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 Tiffoche, C.
Right arrow Articles by Thieulant, M.-L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tiffoche, C.
Right arrow Articles by Thieulant, M.-L.
Endocrinology Vol. 142, No. 9 4106-4119
Copyright © 2001 by The Endocrine Society

ARTICLES

Novel Intronic Promoter in the Rat ER{alpha} Gene Responsible for the Transient Transcription of a Variant Receptor

Christophe Tiffoche, Colette Vaillant, Diane Schausi and Marie-Lise Thieulant

Université de Rennes I, Interactions Cellulaires et Moléculaires, Equipe Information et Programmation Cellulaires, Centre National de la Recherche Scientifique UMR 6026, Campus de Beaulieu, Rennes Cedex 35042, France

Address all correspondence and requests for reprints to: Dr. M.-L. Thieulant, Equipe Information et Programmation Cellulaires, UMR 6026, Bat 13, Campus de Beaulieu, 35042 Rennes cedex, France. E-mail: Marie-Lise.Thieulant{at}univ-rennes1.fr


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
To analyze the molecular origin of an ER variant, the truncated ER product-1, transiently expressed at the proestrus in lactotrope cells, we generated a 2.5-kb sequence of a genomic region upstream and downstream the specific sequence truncated ER product-1. Genomic Southern blot analysis showed that truncated ER product-1 is spliced from a noncoding leader exon localized within the intron 4 of the ER {alpha} gene. Analysis of the promoter sequence revealed the presence of a major transcriptional start site, a canonical TATA box and putative cis regulatory elements for pituitary specific expression as well as an E-responsive element. In transient transfection, the truncated ER product-1 promoter was transcriptionally the most active in the lactotrope cell lines (MMQ). Analysis of truncated ER product-1 functionality showed that: 1) the protein inhibited ER{alpha} binding to the E-responsive element in electromobility shift assays, 2) inhibited the E2 binding to ER{alpha} in binding assays, 3) the truncated ER product-1/ER{alpha} complex antagonized the transcriptional activity elicited by E2, 4) nuclear localization of green fluorescent protein-ER{alpha} was altered in Chinese hamster ovary cell lines stably expressing truncated ER product-1. Collectively, these data demonstrated that the protein exerts full dominant negative activity against ER{alpha}. Moreover, truncated ER product-1/ER {alpha} complex also repressed the activity of all promoters tested to date, suggesting a general inhibitory effect toward transcription. In conclusion, the data suggest that truncated ER product-1 could regulate estrogen signaling via a specific promoter in lactotrope cells.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
THE MECHANISM OF action of E2 is classically mediated by the ER that belongs to the nuclear superfamily of ligand-dependent transcription factors. ER is encoded by two genes, {alpha} and ß (ER{alpha} and ß) (1). Like other members of the superfamily, the ERs have a modular structure consisting of distinct functional domains (2). The DNA-binding domain enables the receptor to bind to its cognate target site. The carboxy-terminal domain contains the ligand-binding domain as well as sequences necessary for nuclear localization, receptor dimerization, and transcriptional function (AF-2) (2, 3, 4). A second activation function, AF-1, is present in the amino-terminal domain of the receptor (4, 5). Although the full transcriptional activation is mediated through both AF-1 and AF-2, only activation through AF-2 requires hormone binding (2, 6). The transactivation mediated by estrogen-occupied receptor is a complex process, involving interactions with transcription factors and other components of the basal transcriptional machinery, and finally modulating gene transcription (6).

Multiple forms are known to exist for the members of the nuclear receptor superfamily. Regarding ER{alpha}, multiple variants have been described in human breast cancer biopsies, cell lines, and pituitary tumors (7, 8, 9, 10). Mutations, deletions, or rearrangements of the ER{alpha} gene as well as alternative splicing, yielding precise deletions of either a single exon or two exons, have been found (7, 8, 9). In addition, human and rat ER{alpha} genes are transcribed from different promoters, yielding different mRNA isoforms (11, 12, 13). Several groups have attempted to establish a relationship between overexpression of ER{alpha} variants and tumor progression. However, their data have been conflicting (14). These ER{alpha} variants do not seem to be restricted to malignant tissues because they have been found in normal tissues expressing the wild-type (WT) ER (7, 8, 9, 15). Interestingly, it has been shown that a non-DNA-binding ER{alpha} isoform, missing exon 3, suppressed E2-stimulated gene expression in breast cancer and that its relative loss may be important in carcinogenesis (16). Alternative translational start sites for ER{alpha} may also account for the formation of truncated ER{alpha} proteins (17, 18). Although the majority of data for ER{alpha} variant expression are at the RNA level, accumulating data support the detection of proteins corresponding to some of the previously identified ER{alpha} variant mRNAs, but their function still remains unclear.

The presence of ER{alpha} mRNA variants or truncated proteins has been previously described in both male and female rat pituitaries (19, 20, 21). After E2 injection, Friend et al. (20) cloned two variants missing exons 1–4 with a unique 31-bp untranslated sequence upstream of exon 5 (named TERP-1 and -2, for the female and male pituitary, respectively). Our laboratory previously identified such an ER{alpha} variant mRNA in a female rat pituitary gland (21). The TERP-1 mRNA appears to be restricted to the pituitary gland on proestrous stage. This mRNA is inducible by E2 or gonadotrophins and localized in lactotrope populations (20, 21). A protein product of 20–24 kDa was described (22, 23). Although TERP-1 lacks the A/B-, DNA binding, and hinge region, it can form heterodimers with WT-ER both in vitro (24) and in vivo (25) and inhibits ER{alpha} binding to its cognate DNA response element (24). On the other hand, contradictory results have been published concerning the transcriptional effect of TERP-1 on WT-ER transactivation (24, 26, 27). TERP-1 physiological function remains unknown.

Because TERP-1 expression is transient and pituitary cell specific, we hypothesized that it could be generated by the use of an alternative promoter. As an initial step to understand how TERP-1 expression may be regulated, we isolated genomic clones encoding a specific promoter. In this report, we demonstrate that TERP-1 originates from an intronic promoter into the ER{alpha} gene. The truncated protein has also been functionally characterized. Specifically, we reexamine the transcriptional control exerted by TERP-1 on WT-ER stimulated activity using various E-responsive element (ERE) reporter gene constructs. We confirm that TERP-1 interferes with the ability of ER{alpha} to bind to a consensus ERE. At transcriptional level, we observed the TERP-1/ER{alpha} complexes have an inhibitory effect on corresponding reporter genes. We demonstrate that the effects of TERP-1 could be due to an alteration of the E2 binding to ER{alpha}. Moreover, the TERP-1/ER{alpha} complex can apparently block general transcriptional activity, suggesting that it could work as a negative regulator of nuclear receptor activity in vivo. Taken together, data herein show that TERP-1 potentially functions as a dominant negative inhibitor of ER{alpha} activity and suggest that TERP-1 could regulate estrogen signaling via a specific promoter in lactotrope cells.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals and cell cultures
Wistar rats (CERJ, Le Genest, France) were housed under a controlled light schedule of 14-h light, 10-h dark and provided with rat pellets and water ad libitum. All animals studies were conducted in accordance with the guidelines for Care and Use of Experimental Animals. Immature female rats (28 d old) were primed with PMSG (12 IU, Sigma, St. Quentin, Fallavier, France) to induce a proestrus 50 h later as previously described (21). Chinese hamster ovary cells (CHO-K1) were maintained in DMEM/F12 supplemented with 10% FCS and antibiotics. Human embryonal kidney 293 cells, MDA-MB-231 cells, and MCF-7 cell lines were grown under similar conditions but in DMEM. MMQ cells (28) were cultured in RPMI 1640 h medium with 7.5% horse serum and 2.5% FCS. GH4-C1 (29) were grown in HAM-F10 with 15% horse serum and 2.5% FCS. Culture media were purchased from Life Technologies (Cergy-Pontoise, France).

Cloning TERP-1
A set of primers was designated according to the published TERP-1 sequence (20) and ER{alpha} sequence (30). The sequences were as follows: TERP-1, 5'-CGGGATCCATTTCTTGAGCTTGTTGAACAGC-3'; exon 8, 5'-GGAATTCCGATTCGCAGAACCTTGTGGG-3'. Single-strand cDNA was reverse transcribed from pituitary total RNA from rats on proestrous stage. PCR cycling was performed as following : 1) 94 C, 45 sec; 60 C, 45 sec, and 72 C, 1 min (5 cycles); and 2) 94 C, 45 sec; 66 C, 45 sec, and 72 C, 1 min (35 cycles). The resulting PCR fragments were subcloned into the EcoRI/BamHI sites of plasmid and carrier DNA (pcDNA) 3.0 (Invitrogen, Gröningen, The Netherlands) and were completely sequenced from both strands.

Cloning of the 5'-flanking region of the TERP-1 exon
We have cloned the 5'-flanking region of the TERP-1 exon with the PCR-based gene walking method (Genome Walker kit; CLONTECH Laboratories, Inc., Palo Alto, CA). The 5'-flanking region of the TERP-1 sequence was amplified from the five rat Genome Walker genomic libraries using Expand High Fidelity Polymerase (Roche Diagnostics, Meylan, France). Two consecutive rounds of PCR using the adaptor primers AP1 and AP2 and the gene-specific reverse primers S-TERP-1 (5'-CTGGTCGCTGTTCAACAAGCTC-3', nucleotides at positions 34–12 of specific TERP-1 sequence) and FL-TERP-1 (5' CTGGTCGCTGTTCAACAAGCTCAAGAAATGG-3', at positions 34–3 of specific TERP-1 sequence) were performed. The first amplification was performed using AP1 and S-TERP-1. The reaction involved 2 sec of denaturation at 94 C, followed by 40 cycles consisting of 2 sec of denaturation at 94 C and 3.5 min of annealing and extension at 68 C. The PCR products were diluted to 1/50 and then subjected to the secondary PCR with the nested primers AP2 and FL-TERP-1 using the same protocol. The PCR products were subcloned either into the BlueScript vector (Stratagene, Amsterdamzuidoost, The Netherlands) for sequencing or pGL3 vector (Promega Corp., Charbonnières, France), respectively.

Southern blot analysis
Ten micrograms of rat genomic DNA were digested overnight with restriction endonucleases, electrophoresed on a 0.8% agarose gel, and subjected to Southern blot analysis using either P-labeled rat ER{alpha} cDNA (exons 4–5) or 726-bp fragment of the 5'-flanking region of the TERP-1 gene (corresponding to residues +34 -692 of the rat DNA library 1 fragment) as probes. Hybridization was carried out for 16 h at 55 C in 1 mM EDTA, 1% BSA, 7% SDS, 0.5 M phosphate buffer, pH 7.2. Filters were washed under high stringency conditions in 0.2x salt sodium citrate, 0.1% SDS for 30 min each at 25 C, 55 C, and 68 C.

Primer extension
A 22-oligonucleotide primer selected to be 57 bp downstream of the translational start codon (ATG) (-57 to -78 bp) was used for primer extension. The primer was end-labeled with 32P using T4 polynucleotide kinase (Roche Diagnostics). Total RNA (50 or 100 µg) from the MMQ lactotrope pituitary cell line or rat liver and spleen was isolated and was hybridized to the primer for 16 h at 50 C in 80% Formamide, 0.4 M NaCl, 1 mM EDTA, and 40 mM PIPES pH 6.4. Extension was initiated with reverse transcriptase (Superscript II, Life Technologies, Inc.), and the reaction was allowed to proceed for 1 h at 42 C. The primer extension reaction products were analyzed on a 4% polyacrylamide-7 M urea gel and their lengths determined by comparison with a sequencing ladder.

Construction of the GFP-ER{alpha} expression vector
The rat ER{alpha} cDNA was generously provided by Dr. M. Muramatsu, isolated from pUC-ER{alpha} (30), and mutated in the first ATG of the ER{alpha} coding region. The pEGFPC3 vector, which encodes a green fluorescent protein (GFP) variant (S65T-GFP) was purchased from CLONTECH Laboratories, Inc. EcoRI and BamHI restriction sites were added to the 5' and 3' ends of ER{alpha} using PCR. In GFP-ER, GFP was fused at the C terminus of the rat ER{alpha}.

Construction of stable cell lines expressing TERP-1
CHO-K1 cells were transfected in 25-cm2 flasks using Fugene-6 (Roche Diagnostics) with 3 µg TERP-1 pcDNA. Twenty-four hours after transfection, the medium was changed, and G-418 (Roche Diagnostics, Inc.) was added to a final concentration of 500 µg/ml. After 24 h, the cells were trypsinized, diluted 1:100 in complete medium plus G418, plated in 25-cm2 flasks, and incubated until cell clones became visible. The clones were characterized by RT-PCR for TERP-1 expression.

Transient transfection assays
For TERP-1/ER heterodimer transactivation activities, 1 x 105 cells were plated on 6-well plates 24 h before transfection in phenol red-free media and charcoal-stripped serum. Precipitates were formed using expression vectors in the amounts indicated in the figures, 3.5 µg of reporter pcDNA (pcDNA3 or pCMV5) to a total of 4.5 µg. Following transfection, cells received fresh medium containing twice charcoal-stripped serum and were incubated with or without E2 (10 nM). For TERP-1 promoter activity, medium with complete serum was used. Cells were transfected with Fugene-6 according to manufacturer’s instructions (Roche Diagnostics). Then 0.7 µg of RDL-4-LUC vector and 0.3 µg ß-galactosidase expression plasmid (pCMV-ßgal, Promega Corp.) as internal control were transfected. For all studies, cells were harvested 30 h after removal of the precipitates. All transfections were done in triplicates. Protein concentrations in cell lysates were determined using the colorimetric protein assay system (Bio-Rad Laboratories, Inc., Ivry/Seine, France). Chloramphenicol acetyl transferase assays (31) and ß-galactosidase assays were done using the method essentially as described (32). Luciferase activity (31) was determined with reagents according to the manufacturer’s instructions (Promega Corp.) using either a luminometer ML 3000 (Dynatech Corp., Guyancourt, France) or a reporter microplate luminometer system (Promega Corp.). According to the experiments, luciferase activities were normalized either per 100-µg proteins or relative to ß-galactosidase activity.

Analysis of TERP-1 by RT-PCR
Tissues or cells were collected and total RNA prepared using Trizol (Life Technologies, Inc.) according to the manufacturer’s instructions. Total RNA was reverse transcribed (400 U; Superscript, Life Technologies, Inc.) as previously described (21). Briefly, amplification conditions were 94 C for 45 sec, annealing for 45 sec, 72 C for 1 min for 30 cycles, and a final cycle of a 5-min extension step. The PCR products were electrophoresed on 1.2% agarose gel, visualized by ethidium bromide staining. The oligonucleotides (Genset, SA, Paris, France) TERP-1, 5'-CCATTTCTTGAGCTTGTTG-3' and exon 8, 5'-CGTTTCAGGGATTCGCAG-3' were used for amplification of TERP-1 mRNA. The oligonucleotides, exon 1, CAGCAGCGAGAAGGGAAACA-3', and exon 4, 5'-GGGCGGGGCTATTCTTCTTA-3' were used for amplification of ER{alpha} mRNA. The oligonucleotides, RL19–1, 5'-GGACCCCAATGAAACCAACG-3' and RL19–2, 5'-CCTTCCTCTTCCCTATGCCC-3' were used for amplification of control RL19 mRNA as previously described (21). In a few experiments, the oligonucleotides, Actin-1, 5'-TGGTGGGTATGGGTCAGA-3', and Actin-2, 5'-CCAGCATAGAGCCACCAA-3' were also used as controls.

In vitro translation
Recombinant ER{alpha} and TERP-1 cDNAs were in vitro transcribed and translated in TNT-T7 coupled rabbit reticulocytes lysates (Promega Corp.) from the T7 promoter following the manufacturer’s guidelines. Typically, the reactions were carried out in 12.5-µl TNT lysate containing 0.5 µg plasmid, 1-µl TNT buffer, 0.5 µl of amino acid mixture minus methionine, 0.5 µl [35S] methionine (1175 Ci/mmol, ICN), 0.5 µl T7 RNA polymerase, and completed up to 25 µl with DNasefree water.

Electrophoretic mobility shift assay
Electrophoretic assays were performed in a 20-µl binding reaction in a binding buffer (10 mM Tris-HCl pH 7.5, 1 mM EDTA, 0.5 mM dithiothreitol, 5% glycerol, and 50 mM NaCl). Each binding reaction contained 10 µl ER{alpha} and/or TERP-1 cotranslated in vitro as indicated above, 0.5 µg of poly(dI-dC), and 20,000 cpm of consensus ERE probe (33) labeled with [{gamma}-32P] ATP (4000 Ci/mmol, ICN) using T4 polynucleotide kinase (Roche Diagnostics). Preincubations containing proteins and/or cold competitor as indicated were performed at 4 C for 20 min. After the incubation step, the probe was added, and binding was conducted for 20 min at room temperature. Protein-DNA complexes and free probe were then separated on a 6% native polyacrylamide gel using 0.5x Tris-borate-EDTA buffer.

Ligand binding analysis
Ligand binding studies were conducted essentially as previously described (34) with the following modifications: ER{alpha} and TERP-1 proteins were synthesized in vitro either alone or cotranslated in a 1:5 ratio. Translation reaction mixtures were diluted 10 times with TET (10 mM Tris-HCl, pH 7.4, 1.5 mM EDTA, 10 mM thioglycerol, containing protease inhibitors, 50 µg/ml leupeptin, 5 µg/ml phenylmethyl sulfonyl fluoride, and 5 µg/ml aprotinin) and 0.1-ml aliquots were incubated at 4 C overnight with [2, 4, 6, 7-3H] 17ß-estradiol (110 Ci/mmol, Amersham Pharmacia Biotech, Orsay, France) ranging from 0.1 to 50 nM. Nonspecific binding was assessed by including a 200-fold excess of unlabeled E2. Binding was measured by the protamine sulfate assay (34).

Western blotting
Pituitary cell extracts (75 µg) were resolved by 13% SDS-PAGE and then electrotransferred to a nitrocellulose membrane (Hybond C, Amersham Pharmacia Biotech). Filters were blocked overnight with PBS containing 2% nonfat milk and 0.1% Tween 20 (Sigma, Inc.), incubated with primary anti-ER{alpha} monoclonal antibody F3 (35) for 2 h at room temperature. Protein bands were visualized by enhanced chemiluminescence (ECL, Amersham Pharmacia Biotech) using a horseradish peroxidase-conjugated goat antirabbit secondary antibody. Filters were washed in PBS-Tween four times for 5 min between each step.

Data analyses
The values were expressed as mean ± SEM of at least three independent experiments. A t test was used to determine statistical significance. Differences between groups were considered to be significant at P < 0.05.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
An intronic promoter in the TERP-1 5'-flanking region
TERP-1 cDNA was cloned by RT-PCR from rat pituitary total RNA at proestrus using PCR primer pairs based on the published sequences of TERP-1 and ER{alpha} exon 8 (20, 30). A schematic representation of TERP-1 structural organization is presented in Fig. 1AGo. Figure 1BGo shows the pattern of PCR products appearing during the estrous cycle, confirming data previously observed using Northern blot analysis (21). According to the high sensitivity of the PCR technique, a very low-level signal was detected on metestrous day. This signal increased up to a maximal level on proestrous day, in which two TERP products were clearly identified. No PCR product was amplified at estrus. One of these amplified products corresponds to TERP-1; the second product, TERP-C, was cloned and sequenced. It contains the TERP-1 specific 31 bp and an additional insertion of 99 bp between exons 5 and 6 of ER{alpha}. Figure 1CGo shows the expression pattern of TERP-1 in two pituitary cell lines: GH4-C1, a mammosomatotrope tumor cell line (29), and MMQ, a pure lactotrope tumor cell line (28). Bands corresponding either to the expected size of ER{alpha} or TERP-1 transcripts were detectable in both lactotrope cell lines (Fig. 1CGo). The two variant TERPs were detected as the major ER in MMQ cells, confirming the specific expression in the lactotrope cells (21). No PCR product was detected for TERP-1 either in gonadotrope cell lines ({alpha}T3–1) or in other tissues tested, (uterus and hypothalamus [data not shown]).



View larger version (26K):
[in this window]
[in a new window]
  		Figure 1. TERP-1 RNA expression. A, Schematic presentation of the ER{alpha} and TERP mRNAs (20 30 ). Exons are represented as open boxes and are numbered 1–8. The top line indicates the functional domains of the ER{alpha}. Both TERPs have a unique 31-bp sequence (dashed box), whereas TERP-C has an additional 99-bp sequence (black box) between exons 5 and 6. The major translational start codons (ATG) and terminations codons (TGA) are indicated. Positions and orientations of specific oligonucleotide primers are shown by horizontal arrows. B and C, Analysis of TERP-1 expression was performed at all the stages of the estrous cycle (B) and in lactotrope cell lines (C) (MMQ and GH4-C1) by RT-PCR amplification. PCR primer pairs used were based on the published sequences of TERP-1 (20 ) and ER{alpha} exon 8 (30 ) as described in Materials and Methods. RL 19 or ß- actin were used as PCR internal controls. P, Proestrus; E, estrus; M, metestrus; D, diestrus.

 
To identify the 5'-flanking sequence of the TERP-1 transcript, three overlapping genomic fragments noted as RDL-1, RDL-3, and RDL-4 were generated by nested PCR. To confirm and complete the sequence, two successive PCRs were performed using primers chosen in the generated 5'-flanking sequence of TERP-1. The sequences of these generated PCR products were the same as the initially generated 5'- sequence (RDL-1) with an additional 500 bp downstream the TERP-1-specific 31 bp. The flanking sequences of TERP-1 are shown in Fig. 2Go. Finally, the nucleotide sequence contains 2001 bp upstream the TERP-1 sequence and 507 bp downstream.



View larger version (62K):
[in this window]
[in a new window]
  		Figure 2. Nucleotide sequence of the 5'-flanking regions of TERP-1. The transcription start site was set at +1 and indicated by an arrow. The sequence shown in bold letters and surrounded by a continuous line indicates the noncoding leader TERP-1 exon. Underlined bold sequences indicate the potential transcriptional regulatory sequences. The 5'-ends of clones RDL 1, 3, and 4 are marked by upward arrows.

 
Primer extension analysis was performed to define precisely the sites of transcription initiation of the Terp-1 gene. A single-stranded cDNA template (+22 bp to +489 bp as indicated in Fig. 3AGo) was labeled and hybridized to total RNA from MMQ lactotrope pituitary cell line and extended with a reverse transcriptase deprived of RNase H activity. Negative controls were performed with total RNA isolated from rat liver and spleen. Several primer-extended products were generated from pituitary but not from liver, spleen, or yeast tRNA (Fig. 3BGo). A major extended product mapped transcription start site at position 99 bp upstream from the ATG initiation codon. This site is referenced in the sequence as position +1 (Fig. 3Go). Three additional minor extended products were found, one of them in the proximal region.



View larger version (42K):
[in this window]
[in a new window]
  		Figure 3. Primer extension analysis. A, Representation of the TERP-1 mRNA and detailed sequences (bp) of exon TERP-1 and a part of exon 5. The major translational start codons (ATG) and terminations codons (TGA) are indicated. Bases are numbered from the major transcription start site designated +1. The relative position of the single-stranded cDNA template used for primer extension is numbered relative to the +1. Position of nucleotide 22 of the probe is indicated by a vertical arrow in the sequence of TERP-1 exon. B, Primer extension reactions with the end-labeled primer (lane 1) were carried out using total RNA isolated from rat liver (lane 3), spleen (lane 4), or rat MMQ pituitary cells (lanes 5 and 6 corresponding to 100 or 50 µg total RNA, respectively), with tRNA (lane 2) as the negative control. The primer extended products were separated on a 6% polyacrylamide, 7 M urea gel. A sequencing ladder (G, A, T, and C) of a different fragment from the primer-extended products was used to determine sizes of the extended products. Position of the major transcription initiation site is indicated by the arrow (+1). Asterisks indicate the minor transcription initiation sites. The experiments were repeated two times with similar results.

 
Analysis of the 5'-flanking region of the gene showed that TERP-1 contains 34 specific bp corresponding to a distinct untranslated leader exon, denoted exon TERP-1. This analysis revealed the presence of a TATA box motif (CATAAA) residing in close proximity (-24 bp to -29 bp) to the major transcription initiation site (Fig. 2Go). A computer-assisted search for consensus transcription factor-binding sites localized a putative palindromic ERE at positions -196/205. For 11 out of 12 bp, this ERE is similar to the EREc (AGGTCAnnnTGACCT) described in the vitellogenin A2 gene of Xenopus laevis (33). Four perfect half-EREs were also found in this sequence. The identification of eight potential Pit-1 sites (36), a pituitary-specific transcription factor, is of particular interest. In addition, the promoter region contains three consensus motifs for pituitary homeobox 1 and 2 factors, Pitx1/2 (37), and four potential binding sites for the specific pituitary factor neural zinc finger factor-1 (NZF-1) (38). Other consensus binding sites for some factors known to regulate eukaryotic promoters were also found. Notably, two potential activator protein-1 (AP-1) recognition sites were also present.

To investigate the exon TERP-1 position, we carried out a Southern blot analysis of the rat genomic DNA using two different 32P-labeled probes corresponding either to the cloned genomic fragment RDL-1 (Fig. 2Go) or to exons 4–5 of rat ER{alpha} cDNA (nucleotides 908-1331). Rat genomic DNA was digested with various restriction endonucleases, separated on agarose gels, and hybridized under high stringency conditions with the two probes. As shown in Fig. 4AGo, it clearly appears that the fragments of the HindIII and BamHI digests hybridized both probes. These results indicate that the TERP-1-exon is localized within intron 4 of ER{alpha} gene (Fig. 4BGo).



View larger version (49K):
[in this window]
[in a new window]
  		Figure 4. Southern blot analysis. A, Rat genomic DNA digested with restriction endonucleases SmaI (lane 1), HindIII (lane 2), EcoRV (lane 3), EcoRI (lane 4), and BamHI (lane 5) was hybridized under high stringency conditions with either 871-bp fragment (RDL-1) containing the 5'-flanking region of the TERP-1 gene (probe A) or rat ER{alpha} cDNA (exons 4–5) (probe B). The arrows show that two hybridized bands of 6,000 and 10,000 bp, respectively, map both TERP-1 and ER{alpha} cDNAs. The positions of the ladder are indicated in kilobase pairs on the right. B, Schematic representation of the rat ER{alpha} mRNA, the intron 4 structure and of the promoter region controlling the expression of the TERP-1. ER{alpha} exons are represented as open boxes and are numbered with arabic numbers. The presence of additional exon TERP-1 within promoter is marked as a dashed box. Hashed line represents intron 4 sequence that has not been analyzed. The continuous line within the intron symbolizes the TERP-1 promoter.

 
Functional analysis of the 5'-flanking region of the TERP-1 exon
To determine whether the genomic region identified above can really function as a promoter, the fragment RDL-4 straddling the region upstream the putative translation start site of TERP-1 was subcloned and inserted into a luciferase reporter vector (pGL3). This construct was designated RDL-4-LUC. Cell lines of pituitary ({alpha}T3–1 and MMQ) and nonpituitary (MCF-7 human mammary tumor cells and CHO-K1 cells) origins were transiently transfected with RDL-4-LUC. Comparisons were made with the {alpha}-LUC promoter (containing the -846/+50 promoter fragment of human {alpha}- subunit) as a pituitary-specific control, and the rainbow trout ER{alpha} gene promoter inserted into a pGL2 vector (containing the -40/-248 ER{alpha} promoter fragment of rainbow trout, rtER{alpha}) as an ER{alpha} control promoter, and the promoterless control luciferase vectors, pGL3 and pGL2, which served as controls for basal levels of luciferase expression. Our results indicate that the 2.0-kb 5'-flanking region of exon TERP-1 was transcriptionally active in all the tested cell lines (Fig. 5Go). Transcriptional activity of RDL-4-LUC in MMQ cells (which contain endogenous ER{alpha} and Pit-1) increased by 30.2 ± 3.0-fold relative to the activity of the promoterless vector, pGL3 (n = 6, P < 0.0001). In contrast, RDL-4-LUC activity in the {alpha}T3–1 gonadotrope cell line (which expresses endogenous ER{alpha} but not Pit-1) showed only a 5.3 ± 0.5-fold increase (n = 6, P < 0.0001) above the pGL3 control. When RDL-4-LUC was transfected into the nonpituitary cell line MCF-7 (ER{alpha} positive), luciferase activity was 8.6 ± 0.4-fold (n = 3, P < 0.0001) higher than pGL3 controls. Transfection of TERP-1 promoter led to a modest but significant 1.8 ± 0.2-fold (n = 3, P < 0.01) increase in luciferase activity over the parent vector pGL3-LUC in CHO cells (ER{alpha} negative) (Fig. 5Go). The activity of the TERP-1 promoter was 25% higher than that of promoter {alpha}LUC in both MMQ and {alpha}T3–1 (n = 3). Comparatively, the activity of the well-characterized rtER{alpha} promoter was 3.0 ± 0.3-fold (n = 2) in MCF-7 cells and 4-fold in MMQ cells. TERP-1 promoter therefore directed luciferase gene expression in the MCF-7 and {alpha}T3–1 cells at a level superior to that of both the rtER{alpha} and {alpha}-LUC. Transient transfection studies thus indicate that the putative TERP-1 promoter is functional. In addition, the data show that maximal cell-specific expression was conferred in lactotrope cells, suggesting that regulatory elements for cell-specific expression are present within the studied promoter sequence.



View larger version (13K):
[in this window]
[in a new window]
  		Figure 5. Expression of TERP-1 promoter construct in lactotrope cell line (MMQ), mouse gonadotrope cell line ({alpha}T3–1), CHO cell line, and human breast tumor cell line (MCF-7). The RDL-4 region of TERP-1 (2.0 kb) was fused to the pGL3 vector (TERP-1-LUC) or to the promoterless and enhancerless pGL3 luciferase (basic) vector were transiently expressed into MMQ, {alpha}T3–1, CHO, and MCF-7 cells. The luciferase activity of TERP-1-LUC was normalized with ß-galactosidase activities. Results are expressed as fold activity over the activity of the basic vector. Each value represents the mean of three to six independent experiments. The rtER transfected into {alpha}T3–1 and MCF-7 cells as a reference control plasmid yielded a mean activity of 3.0–4.0. * P < 0.0001, ** P < 0.01, compared with pGL3 control.

 
Characterization of TERP-1
Fig. 6AGo shows the in vitro translation protein products from ER{alpha} and TERP-1 cDNA. [35S]-methionine-labeled protein products migrated at the expected sizes of 66 kDa for WT-ER and 24 and 22 kDa for TERP-1. The same pattern was observed when TERP-1 and ER{alpha} were cotranslated in a 1:1 ratio. To identify TERP-1 in vivo, Western blot analyses were performed using F3 (35), a monoclonal antibody specifically directed against the F-domain of ER{alpha}. Notably, a protein of approximately 24 kDa was detected in whole-cell extracts of pituitaries from 28-d-old PMSG-primed female rats, a treatment inducing a proestrous-like (21) (Fig. 6BGo, lane 1). The 24-kDa protein was not detected in control rats (lane 2) or by ER 715, a polyclonal antibody directed against the D domain of ER{alpha} (39) (data not shown). The 32- to 40-kDa bands observed in these studies were nonspecific because they were also present in the spleen used as a negative ER{alpha} control. Consequently, the data strongly suggest that the in vivo translation product of TERP-1 mRNA is the 24-kDa protein. Consequently, the methionine 393 of ER{alpha} is the apparent translation initiation site of TERP-1. The 22-kDa protein observed in in vitro-translated preparations is presumably translated from a second in-frame start codon corresponding to methionine 401, leading in vitro to the production of two alternative forms of the TERP-1 (Fig. 6AGo).



View larger version (27K):
[in this window]
[in a new window]
  		Figure 6. In vivo and in vitro TERP-1 expression. A, Analysis of TERP-1 protein products was performed using in vitro-translated proteins synthesized from WT-ER, TERP-1 cDNAs alone, or cotranslated with WT-ER. Proteins were labeled with (35S) methionine for 2 h, analyzed by a 13% SDS-PAGE, and autoradiographed. Lane 1, WT-ER; lane 2, TERP-1; and lane 3, cotranslation of WT-ER with TERP-1. Arrows indicate the positions of 66-, 24-, and 22-kDa proteins, respectively. B, Analysis of TERP-1 products was performed in vivo in pituitary whole-cell extracts prepared from PMSG-primed 28-d-old female rats. Proteins were separated by a 13% SDS-PAGE. ER{alpha} and TERP-1 were detected by Western blotting using the ER{alpha}-monoclonal antibody F3 (35 ). The bands of 32 and 40 kDa are nonspecific and are indicated by asterisks. Lane 1, PMSG-primed pituitary; lane 2, control pituitary. Arrows indicate the positions of 66- and 24-kDa proteins, respectively.

 
TERP-1 inhibits WT-ER DNA binding activity in vitro
To investigate the molecular mechanism underlying ER{alpha} and TERP-1 interaction, we examined the possibility for TERP-1 to specifically modulate the ability of WT-ER to bind to its cognate DNA-binding motif, a consensus ERE, using EMSAs. The ERE used in these experiments contained a 35-bp palindromic sequence identified in the Xenopus vitellogenin A2 promoter (33). ER{alpha} and TERP-1 proteins were produced by in vitro translation, incubated with a 32P-labeled double-stranded ERE oligonucleotide and analyzed in EMSAs. The translational efficiencies of the proteins were determined to be equivalent by performing parallel reactions in the presence of (35S) methionine (data not shown). As expected, the WT-ER translated in lysate was able to form specific ER-ERE complexes, which were completely displaced with an excess of unlabeled ERE. Neither an excess of ARE nor an excess of GRE inhibited WT-ER binding (Fig. 7AGo). Control experiments showed that TERP-1 did not bind this element (data not shown).



View larger version (53K):
[in this window]
[in a new window]
  		Figure 7. In vitro-translated TERP-1 impairs ability of WT-ER to bind its cognate DNA sequence. EMSA analysis was used to study the effect of in vitro-translated ER{alpha} and/or ER/TERP-1 on DNA binding. 32P-labeled ERE probes were incubated in the presence of E2 (10 nM) with WT-ER and/or TERP-1 cotranslated in vitro. The DNA-protein complexes were separated on a 6% nondenaturing gel as described in Material and Methods. Arrows point to the specific complex formed between the WT-ER and ERE. A, Lane 1, free probe; lane 2, unprogrammed lysate; lane 4, WT-ER. Specificity of WT-ER binding was controlled by addition of a 100-fold excess of unlabeled GRE, ARE, or ERE (lanes 5–7, respectively). Lane 3 shows that COUP-TF did not bind ERE. B, Lane 1, WT-ER; lanes 2–5, lysates with a constant amount of the WT-ER and increasing amounts of the TERP-1 cotranslated together in ratios (TERP-1/ER) of 0.5:1.0, 1:1, 5:1, and 10:1, respectively. To maintain constant level of plasmid, control lysate transcribed and translated with vector alone was added when necessary. C, EMSA analysis was performed as in B but proteins were synthesized by in vitro cotranslation in molar ratios of TERP-1 and WT-ER (TERP-1/ER) varying from 0 to 10 (0.5:1.0, 1:1, 5:1, and 10:1, lanes 1–5).

 
In a first set of experiments, a constant amount of WT-ER was cotranslated with increasing amounts of TERP-1. As the amount of TERP-1 increased, the binding of WT-ER to its response element was progressively inhibited, as shown by a decrease in the intensity of the ER-ERE complexes (Fig. 7BGo). On the other hand, when increasing ratios of WT-ER were used, the binding activity was maintained constant (data not shown).

The ability of TERP-1 to antagonize the binding of WT-ER to ERE was further evaluated in experiments in which different molar ratios (0.5:1.0, 1:1, 5:1, and 10:1) of TERP-1 and WT-ER were used. At a 1:1 molar ratio of the two plasmids, TERP-1 highly inhibited the formation of WT-ER-ERE complexes (Fig. 7CGo, lane 3). These results demonstrate that TERP-1 prevents WT-ER from binding to its response element under standard experimental conditions when cotranslated with ER{alpha}.

TERP-1 inhibits E2 binding to WT-ER in vitro
To ascertain whether TERP-1 was able to interfere with E2 binding to WT-ER, binding studies were performed on in vitro-translated TERP-1, ER{alpha} or cotranslated TERP-1 and ER{alpha}. As shown in Fig. 8AGo, when TERP-1 and WT-ER were cotranslated in a ratio of 5:1, the presence of TERP-1 inhibited by 90% the E2 specific binding to WT-ER. The number of binding sites was very low for TERP-1 translated alone and represented only 10% of the capacity of WT-ER binding. Scatchard plots revealed a dissociation constant (Kd) of 0.18 nM for ER{alpha} similar to that measured in uterine cell extracts (Kd = 0.22 nM) whereas TERP-1 showed a Kd of 0.8 ± 0.27 nM (n = 3) (Fig. 8BGo).



View larger version (15K):
[in this window]
[in a new window]
  		Figure 8. Binding properties of ER/TERP-1. Binding analysis was performed using in vitro-translated proteins synthesized from WT-ER, TERP-1 cDNAs alone, or cotranslated with WT-ER. Proteins were incubated with (2, 4, 6, 7, -3H)-17ß-E2. Specific binding was determined in the absence (total binding) or presence of a 200-fold excess of unlabeled E2 (nonspecific binding). A, Plots of ER{alpha} and ER{alpha}/TERP-1 cotranslated are shown. B, Scatchard analysis of the data for WT-ER and TERP-1 translated alone was performed. The figure is representative of three separate experiments.

 
TERP-1 effects on WT-ER-stimulated activity of ERE-reporter gene constructs
To determine whether TERP-1 interferes with ER-mediated transcriptional activation, transient cotransfection studies were performed in CHO cell lines using expression vectors for the WT-ER and TERP-1 and ER-responsive reporter plasmids. Preliminary experiments were conducted to determine the maximal level of WT-ER expression vector to transfect to avoid sequestering. WT-ER showed maximal activity over a concentration range between 10 ng and 50 ng/well of transfected plasmid (data not shown). When the amount of ER{alpha} expression vector increased up to 250 ng/well, only a 20% decrease of the maximal activity was observed. TERP-1 had no constitutive or E2-stimulated effects on reporter gene transcription when expressed alone in the CHO cell line (data not shown).

WT-ER plasmid was cotransfected with TERP-1 at different molar ratios using various ERE reporter genes: (ERE)2-tk-CAT, p(ERE)2-TATA-CAT, ERE-tk-LUC, and ERE-SV-LUC. Data displayed in Fig. 9AGo show that TERP-1 inhibits E2-dependent transcription of all ERE gene reporters tested. When WT-ER and TERP-1 were coexpressed at a molar ratio 5:1, E2-dependent transcription from the (ERE)2-TATA-CAT reporter was inhibited by 48 ± 4% (n = 3). Coexpression of a 10-, 15-, or 20-fold molar excess of TERP-1 over WT-ER did not induce a further decrease. A similar pattern of response was reproduced with the (ERE)2-tk-CAT reporter. Inhibition was even more marked on simple ERE-tk-LUC or ERE-SV-LUC.



View larger version (35K):
[in this window]
[in a new window]
  		Figure 9. Inhibition of hormone-stimulated ER{alpha} or AR transcription by TERP-1. A, CHO cells were transfected with WT-ER and TERP-1 expression vectors at the ratios indicated below the figure and (ERE)2-tk-CAT (•), (ERE)2-TATA-CAT ({blacksquare}), ERE-tk-LUC ({diamondsuit}), or ERE-SV-LUC ({blacktriangleup}) plasmids. Twenty-five nanograms WT-ER expression vector and 25–500 ng/dish of TERP-1 expression vector were employed. The total amount of plasmid was equalized in each well at 4.5 µg using the empty expression vector. Transfected cells were treated for 48 h with 10 nM E2. LUC or CAT activities were normalized for protein. The results are expressed vs. ER{alpha} activity, which was considered 100. The TERP-1 plasmid by itself with or without hormone did not induce reporter gene activity (data not shown). The values correspond to the average of two to three independent experiments that showed no significant difference. B, MDA-MB-231, which are AR positive, were transfected with (ARE)4-tk-LUC (100 ng) and increasing concentrations (20–60 ng) of TERP-1 plasmid. The total amount of plasmid was equalized at 250 ng in each well using the empty expression vector. The cells were treated or not with DHT and harvested after 36 h. LUC activity was normalized for protein. The activity obtained with (ARE)4-tk-LUC in presence of DHT was considered as 100. Diagram represents the mean ± SEM for one experiment in triplicate and is representative of other transfections performed.

 
Surprisingly, the extent of the decrease was always more important in the absence than in the presence of the hormone, suggesting that the TERP-1-induced decrease was not E2 dependent. Therefore, we checked whether these inhibitions could be the result of nonspecific effects on the transcriptional machinery. Cotransfection of TERP-1 and ER{alpha} significantly repressed the activity of either tk (Fig. 10AGo) or CMV promoters (Fig. 10BGo), resulting in a dose-dependent decrease of promoter activities. Therefore, the TERP-1/ER{alpha} complex could have a more general inhibitory effect toward transcription.



View larger version (56K):
[in this window]
[in a new window]
  		Figure 10. Inhibition of tk-CAT or CMV-ßGAL activities by TERP-1/ER. CHO cells were transfected with WT-ER and TERP-1 expression vectors at the ratios indicated below the figure and either tk-CAT (A) or CMV-ßGAL (B) plasmids in the presence (hashed bars) or not (open bars) of E2. CAT or GAL activities were determined in cell extracts after 48 h and normalized for protein. Diagram represents the mean ± SEM for one experiment in triplicate and is representative of two independent experiments.

 
To assess the specificity of the TERP-1 inhibitor effect, we tested the effect of TERP-1 on endogenous AR in MDA-MB 231 cell lines. Cotransfection of an (ARE)4-tk-LUC reporter construct and 20–100 ng of TERP-1 plasmid showed that the activation of AR was efficiently inhibited by TERP-1 (Fig. 9BGo). Similar results were observed in CHO cell lines using hAR expression vector, suggesting a repressing activity toward various nuclear receptors.

Qualitative analysis of ER{alpha} subcellular localization in stable cell lines expressing TERP-1
To follow the subcellular localization of the ER{alpha} in living cells expressing TERP-1, we tagged the amino terminus of the rat ER{alpha} with the S65T variant of GFP (pGFP-ER). The TERP-1 vector was stably transfected in ER{alpha} negative CHO cell lines (CHO-TERP-1). In a first set of experiments, original CHO cells were transfected with either pGFP or pGFP-ER to verify the localization of GFP alone or GFP-ER in the absence of TERP-1. As expected, fluorescence in CHO cells transiently transfected with GFP alone appears uniformly distributed throughout the cytoplasm and nucleus (Fig. 11AGo) whereas CHO cells transfected with pGFP-ER show green fluorescence restricted to the nucleus (Fig. 11BGo). Thus, GFP-ER localization is consistent with previously reported data for the endogenous ER{alpha}. The same distribution of ER{alpha} was observed in CHO cells using immunocytochemistry clearly demonstrating the validity of our conditions. These data also show that the presence of the GFP tag has little effect on the localization of WT-ER. In the second set of experiments, TERP-1 localization was assessed by immunocytochemistry in CHO-TERP-1 cells. Results (not shown) confirmed those observed after transient transfection by Resnick et al. (24) demonstrating that TERP-1 is present in both the cytoplasm and nucleus of cells. Then CHO-TERP-1 cells were transiently transfected with pGFP-ER. Surprisingly, part of the fluorescent GFP-ER receptor was diffusely distributed throughout the cytoplasm (Fig. 11Go, D through F). As a control, GFP-COUP-TFI transiently introduced in the stable line CHO-TERP-1 showed an exclusive nuclear localization (Fig. 11FGo). Similar data were observed with pGFP-GR (data not shown). These results suggest a retentional activity of TERP-1 toward the ER{alpha} in the cytoplasm.



View larger version (54K):
[in this window]
[in a new window]
  		Figure 11. Effect of TERP-1 on the intracellular distribution of GFP-ER{alpha}. Original CHO cell lines and CHO-TERP-1 cell lines stably expressing TERP-1 were transiently transfected with either pGFP, pGFP-ER, or GFP-COUP-TFI and cultured on coverslips overnight. The next day media were changed, and the cells were maintained with complete serum for 24 h. Cells were visualized by epifluorescence using a standard FITC filter set. Frame A shows fluorescence of pGFP alone; frame B shows the exclusive nuclear localization of GFP-ER. In frames C-E, three fields show the distribution of pGFP-ER in CHO-TERP-1 cell lines transiently transfected with pGFP-ER. As a control, frame E shows CHO-TERP-1 cell lines transfected with GFP-COUP-TFI.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
In this report, we have hypothesized that the TERP-1 variant transiently expressed in lactotrope cells could be generated using an alternative promoter. Indeed, we have characterized for the first time an alternative intronic promoter in the rat ER{alpha} gene, which controls the expression of TERP-1 mRNA, and analyzed the promoter region. We also provide functional evidence demonstrating that TERP-1 can antagonize ER{alpha}.

We have isolated three overlapping genomic clones encoding 2.5 kb in the flanking regions upstream and downstream the sequence TERP-1. Genomic Southern blot analysis showed that TERP-1 sequence corresponds to a noncoding leader exon localized within the intron 4 of ER{alpha} gene. Therefore, we demonstrate that TERP-1 is alternatively spliced from an intronic promoter within the ER{alpha} gene. The existence of a tight transcriptional control of the expression of the TERP-1 mRNA through this promoter could explain the fluctuations observed in the pattern of expression during the estrous cycle (21, 22). The triggering mechanism for the choice between the "classical" alternate exons 1 (10, 11, 12) and exon TERP-1 remains unknown. However, transient and cell-specific expression of the variant suggests that a distinct activating signal is necessary to drive the TERP-1 promoter during the estrous cycle.

Analysis of the 5'-flanking region of the gene showed a major transcription initiation site and revealed the presence of a TATA box motif residing near the transcription initiation site. The presence of consensus TATA sequences is unusual among the steroid receptor family. Consensus binding sites for the POU-homeodomain transcription factor, Pit-1, may play an important role in the cell-specific expression of TERP-1. In addition, the promoter region contains three consensus motifs for bicoid-related homeobox 1 and/or 2 factors (Pitx-1/2), and four potential binding sites for the pituitary NZF-1, which have been shown to be involved in pituitary-specific expression and development (36, 37, 38). The presence of these sites has not been yet described in the promoter of nuclear receptor and could suggest an involvement of TERP-1 in differentiation of the lactotrope and somatotrope lineages. Of note, expression of Pit-1 and of the latter three factors is maintained in the mature pituitary gland (38, 40). Most E-responsive genes, including the classical promoter of the ER{alpha} gene, lack the canonical ERE and contain one or more imperfect EREs or multiple copies of ERE half-sites. The presence of consensus palindromic ERE sequence suggests that estrogen may have a direct effect on the TERP-1 promoter and play a role in the regulation of TERP-1. Several half-EREs are also present in the TERP-1 promoter sequence and could be active individually or cooperatively (41, 42, 43). Moreover, cooperative interactions between the Pit-1 protein and the ER{alpha} have been shown to be important in the regulation of PRL gene transcription (44) and could also be involved in the regulation of such a promoter.

The 2.0-kb promoter is functional and directs strong basal expression in pituitary cells. It is transcriptionally more active in lactotrope than in gonadotrope cell lines. The transcriptional activity of TERP-1 promoter is even slightly higher in MCF-7 cells than in gonadotrope cell lines. The higher basal activity observed in MMQ, compared with {alpha}T3–1, cells could be explained by the absence of any endogenous Pit-1 in the latter cells. In addition, the promoter activity in ER{alpha}-positive {alpha}T3–1 and MCF-7 cells is probably attributable to the presence of the EREc. These results once again agree with the cell-specific distribution of TERP-1 mRNA previously observed in rat lactotrope cells (21); we also observed a high expression of TERP-1 mRNA in lactotrope cell lines as described by Schreihofer et al. (45). However, in contrast to their findings, we did not observe any expression in gonadotrope {alpha}-T3 cell line. Taken together, these results suggest a cell-specific pituitary regulation of the TERP-1 promoter.

The expression of the truncated protein correlates with our previous data showing TERP-1 mRNA expression in pituitary cells on proestrous day (21). Studies from Friend et al. (22) and Mitchner et al. (23) have also described a 20- to 24-kDa TERP-1 protein in female estrogen-treated pituitaries. In this report, we characterized a 24-kDa protein in a proestrous-like situation. The methionine 393 is the apparent translation initiation site of TERP-1 in vivo. In contrast, Mitchner et al. (23) have suggested that the methionine 426 could be the initiation site of TERP-1 translation. The reasons for this discrepancy may be attributed to different experimental procedures.

Previous experiments have clearly demonstrated heterodimerization between ER{alpha} and TERP-1 both in vitro (24) and in vivo (25). Although TERP-1 is unable to bind to an ERE by itself, our experiments clearly showed that an equimolar concentration of TERP-1 effectively inhibits WT-ER binding to EREc, confirming the recent findings of Resnik et al. (24). Because contradictory results have been published concerning transcriptional effect of TERP-1 on WT-ER transactivation (24, 26, 27), we reexamined the transcriptional control exerted by TERP-1 on WT-ER stimulated activity of various ERE-reporter gene constructs. We used reporters containing various promoters to compare the effects of TERP-1/ER{alpha} heterodimer on single or multiple tandem copies of EREs. In this work, we confirm in CHO cells that TERP-1 is able to repress ER{alpha}-mediated activation. In addition, the inhibition varies according to the strength of the promoter and the presence of single or double EREc.

To obtain a transrepression mediated by TERP-1, molar ratios 5:1 (TERP-1/ER{alpha}) of the expression plasmids are required. No effect was obtained at lower ratios (equimolar or submolar ratios). However, the biphasic action of TERP-1 on transcription obtained by Schreihofer et al. (26) was not observed under the conditions employed in the present study, action involving transcriptional repression only when WT-ER and TERP-1 are coexpressed at high concentrations of transfected vectors. The reasons for these discrepancies could be due either to the total amount of expression plasmids employed or to the two involved cell types (CHO vs. COS cell lines). On the other hand, Mitchner et al. (27) were unable to demonstrate any effect of TERP-1 overexpression in GH3 cells on either PRL or a triple-vitellogenin ERE/luciferase construct. Under the conditions used in the present study, maximal transcriptional activity required approximately 25 ng ER{alpha} expression plasmid. At the efficient ratio 5:1, the TERP-1 plasmid represents only 125 ng. Transfection of 10-fold higher levels of WT-ER resulted in no significant change in transcriptional activity. Therefore, it is unlikely that the transcriptional repression was due to sequestering of cellular factors. In addition, these data were extended using cell lines expressing normal AR and showed similar results. All together, the data demonstrate that TERP-1 exerts full dominant-negative activity. Because such a transcriptional repression was also observed using ERß (24), TERP-1 therefore might be a natural antagonist of nuclear receptors. Furthermore, an important finding of this study is that the TERP-1/ER{alpha} heterodimer can repress the basal activity of all promoters tested to date, suggesting that it could exert a more general inhibitory effect toward transcription.

The exact mechanism responsible for the inhibitory activity of TERP-1 in the presence of ER{alpha} is still unclear. TERP-1 could suppress estrogenic action by squelching the WT receptor, preventing either ER binding to DNA, WT-ER transcriptional activity, or nuclear transport. Our findings concerning the inhibition of E2 binding to WT-ER by TERP-1 are also of great importance and contribute to the repressive effect that we observed. ER{alpha} partition between the cytoplasmic and nuclear compartments in living cells stably transfected with TERP-1 strengthens these results. TERP-1 could also interact with the plasma membrane ER, a receptor for which strong evidence of its existence has now emerged (46), and able to mediate estrogen-stimulated PRL release from rat pituitary tumor cells (47). Another possible hypothesis would be that TERP-1/ER heterodimer may inhibit transcription by squelching receptor coactivators and/or general transcription factors. TERP-1 was recently shown to be able to bind SRC-1, steroid receptor coactivator (24). Therefore, TERP-1 suppression of ER{alpha} transcription could occur both by formation of inactive heterodimers unable to bind the hormone and by competition for transcription factors.

The results presented in this report provide important insights into the pituitary regulation of ER{alpha} gene expression during the reproductive cycle. Because TERP-1 expression is tightly regulated throughout the estrous cycle and is controlled through a specific promoter, we propose that expression of the TERP-1 isoform represents a novel mechanism for the modulation of lactotrope hormonal responsiveness through protein-protein interaction. The relative ratio of the ER{alpha} and TERP-1 proteins in vivo may be a critical factor regulating lactotrope cell responsiveness to estrogenic stimuli. At proestrus, the level of TERP mRNA is 45% of the normal ER{alpha} mRNA in the pituitary gland (21). TERP-1 could modulate ER{alpha} activity, exerting a protective role against the high level of E2 on proestrus suppressing E2 binding to WT-ER. On the other hand, our results do not rule out that an E2-independent pathway could regulate target genes via the TERP-1/ER{alpha} heterodimer. Lactotrope cells exhibit unique features that distinguish them from other anterior pituitary cells: only lactotrope cells indeed exhibit a change in proliferation rates through out the estrous cycle with a peak on the estrous morning (48). Moreover, the progesterone receptor induction, a typical estrogenic response, is not observed in lactotrope cells (49, 50). Therefore, TERP-1 could exert either a role in the control of lactotrope proliferation or a repressing role toward the progesterone receptor.

In conclusion, the interest of our findings is that TERP-1 is a natural variant expressed in a cell type-specific manner, transiently during the proestrous stage and regulated by an intronic promoter within rat ER{alpha} gene. The identification and partial characterization of TERP-1 promoter is a first step in the study of the mechanisms involved in the transcriptional regulation of ER{alpha}/TERP-1 expression and their role in the pituitary gland. Further study is required to determine the precise elements involved in tissue-specific expression as well as the identification of factors that bind to these response elements.


   Acknowledgments
 
We are very grateful to Dr. R. A. Maurer (HSU, Oregon) for the {alpha}-LUC promoter (containing the -846/+50 promoter fragment of human {alpha}-subunit), Dr. P. Chambon (IGBMC, University L. Pasteur, Strasbourg, France), and Dr. J. D. Parlow (National Hormone and Pituitary Program, Harbor-UCLA Medical Centre) for the gifts of monoclonal antibody F3 and polyclonal antibody ER 715, respectively. We thank Dr. C. H. Le Dréan (University of Rennes 1, France) and Dr. C. L. Hew (Research Institute, Toronto, Canada), for kindly providing us with the p(ERE)2 [(ERE)2-TATA-CAT]. We thank Dr. W. Wahli (IBM, University of Lausanne) for the ERE-tk-LUC, Dr. O. Jänne (University of Helsinki) for the p(ARE)4-tk-LUC construct and hAR expression vector, Dr. Lefebvre (INSERM U 459, University of Lille) for GFP-GR, and Dr. M. Muramatsu (University of Tokyo, Japan) for the rat ER cDNA. We thank our colleagues Dr. F. Pakdel for the (ERE)2-tk-CAT and the trER promoter and Dr. G. Salbert for the plasmid GFP-COUP-TF. We also thank Dr. P. Mellon (University of California, La Jolla, CA) for supply of {alpha}T3-1 pituitary cells. This work was supported by grants from the Association Recherche pour le Cancer (Agreement No. 9898) and from the Fondation Langlois.


   Footnotes
 
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank Data Bank with accession number AF169237.

Abbreviations: AF-2, Transcriptional function; AP-1, activator protein-1; ARE, androgen-responsive element; GRE, glucocorticoid-responsive element; GFP, green fluorescent protein; ATG, translational start codon; CHO, Chinese hamster ovary; ERE, E-responsive element; NZF-1, neural zinc finger factor-1; RDL, rat DNA library; TERR-1, truncated ER product-1; pcDNA, plasmid and carrier DNA; TERP, truncated ER product; WT, wild-type; WT-ER, wild-type ER.

Received January 25, 2001.

Accepted for publication May 28, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Kuiper GGJM, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA 1996 Cloning of a novel estrogen receptor expressed in rat prostate and ovary. Proc Natl Acad Sci USA 93:5925–5930[Abstract/Free Full Text]
  2. Parker MG 1995 Structure and function of estrogen receptors. Vitam Horm 51:267–287[Medline]
  3. Tora L, White J, Brou C, et al. 1989 The human estrogen receptor has two independent nonacidic transcriptional activation functions. Cell 59:477–487[CrossRef][Medline]
  4. Tzukerman MT, Esty A, Santiso-Mere MD, et al. 1994 Human estrogen receptor transactivational capacity is determined by both cellular and promoter context and mediated by two functionally distinct intramolecular regions. Mol Endocrinol 8:21–30[Abstract]
  5. Pierrat B, Heery DM, Chambon P, Losson R 1994 A highly conserved region in the hormone-binding domain of the human estrogen receptor functions as an efficient transactivation domain in yeast. Gene 143:193–200[CrossRef][Medline]
  6. McKenna NJ, Lanz RB, O’Malley BW 1999 Nuclear receptor coregulators: cellular and molecular biology. Endocr Rev 20:321–344[Abstract/Free Full Text]
  7. Dowsett M, Daffada A, Chan CM, Johnston SR 1997 Oestrogen receptor mutants and variants in breast cancer. Eur J Cancer 33:1177–1183
  8. Murphy LC, Dotzlaw H, Leygue E, Douglas D, Coutts A, Watson PH 1997 Minireview: estrogen receptor variants and mutations. J Steroid Biochem Mol Biol 62:363–372[CrossRef][Medline]
  9. Hopp TA, Fuqua SAW, Borg A 1998 Estrogen receptor variants. J Mam Gland Biol Neoplasia 3:73–83
  10. Shupnik MA, Pitt LK, Soh AY, Anderson A, Lopes MB, Laws Jr ER 1998 Selective expression of estrogen receptor {alpha} and ß isoforms in human pituitary tumors. J Clin Endocrinol Metab 83:3965–3972[Abstract/Free Full Text]
  11. Grandien K, Berkenstam A, Gustafsson J-A 1997 The estrogen receptor gene: promoter organization and expression. Int J Biochem Cell Biol 29:1343–1369[CrossRef][Medline]
  12. Kato J, Hirata S, Koh T, Yamada-Mouri N, Hoshi K, Okinaga S 1998 The multiple untranslated first exons and promoters system of the estrogen receptor gene in the brain and peripheral tissues of the rat and monkey and the developing rat cerebral cortex. J Steroid Biochem Mol Biol 65:281–293[CrossRef][Medline]
  13. Flouriot G, Griffin C, Kenealy M, Sonntag-Buck V, Gannon F 1998 Differentially expressed messenger RNA isoforms of the human estrogen receptor {alpha} gene are generated by alternative splicing and promoter usage. Mol Endocrinol 12:1939–1954[Abstract/Free Full Text]
  14. Murphy LC, Dotzlaw H, Leygue E, Coutts A, Watson P 1998 The pathophysiological role of estrogen receptor variants in human breast cancer. J Steroid Biochem Mol Biol 65:175–180[CrossRef][Medline]
  15. Yang J, Liu A, Chiou SK, Guzman R, Nandi S 2000 Estrogen receptor variants are present in many normal tissues. Int J Mol Med 5:223–227[Medline]
  16. Erenburg I, Schachter B, Mira Y, Lopez R, Ossowski L 1997 Loss of an estrogen receptor isoform (ER{alpha} {delta}3) in breast cancer and the consequences of its reexpression: interference with estrogen-stimulated properties of malignant transformation. Mol Endocrinol 11:2004–2015[Abstract/Free Full Text]
  17. Claret F-X, Chapel S, Garcés J, et al. 1994 Two functional forms of the xenopus laevis estrogen receptor translated from a single mRNA species. J Biol Chem 269:14047–14055[Abstract/Free Full Text]
  18. Barraille P, Chinestra P, Bayard F, Faye JC 1999 Alternative initiation of translation accounts for a 67/45 kDa dimorphism of the human estrogen receptor ER{alpha}. Biochem Biophys Res Commun 257:84–88[CrossRef][Medline]
  19. Roisne S, Duval J, Thieulant ML 1992 Multiple forms of affinity labeled estrogen receptors in rat distinct pituitary cells. Endocrinology 131:1503–1510[Abstract]
  20. Friend KE, Ang LW, Shupnik MA 1995 Estrogen regulates the expression of several different estrogen receptor mRNA isoforms in rat pituitary. Proc Natl Acad Sci USA 92:4367–4371[Abstract/Free Full Text]
  21. Demay F, Tiffoche C, Thieulant M-L 1996 Sex- and cell-specific expression of an estrogen receptor isoform in the pituitary gland. Neuroendocrinology 63:522–529[Medline]
  22. Friend KE, Resnick EM, Ang LW, Shupnik MA 1997 Specific modulation of estrogen receptor mRNA isoforms in rat pituitary throughout the estrous cycle and in response to steroid hormones. Mol Cell Endocrinol 131:147–155[CrossRef][Medline]
  23. Mitchner NA, Garlick C, Ben-Jonathan N 1998 Cellular distribution and gene regulation of estrogen receptors {alpha} and ß in the rat pituitary gland. Endocrinology 139:3976–3983[Abstract/Free Full Text]
  24. Resnik EM, Schreihofer DA, Periasamy A, Shupnik MA 2000 Truncated estrogen receptor product-1 suppresses estrogen receptor transactivation by dimerization with estrogen receptors {alpha} and ß. J Biol Chem 275:7158–7166[Abstract/Free Full Text]
  25. Tiffoche C, Vaillant C, Thieulant M-L 1999 Novel promoter regulating a variant of the estrogen receptor transiently expressed in rat pituitary. Proceedings of the 81st Annual Meeting of The Endocrine Society, San Diego, CA, p 523 (Abstract)
  26. Schreihofer DA, Resnick EM, Soh AY, Shupnik MA 1999 Transcriptional Regulation by a naturally occurring truncated rat estrogen receptor (ER), truncated ER product-1 (TERP-1). Mol Endocrinol 13:320–329[Abstract/Free Full Text]
  27. Mitchner NA, Garlick C, Steinmetz RW, Ben-Jonathan N 1999 Differential regulation and action of estrogen receptors {alpha} and ß in GH3 cells. Endocrinology 140:2651–2658[Abstract/Free Full Text]
  28. Judd UM, Login IS, Kovacs K, et al. 1988 Characterization of the MMQ cell, a prolactin-secreting clonal cell line that is responsive to dopamine. Endocrinology 123:2341–2350[Abstract]
  29. Scammell JG, Burrage TG, Dannies PS 1986 Hormonal induction of secretory granules in a pituitary cell line. Endocrinology 119:1543–1548[Abstract]
  30. Koike S, Sakai M, Muramatsu M 1987 Molecular cloning and characterization of rat estrogen receptor cDNA. Nucleic Acids Res 15:2499–2513[Abstract/Free Full Text]
  31. Madigou T, Tiffoche C, Lazennec G, Pelletier J, Thieulant M-L 1996 The sheep estrogen receptor: cloning and regulation of expression in the hypothalamo-pituitary axis. Mol Cell Endocrinol 121:153–163[CrossRef][Medline]
  32. Herbommel P, Bourachot B, Yaniv M 1984 Two distinct enhancers with different cell specificities coexist in the regulatory region of polyoma. Cell 39:653–662[CrossRef]