part of TIF1, a putative mediator of the ligand-dependent activation function (AF-2) of nuclear receptor, is fused to B-raf in the oncogenic protein T18. EMBO J 14:2020–2033[Medline]
  18. Cavailles V, Dauvois S, L’Horset F, Lopez G, Hoare S, Kushner PJ, Parker MG 1995 Nuclear factor RIP140 modulates transcriptional activation by the estrogen receptor. EMBO J 14:3741–3751[Medline]
  19. Kamei Y, Xu L, Heinzel T, Torchia J, Kurokawa R, Gloss B, Lin S-C, Heyman RA, Rose DW, Glass CK, Rosenfeld MG 1996 A CBP integrator complex mediates transcriptional activation and AP-1 inhibition by nuclear receptors. Cell 85:403–414[CrossRef][Medline]
  20. Chakravarti D, La Morte VJ, Nelson MC, Nakajima T, Schulman IG, Juguilon H, Montminy M, Avans RM 1996 Role of CBP/p300 in nuclear receptor signalling. Nature 383:99–103[CrossRef][Medline]
  21. Hanstein B, Eckner R, DiRenzo J, Halachmi S, Liu H, Searcy B, Kurokawa R, Brown M 1996 p300 is a component of an estrogen receptor coactivator complex. Proc Natl Acad Sci USA 93:11540–11545[Abstract/Free Full Text]
  22. Jonat C, Rahmsdorf HJ, Park K-K, Cato ACB, Gebel S, Ponta H, Herrlich P 1990 Antitumor promotion and antiinflammation: down-modulation of AP-1 (Fos/Jun) activity by glucocorticoid hormone. Cell 62:1189–1204[CrossRef][Medline]
  23. Mukaida N, Morita M, Ishikawa Y, Rice N, Okamoto S, Kasahara T, Matsushima K 1994 Novel mechanism of glucocorticoid-mediated gene repression. J Biol Chem 269:13289–13295[Abstract/Free Full Text]
  24. O’Brien RM, Lucas PC, Yamasaki T, Noisin EL, Granner DK 1994 Potential convergence of insulin and cAMP signal transduction systems at the phosphoenolpyruvate carboxykinase (PEPCK) gene promoter through CCAAT/enhancer binding protein (C/EBP). J Biol Chem 269:30419–30428[Abstract/Free Full Text]
  25. Lee SK, Kim HJ, Na SY, Kim TS, Choi HS, Im SY, Lee JW 1998 Steroid receptor coactivator-1 coactivates activating protein-1-mediated transactivations through interaction with the c-Jun and c-Fos subunits. J Biol Chem 273:16651–16654[Abstract/Free Full Text]
  26. Ding XF, Anderson CM, Ma H, Hong H, Uht RM, Kushner PJ, Stallcup MR 1998 Nuclear receptor-binding sites of coactivators glucocorticoid receptor interacting protein 1 (GRIP1) and steroid receptor coactivator 1 (SRC-1): multiple motifs with different binding specificities. Mol Endocrinol 12:302–313[Abstract/Free Full Text]
  27. Meyer M-E, Gronemeyer H, Turcotte B, Bocquel M-T, Tasset D, Chambon P 1989 Steroid hormone receptors compete for factors that mediate their enhancer function. Cell 57:433–442[CrossRef][Medline]
  28. Watts CK, King RJ 1994 Overexpression of estrogen receptor in HTB 96 human osteosarcoma cells results in estrogen-induced growth inhibition and receptor cross talk. J Bone Miner Res 9:1251–1258[Medline]
  29. Uht RM, Anderson CM, Webb P, Kushner PJ 1997 Transcriptional activities of estrogen and glucocorticoid receptors are functionally integrated at the AP-1 response element. Endocrinology 138:2900–2908[Abstract/Free Full Text]
  30. Liesegang P, Romalo G, Sudmann M, Wolf L, Shcweikert HU 1994 Human osteoblast-like cells contain specific, saturable, high-affinity glucocorticoid, androgen, estrogen and 1 {alpha}, 25-dihydroxycholecalciferol receptors. J Androl 15:194–199[Abstract/Free Full Text]
  31. Ewing TM, Murphy LJ, Ng ML, Pang GYN, Lee CSL, Watts CKW, Sutherland RL 1989 Regulation of epidermal growth factor receptor by progestins and glucocorticoids in human breast cell lines. Int J Cancer 44:744–752[Medline]
  32. Zhou F, Bouillard B, Pharaboz-Joly MO, Andre J 1989 Non-classical antiestrogenic actions of dexamethasone in variant MCF-7 human breast cancer cells in culture. Mol Cell Endocrinol 66:189–197[CrossRef][Medline]
  33. Bigsby RM 1993 Progesterone and dexamethasone inhibition of estrogen induced synthesis of DNA and complement in rat uterine epithelium: effects of antiprogesterone compounds. J Steroid Biochem Mol Biol 45:295–301[CrossRef][Medline]
  34. Shuto T, Kukita T, Hirata M, Jimi E, Koga T 1994 Dexamethasone stimulates osteoclast-like cell formation by inhibiting granulocyte-macrophage colony-stimulating factor production in mouse bone marrow cultures. Endocrinology 134:1121–1126[Abstract]
  35. Pottinger TG, Pickering AD 1990 The effect of cortisol administration on hepatic and plasma estradiol-binding capacity in immature female rainbow trout (Oncorhynchus mykiss). Gen Comp Endocrinol 80:264–273[CrossRef][Medline]
  36. Carragher JF, Sumpter JP, Pottinger TG, Pickering AD 1989 The deleterious effects of cortisol implantation on reproductive function in two species of trout, salmo trutta L. and Salmon gairdneri Richardson. Gen Comp Endocrinol 76:310–321[CrossRef][Medline]
  37. Lethimonier C, Flouriot F, Valotaire Y, Kah O, Ducouret B 2000 Transcriptional interference between glucocorticoid receptor and estradiol receptor mediates the inhibitory effect of cortisol on fish vitellogenesis. Biol Reprod 62:1763–1771[Abstract/Free Full Text]
  38. Le Dréan Y, Lazennec G, Kern L, Saligaut D, Pakdel F, Valotaire Y 1995 Characterization of an estrogen-responsive element implicated in regulation of the rainbow trout estrogen receptor gene. J Mol Endocrinol 15:37–47[Abstract]
  39. Lazennec G, Kern L, Salbert G, Saligaut D, Valotaire Y 1996 Cooperation between the human estrogen receptor ER and MCF-7 cell-specific transcription factors elicits high activity of an estrogen inducible enhancer from the trout ER gene promoter. Mol Endocrinol 10:1116–1126[Abstract]
  40. Lazennec G, Kern L, Valotaire Y, Salbert G 1997 The nuclear orphan receptors COUP-TF and ARP-1 positively regulate the trout estrogen receptor gene through enhancing autoregulation. Mol Cell Biol 17:5053–5066[Abstract]
  41. Petit F, Métivier R, Valotaire Y, Pakdel F 1999 Synergism between a half-site and an imperfect estrogen-responsive element, and cooperation with COUP-TFI are required for estrogen receptor (ER) to achieve a maximal estrogen-stimulation of rainbow trout ER gene. Eur J Biochem 259:385–395[Medline]