Estrogen Regulates Transcription of the Ovine Oxytocin Receptor Gene through GC-Rich SP1 Promoter Elements
http://www.100md.com
《内分泌学杂志》
Center for Animal Biotechnology and Genomics (J.G.W.F., T.E.S., F.W.B.) and Departments of Animal Science (T.E.S., F.W.B.) and Veterinary Physiology and Pharmacology (S.H.S., F.W.B.), Texas A&M University, College Station, Texas 77843
Abstract
Establishment of pregnancy in ruminants results from paracrine signaling by interferon (IFNT) from the conceptus to uterine endometrial luminal epithelia (LE) that prevents release of luteolytic prostaglandin F2 pulses. In cyclic and pregnant ewes, progesterone down-regulates progesterone receptor (PGR) gene expression in LE. In cyclic ewes, loss of PGR allows for increases in estrogen receptor (ESR1) and then oxytocin receptor (OXTR) gene expression followed by oxytocin-induced prostaglandin F2 pulses. In pregnant ewes, IFNT inhibits transcription of the ESR1 gene, which presumably inhibits OXTR gene transcription. Alternatively, IFNT may directly inhibit OXTR gene transcription. The 5' promoter/enhancer region of the ovine OXTR gene was cloned and found to contain predicted binding sites for activator protein 1, SP1, and PGR, but not for ESR1. Deletion analysis showed that the basal promoter activity was dependent on the region from –144 to –4 bp that contained only SP1 sites. IFNT did not affect activity of the OXTR promoter. In cells transfected with ESR1, E2, and ICI 182,780 increased promoter activity due to GC-rich SP1 binding sites at positions –104 and –64. Mutation analyses showed that the proximal SP1 sites mediated ESR1 action as well as basal activity of the promoter. In response to progesterone, progesterone receptor B also increased OXTR promoter activity. SP1 protein was constitutively expressed and abundant in the LE of the ovine uterus. These results support the hypothesis that the antiluteolytic effects of IFNT are mediated by direct inhibition or silencing of ESR1 gene transcription, thereby precluding ESR1/SP1 from stimulating OXTR gene transcription.
Introduction
MATERNAL RECOGNITION OF pregnancy is the physiological process whereby the conceptus signals its presence to the maternal system and prolongs lifespan of the corpus luteum (CL) (1). Sheep experience uterine-dependent estrous cycles until establishment of pregnancy (2). The estrous cycle is dependent on the uterus, because it releases prostaglandin F2 (PGF) in a pulsatile manner to induce luteolysis during late diestrus. The luteolytic pulses of PGF are produced by the endometrial luminal epithelium (LE) and superficial ductal glandular epithelium (sGE) and are generated by oxytocin binding to oxytocin receptors (OXTR) on those epithelia (3, 4, 5, 6). Endometrial epithelial expression of OXTR is regulated primarily by receptors for progesterone [progesterone receptor (PGR)] and estrogen [estrogen receptor (ESR1)] (7, 8, 9). During proestrus and estrus (d –3 to 0), estrogen from ovarian follicles increases ESR1, PGR, and OXTR expression. During diestrus (d 3–15), progesterone levels increase and act via PGR to "block" expression of ESR1 and OXTR in LE and glandular epithelium (GE) between d 5 and 11 after onset of estrus (10, 11, 12). However, continuous exposure of the uterus to progesterone down-regulates expression of PGR in LE/sGE after d 11 and GE after d 13, allowing for rapid increases in expression of ESR1 on d 12–13 and then OXTR on d 14 in those epithelia (9, 10, 13, 14). ESR1, presumably activated by estrogen from ovarian follicles or possibly growth factors from the stroma, stimulates transcription of the OXTR gene in the ovine endometrium (15, 16, 17, 18). Oxytocin, secreted from as early as d 9 from posterior pituitary and/or the CL, binds to OXTR to induce pulsatile release of luteolytic PGF between d 14–16 (3). In response to four to five luteolytic pulses of PGF over a 25-h period, the CL undergoes functional and structural regression. The systemic loss in progesterone between d 15–16 allows the ewe to return to estrus, complete the 17-d estrous cycle, and experience another opportunity to mate and establish pregnancy.
Interferon (IFNT), the pregnancy recognition signal in ruminants, is expressed by conceptus trophectoderm between d 10 and 21 of gestation (19) and acts in a paracrine antiluteolytic manner on the endometrial LE and sGE to inhibit OXTR gene expression (10, 13, 20). IFNT does not inhibit basal production of PGF, which is higher in pregnant than cyclic ewes, and the conceptus and IFNT do not decrease expression of PTGS2 [prostaglandin-endoperoxide synthase 2 (prostaglandin G/H synthase and cyclooxygenase)], the rate-limiting enzyme in prostaglandin production, in endometrial epithelia of pregnant ewes (5, 21). The antiluteolytic effects of IFNT to inhibit OXTR gene expression in the endometrial LE and sGE have been hypothesized to involve: 1) direct stabilization of epithelial PGR expression that, in turn, extends progesterone inhibition of ESR1 and OXTR expression; 2) direct inhibition or silencing of ESR1 gene transcription that, in turn, indirectly inhibits OXTR gene expression; and 3) direct inhibition of OXTR gene transcription. IFNT does not inhibit PGR loss in the endometrial epithelia (12, 16), but rather appears to indirectly inhibit OXTR gene transcription by silencing ESR1 gene transcription in the endometrial LE (15, 16, 22). Indeed, the 2.7-kb 5' promoter/enhancer region of the ovine ESR1 gene was found to contain four interferon regulatory factor elements (IRFEs) and one interferon-stimulated response element (ISRE) that were functional in binding interferon regulatory factor (IRF)2 (23). IRF2 is a transcriptional repressor that is expressed specifically in the endometrial LE and sGE of the ovine uterus and increases in expression from d 10–16 of pregnant but not cyclic ewes (24). In transfection assays, IFNT inhibited transcriptional activity of the ovine ESR1 promoter, and analyses of sequential 5'-deletion mutants of the ovine ESR1 promoter indicated that the effects of IFNT may be mediated by IRFEs as well as other elements (23). Physiological evidence indicates that estrogen stimulates OXTR gene expression and that IFNT does not directly inhibit OXTR gene expression in the sheep uterus (15, 25). Direct effects of estrogen and IFNT on the ovine OXTR gene have not been studied previously, because the gene had not been cloned. The bovine OXTR gene promoter region was cloned and found to contain an IRFE, ESR1 response element (ERE) half sites, and SP1 sites, and could be transactivated by estrogen if cells were cotransfected with ESR1 and steroid receptor coactivator 1 (26). Curiously, IRF2 overexpression increased activity of the bovine OXTR promoter, but a direct effect of IFNT on promoter activity was not reported. In all mammals, estrogen is considered a key regulator of OXTR gene expression; however, all of the studied OXTR genes lack a complete ERE, suggesting estrogen induction of OXTR gene transcription may be indirect rather than due to direct ESR1/ERE interactions (27).
Our working hypothesis is that estrogen acts indirectly to regulate OXTR gene expression in the endometrium of the ovine uterus and that the antiluteolytic effects of IFNT are manifest on silencing or inhibition of ESR1 gene transcription in LE and sGE subsequent to PGR down-regulation, which then precludes the ability of estrogen or perhaps growth factors to activate ESR1 and stimulate OXTR gene expression. These actions of IFNT prevent formation of OXTR and abrogate uterine release of luteolytic pulses of PGF (4, 8, 23, 28). The objective of the present studies was to determine if estrogen and/or IFNT regulate OXTR promoter activity. The promoter region of the ovine OXTR gene was cloned, sequenced, and analyzed for functional cis-elements mediating effects of steroid hormones and IFNT. Relevant findings are that IFNT does not affect OXTR promoter activity nor inhibit estrogen induction of OXTR promoter activity. Two GC-rich SP1 binding sites, located within 140 bp of the translational start site, regulated basal activity of the promoter as well as hormonal responsiveness to 17-estradiol (E2), ICI 182,780, and progesterone. Thus, estrogen regulation of ovine OXTR promoter activity involves ESR1/SP1 interactions.
Materials and Methods
Cells and reagents
The 2fTGH (parental) and U3A (STAT1-deficient 2fTGH) immortalized cells (29) were maintained in DMEM-F12 medium (Sigma-Aldrich Corp., St. Louis, MO) supplemented with 5% fetal bovine serum (FBS) (Hyclone, Logan, UT) and penicillin/streptomycin sulfate/amphotericin B solution (Invitrogen, Carlsbad, CA). Recombinant ovine IFNT (108 antiviral units/mg) was prepared and assayed as described previously (30). Restriction endonucleases, T4 DNA ligase, T4 DNA kinase, and recombinant human SP1 protein were purchased from Promega (Madison, WI). Vent Taq polymerase (New England Biolabs, Beverly, MA), AmpliTaq polymerase (Applied Biosystems, Foster City, CA), and ExTaq polymerase (Takara, Kyoto, Japan) were used. R5020 was purchased from PerkinElmer Life Sciences (Boston, MA), and E2 was from Sigma-Aldrich Corp. Plasmid DNAs were purified by the alkaline lysis method according to the manufacturer’s instructions (Qiagen, Valencia, CA).
Cloning of the 5' upstream region and luciferase constructs
A 273-bp DNA fragment of the 5' end of the coding region of the ovine OXTR gene was PCR amplified in 25-μl reactions containing 40 ng ovine genomic DNA (a gift from Dr. C. Gill, Texas A&M University, College Station, TX), PCR Optimized Buffer N (Invitrogen), 125 μM dNTPs, 2.5 μM forward primer (5'-GGAGGCGGTCAACGGGAG-3'), 2.5 μM reverse primer (5'-CGTGATGTCCCACAGAAGC-3'), and 1 U AmpliTaq polymerase using an Eppendorf Mastercycler thermocycler with conditions of: 1) 94 C for 5 min; 2) 94 C for 30 sec, 53 C for 30 sec, and 72 C for 30 sec for 35 cycles; and 3) 72 C for 7 min. The product was cloned into pCRII Dual using the TA Cloning kit (Invitrogen). The cloned 273-bp OXTR fragment was excised with EcoRI, random prime labeled with [-32P]dCTP, and used to screen an ovine genomic library cloned in Bluestar (Novagen, Madison, WI), which was kindly provided by Dr. J. C. DeMartini (Colorado State University, Fort Collins, CO). A recombinant pBluestar phagemid DNA, containing approximately 9.8 kb of ovine genomic DNA, was excised from plaque-purified isolates by Cre-Lox excision according to the manufacturer’s instructions. Physical mapping, Southern blotting with the 273-bp OXTR probe, and sequencing with T7 and T3 primers and several OXTR-specific primers were done to identify the OXTR coding and upstream sequences in the clone. The 5' terminus of the coding sequence and the contiguous upstream sequences were sequenced on both strands by the dideoxy chain termination method. The OXTR upstream region was immediately adjacent to the NotI site in the vector multiple cloning site. The upstream region was excised as an 826-bp fragment using the vector NotI and a NotI site at –4 relative to the ATG (+1) of the OXTR coding sequences and cloned in pCRII Dual in both orientations as determined by physical mapping of the unique BglII site at –144. The OXTR(–830/–4) upstream region then was directionally subcloned into the luciferase vector pGL3Basic (Promega). Computer-assisted prediction of transcription factor binding sites in the promoter sequence was performed with TESS (Transcription Element Search System) using TRANSFAC version 4.0 (30) and MatInspector using Matrix Family Library Version 2.4 (Genomatix, Munich, Germany) (31).
Truncations of the upstream region at approximately 100-bp intervals were made by PCR amplification in 50-μl reactions containing 1 ng pCRII-OXTR(–830/–4), 25 mM TAPS (N-[Tris(hydroxymethyl)methyl]-3-aminopropane-sulfonic acid, pH 9.3 at room temperature), 50 mM KCl, 2 mM MgCl2, 200 μM dNTPs, 0.2 μM each forward primer (Table 1), 0.2 μM T7 primer as reverse primer, and 1.25 U ExTaq in a Perkin-Elmer 9700 thermocycler using conditions of: 1) 94 C for 2 min; 2) 94 C for 1 min, 50 C for 1 min, 72 C for 1 min for 30 cycles; and 3) 72 C for 10 min. The truncated PCR products were cloned into pCRII Dual, physically mapped to determine orientation, and directionally subcloned into the pGL3Basic vector. The OXTR(–144/–4) LUC truncated clone was made by digestion of OXTR(–220/–4) LUC plasmid DNA with BglII, reinserting the BglII (–144/–4) fragment into the luciferase vector, and physically mapping the construct to confirm proper orientation. Point mutations were introduced into the SP1 sites at –104 and –64 of the OXTR(–220/–4) truncated pCRII subclone by PCR-directed mutagenesis using the primers listed in Table 1 and Vent Taq polymerase and ExTaq (31). A OXTR(–220/–4) pCRII clone containing an introduced mutation at the –64 SP1 site was used as the template for the construction of OXTR–220 constructs with mutated SP1 sites at both –104 and –64. Mutated DNAs were confirmed by sequencing, and fragments containing only the desired point mutations were directionally subcloned into the luciferase vector as described above.
Transient transfections and luciferase assays
Cells were subcultured into 12-well plates (67–75% confluent) and transiently transfected as described previously (23) with the following modifications. Luciferase constructs (500 ng/well) were cotransfected with either pEF1-Myc-His-LacZ (500 ng/well; Invitrogen) or expression plasmids for ovine IRF1, ovine IRF2, human ESR1wt, human ESR1IIc, or human progesterone receptor B (PGR-B) (500 ng/well except where noted). The ovine IRF1 and IRF2 mammalian overexpression plasmids have been described previously (24). The ESR1wt (wild-type human ESR1) and ESR1IIc (a DNA binding domain mutant construct) overexpression plasmids have been described previously (32, 33). The PGR-B plasmid, pSV40-hPGR-B that overexpresses the B form of human PGR, was kindly provided by Dr. M.-j. Tsai (Baylor College of Medicine, Houston, TX). Transfected cells were grown in 10% FBS for approximately 14–16 h before treatment for 24 h in serum-free medium. Phenol red-free DMEM-F12 medium and dextran-coated charcoal-stripped FBS were substituted in experiments when testing for effects of steroids. Steroid agonists and antagonists were dissolved in 100% ethanol. For the control, cells were treated with the same volume of 100% ethanol alone. Luciferase and protein assays were done as described previously (23, 24). Transfection assays were repeated a minimum of three times.
EMSA
Double-stranded oligonucleotide primers listed in Table 1 were end-labeled to high specific activity with [-32P]ATP and T4 DNA kinase by standard methods. Purified SP1 protein (54 ng/reaction) was incubated in 10 μl reactions containing 4% glycerol, 1 mM MgCl2, 0.5 mM EDTA, 50 mM NaCl, 10 mM Tris-HCl (pH 7.5), 0.5 mM DTT, 0.05 mg/ml poly(dI-dC) at room temperature for 15 min. A 100-fold excess of unlabeled specific or nonspecific competitor oligonucleotides was included in some reactions. The nonspecific competitor was a 22-nt oligonucleotide with the same base composition as the SP1 consensus oligonucleotide to minimize charge differences, but having a different sequence (Table 1). Radiolabeled probes (10 fmol per reaction) were added to reactions that were incubated at room temperature for an additional 15 min. Binding reactions were electrophoresed in 5% native polyacrylamide gels in 1x TBE. Imaging of dried gels was done with a Molecular Dynamics Typhoon phosphorimager.
Immunohistochemistry
Mature crossbred Suffolk ewes (Ovis aries) were observed daily for estrus in the presence of vasectomized rams and used in experiments only after they had exhibited at least two estrous cycles of normal duration (16–18 d). All experimental and surgical procedures were in compliance with the Guide for the Care and Use of Agriculture Animals and approved by the University Laboratory Animal Care and Use Committee of Texas A&M University.
At estrus (d 0), ewes were mated to either an intact or vasectomized ram as described previously (34) and then hysterectomized (n = 5 ewes per day) on either d 10, 12, 14, or 16 of the estrous cycle or d 10, 12, 14, 16, 18, or 20 of pregnancy. Pregnancy was confirmed on d 10–16 after mating by the presence of a morphologically normal conceptus(es) in the uterus. At hysterectomy, several sections (0.5 cm) from the mid-portion of each uterine horn ipsilateral to the CL were fixed in fresh 4% paraformaldehyde in PBS (pH 7.2). After 24 h, fixed tissues were changed to 70% ethanol for 24 h and then dehydrated and embedded in Paraplast-Plus (Oxford Labware, St. Louis, MO). In monovulatory pregnant ewes, uterine tissue samples were marked as either contralateral or ipsilateral to the ovary bearing the CL. No tissues from the contralateral uterine horn were used for study.
Immunoreactive SP1 proteins were localized in cross-sections (5 μm) of the uterus using a rabbit polyclonal antibody to human SP1 (catalog SC-59; Santa Cruz Biotechnology Inc., Santa Cruz, CA) at a final working concentration of 0.1 μg/ml and a VectaStain Rabbit IgG Elite ABC kit (Vector Laboratories, Burlingame, CA) using methods described previously (35). Antigen retrieval using boiling citrate buffer was performed as described previously (35, 36). The chromagen used for peroxidase localization was 3,3'-diaminobenzidine tetrahydrochloride from Sigma Chemical Co. (St. Louis, MO). Negative controls were performed in which the primary antibody was substituted with the same concentration of purified normal rabbit IgG from Sigma Chemical Co. Multiple tissue sections from each ewe were processed as sets within an experiment. Sections were not counterstained before affixing coverslips. Representative photomicrographs of uterine tissues were taken using a Nikon Eclipse 1000 photomicroscope (Nikon Instruments Inc., Lewisville, TX) fitted with a Nikon DXM1200 digital camera. Digital images were captured and assembled using Adobe Photoshop (Adobe Systems, Seattle, WA).
Statistical analyses
The effects of steroid hormones, ICI 182,780, or IFNT on the activity of promoter reporter constructs in transient transfection assays were analyzed by least squares ANOVA using the General Linear Models procedure of the Statistical Analysis System (Cary, NC). A P value of 0.10 or less was considered statistically significant. Unless denoted, data are reported as mean with SD. Effects of treatment were determined using orthogonal or nonorthogonal contrasts using the PDIFF option of General Linear Models.
Results
Isolation and sequence analysis of the OXTR 5' upstream region
A phagemid DNA containing approximately 9.8 kbp of ovine genomic DNA was isolated from a ovine genomic library by conventional screening. Southern blotting, physical mapping, and sequence analysis with several OXTR-specific primers revealed that the clone contained a 1059-bp region that included the 5' end of the coding sequence of the ovine OXTR gene contiguous to 830 nt of upstream sequence. The 1059-bp clone was sequenced in both directions and deposited into GenBank (accession AY163261). Sequence comparisons of the ovine upstream region with the previously characterized Bos taurus OXTR promoter region and adjacent coding regions (GenBank AF100633) indicated that the two sequences were 91% identical. Although the coding sequence displayed high homology with the OXTR cDNA from other species (human, dog, pig, gorilla, rat, mouse, chicken), the promoter region had very little or no homology to those of other species except for bovine.
Bioinformatic analyses found that the ovine OXTR promoter region contained multiple putative binding sites for the transcription factors SP1 and activator protein 1 (AP-1), which regulates target genes by steroid hormone receptors (see Refs.37 and 38 for review). Putative nonconsensus SP1 sites were found at –748, –543, and –444 relative to the translational start site (ATG +1) in addition to consensus sites at –104 and –64 (Fig. 1). The putative –64 SP1 binding site is identical to the consensus SP1 binding motif (5'-CCCCGCCCC-3', sense strand). The –104 putative binding site (5'-CCCCACCCCGCCCC-3', sense strand) also contains a consensus binding motif that is immediately adjacent to and overlaps a sequence that resembles a nonconsensus SP1 binding site (5'-CCCCACCCC-3') (Fig. 1). Nonconsensus SP1 sites having the sequence 5'-CCCCACCCC-3' have been reported for the long terminal repeats of the human endogenous retrovirus H family of human retrovirus-like elements (39). Alternatively, part of the sequence of the OXTR –104 SP1 site may contain a CACCC box at the 5' end. The CACCC elements bind SP1 and related Sp/Kruppel-like factor transcription factors (40) and are involved in regulation of both basal (41) and induced levels of transcription in other systems (42). A consensus AP-1 binding site at –680 and nonconsensus AP-1 sites at –574, –329, and –221 were identified, which also mediate hormone actions (38). No full progesterone response elements (PREs) were found, but several putative PRE half sites were detected at –639, –486, –477, and –187. No full EREs or ERE half sites were found by computer-assisted analysis of the ovine OXTR upstream region. Furthermore, elements mediating effects of type I or II IFNs or STATs, including ISRE, IRFE, or activation sites (GAS) (43), were also not found by bioinformatic analyses.
IFNT does not regulate OXTR promoter activity
To determine whether IFNT regulated activity of the OXTR promoter, transient transfection assays were conducted in 2fTGH parental and U3A (STAT1 null) cells, which are responsive to IFNT (44, 45, 46) in a STAT-dependent (2fTGH) and STAT1-independent (U3A) manner. These cells have been extensively used by our laboratory as a model for the endometrial stroma and LE, respectively (44, 45, 46). Cells were transfected with the full-length OXTR(–830/–4) and sequential 5'-deletion reporter constructs and treated with different doses of recombinant ovine IFNT (103–106 antiviral units). No dose-dependent effects of IFNT on promoter activity were detected (P > 0.10) with any of the OXTR promoter-reporter constructs in either 2fTGH or U3A cells (data not shown). Cells were also cotransfected with the OXTR(–830/–4) promoter-reporter construct and either pEF1-Myc/His-LacZ, ovine IRF1, or ovine IRF2 overexpression vectors. Relative to the LacZ control, overexpression of ovine IRF1 or ovine IRF2 did not affect (P > 0.10) activity of the OXTR promoter constructs in either cell type (data not shown). These results are consistent with the lack of IFN-responsive elements (GAS, ISRE, or IRFE) in the ovine OXTR promoter.
In 2fTGH cells transfected with the OXTR(–830/–4) construct and ESR1wt, E2 (10–8 M) increased (P < 0.01) luciferase activity, and this induction was not affected (P > 0.10) by IFNT treatment (Fig. 2). Treatment with IFNT lowered (P < 0.05) the basal activity of the ovine OXTR promoter; however, the fold induction by E2 was the same in untreated and IFNT-treated cells. Similar results were obtained using U3A cells (data not shown). The basal activity of several promoter-reporter constructs appears to be nonspecifically affected by IFNT, presumably due to effects on cell metabolism due to induction of an antiviral state (data not shown). These results indicate that IFNT does not inhibit E2-induced transactivation in cells transfected with OXTR promoter-derived constructs.
Estrogen and antiestrogen stimulate OXTR promoter activity
In 2fTGH and U3A cells, basal activity of the OXTR promoter changed with length of the promoter-reporter construct (Figs. 3 and 4). The full-length OXTR(–830/–4)-LUC construct had considerably lower basal activity compared (P < 0.01) to all other 5'-deletion constructs. Basal activity of the OXTR(–340/–4) construct was approximately 1.5-fold greater (P < 0.05) than the OXTR(–220/–4) construct containing the minimal promoter.
2fTGH cells were cotransfected with OXTR promoter-reporter constructs and ESR1wt (Fig. 3A) or ESR1IIc (Fig. 4B), a DNA binding domain mutant derived from ESR1wt, and treated with vehicle as a control or 10–8 M E2 for 24 h. In cells transfected with ESR1wt, E2 treatment increased (P < 0.05) activity of the OXTR(–830/–4) promoter-LUC construct as well as all 5' deletions (Fig. 3A). Similarly, E2 treatment increased (P < 0.01) activity of all OXTR promoter-reporter constructs in cells cotransfected with ESR1IIc (Fig. 4A). The fold increase in OXTR promoter activity was consistently greater in cells transfected with ESR1IIc than ESR1wt. Dose response studies found that E2 at a concentration of greater than 10–9 M increased activity of the full-length OXTR(–830/–4) promoter as well as the minimal OXTR(–220/–4) promoter in cells cotransfected with ESR1wt (Fig. 3B) or ESR1IIc (Fig. 4B). These results indicate that cis elements in the minimal promoter (–220/–4) are sufficient to mediate ligand-activated ESR1 effects. Furthermore, ESR1 induction of the OXTR promoter does not require DNA binding, suggesting protein-protein interactions through another transcription factor. Estrogen induces activity of many genes with GC-rich promoters in breast cancer cells transfected with ESR1 or ESR1IIc, even though the genes have no EREs in their promoters (see Refs.37 and 47).
ICI 182,780 is a pure antiestrogen that inhibits liganded ESR1 from transactivating estrogen-responsive genes containing an ERE(s) (48). However, ICI can exhibit partial agonist activity in some cells and induce ESR1 activation of some estrogen-responsive genes containing consensus GC-rich motifs in the promoter (49). To examine the effects of E2 and ICI 182,780 on ovine OXTR promoter activity, 2fTGH and U3A cells were cotransfected with ESR1wt or ESR1IIc and OXTR(–830/–4)-LUC, OXTR(–220/–4)-LUC, 3xERE-TATA-LUC, or 3xSP1-LUC reporters. Cells were then treated for 24 h with E2 (10–8 M) or ICI 182,780 (10–6 M). In 2fTGH cells, E2 stimulated (P < 0.01) activity of the OXTR reporters in cells transfected with ESR1wt or ESR1IIc (Fig. 5). E2 also induced (P < 0.01) activity of 3xERE-TATA-LUC in cells cotransfected with ESR1wt but not ESR1IIc, consistent with the requirement for DNA bound ESR1. ICI stimulated (P < 0.01) OXTR promoter activity in cells cotransfected with ESR1wt and ESR1IIc. As expected, ICI did not affect activity of the 3xERE-TATA-LUC in 2fTGH cells cotransfected with ESR1IIc. Both E2 and ICI stimulated (P < 0.01) activity of the 3xSP1-LUC reporter in 2fTGH cells transfected with either ESR1wt or ESR1IIc. These results are consistent with E2 and ICI activation of ovine OXTR promoter activity through an ESR1/SP1-mediated pathway as previously observed for several hormone-responsive genes with GC-rich promoters (see Refs.37 and 47).
Minimal OXTR promoter contains functional SP1 elements
Deletion analysis showed that the basal promoter activity and minimal hormone-responsive region was between –220 to –4 bp in the OXTR promoter. This region contains an AP-1 site, PRE half site, and three SP1 sites (Fig. 1). The functionality of the SP1 elements was determined by EMSA using recombinant human SP1 protein (Fig. 6). As illustrated in Fig. 6A, the predicted SP1 binding site at –64 in the OXTR promoter bound SP1 protein (lane 6). As a positive control, SP1 protein-DNA binding was observed with an oligonucleotide containing a consensus SP1 binding site (lane 2). SP1 binding to the –64 OXTR oligonucleotide was specific, because it was competitively decreased by incubation with a 100-fold excess of unlabeled SP1 consensus oligonucleotide (lane 5) or the unlabeled –64 OXTR oligonucleotide itself (lane 3). Mutation of two critical C residues in the –64 OXTR oligonucleotide rendered it unable to bind SP1 protein (lane 7) or compete for binding of SP1 protein to the wild-type –64 OXTR oligonucleotide (lane 4). In other experiments, 100-fold excess of unlabeled wt –64 OXTR oligonucleotide prevented binding of SP1 protein to the consensus SP1 oligonucleotide, whereas the mutant –64 OXTR oligonucleotide was unable to compete (data not shown). These results demonstrate that the ovine OXTR gene promoter contains a high affinity SP1 binding site at position –64.
As illustrated in Fig. 6B, the GC-rich sites at –104 in the ovine OXTR promoter also bound SP1 protein (lane 2) in EMSAs. Incubation with a 100-fold excess of the wild-type –104 OXTR oligonucleotide (lane 3) or consensus SP1 oligonucleotide (lane 4), but not a nonspecific oligonucleotide (lane 5), competitively decreased SP1 protein binding to the radiolabeled –104 OXTR oligonucleotide. The sequence of the –104 SP1 binding site in the OXTR promoter contains a 5' nonconsensus GC-rich SP1 binding site adjacent to a 3' consensus site (Fig. 1). Mutation of the 3' consensus GC-rich site prevented SP1 protein binding (mut1) (lane 6), whereas mutation of the 5' nonconsensus binding site in the –104 OXTR oligonucleotide (mut2) did not affect SP1 binding (lane 7). Furthermore, SP1 binding was not observed when both the 5' and 3' predicted SP1 elements were mutated together (mut4) (lane 8). Collectively, EMSA results indicate that two GC-rich sites in the ovine OXTR promoter at –64 and –104 bp bind SP1 protein.
To determine whether the GC-rich sites at –104 and –64 are critical for responsiveness to ESR1, these sites in the –220 OXTR-LUC reporter construct were mutated (see Fig. 1). Due to the complex structure of the –104 SP1 element, mutation of either the 5' nonconsensus SP1 site or the 3' consensus SP1 element was performed, because it could alter the integrity of the overlapping site and affect the ability of the –104 sequence to bind SP1. Therefore, in addition to mutations of the –104 site tested in initial gel shift analyses, some CA transversions that mutated either the consensus or nonconsensus portion of the site, without altering the other overlapping moiety, were also introduced. The oligonucleotides used to create the mutations were tested initially for their ability to bind SP1 protein by EMSA (Fig. 7). Mutation of the 5' nonconsensus site did not affect SP1 binding regardless of whether the overlapping 3' consensus portion remained wild-type (mut3; lane 5) or was slightly altered (mut2; lane 4). Mutation of the 3' consensus sequence alone (mut1, lane 3) substantially reduced SP1 binding compared with the wild-type –104 OXTR oligonucleotide (lane 2), but did not eliminate binding entirely. Mutation of both 5' and 3' sites (mut4) completely prevented SP1 binding (lane 6). In competition experiments (Fig. 7B), excess unlabeled wt –104 oligonucleotide and the mut1, mut2, and mut3 oligonucleotides reduced or prevented SP1 binding to the radiolabeled wt –104 oligonucleotide (lanes 6, 7, and 8). As expected, competition with the fully mutated mut4 oligonucleotide did not affect SP1 binding to the radiolabeled –104 OXTR wt oligonucleotide (lane 9). Collectively, these results indicate that SP1 binding to the –104 OXTR sequence depends primarily on the 3' consensus GC-rich element. However, the 5' nonconsensus GC-rich element appears to weakly bind SP1 protein, suggesting that the two overlapping motifs at –104 may act cooperatively to bind SP1.
Transient transfection assays were then conducted to determine the relative contributions of the –104 and –64 SP1 binding sites on hormonal responsiveness of the ovine OXTR promoter (Fig. 8). Wild-type or mutant OXTR(–220/–4) promoter-reporter constructs were cotransfected with ESR1IIc into 2fTGH cells and left unstimulated or treated with E2 (10–8 M). Mutation of either the 5' nonconsensus site (constructs 2 and 3), the 3' consensus site (construct 4), or both sites (construct 5) at –104 reduced (P < 0.05) basal luciferase expression in unstimulated cells by 25–40% compared with the wild-type reporter (construct 1) depending on the location of the mutation. Mutation of the –64 SP1 site reduced (P < 0.01) basal luciferase expression to very low levels, regardless of whether the –104 site was wt (construct 6), double mutated (constructs 7 and 9), or point mutated (construct 8). Indeed, basal luciferase levels of the –64 SP1 mutants were about 10% of wild-type parental levels. Mutation of the 5' consensus SP1 site or both the nonconsensus and consensus sites at –104 and the –64 SP1 site (construct 10) further reduced (P < 0.10) basal activity to about 2% of activity observed for the wt construct.
Estradiol-induced activity, in all cells transfected with the minimal OXTR(–220/–4) promoter, was lower (P < 0.10) in constructs containing a mutation in the –104 and –64 SP1 binding sites (Fig. 8). Mutation of both sites at position –104 further reduced stimulation by estradiol, and this was consistent with the EMSA data suggesting that both the left and right moieties are involved in binding SP1 protein. However, mutation of the –64 SP1 site decreased (P < 0.10) stimulation by estradiol, regardless of whether the –104 SP1 site was wt or mutated. Overall, these results suggest that both the –104 and –64 SP1 sites in the proximal OXTR promoter regulates basal levels of OXTR gene transcription, and that both SP1 binding sites, especially the consensus sequences, regulate E2 stimulation of OXTR gene expression.
Progesterone stimulates OXTR promoter activity
During metestrus and estrus in cyclic ewes, PGR return to the OXTR-expressing endometrial epithelia as systemic progesterone declines to undetectable levels (9, 10, 12, 13, 14). The prevailing theory is that PGR do not directly inhibit OXTR gene expression, but rather the progesterone block to OXTR formation is indirect via progesterone inhibition of ESR1 expression (4, 8, 28). To determine whether the OXTR promoter is regulated by PGR, 2fTGH cells were cotransfected with human PGR-B and OXTR promoter-LUC constructs and then treated with R5020 (10–8 M), a nonmetabolizable form of progesterone (Fig. 9). As observed previously, basal activity of the OXTR promoter constructs increased (P < 0.05) by deletion of the region from –830 to –711. Treatment with R5020 stimulated (P < 0.05) the activity of all OXTR promoter-reporter constructs. Next, the same experiment was repeated with the reporter constructs and PGR-B, but cells were treated with a range of R5020 doses. Dose-dependent effects of R5020 (10–14–10–5 M) on activity of the –830 and –220 OXTR promoter constructs were observed in 2fTGH cells cotransfected with PGR-B, with concentrations as low as 10–10 M R5020 stimulated transactivation in cells transfected with OXTR constructs (data not shown).
Deletion of the AP-1 and PRE half site does not affect E2 or R5020 stimulation of OXTR promoter activity
The minimal OXTR(–220/–4) promoter contains an AP-1 site at –220 and a PRE half site at position –187. To determine whether these sites are involved in hormone responsiveness, the sites were sequentially removed by 5' deletion. The OXTR promoter constructs were then transfected into 2fTGH cells along with ESR1IIc or PGR-B and treated with E2 (10–8 M) or R5020 (10–8 M) for 24 h, respectively (Fig. 10). Deletion of the AP-1 site and PRE half site did not affect (P > 0.10) the stimulatory effects of E2 and R5020 on OXTR promoter activity. These results indicate that the SP1 binding sites at –104 and –64 are the only cis elements required for basal activity of the OXTR promoter and its responsiveness to stimulation by ligand-activated ESR1 and PGR-B.
Immunolocalization of SP1 protein in the endometrium of cyclic and pregnant ewes
To confirm that SP1 was expressed in the endometrium, SP1 protein was studied in the uterus of cyclic and early pregnant ewes by immunohistochemistry (Fig. 11). Immunoreactive SP1 protein was observed in all endometrial cell types of uteri from ewes on d 10–16 of the cycle and d 10–20 of pregnancy. Although all endometrial cell types expressed SP1 protein, it was most abundant in the endometrial LE and GE. No differences in SP1 protein abundance were detected between the endometrial cells of cyclic and pregnant ewes or between LE of intercaruncular and caruncular areas of the endometrium (data not shown). In pregnant ewes, SP1 protein was also detected in conceptus trophectoderm.
Discussion
The gene for OXTR exists as a single copy in all studied species, and the promoter region of the OXTR genes of the human (50), rat (51, 52), bovine (53, 54), mouse (55), and vole (56) are available. Comparative studies of the promoter regions of homologous genes from several species often reveal conserved and important regulatory elements. However, the promoter regions of the ovine and bovine OXTR genes are not well conserved with other species, despite significant sequence conservation found in the coding sequence. The human OXTR gene promoter can be regulated by proinflammatory cytokines, such as IL-1 and IL-6 via activation of the binding elements for nuclear factor-IL-6 or STAT3 (57). Serum increases activity of the human OXTR promoter through a protein kinase C-dependent pathway that can be synergistically augmented by dexamethasone (58). In rabbit amnion cells, in vitro treatment with forskolin and/or cortisol activates OXTR gene transcription (59). Protein kinase A, protein kinase C, and nerve growth factor-dependent OXTR up-regulation is also observed in MCF-7 and SK-N-SH cells transfected with the rat OXTR gene promoter (60). In the human, basal promoter activity is conferred by the 85-bp flanking the 5' end of the human OXTR gene. In this sequence, there is a consensus ETS binding sequence. GABP (GA binding protein transcription factor), which binds to the ETS element, cooperates with AP-1 (FOS/JUN) to activate human OXTR promoter transcription in Hs578T cells (61). However, elements that regulate activity of the human and rat OXTR promoter/enhancer regions are not present in the ovine and bovine OXTR promoters, suggesting that the identified pathways regulating OXTR expression in other species may not be the same for ruminants.
In every studied mammal, estrogen is considered a key regulator of OXTR gene expression (27). Similarly in sheep, in vivo administration of estrogen increases OXTR mRNA and the number of oxytocin binding sites in the endometrium (15, 16, 22, 62). The classical pathway for estrogen regulation of target genes involves ligand-activated ESR1/ERE interactions or ERE half sites (37, 47). In the rat OXTR gene, a palindromic ERE was identified about 4 kb upstream of the transcriptional start site (52). This sequence was hormone-responsive in transfection studies in human MCF-7 breast cancer cells, but only when the 3.3-kb fragment between the ERE and the basal promoter of the rat OXTR gene was truncated. In the wild-type construct, this ERE was not activated by estrogen. However, the human OXTR gene does not have a palindromic ERE in the area comparable to that of the rat gene (27). Therefore, estrogen induction of ovine OXTR gene transcription may be indirect rather than due to direct ESR1/ERE interactions.
The present study found that an estrogen agonist (E2) and antagonist (ICI 182,780) transactivated the ovine OXTR promoter in cells transfected with ESR1. Deletion and mutation analyses found that induction of ovine OXTR promoter activity by E2-activated ESR1 is dependent on GC-rich SP1 binding sites at –104 and –64 in the minimal promoter, and basal activity of this promoter is also regulated by the same GC-rich SP1 elements adjacent to the translational start site of the OXTR gene. One caveat of the present studies is that human fibrosarcoma cells were used for transfection assays instead of an ovine endometrial epithelial cell line the endogenous expresses the OXTR gene. Indeed, the GC-rich motifs in several gene promoters are cis-acting elements for an increasing number of ligand-activated nuclear and orphan receptors that interact with SP1 and related proteins (see Refs.37 and 47). In the present study, SP1 transcription factor was constitutively expressed in nearly all cell types of the endometrium from both cyclic and pregnant ewes and was particularly abundant in LE and sGE. Therefore, the primary determinant of OXTR gene regulation in the endometrial epithelia is expression of the ESR1 gene. Administration of estrogen on d 11 or 12 postestrus induces structural and functional luteolysis in cyclic ewes with the following temporal events: 1) increase ESR1 mRNA and protein in endometrial epithelia between 12–24 h; 2) moderate increase in endometrial OXTR density between 12–36 h; 3) large increase in OXTR density between 36–48 h; 4) a decline in concentrations of progesterone in plasma after 36 h; and 5) decrease in CL weight by 48 h (15, 17). Similar temporospatial alterations in ESR1 and OXTR gene expression occur in the endometrium of cyclic ewes between d 12–16 postestrus (9, 10, 12). Although exogenous estrogen can clearly elicit development of the endometrial luteolytic mechanism and induce OXTR promoter activity, the precise physiological role of estrogen in the luteolytic mechanism in vivo during a natural estrous cycle has not been adequately investigated. Estrogen may not be needed to increase OXTR gene expression, because ESR1 can be activated in a ligand-dependent manner by estrogens or a ligand-independent manner by growth factors (63), such as IGF-I and IGF-II that are abundant in the ovine endometrial stroma (64). Furthermore, ICI 182,780, a pure antiestrogen, will not be useful for in vivo investigations, because it activated ESR1 and induced OXTR promoter activity in the present study. Similarly, the antiestrogens 4'-hydroxytamoxifen (4-OHT) and ICI 182,780 activate reporter gene activity in cells transfected with constructs containing GC-rich promoters with SP1 and SP3 binding sites (65). In addition to the OXTR, many other endometrial genes that are regulated by hormone and orphan receptors are likely to involve SP1 and related transcription factors.
Development of the endometrial luteolytic mechanism and uterine release of luteolytic pulses of PGF occur in the presence of high circulating levels of progesterone. However, the endometrial epithelia are not responsive to receptor-dependent actions of progesterone, because they are PGR-negative between d 11–15 of the estrous cycle. As progesterone levels decrease and estrogen levels increase during proestrus (d 15–17), PGR protein returns to the endometrial epithelia of the ovine uterus (10, 12). In the present study, PGR also increased activity of the ovine OXTR promoter through GC-rich SP1 sites rather than the AP-1 site or PRE half sites also present in the promoter of the OXTR gene. Progestin-dependent regulation of OXTR through GC-rich sites is not unprecedented becausePGR/SP1 activation of glycodelin A and CDKN1A (cyclin-dependent kinase inhibitor 1A or p21) in endometrial and breast cancer cells, respectively, is also due to specific GC-rich SP1 binding sites in their promoters (66, 67). Indeed, uterine OXTR mRNA and protein increases in the endometrial epithelia and stroma after d 15 and is maximal at estrus, which is coincident with PGR mRNA and protein as well as SP1 expression in the same cells (10, 12, 13). Therefore, the progesterone block to OXTR formation must be due to inhibition of ESR1 gene transcription via direct or indirect actions that does not involve SP1.
In the present study, IFNT did not affect OXTR promoter activity or E2 induction of OXTR promoter activity. These results were consistent with the lack of classical interferon regulatory elements (ISRE, IRFE, or GAS) in the ovine OXTR promoter and the lack of IRF1 and IRF2 effects on OXTR promoter activity in transient transfection assays. Collectively, results of the present and previous studies continue to support our theory that the antiluteolytic effects of IFNT from the conceptus are to inhibit or silence ESR1 gene transcription, which in turn precludes ESR1 stimulation of OXTR gene transcription and, therefore, oxytocin-induced release of luteolytic pulses of PGF by uterine epithelia. Future experiments are needed to address how progesterone down-regulates PGR gene expression, what factors critically regulate PGR inhibition of ESR1 gene transcription in the endometrial epithelia of the uterus, and if SP1 and related transcription factors regulate other endometrial genes in the ovine uterus.
Acknowledgments
We thank Haijun Gao for assistance with immunohistochemical analysis of SP1 protein and other members of our laboratories who contributed to the research presented in this paper.
Footnotes
Financial support for these studies was obtained from National Institutes of Health R01 Grant HD32534 and National Institute on Environmental Health Sciences 5 P30 ES09106-03.
First Published Online October 27, 2005
Abbreviations: AP-1, Activator protein 1; CL, corpus luteum; E3, 17-estradiol; ERE, ESR1 response element; ESR1, estrogen receptor ; FBS, fetal bovine serum; GAS, activation site; GE, glandular epithelium; IFNT, interferon ; IRF, interferon regulatory factor; IRFE, interferon regulatory factor element; ISRE, interferon-stimulated response element; LE, luminal epithelium; OXTR, oxytocin receptor; PGF, prostaglandin F2; PGR, progesterone receptor; PGR-B, progesterone receptor B form; PRE, progesterone response element; sGE, superficial ductal glandular epithelium.
Accepted for publication October 19, 2005.
References
Bazer FW, Thatcher WW, Hansen PJ, Mirando MA, Ott TL, Plante C 1991 Physiological mechanisms of pregnancy recognition in ruminants. J Reprod Fertil Suppl 43:39–47
Bazer FW 1992 Mediators of maternal recognition of pregnancy in mammals. Proc Soc Exp Biol Med 199:373–384
McCracken JA, Custer EE, Lamsa JC 1999 Luteolysis: a neuroendocrine-mediated event. Physiol Rev 79:263–323
Spencer TE, Bazer FW 2002 Biology of progesterone action during pregnancy recognition and maintenance of pregnancy. Front Biosci 7:d1879–d1898
Charpigny G, Reinaud P, Tamby JP, Creminon C, Martal J, Maclouf J, Guillomot M 1997 Expression of cyclooxygenase-1 and -2 in ovine endometrium during the estrous cycle and early pregnancy. Endocrinology 138:2163–2171
Gray C, Bartol FF, Taylor KM, Wiley AA, Ramsey WS, Ott TL, Bazer FW, Spencer TE 2000 Ovine uterine gland knock-out model: effects of gland ablation on the estrous cycle. Biol Reprod 62:448–456
Spencer TE, Ott TL, Bazer FW 1996 -Interferon: pregnancy recognition signal in ruminants. Proc Soc Exp Biol Med 213:215–229
Spencer TE, Bazer FW 2004 Conceptus signals for establishment and maintenance of pregnancy. Reprod Biol Endocrinol 2:49
Ayad VJ, Matthews EL, Wathes DC, Parkinson TJ, Wild ML 1991 Autoradiographic localization of oxytocin receptors in the endometrium during the oestrous cycle of the ewe. J Endocrinol 130:199–206
Wathes DC, Hamon M 1993 Localization of oestradiol, progesterone and oxytocin receptors in the uterus during the oestrous cycle and early pregnancy of the ewe. J Endocrinol 138:479–492
McCracken J, Schramm W, Okulicz WC 1984 Hormone receptor control of pulsatile secretion of PGF-2 from the ovine uterus during luteolysis and its abrogation in early pregnancy. Anim Reprod Sci 7:31–55
Spencer TE, Bazer FW 1995 Temporal and spatial alterations in uterine estrogen receptor and progesterone receptor gene expression during the estrous cycle and early pregnancy in the ewe. Biol Reprod 53:1527–1543
Stevenson KR, Riley PR, Stewart HJ, Flint AP, Wathes DC 1994 Localization of oxytocin receptor mRNA in the ovine uterus during the oestrous cycle and early pregnancy. J Mol Endocrinol 12:93–105
Wathes DC, Mann GE, Payne JH, Riley PR, Stevenson KR, Lamming GE 1996 Regulation of oxytocin, oestradiol and progesterone receptor concentrations in different uterine regions by oestradiol, progesterone and oxytocin in ovariectomized ewes. J Endocrinol 151:375–393
Spencer TE, Becker WC, George P, Mirando MA, Ogle TF, Bazer FW 1995 Ovine interferon- inhibits estrogen receptor up-regulation and estrogen-induced luteolysis in cyclic ewes. Endocrinology 136:4932–4944
Spencer TE, Becker WC, George P, Mirando MA, Ogle TF, Bazer FW 1995 Ovine interferon- regulates expression of endometrial receptors for estrogen and oxytocin but not progesterone. Biol Reprod 53:732–745
Hixon JE, Flint AP 1987 Effects of a luteolytic dose of oestradiol benzoate on uterine oxytocin receptor concentrations, phosphoinositide turnover and prostaglandin F-2 secretion in sheep. J Reprod Fertil 79:457–467
Vallet JL, Lamming GE, Batten M 1990 Control of endometrial oxytocin receptor and uterine response to oxytocin by progesterone and oestradiol in the ewe. J Reprod Fertil 90:625–634
Roberts RM, Ealy AD, Alexenko AP, Han CS, Ezashi T 1999 Trophoblast interferons. Placenta 20:259–264
Wathes DC, Lamming GE 1995 The oxytocin receptor, luteolysis and the maintenance of pregnancy. J Reprod Fertil Suppl 49:53–67
Kim S, Choi Y, Spencer TE, Bazer FW 2003 Effects of the estrous cycle, pregnancy and interferon on expression of cyclooxygenase two (COX-2) in ovine endometrium. Reprod Biol Endocrinol 1:58
Spencer TE, Bazer FW 1996 Ovine interferon suppresses transcription of the estrogen receptor and oxytocin receptor genes in the ovine endometrium. Endocrinology 137:1144–1147
Fleming JA, Choi Y, Johnson GA, Spencer TE, Bazer FW 2001 Cloning of the ovine estrogen receptor- promoter and functional regulation by ovine interferon-. Endocrinology 142:2879–2887
Choi Y, Johnson GA, Burghardt RC, Berghman LR, Joyce MM, Taylor KM, Stewart MD, Bazer FW, Spencer TE 2001 Interferon regulatory factor-two restricts expression of interferon-stimulated genes to the endometrial stroma and glandular epithelium of the ovine uterus. Biol Reprod 65:1038–1049
Vallet JL, Bazer FW 1989 Effect of ovine trophoblast protein-1, oestrogen and progesterone on oxytocin-induced phosphatidylinositol turnover in endometrium of sheep. J Reprod Fertil 87:755–761
Telgmann R, Bathgate RA, Jaeger S, Tillmann G, Ivell R 2003 Transcriptional regulation of the bovine oxytocin receptor gene. Biol Reprod 68:1015–1026
Kimura T, Saji F, Nishimori K, Ogita K, Nakamura H, Koyama M, Murata Y 2003 Molecular regulation of the oxytocin receptor in peripheral organs. J Mol Endocrinol 30:109–115
Spencer TE, Johnson GA, Burghardt RC, Bazer FW 2004 Progesterone and placental hormone actions on the uterus: insights from domestic animals. Biol Reprod 71:2–10
Pellegrini S, John J, Shearer M, Kerr IM, Stark GR 1989 Use of a selectable marker regulated by interferon to obtain mutations in the signaling pathway. Mol Cell Biol 9:4605–4612
Van Heeke G, Ott TL, Strauss A, Ammaturo D, Bazer FW 1996 High yield expression and secretion of the ovine pregnancy recognition hormone interferon- by Pichia pastoris. J Interferon Cytokine Res 16:119–126
Ausebel FM, Brent R, Kinston RE, Moore DD, Seidmen JG, Smith HA, Struhl KA 1998 Current protocols in molecular biology. New York: John Wiley and Sons
Stoner M, Wang F, Wormke M, Nguyen T, Samudio I, Vyhlidal C, Marme D, Finkenzeller G, Safe S 2000 Inhibition of vascular endothelial growth factor expression in HEC1A endometrial cancer cells through interactions of estrogen receptor and Sp3 proteins. J Biol Chem 275:22769–22779
Sun G, Porter W, Safe S 1998 Estrogen-induced retinoic acid receptor 1 gene expression: role of estrogen receptor-Sp1 complex. Mol Endocrinol 12:882–890
Spencer TE, Bartol FF, Bazer FW, Johnson GA, Joyce MM 1999 Identification and characterization of glycosylation-dependent cell adhesion molecule 1-like protein expression in the ovine uterus. Biol Reprod 60:241–250
Taylor KM, Gray CA, Joyce MM, Stewart MD, Bazer FW, Spencer TE 2000 Neonatal ovine uterine development involves alterations in expression of receptors for estrogen, progesterone, and prolactin. Biol Reprod 63:1192–1204
Gray CA, Taylor KM, Bazer FW, Spencer TE 2000 Mechanisms regulating norgestomet inhibition of endometrial gland morphogenesis in the neonatal ovine uterus. Mol Reprod Dev 57:67–78
Safe S, Kim K 2004 Nuclear receptor-mediated transactivation through interaction with Sp proteins. Prog Nucleic Acids Res Mol Biol 77:1–36
Kushner PJ, Agard DA, Greene GL, Scanlan TS, Shiau AK, Uht RM, Webb P 2000 Estrogen receptor pathways to AP-1. J Steroid Biochem Mol Biol 74:311–317
Sjottem E, Anderssen S, Johansen T 1996 The promoter activity of long terminal repeats of the HERV-H family of human retrovirus-like elements is critically dependent on Sp1 family proteins interacting with a GC/GT box located immediately 3' to the TATA box. J Virol 70:188–198
Black AR, Black JD, Azizkhan-Clifford J 2001 Sp1 and kruppel-like factor family of transcription factors in cell growth regulation and cancer. J Cell Physiol 188:143–160
Wang X, Talamantez JL, Adamo ML 1998 A CACCC box in the proximal exon 2 promoter of the rat insulin-like growth factor I gene is required for basal promoter activity. Endocrinology 139:1054–1066
Marini MG, Asunis I, Porcu L, Salgo MG, Loi MG, Brucchietti A, Cao A, Moi P 2004 The distal -globin CACCC box is required for maximal stimulation of the -globin gene by EKLF. Br J Haematol 127:114–117
Stark GR, Kerr IM, Williams BR, Silverman RH, Schreiber RD 1998 How cells respond to interferons. Annu Rev Biochem 67:227–264
Stewart MD, Johnson GA, Bazer FW, Spencer TE 2001 Interferon- (IFN) regulation of IFN-stimulated gene expression in cell lines lacking specific IFN-signaling components. Endocrinology 142:1786–1794
Stewart MD, Choi Y, Johnson GA, Yu-Lee LY, Bazer FW, Spencer TE 2002 Roles of Stat1, Stat2, and interferon regulatory factor-9 (IRF-9) in interferon regulation of IRF-1. Biol Reprod 66:393–400
Kim S, Choi Y, Bazer FW, Spencer TE 2003 Identification of genes in the ovine endometrium regulated by interferon independent of signal transducer and activator of transcription 1. Endocrinology 144:5203–5214
Safe S 2001 Transcriptional activation of genes by 17 -estradiol through estrogen receptor-Sp1 interactions. Vitam Horm 62:231–252
Frasor J, Stossi F, Danes JM, Komm B, Lyttle CR, Katzenellenbogen BS 2004 Selective estrogen receptor modulators: discrimination of agonistic versus antagonistic activities by gene expression profiling in breast cancer cells. Cancer Res 64:1522–1533
Kim K, Thu N, Saville B, Safe S 2003 Domains of estrogen receptor (ER) required for ER/Sp1-mediated activation of GC-rich promoters by estrogens and antiestrogens in breast cancer cells. Mol Endocrinol 17:804–817
Inoue T, Kimura T, Azuma C, Inazawa J, Takemura M, Kikuchi T, Kubota Y, Ogita K, Saji F 1994 Structural organization of the human oxytocin receptor gene. J Biol Chem 269:32451–32456
Rozen F, Russo C, Banville D, Zingg HH 1995 Structure, characterization, and expression of the rat oxytocin receptor gene. Proc Natl Acad Sci USA 92:200–204
Bale TL, Dorsa DM 1997 Cloning, novel promoter sequence, and estrogen regulation of a rat oxytocin receptor gene. Endocrinology 138:1151–1158
Ivell R, Bathgate R, Rust W, Balvers M, Morley S 1995 Structure and organization of the bovine oxytocin receptor gene. Adv Exp Med Biol 395:295–300
Bathgate R, Rust W, Balvers M, Hartung S, Morley S, Ivell R 1995 Structure and expression of the bovine oxytocin receptor gene. DNA Cell Biol 14:1037–1048
Kubota Y, Kimura T, Hashimoto K, Tokugawa Y, Nobunaga K, Azuma C, Saji F, Murata Y 1996 Structure and expression of the mouse oxytocin receptor gene. Mol Cell Endocrinol 124:25–32
Young LJ, Waymire KG, Nilsen R, Macgregor GR, Wang Z, Insel TR 1997 The 5' flanking region of the monogamous prairie vole oxytocin receptor gene directs tissue-specific expression in transgenic mice. Ann NY Acad Sci 807:514–517
Schmid B, Wong S, Mitchell BF 2001 Transcriptional regulation of oxytocin receptor by interleukin-1 and interleukin-6. Endocrinology 142:1380–1385
Copland JA, Jeng YJ, Strakova Z, Ives KL, Hellmich MR, Soloff MS 1999 Demonstration of functional oxytocin receptors in human breast Hs578T cells and their up-regulation through a protein kinase C-dependent pathway. Endocrinology 140:2258–2267
Jeng YJ, Lolait SJ, Soloff MS 1998 Induction of oxytocin receptor gene expression in rabbit amnion cells. Endocrinology 139:3449–3455
Bale TL, Dorsa DM 1998 NGF, cyclic AMP, and phorbol esters regulate oxytocin receptor gene transcription in SK-N-SH and MCF7 cells. Brain Res Mol Brain Res 53:130–137
Hoare S, Copland JA, Wood TG, Jeng YJ, Izban MG, Soloff MS 1999 Identification of a GABP / binding site involved in the induction of oxytocin receptor gene expression in human breast cells, potentiation by c-Fos/c-Jun. Endocrinology 140:2268–2279
Spencer TE, Mirando MA, Mayes JS, Watson GH, Ott TL, Bazer FW 1996 Effects of interferon- and progesterone on oestrogen-stimulated expression of receptors for oestrogen, progesterone and oxytocin in the endometrium of ovariectomized ewes. Reprod Fertil Dev 8:843–853
Smith CL 1998 Cross-talk between peptide growth factor and estrogen receptor signaling pathways. Biol Reprod 58:627–632
Stevenson KR, Gilmour RS, Wathes DC 1994 Localization of insulin-like growth factor-I (IGF-I) and -II messenger ribonucleic acid and type 1 IGF receptors in the ovine uterus during the estrous cycle and early pregnancy. Endocrinology 134:1655–1664
Saville B, Wormke M, Wang F, Nguyen T, Enmark E, Kuiper G, Gustafsson JA, Safe S 2000 Ligand-, cell-, and estrogen receptor subtype (/)-dependent activation at GC-rich (Sp1) promoter elements. J Biol Chem 275:5379–5387
Gao J, Mazella J, Seppala M, Tseng L 2001 Ligand activated hPR modulates the glycodelin promoter activity through the Sp1 sites in human endometrial adenocarcinoma cells. Mol Cell Endocrinol 176:97–102
Owen GI, Richer JK, Tung L, Takimoto G, Horwitz KB 1998 Progesterone regulates transcription of the p21(WAF1) cyclin-dependent kinase inhibitor gene through Sp1 and CBP/p300. J Biol Chem 273:10696–10701(JoAnn G. W. Fleming, Thomas E. Spencer, )
Abstract
Establishment of pregnancy in ruminants results from paracrine signaling by interferon (IFNT) from the conceptus to uterine endometrial luminal epithelia (LE) that prevents release of luteolytic prostaglandin F2 pulses. In cyclic and pregnant ewes, progesterone down-regulates progesterone receptor (PGR) gene expression in LE. In cyclic ewes, loss of PGR allows for increases in estrogen receptor (ESR1) and then oxytocin receptor (OXTR) gene expression followed by oxytocin-induced prostaglandin F2 pulses. In pregnant ewes, IFNT inhibits transcription of the ESR1 gene, which presumably inhibits OXTR gene transcription. Alternatively, IFNT may directly inhibit OXTR gene transcription. The 5' promoter/enhancer region of the ovine OXTR gene was cloned and found to contain predicted binding sites for activator protein 1, SP1, and PGR, but not for ESR1. Deletion analysis showed that the basal promoter activity was dependent on the region from –144 to –4 bp that contained only SP1 sites. IFNT did not affect activity of the OXTR promoter. In cells transfected with ESR1, E2, and ICI 182,780 increased promoter activity due to GC-rich SP1 binding sites at positions –104 and –64. Mutation analyses showed that the proximal SP1 sites mediated ESR1 action as well as basal activity of the promoter. In response to progesterone, progesterone receptor B also increased OXTR promoter activity. SP1 protein was constitutively expressed and abundant in the LE of the ovine uterus. These results support the hypothesis that the antiluteolytic effects of IFNT are mediated by direct inhibition or silencing of ESR1 gene transcription, thereby precluding ESR1/SP1 from stimulating OXTR gene transcription.
Introduction
MATERNAL RECOGNITION OF pregnancy is the physiological process whereby the conceptus signals its presence to the maternal system and prolongs lifespan of the corpus luteum (CL) (1). Sheep experience uterine-dependent estrous cycles until establishment of pregnancy (2). The estrous cycle is dependent on the uterus, because it releases prostaglandin F2 (PGF) in a pulsatile manner to induce luteolysis during late diestrus. The luteolytic pulses of PGF are produced by the endometrial luminal epithelium (LE) and superficial ductal glandular epithelium (sGE) and are generated by oxytocin binding to oxytocin receptors (OXTR) on those epithelia (3, 4, 5, 6). Endometrial epithelial expression of OXTR is regulated primarily by receptors for progesterone [progesterone receptor (PGR)] and estrogen [estrogen receptor (ESR1)] (7, 8, 9). During proestrus and estrus (d –3 to 0), estrogen from ovarian follicles increases ESR1, PGR, and OXTR expression. During diestrus (d 3–15), progesterone levels increase and act via PGR to "block" expression of ESR1 and OXTR in LE and glandular epithelium (GE) between d 5 and 11 after onset of estrus (10, 11, 12). However, continuous exposure of the uterus to progesterone down-regulates expression of PGR in LE/sGE after d 11 and GE after d 13, allowing for rapid increases in expression of ESR1 on d 12–13 and then OXTR on d 14 in those epithelia (9, 10, 13, 14). ESR1, presumably activated by estrogen from ovarian follicles or possibly growth factors from the stroma, stimulates transcription of the OXTR gene in the ovine endometrium (15, 16, 17, 18). Oxytocin, secreted from as early as d 9 from posterior pituitary and/or the CL, binds to OXTR to induce pulsatile release of luteolytic PGF between d 14–16 (3). In response to four to five luteolytic pulses of PGF over a 25-h period, the CL undergoes functional and structural regression. The systemic loss in progesterone between d 15–16 allows the ewe to return to estrus, complete the 17-d estrous cycle, and experience another opportunity to mate and establish pregnancy.
Interferon (IFNT), the pregnancy recognition signal in ruminants, is expressed by conceptus trophectoderm between d 10 and 21 of gestation (19) and acts in a paracrine antiluteolytic manner on the endometrial LE and sGE to inhibit OXTR gene expression (10, 13, 20). IFNT does not inhibit basal production of PGF, which is higher in pregnant than cyclic ewes, and the conceptus and IFNT do not decrease expression of PTGS2 [prostaglandin-endoperoxide synthase 2 (prostaglandin G/H synthase and cyclooxygenase)], the rate-limiting enzyme in prostaglandin production, in endometrial epithelia of pregnant ewes (5, 21). The antiluteolytic effects of IFNT to inhibit OXTR gene expression in the endometrial LE and sGE have been hypothesized to involve: 1) direct stabilization of epithelial PGR expression that, in turn, extends progesterone inhibition of ESR1 and OXTR expression; 2) direct inhibition or silencing of ESR1 gene transcription that, in turn, indirectly inhibits OXTR gene expression; and 3) direct inhibition of OXTR gene transcription. IFNT does not inhibit PGR loss in the endometrial epithelia (12, 16), but rather appears to indirectly inhibit OXTR gene transcription by silencing ESR1 gene transcription in the endometrial LE (15, 16, 22). Indeed, the 2.7-kb 5' promoter/enhancer region of the ovine ESR1 gene was found to contain four interferon regulatory factor elements (IRFEs) and one interferon-stimulated response element (ISRE) that were functional in binding interferon regulatory factor (IRF)2 (23). IRF2 is a transcriptional repressor that is expressed specifically in the endometrial LE and sGE of the ovine uterus and increases in expression from d 10–16 of pregnant but not cyclic ewes (24). In transfection assays, IFNT inhibited transcriptional activity of the ovine ESR1 promoter, and analyses of sequential 5'-deletion mutants of the ovine ESR1 promoter indicated that the effects of IFNT may be mediated by IRFEs as well as other elements (23). Physiological evidence indicates that estrogen stimulates OXTR gene expression and that IFNT does not directly inhibit OXTR gene expression in the sheep uterus (15, 25). Direct effects of estrogen and IFNT on the ovine OXTR gene have not been studied previously, because the gene had not been cloned. The bovine OXTR gene promoter region was cloned and found to contain an IRFE, ESR1 response element (ERE) half sites, and SP1 sites, and could be transactivated by estrogen if cells were cotransfected with ESR1 and steroid receptor coactivator 1 (26). Curiously, IRF2 overexpression increased activity of the bovine OXTR promoter, but a direct effect of IFNT on promoter activity was not reported. In all mammals, estrogen is considered a key regulator of OXTR gene expression; however, all of the studied OXTR genes lack a complete ERE, suggesting estrogen induction of OXTR gene transcription may be indirect rather than due to direct ESR1/ERE interactions (27).
Our working hypothesis is that estrogen acts indirectly to regulate OXTR gene expression in the endometrium of the ovine uterus and that the antiluteolytic effects of IFNT are manifest on silencing or inhibition of ESR1 gene transcription in LE and sGE subsequent to PGR down-regulation, which then precludes the ability of estrogen or perhaps growth factors to activate ESR1 and stimulate OXTR gene expression. These actions of IFNT prevent formation of OXTR and abrogate uterine release of luteolytic pulses of PGF (4, 8, 23, 28). The objective of the present studies was to determine if estrogen and/or IFNT regulate OXTR promoter activity. The promoter region of the ovine OXTR gene was cloned, sequenced, and analyzed for functional cis-elements mediating effects of steroid hormones and IFNT. Relevant findings are that IFNT does not affect OXTR promoter activity nor inhibit estrogen induction of OXTR promoter activity. Two GC-rich SP1 binding sites, located within 140 bp of the translational start site, regulated basal activity of the promoter as well as hormonal responsiveness to 17-estradiol (E2), ICI 182,780, and progesterone. Thus, estrogen regulation of ovine OXTR promoter activity involves ESR1/SP1 interactions.
Materials and Methods
Cells and reagents
The 2fTGH (parental) and U3A (STAT1-deficient 2fTGH) immortalized cells (29) were maintained in DMEM-F12 medium (Sigma-Aldrich Corp., St. Louis, MO) supplemented with 5% fetal bovine serum (FBS) (Hyclone, Logan, UT) and penicillin/streptomycin sulfate/amphotericin B solution (Invitrogen, Carlsbad, CA). Recombinant ovine IFNT (108 antiviral units/mg) was prepared and assayed as described previously (30). Restriction endonucleases, T4 DNA ligase, T4 DNA kinase, and recombinant human SP1 protein were purchased from Promega (Madison, WI). Vent Taq polymerase (New England Biolabs, Beverly, MA), AmpliTaq polymerase (Applied Biosystems, Foster City, CA), and ExTaq polymerase (Takara, Kyoto, Japan) were used. R5020 was purchased from PerkinElmer Life Sciences (Boston, MA), and E2 was from Sigma-Aldrich Corp. Plasmid DNAs were purified by the alkaline lysis method according to the manufacturer’s instructions (Qiagen, Valencia, CA).
Cloning of the 5' upstream region and luciferase constructs
A 273-bp DNA fragment of the 5' end of the coding region of the ovine OXTR gene was PCR amplified in 25-μl reactions containing 40 ng ovine genomic DNA (a gift from Dr. C. Gill, Texas A&M University, College Station, TX), PCR Optimized Buffer N (Invitrogen), 125 μM dNTPs, 2.5 μM forward primer (5'-GGAGGCGGTCAACGGGAG-3'), 2.5 μM reverse primer (5'-CGTGATGTCCCACAGAAGC-3'), and 1 U AmpliTaq polymerase using an Eppendorf Mastercycler thermocycler with conditions of: 1) 94 C for 5 min; 2) 94 C for 30 sec, 53 C for 30 sec, and 72 C for 30 sec for 35 cycles; and 3) 72 C for 7 min. The product was cloned into pCRII Dual using the TA Cloning kit (Invitrogen). The cloned 273-bp OXTR fragment was excised with EcoRI, random prime labeled with [-32P]dCTP, and used to screen an ovine genomic library cloned in Bluestar (Novagen, Madison, WI), which was kindly provided by Dr. J. C. DeMartini (Colorado State University, Fort Collins, CO). A recombinant pBluestar phagemid DNA, containing approximately 9.8 kb of ovine genomic DNA, was excised from plaque-purified isolates by Cre-Lox excision according to the manufacturer’s instructions. Physical mapping, Southern blotting with the 273-bp OXTR probe, and sequencing with T7 and T3 primers and several OXTR-specific primers were done to identify the OXTR coding and upstream sequences in the clone. The 5' terminus of the coding sequence and the contiguous upstream sequences were sequenced on both strands by the dideoxy chain termination method. The OXTR upstream region was immediately adjacent to the NotI site in the vector multiple cloning site. The upstream region was excised as an 826-bp fragment using the vector NotI and a NotI site at –4 relative to the ATG (+1) of the OXTR coding sequences and cloned in pCRII Dual in both orientations as determined by physical mapping of the unique BglII site at –144. The OXTR(–830/–4) upstream region then was directionally subcloned into the luciferase vector pGL3Basic (Promega). Computer-assisted prediction of transcription factor binding sites in the promoter sequence was performed with TESS (Transcription Element Search System) using TRANSFAC version 4.0 (30) and MatInspector using Matrix Family Library Version 2.4 (Genomatix, Munich, Germany) (31).
Truncations of the upstream region at approximately 100-bp intervals were made by PCR amplification in 50-μl reactions containing 1 ng pCRII-OXTR(–830/–4), 25 mM TAPS (N-[Tris(hydroxymethyl)methyl]-3-aminopropane-sulfonic acid, pH 9.3 at room temperature), 50 mM KCl, 2 mM MgCl2, 200 μM dNTPs, 0.2 μM each forward primer (Table 1), 0.2 μM T7 primer as reverse primer, and 1.25 U ExTaq in a Perkin-Elmer 9700 thermocycler using conditions of: 1) 94 C for 2 min; 2) 94 C for 1 min, 50 C for 1 min, 72 C for 1 min for 30 cycles; and 3) 72 C for 10 min. The truncated PCR products were cloned into pCRII Dual, physically mapped to determine orientation, and directionally subcloned into the pGL3Basic vector. The OXTR(–144/–4) LUC truncated clone was made by digestion of OXTR(–220/–4) LUC plasmid DNA with BglII, reinserting the BglII (–144/–4) fragment into the luciferase vector, and physically mapping the construct to confirm proper orientation. Point mutations were introduced into the SP1 sites at –104 and –64 of the OXTR(–220/–4) truncated pCRII subclone by PCR-directed mutagenesis using the primers listed in Table 1 and Vent Taq polymerase and ExTaq (31). A OXTR(–220/–4) pCRII clone containing an introduced mutation at the –64 SP1 site was used as the template for the construction of OXTR–220 constructs with mutated SP1 sites at both –104 and –64. Mutated DNAs were confirmed by sequencing, and fragments containing only the desired point mutations were directionally subcloned into the luciferase vector as described above.
Transient transfections and luciferase assays
Cells were subcultured into 12-well plates (67–75% confluent) and transiently transfected as described previously (23) with the following modifications. Luciferase constructs (500 ng/well) were cotransfected with either pEF1-Myc-His-LacZ (500 ng/well; Invitrogen) or expression plasmids for ovine IRF1, ovine IRF2, human ESR1wt, human ESR1IIc, or human progesterone receptor B (PGR-B) (500 ng/well except where noted). The ovine IRF1 and IRF2 mammalian overexpression plasmids have been described previously (24). The ESR1wt (wild-type human ESR1) and ESR1IIc (a DNA binding domain mutant construct) overexpression plasmids have been described previously (32, 33). The PGR-B plasmid, pSV40-hPGR-B that overexpresses the B form of human PGR, was kindly provided by Dr. M.-j. Tsai (Baylor College of Medicine, Houston, TX). Transfected cells were grown in 10% FBS for approximately 14–16 h before treatment for 24 h in serum-free medium. Phenol red-free DMEM-F12 medium and dextran-coated charcoal-stripped FBS were substituted in experiments when testing for effects of steroids. Steroid agonists and antagonists were dissolved in 100% ethanol. For the control, cells were treated with the same volume of 100% ethanol alone. Luciferase and protein assays were done as described previously (23, 24). Transfection assays were repeated a minimum of three times.
EMSA
Double-stranded oligonucleotide primers listed in Table 1 were end-labeled to high specific activity with [-32P]ATP and T4 DNA kinase by standard methods. Purified SP1 protein (54 ng/reaction) was incubated in 10 μl reactions containing 4% glycerol, 1 mM MgCl2, 0.5 mM EDTA, 50 mM NaCl, 10 mM Tris-HCl (pH 7.5), 0.5 mM DTT, 0.05 mg/ml poly(dI-dC) at room temperature for 15 min. A 100-fold excess of unlabeled specific or nonspecific competitor oligonucleotides was included in some reactions. The nonspecific competitor was a 22-nt oligonucleotide with the same base composition as the SP1 consensus oligonucleotide to minimize charge differences, but having a different sequence (Table 1). Radiolabeled probes (10 fmol per reaction) were added to reactions that were incubated at room temperature for an additional 15 min. Binding reactions were electrophoresed in 5% native polyacrylamide gels in 1x TBE. Imaging of dried gels was done with a Molecular Dynamics Typhoon phosphorimager.
Immunohistochemistry
Mature crossbred Suffolk ewes (Ovis aries) were observed daily for estrus in the presence of vasectomized rams and used in experiments only after they had exhibited at least two estrous cycles of normal duration (16–18 d). All experimental and surgical procedures were in compliance with the Guide for the Care and Use of Agriculture Animals and approved by the University Laboratory Animal Care and Use Committee of Texas A&M University.
At estrus (d 0), ewes were mated to either an intact or vasectomized ram as described previously (34) and then hysterectomized (n = 5 ewes per day) on either d 10, 12, 14, or 16 of the estrous cycle or d 10, 12, 14, 16, 18, or 20 of pregnancy. Pregnancy was confirmed on d 10–16 after mating by the presence of a morphologically normal conceptus(es) in the uterus. At hysterectomy, several sections (0.5 cm) from the mid-portion of each uterine horn ipsilateral to the CL were fixed in fresh 4% paraformaldehyde in PBS (pH 7.2). After 24 h, fixed tissues were changed to 70% ethanol for 24 h and then dehydrated and embedded in Paraplast-Plus (Oxford Labware, St. Louis, MO). In monovulatory pregnant ewes, uterine tissue samples were marked as either contralateral or ipsilateral to the ovary bearing the CL. No tissues from the contralateral uterine horn were used for study.
Immunoreactive SP1 proteins were localized in cross-sections (5 μm) of the uterus using a rabbit polyclonal antibody to human SP1 (catalog SC-59; Santa Cruz Biotechnology Inc., Santa Cruz, CA) at a final working concentration of 0.1 μg/ml and a VectaStain Rabbit IgG Elite ABC kit (Vector Laboratories, Burlingame, CA) using methods described previously (35). Antigen retrieval using boiling citrate buffer was performed as described previously (35, 36). The chromagen used for peroxidase localization was 3,3'-diaminobenzidine tetrahydrochloride from Sigma Chemical Co. (St. Louis, MO). Negative controls were performed in which the primary antibody was substituted with the same concentration of purified normal rabbit IgG from Sigma Chemical Co. Multiple tissue sections from each ewe were processed as sets within an experiment. Sections were not counterstained before affixing coverslips. Representative photomicrographs of uterine tissues were taken using a Nikon Eclipse 1000 photomicroscope (Nikon Instruments Inc., Lewisville, TX) fitted with a Nikon DXM1200 digital camera. Digital images were captured and assembled using Adobe Photoshop (Adobe Systems, Seattle, WA).
Statistical analyses
The effects of steroid hormones, ICI 182,780, or IFNT on the activity of promoter reporter constructs in transient transfection assays were analyzed by least squares ANOVA using the General Linear Models procedure of the Statistical Analysis System (Cary, NC). A P value of 0.10 or less was considered statistically significant. Unless denoted, data are reported as mean with SD. Effects of treatment were determined using orthogonal or nonorthogonal contrasts using the PDIFF option of General Linear Models.
Results
Isolation and sequence analysis of the OXTR 5' upstream region
A phagemid DNA containing approximately 9.8 kbp of ovine genomic DNA was isolated from a ovine genomic library by conventional screening. Southern blotting, physical mapping, and sequence analysis with several OXTR-specific primers revealed that the clone contained a 1059-bp region that included the 5' end of the coding sequence of the ovine OXTR gene contiguous to 830 nt of upstream sequence. The 1059-bp clone was sequenced in both directions and deposited into GenBank (accession AY163261). Sequence comparisons of the ovine upstream region with the previously characterized Bos taurus OXTR promoter region and adjacent coding regions (GenBank AF100633) indicated that the two sequences were 91% identical. Although the coding sequence displayed high homology with the OXTR cDNA from other species (human, dog, pig, gorilla, rat, mouse, chicken), the promoter region had very little or no homology to those of other species except for bovine.
Bioinformatic analyses found that the ovine OXTR promoter region contained multiple putative binding sites for the transcription factors SP1 and activator protein 1 (AP-1), which regulates target genes by steroid hormone receptors (see Refs.37 and 38 for review). Putative nonconsensus SP1 sites were found at –748, –543, and –444 relative to the translational start site (ATG +1) in addition to consensus sites at –104 and –64 (Fig. 1). The putative –64 SP1 binding site is identical to the consensus SP1 binding motif (5'-CCCCGCCCC-3', sense strand). The –104 putative binding site (5'-CCCCACCCCGCCCC-3', sense strand) also contains a consensus binding motif that is immediately adjacent to and overlaps a sequence that resembles a nonconsensus SP1 binding site (5'-CCCCACCCC-3') (Fig. 1). Nonconsensus SP1 sites having the sequence 5'-CCCCACCCC-3' have been reported for the long terminal repeats of the human endogenous retrovirus H family of human retrovirus-like elements (39). Alternatively, part of the sequence of the OXTR –104 SP1 site may contain a CACCC box at the 5' end. The CACCC elements bind SP1 and related Sp/Kruppel-like factor transcription factors (40) and are involved in regulation of both basal (41) and induced levels of transcription in other systems (42). A consensus AP-1 binding site at –680 and nonconsensus AP-1 sites at –574, –329, and –221 were identified, which also mediate hormone actions (38). No full progesterone response elements (PREs) were found, but several putative PRE half sites were detected at –639, –486, –477, and –187. No full EREs or ERE half sites were found by computer-assisted analysis of the ovine OXTR upstream region. Furthermore, elements mediating effects of type I or II IFNs or STATs, including ISRE, IRFE, or activation sites (GAS) (43), were also not found by bioinformatic analyses.
IFNT does not regulate OXTR promoter activity
To determine whether IFNT regulated activity of the OXTR promoter, transient transfection assays were conducted in 2fTGH parental and U3A (STAT1 null) cells, which are responsive to IFNT (44, 45, 46) in a STAT-dependent (2fTGH) and STAT1-independent (U3A) manner. These cells have been extensively used by our laboratory as a model for the endometrial stroma and LE, respectively (44, 45, 46). Cells were transfected with the full-length OXTR(–830/–4) and sequential 5'-deletion reporter constructs and treated with different doses of recombinant ovine IFNT (103–106 antiviral units). No dose-dependent effects of IFNT on promoter activity were detected (P > 0.10) with any of the OXTR promoter-reporter constructs in either 2fTGH or U3A cells (data not shown). Cells were also cotransfected with the OXTR(–830/–4) promoter-reporter construct and either pEF1-Myc/His-LacZ, ovine IRF1, or ovine IRF2 overexpression vectors. Relative to the LacZ control, overexpression of ovine IRF1 or ovine IRF2 did not affect (P > 0.10) activity of the OXTR promoter constructs in either cell type (data not shown). These results are consistent with the lack of IFN-responsive elements (GAS, ISRE, or IRFE) in the ovine OXTR promoter.
In 2fTGH cells transfected with the OXTR(–830/–4) construct and ESR1wt, E2 (10–8 M) increased (P < 0.01) luciferase activity, and this induction was not affected (P > 0.10) by IFNT treatment (Fig. 2). Treatment with IFNT lowered (P < 0.05) the basal activity of the ovine OXTR promoter; however, the fold induction by E2 was the same in untreated and IFNT-treated cells. Similar results were obtained using U3A cells (data not shown). The basal activity of several promoter-reporter constructs appears to be nonspecifically affected by IFNT, presumably due to effects on cell metabolism due to induction of an antiviral state (data not shown). These results indicate that IFNT does not inhibit E2-induced transactivation in cells transfected with OXTR promoter-derived constructs.
Estrogen and antiestrogen stimulate OXTR promoter activity
In 2fTGH and U3A cells, basal activity of the OXTR promoter changed with length of the promoter-reporter construct (Figs. 3 and 4). The full-length OXTR(–830/–4)-LUC construct had considerably lower basal activity compared (P < 0.01) to all other 5'-deletion constructs. Basal activity of the OXTR(–340/–4) construct was approximately 1.5-fold greater (P < 0.05) than the OXTR(–220/–4) construct containing the minimal promoter.
2fTGH cells were cotransfected with OXTR promoter-reporter constructs and ESR1wt (Fig. 3A) or ESR1IIc (Fig. 4B), a DNA binding domain mutant derived from ESR1wt, and treated with vehicle as a control or 10–8 M E2 for 24 h. In cells transfected with ESR1wt, E2 treatment increased (P < 0.05) activity of the OXTR(–830/–4) promoter-LUC construct as well as all 5' deletions (Fig. 3A). Similarly, E2 treatment increased (P < 0.01) activity of all OXTR promoter-reporter constructs in cells cotransfected with ESR1IIc (Fig. 4A). The fold increase in OXTR promoter activity was consistently greater in cells transfected with ESR1IIc than ESR1wt. Dose response studies found that E2 at a concentration of greater than 10–9 M increased activity of the full-length OXTR(–830/–4) promoter as well as the minimal OXTR(–220/–4) promoter in cells cotransfected with ESR1wt (Fig. 3B) or ESR1IIc (Fig. 4B). These results indicate that cis elements in the minimal promoter (–220/–4) are sufficient to mediate ligand-activated ESR1 effects. Furthermore, ESR1 induction of the OXTR promoter does not require DNA binding, suggesting protein-protein interactions through another transcription factor. Estrogen induces activity of many genes with GC-rich promoters in breast cancer cells transfected with ESR1 or ESR1IIc, even though the genes have no EREs in their promoters (see Refs.37 and 47).
ICI 182,780 is a pure antiestrogen that inhibits liganded ESR1 from transactivating estrogen-responsive genes containing an ERE(s) (48). However, ICI can exhibit partial agonist activity in some cells and induce ESR1 activation of some estrogen-responsive genes containing consensus GC-rich motifs in the promoter (49). To examine the effects of E2 and ICI 182,780 on ovine OXTR promoter activity, 2fTGH and U3A cells were cotransfected with ESR1wt or ESR1IIc and OXTR(–830/–4)-LUC, OXTR(–220/–4)-LUC, 3xERE-TATA-LUC, or 3xSP1-LUC reporters. Cells were then treated for 24 h with E2 (10–8 M) or ICI 182,780 (10–6 M). In 2fTGH cells, E2 stimulated (P < 0.01) activity of the OXTR reporters in cells transfected with ESR1wt or ESR1IIc (Fig. 5). E2 also induced (P < 0.01) activity of 3xERE-TATA-LUC in cells cotransfected with ESR1wt but not ESR1IIc, consistent with the requirement for DNA bound ESR1. ICI stimulated (P < 0.01) OXTR promoter activity in cells cotransfected with ESR1wt and ESR1IIc. As expected, ICI did not affect activity of the 3xERE-TATA-LUC in 2fTGH cells cotransfected with ESR1IIc. Both E2 and ICI stimulated (P < 0.01) activity of the 3xSP1-LUC reporter in 2fTGH cells transfected with either ESR1wt or ESR1IIc. These results are consistent with E2 and ICI activation of ovine OXTR promoter activity through an ESR1/SP1-mediated pathway as previously observed for several hormone-responsive genes with GC-rich promoters (see Refs.37 and 47).
Minimal OXTR promoter contains functional SP1 elements
Deletion analysis showed that the basal promoter activity and minimal hormone-responsive region was between –220 to –4 bp in the OXTR promoter. This region contains an AP-1 site, PRE half site, and three SP1 sites (Fig. 1). The functionality of the SP1 elements was determined by EMSA using recombinant human SP1 protein (Fig. 6). As illustrated in Fig. 6A, the predicted SP1 binding site at –64 in the OXTR promoter bound SP1 protein (lane 6). As a positive control, SP1 protein-DNA binding was observed with an oligonucleotide containing a consensus SP1 binding site (lane 2). SP1 binding to the –64 OXTR oligonucleotide was specific, because it was competitively decreased by incubation with a 100-fold excess of unlabeled SP1 consensus oligonucleotide (lane 5) or the unlabeled –64 OXTR oligonucleotide itself (lane 3). Mutation of two critical C residues in the –64 OXTR oligonucleotide rendered it unable to bind SP1 protein (lane 7) or compete for binding of SP1 protein to the wild-type –64 OXTR oligonucleotide (lane 4). In other experiments, 100-fold excess of unlabeled wt –64 OXTR oligonucleotide prevented binding of SP1 protein to the consensus SP1 oligonucleotide, whereas the mutant –64 OXTR oligonucleotide was unable to compete (data not shown). These results demonstrate that the ovine OXTR gene promoter contains a high affinity SP1 binding site at position –64.
As illustrated in Fig. 6B, the GC-rich sites at –104 in the ovine OXTR promoter also bound SP1 protein (lane 2) in EMSAs. Incubation with a 100-fold excess of the wild-type –104 OXTR oligonucleotide (lane 3) or consensus SP1 oligonucleotide (lane 4), but not a nonspecific oligonucleotide (lane 5), competitively decreased SP1 protein binding to the radiolabeled –104 OXTR oligonucleotide. The sequence of the –104 SP1 binding site in the OXTR promoter contains a 5' nonconsensus GC-rich SP1 binding site adjacent to a 3' consensus site (Fig. 1). Mutation of the 3' consensus GC-rich site prevented SP1 protein binding (mut1) (lane 6), whereas mutation of the 5' nonconsensus binding site in the –104 OXTR oligonucleotide (mut2) did not affect SP1 binding (lane 7). Furthermore, SP1 binding was not observed when both the 5' and 3' predicted SP1 elements were mutated together (mut4) (lane 8). Collectively, EMSA results indicate that two GC-rich sites in the ovine OXTR promoter at –64 and –104 bp bind SP1 protein.
To determine whether the GC-rich sites at –104 and –64 are critical for responsiveness to ESR1, these sites in the –220 OXTR-LUC reporter construct were mutated (see Fig. 1). Due to the complex structure of the –104 SP1 element, mutation of either the 5' nonconsensus SP1 site or the 3' consensus SP1 element was performed, because it could alter the integrity of the overlapping site and affect the ability of the –104 sequence to bind SP1. Therefore, in addition to mutations of the –104 site tested in initial gel shift analyses, some CA transversions that mutated either the consensus or nonconsensus portion of the site, without altering the other overlapping moiety, were also introduced. The oligonucleotides used to create the mutations were tested initially for their ability to bind SP1 protein by EMSA (Fig. 7). Mutation of the 5' nonconsensus site did not affect SP1 binding regardless of whether the overlapping 3' consensus portion remained wild-type (mut3; lane 5) or was slightly altered (mut2; lane 4). Mutation of the 3' consensus sequence alone (mut1, lane 3) substantially reduced SP1 binding compared with the wild-type –104 OXTR oligonucleotide (lane 2), but did not eliminate binding entirely. Mutation of both 5' and 3' sites (mut4) completely prevented SP1 binding (lane 6). In competition experiments (Fig. 7B), excess unlabeled wt –104 oligonucleotide and the mut1, mut2, and mut3 oligonucleotides reduced or prevented SP1 binding to the radiolabeled wt –104 oligonucleotide (lanes 6, 7, and 8). As expected, competition with the fully mutated mut4 oligonucleotide did not affect SP1 binding to the radiolabeled –104 OXTR wt oligonucleotide (lane 9). Collectively, these results indicate that SP1 binding to the –104 OXTR sequence depends primarily on the 3' consensus GC-rich element. However, the 5' nonconsensus GC-rich element appears to weakly bind SP1 protein, suggesting that the two overlapping motifs at –104 may act cooperatively to bind SP1.
Transient transfection assays were then conducted to determine the relative contributions of the –104 and –64 SP1 binding sites on hormonal responsiveness of the ovine OXTR promoter (Fig. 8). Wild-type or mutant OXTR(–220/–4) promoter-reporter constructs were cotransfected with ESR1IIc into 2fTGH cells and left unstimulated or treated with E2 (10–8 M). Mutation of either the 5' nonconsensus site (constructs 2 and 3), the 3' consensus site (construct 4), or both sites (construct 5) at –104 reduced (P < 0.05) basal luciferase expression in unstimulated cells by 25–40% compared with the wild-type reporter (construct 1) depending on the location of the mutation. Mutation of the –64 SP1 site reduced (P < 0.01) basal luciferase expression to very low levels, regardless of whether the –104 site was wt (construct 6), double mutated (constructs 7 and 9), or point mutated (construct 8). Indeed, basal luciferase levels of the –64 SP1 mutants were about 10% of wild-type parental levels. Mutation of the 5' consensus SP1 site or both the nonconsensus and consensus sites at –104 and the –64 SP1 site (construct 10) further reduced (P < 0.10) basal activity to about 2% of activity observed for the wt construct.
Estradiol-induced activity, in all cells transfected with the minimal OXTR(–220/–4) promoter, was lower (P < 0.10) in constructs containing a mutation in the –104 and –64 SP1 binding sites (Fig. 8). Mutation of both sites at position –104 further reduced stimulation by estradiol, and this was consistent with the EMSA data suggesting that both the left and right moieties are involved in binding SP1 protein. However, mutation of the –64 SP1 site decreased (P < 0.10) stimulation by estradiol, regardless of whether the –104 SP1 site was wt or mutated. Overall, these results suggest that both the –104 and –64 SP1 sites in the proximal OXTR promoter regulates basal levels of OXTR gene transcription, and that both SP1 binding sites, especially the consensus sequences, regulate E2 stimulation of OXTR gene expression.
Progesterone stimulates OXTR promoter activity
During metestrus and estrus in cyclic ewes, PGR return to the OXTR-expressing endometrial epithelia as systemic progesterone declines to undetectable levels (9, 10, 12, 13, 14). The prevailing theory is that PGR do not directly inhibit OXTR gene expression, but rather the progesterone block to OXTR formation is indirect via progesterone inhibition of ESR1 expression (4, 8, 28). To determine whether the OXTR promoter is regulated by PGR, 2fTGH cells were cotransfected with human PGR-B and OXTR promoter-LUC constructs and then treated with R5020 (10–8 M), a nonmetabolizable form of progesterone (Fig. 9). As observed previously, basal activity of the OXTR promoter constructs increased (P < 0.05) by deletion of the region from –830 to –711. Treatment with R5020 stimulated (P < 0.05) the activity of all OXTR promoter-reporter constructs. Next, the same experiment was repeated with the reporter constructs and PGR-B, but cells were treated with a range of R5020 doses. Dose-dependent effects of R5020 (10–14–10–5 M) on activity of the –830 and –220 OXTR promoter constructs were observed in 2fTGH cells cotransfected with PGR-B, with concentrations as low as 10–10 M R5020 stimulated transactivation in cells transfected with OXTR constructs (data not shown).
Deletion of the AP-1 and PRE half site does not affect E2 or R5020 stimulation of OXTR promoter activity
The minimal OXTR(–220/–4) promoter contains an AP-1 site at –220 and a PRE half site at position –187. To determine whether these sites are involved in hormone responsiveness, the sites were sequentially removed by 5' deletion. The OXTR promoter constructs were then transfected into 2fTGH cells along with ESR1IIc or PGR-B and treated with E2 (10–8 M) or R5020 (10–8 M) for 24 h, respectively (Fig. 10). Deletion of the AP-1 site and PRE half site did not affect (P > 0.10) the stimulatory effects of E2 and R5020 on OXTR promoter activity. These results indicate that the SP1 binding sites at –104 and –64 are the only cis elements required for basal activity of the OXTR promoter and its responsiveness to stimulation by ligand-activated ESR1 and PGR-B.
Immunolocalization of SP1 protein in the endometrium of cyclic and pregnant ewes
To confirm that SP1 was expressed in the endometrium, SP1 protein was studied in the uterus of cyclic and early pregnant ewes by immunohistochemistry (Fig. 11). Immunoreactive SP1 protein was observed in all endometrial cell types of uteri from ewes on d 10–16 of the cycle and d 10–20 of pregnancy. Although all endometrial cell types expressed SP1 protein, it was most abundant in the endometrial LE and GE. No differences in SP1 protein abundance were detected between the endometrial cells of cyclic and pregnant ewes or between LE of intercaruncular and caruncular areas of the endometrium (data not shown). In pregnant ewes, SP1 protein was also detected in conceptus trophectoderm.
Discussion
The gene for OXTR exists as a single copy in all studied species, and the promoter region of the OXTR genes of the human (50), rat (51, 52), bovine (53, 54), mouse (55), and vole (56) are available. Comparative studies of the promoter regions of homologous genes from several species often reveal conserved and important regulatory elements. However, the promoter regions of the ovine and bovine OXTR genes are not well conserved with other species, despite significant sequence conservation found in the coding sequence. The human OXTR gene promoter can be regulated by proinflammatory cytokines, such as IL-1 and IL-6 via activation of the binding elements for nuclear factor-IL-6 or STAT3 (57). Serum increases activity of the human OXTR promoter through a protein kinase C-dependent pathway that can be synergistically augmented by dexamethasone (58). In rabbit amnion cells, in vitro treatment with forskolin and/or cortisol activates OXTR gene transcription (59). Protein kinase A, protein kinase C, and nerve growth factor-dependent OXTR up-regulation is also observed in MCF-7 and SK-N-SH cells transfected with the rat OXTR gene promoter (60). In the human, basal promoter activity is conferred by the 85-bp flanking the 5' end of the human OXTR gene. In this sequence, there is a consensus ETS binding sequence. GABP (GA binding protein transcription factor), which binds to the ETS element, cooperates with AP-1 (FOS/JUN) to activate human OXTR promoter transcription in Hs578T cells (61). However, elements that regulate activity of the human and rat OXTR promoter/enhancer regions are not present in the ovine and bovine OXTR promoters, suggesting that the identified pathways regulating OXTR expression in other species may not be the same for ruminants.
In every studied mammal, estrogen is considered a key regulator of OXTR gene expression (27). Similarly in sheep, in vivo administration of estrogen increases OXTR mRNA and the number of oxytocin binding sites in the endometrium (15, 16, 22, 62). The classical pathway for estrogen regulation of target genes involves ligand-activated ESR1/ERE interactions or ERE half sites (37, 47). In the rat OXTR gene, a palindromic ERE was identified about 4 kb upstream of the transcriptional start site (52). This sequence was hormone-responsive in transfection studies in human MCF-7 breast cancer cells, but only when the 3.3-kb fragment between the ERE and the basal promoter of the rat OXTR gene was truncated. In the wild-type construct, this ERE was not activated by estrogen. However, the human OXTR gene does not have a palindromic ERE in the area comparable to that of the rat gene (27). Therefore, estrogen induction of ovine OXTR gene transcription may be indirect rather than due to direct ESR1/ERE interactions.
The present study found that an estrogen agonist (E2) and antagonist (ICI 182,780) transactivated the ovine OXTR promoter in cells transfected with ESR1. Deletion and mutation analyses found that induction of ovine OXTR promoter activity by E2-activated ESR1 is dependent on GC-rich SP1 binding sites at –104 and –64 in the minimal promoter, and basal activity of this promoter is also regulated by the same GC-rich SP1 elements adjacent to the translational start site of the OXTR gene. One caveat of the present studies is that human fibrosarcoma cells were used for transfection assays instead of an ovine endometrial epithelial cell line the endogenous expresses the OXTR gene. Indeed, the GC-rich motifs in several gene promoters are cis-acting elements for an increasing number of ligand-activated nuclear and orphan receptors that interact with SP1 and related proteins (see Refs.37 and 47). In the present study, SP1 transcription factor was constitutively expressed in nearly all cell types of the endometrium from both cyclic and pregnant ewes and was particularly abundant in LE and sGE. Therefore, the primary determinant of OXTR gene regulation in the endometrial epithelia is expression of the ESR1 gene. Administration of estrogen on d 11 or 12 postestrus induces structural and functional luteolysis in cyclic ewes with the following temporal events: 1) increase ESR1 mRNA and protein in endometrial epithelia between 12–24 h; 2) moderate increase in endometrial OXTR density between 12–36 h; 3) large increase in OXTR density between 36–48 h; 4) a decline in concentrations of progesterone in plasma after 36 h; and 5) decrease in CL weight by 48 h (15, 17). Similar temporospatial alterations in ESR1 and OXTR gene expression occur in the endometrium of cyclic ewes between d 12–16 postestrus (9, 10, 12). Although exogenous estrogen can clearly elicit development of the endometrial luteolytic mechanism and induce OXTR promoter activity, the precise physiological role of estrogen in the luteolytic mechanism in vivo during a natural estrous cycle has not been adequately investigated. Estrogen may not be needed to increase OXTR gene expression, because ESR1 can be activated in a ligand-dependent manner by estrogens or a ligand-independent manner by growth factors (63), such as IGF-I and IGF-II that are abundant in the ovine endometrial stroma (64). Furthermore, ICI 182,780, a pure antiestrogen, will not be useful for in vivo investigations, because it activated ESR1 and induced OXTR promoter activity in the present study. Similarly, the antiestrogens 4'-hydroxytamoxifen (4-OHT) and ICI 182,780 activate reporter gene activity in cells transfected with constructs containing GC-rich promoters with SP1 and SP3 binding sites (65). In addition to the OXTR, many other endometrial genes that are regulated by hormone and orphan receptors are likely to involve SP1 and related transcription factors.
Development of the endometrial luteolytic mechanism and uterine release of luteolytic pulses of PGF occur in the presence of high circulating levels of progesterone. However, the endometrial epithelia are not responsive to receptor-dependent actions of progesterone, because they are PGR-negative between d 11–15 of the estrous cycle. As progesterone levels decrease and estrogen levels increase during proestrus (d 15–17), PGR protein returns to the endometrial epithelia of the ovine uterus (10, 12). In the present study, PGR also increased activity of the ovine OXTR promoter through GC-rich SP1 sites rather than the AP-1 site or PRE half sites also present in the promoter of the OXTR gene. Progestin-dependent regulation of OXTR through GC-rich sites is not unprecedented becausePGR/SP1 activation of glycodelin A and CDKN1A (cyclin-dependent kinase inhibitor 1A or p21) in endometrial and breast cancer cells, respectively, is also due to specific GC-rich SP1 binding sites in their promoters (66, 67). Indeed, uterine OXTR mRNA and protein increases in the endometrial epithelia and stroma after d 15 and is maximal at estrus, which is coincident with PGR mRNA and protein as well as SP1 expression in the same cells (10, 12, 13). Therefore, the progesterone block to OXTR formation must be due to inhibition of ESR1 gene transcription via direct or indirect actions that does not involve SP1.
In the present study, IFNT did not affect OXTR promoter activity or E2 induction of OXTR promoter activity. These results were consistent with the lack of classical interferon regulatory elements (ISRE, IRFE, or GAS) in the ovine OXTR promoter and the lack of IRF1 and IRF2 effects on OXTR promoter activity in transient transfection assays. Collectively, results of the present and previous studies continue to support our theory that the antiluteolytic effects of IFNT from the conceptus are to inhibit or silence ESR1 gene transcription, which in turn precludes ESR1 stimulation of OXTR gene transcription and, therefore, oxytocin-induced release of luteolytic pulses of PGF by uterine epithelia. Future experiments are needed to address how progesterone down-regulates PGR gene expression, what factors critically regulate PGR inhibition of ESR1 gene transcription in the endometrial epithelia of the uterus, and if SP1 and related transcription factors regulate other endometrial genes in the ovine uterus.
Acknowledgments
We thank Haijun Gao for assistance with immunohistochemical analysis of SP1 protein and other members of our laboratories who contributed to the research presented in this paper.
Footnotes
Financial support for these studies was obtained from National Institutes of Health R01 Grant HD32534 and National Institute on Environmental Health Sciences 5 P30 ES09106-03.
First Published Online October 27, 2005
Abbreviations: AP-1, Activator protein 1; CL, corpus luteum; E3, 17-estradiol; ERE, ESR1 response element; ESR1, estrogen receptor ; FBS, fetal bovine serum; GAS, activation site; GE, glandular epithelium; IFNT, interferon ; IRF, interferon regulatory factor; IRFE, interferon regulatory factor element; ISRE, interferon-stimulated response element; LE, luminal epithelium; OXTR, oxytocin receptor; PGF, prostaglandin F2; PGR, progesterone receptor; PGR-B, progesterone receptor B form; PRE, progesterone response element; sGE, superficial ductal glandular epithelium.
Accepted for publication October 19, 2005.
References
Bazer FW, Thatcher WW, Hansen PJ, Mirando MA, Ott TL, Plante C 1991 Physiological mechanisms of pregnancy recognition in ruminants. J Reprod Fertil Suppl 43:39–47
Bazer FW 1992 Mediators of maternal recognition of pregnancy in mammals. Proc Soc Exp Biol Med 199:373–384
McCracken JA, Custer EE, Lamsa JC 1999 Luteolysis: a neuroendocrine-mediated event. Physiol Rev 79:263–323
Spencer TE, Bazer FW 2002 Biology of progesterone action during pregnancy recognition and maintenance of pregnancy. Front Biosci 7:d1879–d1898
Charpigny G, Reinaud P, Tamby JP, Creminon C, Martal J, Maclouf J, Guillomot M 1997 Expression of cyclooxygenase-1 and -2 in ovine endometrium during the estrous cycle and early pregnancy. Endocrinology 138:2163–2171
Gray C, Bartol FF, Taylor KM, Wiley AA, Ramsey WS, Ott TL, Bazer FW, Spencer TE 2000 Ovine uterine gland knock-out model: effects of gland ablation on the estrous cycle. Biol Reprod 62:448–456
Spencer TE, Ott TL, Bazer FW 1996 -Interferon: pregnancy recognition signal in ruminants. Proc Soc Exp Biol Med 213:215–229
Spencer TE, Bazer FW 2004 Conceptus signals for establishment and maintenance of pregnancy. Reprod Biol Endocrinol 2:49
Ayad VJ, Matthews EL, Wathes DC, Parkinson TJ, Wild ML 1991 Autoradiographic localization of oxytocin receptors in the endometrium during the oestrous cycle of the ewe. J Endocrinol 130:199–206
Wathes DC, Hamon M 1993 Localization of oestradiol, progesterone and oxytocin receptors in the uterus during the oestrous cycle and early pregnancy of the ewe. J Endocrinol 138:479–492
McCracken J, Schramm W, Okulicz WC 1984 Hormone receptor control of pulsatile secretion of PGF-2 from the ovine uterus during luteolysis and its abrogation in early pregnancy. Anim Reprod Sci 7:31–55
Spencer TE, Bazer FW 1995 Temporal and spatial alterations in uterine estrogen receptor and progesterone receptor gene expression during the estrous cycle and early pregnancy in the ewe. Biol Reprod 53:1527–1543
Stevenson KR, Riley PR, Stewart HJ, Flint AP, Wathes DC 1994 Localization of oxytocin receptor mRNA in the ovine uterus during the oestrous cycle and early pregnancy. J Mol Endocrinol 12:93–105
Wathes DC, Mann GE, Payne JH, Riley PR, Stevenson KR, Lamming GE 1996 Regulation of oxytocin, oestradiol and progesterone receptor concentrations in different uterine regions by oestradiol, progesterone and oxytocin in ovariectomized ewes. J Endocrinol 151:375–393
Spencer TE, Becker WC, George P, Mirando MA, Ogle TF, Bazer FW 1995 Ovine interferon- inhibits estrogen receptor up-regulation and estrogen-induced luteolysis in cyclic ewes. Endocrinology 136:4932–4944
Spencer TE, Becker WC, George P, Mirando MA, Ogle TF, Bazer FW 1995 Ovine interferon- regulates expression of endometrial receptors for estrogen and oxytocin but not progesterone. Biol Reprod 53:732–745
Hixon JE, Flint AP 1987 Effects of a luteolytic dose of oestradiol benzoate on uterine oxytocin receptor concentrations, phosphoinositide turnover and prostaglandin F-2 secretion in sheep. J Reprod Fertil 79:457–467
Vallet JL, Lamming GE, Batten M 1990 Control of endometrial oxytocin receptor and uterine response to oxytocin by progesterone and oestradiol in the ewe. J Reprod Fertil 90:625–634
Roberts RM, Ealy AD, Alexenko AP, Han CS, Ezashi T 1999 Trophoblast interferons. Placenta 20:259–264
Wathes DC, Lamming GE 1995 The oxytocin receptor, luteolysis and the maintenance of pregnancy. J Reprod Fertil Suppl 49:53–67
Kim S, Choi Y, Spencer TE, Bazer FW 2003 Effects of the estrous cycle, pregnancy and interferon on expression of cyclooxygenase two (COX-2) in ovine endometrium. Reprod Biol Endocrinol 1:58
Spencer TE, Bazer FW 1996 Ovine interferon suppresses transcription of the estrogen receptor and oxytocin receptor genes in the ovine endometrium. Endocrinology 137:1144–1147
Fleming JA, Choi Y, Johnson GA, Spencer TE, Bazer FW 2001 Cloning of the ovine estrogen receptor- promoter and functional regulation by ovine interferon-. Endocrinology 142:2879–2887
Choi Y, Johnson GA, Burghardt RC, Berghman LR, Joyce MM, Taylor KM, Stewart MD, Bazer FW, Spencer TE 2001 Interferon regulatory factor-two restricts expression of interferon-stimulated genes to the endometrial stroma and glandular epithelium of the ovine uterus. Biol Reprod 65:1038–1049
Vallet JL, Bazer FW 1989 Effect of ovine trophoblast protein-1, oestrogen and progesterone on oxytocin-induced phosphatidylinositol turnover in endometrium of sheep. J Reprod Fertil 87:755–761
Telgmann R, Bathgate RA, Jaeger S, Tillmann G, Ivell R 2003 Transcriptional regulation of the bovine oxytocin receptor gene. Biol Reprod 68:1015–1026
Kimura T, Saji F, Nishimori K, Ogita K, Nakamura H, Koyama M, Murata Y 2003 Molecular regulation of the oxytocin receptor in peripheral organs. J Mol Endocrinol 30:109–115
Spencer TE, Johnson GA, Burghardt RC, Bazer FW 2004 Progesterone and placental hormone actions on the uterus: insights from domestic animals. Biol Reprod 71:2–10
Pellegrini S, John J, Shearer M, Kerr IM, Stark GR 1989 Use of a selectable marker regulated by interferon to obtain mutations in the signaling pathway. Mol Cell Biol 9:4605–4612
Van Heeke G, Ott TL, Strauss A, Ammaturo D, Bazer FW 1996 High yield expression and secretion of the ovine pregnancy recognition hormone interferon- by Pichia pastoris. J Interferon Cytokine Res 16:119–126
Ausebel FM, Brent R, Kinston RE, Moore DD, Seidmen JG, Smith HA, Struhl KA 1998 Current protocols in molecular biology. New York: John Wiley and Sons
Stoner M, Wang F, Wormke M, Nguyen T, Samudio I, Vyhlidal C, Marme D, Finkenzeller G, Safe S 2000 Inhibition of vascular endothelial growth factor expression in HEC1A endometrial cancer cells through interactions of estrogen receptor and Sp3 proteins. J Biol Chem 275:22769–22779
Sun G, Porter W, Safe S 1998 Estrogen-induced retinoic acid receptor 1 gene expression: role of estrogen receptor-Sp1 complex. Mol Endocrinol 12:882–890
Spencer TE, Bartol FF, Bazer FW, Johnson GA, Joyce MM 1999 Identification and characterization of glycosylation-dependent cell adhesion molecule 1-like protein expression in the ovine uterus. Biol Reprod 60:241–250
Taylor KM, Gray CA, Joyce MM, Stewart MD, Bazer FW, Spencer TE 2000 Neonatal ovine uterine development involves alterations in expression of receptors for estrogen, progesterone, and prolactin. Biol Reprod 63:1192–1204
Gray CA, Taylor KM, Bazer FW, Spencer TE 2000 Mechanisms regulating norgestomet inhibition of endometrial gland morphogenesis in the neonatal ovine uterus. Mol Reprod Dev 57:67–78
Safe S, Kim K 2004 Nuclear receptor-mediated transactivation through interaction with Sp proteins. Prog Nucleic Acids Res Mol Biol 77:1–36
Kushner PJ, Agard DA, Greene GL, Scanlan TS, Shiau AK, Uht RM, Webb P 2000 Estrogen receptor pathways to AP-1. J Steroid Biochem Mol Biol 74:311–317
Sjottem E, Anderssen S, Johansen T 1996 The promoter activity of long terminal repeats of the HERV-H family of human retrovirus-like elements is critically dependent on Sp1 family proteins interacting with a GC/GT box located immediately 3' to the TATA box. J Virol 70:188–198
Black AR, Black JD, Azizkhan-Clifford J 2001 Sp1 and kruppel-like factor family of transcription factors in cell growth regulation and cancer. J Cell Physiol 188:143–160
Wang X, Talamantez JL, Adamo ML 1998 A CACCC box in the proximal exon 2 promoter of the rat insulin-like growth factor I gene is required for basal promoter activity. Endocrinology 139:1054–1066
Marini MG, Asunis I, Porcu L, Salgo MG, Loi MG, Brucchietti A, Cao A, Moi P 2004 The distal -globin CACCC box is required for maximal stimulation of the -globin gene by EKLF. Br J Haematol 127:114–117
Stark GR, Kerr IM, Williams BR, Silverman RH, Schreiber RD 1998 How cells respond to interferons. Annu Rev Biochem 67:227–264
Stewart MD, Johnson GA, Bazer FW, Spencer TE 2001 Interferon- (IFN) regulation of IFN-stimulated gene expression in cell lines lacking specific IFN-signaling components. Endocrinology 142:1786–1794
Stewart MD, Choi Y, Johnson GA, Yu-Lee LY, Bazer FW, Spencer TE 2002 Roles of Stat1, Stat2, and interferon regulatory factor-9 (IRF-9) in interferon regulation of IRF-1. Biol Reprod 66:393–400
Kim S, Choi Y, Bazer FW, Spencer TE 2003 Identification of genes in the ovine endometrium regulated by interferon independent of signal transducer and activator of transcription 1. Endocrinology 144:5203–5214
Safe S 2001 Transcriptional activation of genes by 17 -estradiol through estrogen receptor-Sp1 interactions. Vitam Horm 62:231–252
Frasor J, Stossi F, Danes JM, Komm B, Lyttle CR, Katzenellenbogen BS 2004 Selective estrogen receptor modulators: discrimination of agonistic versus antagonistic activities by gene expression profiling in breast cancer cells. Cancer Res 64:1522–1533
Kim K, Thu N, Saville B, Safe S 2003 Domains of estrogen receptor (ER) required for ER/Sp1-mediated activation of GC-rich promoters by estrogens and antiestrogens in breast cancer cells. Mol Endocrinol 17:804–817
Inoue T, Kimura T, Azuma C, Inazawa J, Takemura M, Kikuchi T, Kubota Y, Ogita K, Saji F 1994 Structural organization of the human oxytocin receptor gene. J Biol Chem 269:32451–32456
Rozen F, Russo C, Banville D, Zingg HH 1995 Structure, characterization, and expression of the rat oxytocin receptor gene. Proc Natl Acad Sci USA 92:200–204
Bale TL, Dorsa DM 1997 Cloning, novel promoter sequence, and estrogen regulation of a rat oxytocin receptor gene. Endocrinology 138:1151–1158
Ivell R, Bathgate R, Rust W, Balvers M, Morley S 1995 Structure and organization of the bovine oxytocin receptor gene. Adv Exp Med Biol 395:295–300
Bathgate R, Rust W, Balvers M, Hartung S, Morley S, Ivell R 1995 Structure and expression of the bovine oxytocin receptor gene. DNA Cell Biol 14:1037–1048
Kubota Y, Kimura T, Hashimoto K, Tokugawa Y, Nobunaga K, Azuma C, Saji F, Murata Y 1996 Structure and expression of the mouse oxytocin receptor gene. Mol Cell Endocrinol 124:25–32
Young LJ, Waymire KG, Nilsen R, Macgregor GR, Wang Z, Insel TR 1997 The 5' flanking region of the monogamous prairie vole oxytocin receptor gene directs tissue-specific expression in transgenic mice. Ann NY Acad Sci 807:514–517
Schmid B, Wong S, Mitchell BF 2001 Transcriptional regulation of oxytocin receptor by interleukin-1 and interleukin-6. Endocrinology 142:1380–1385
Copland JA, Jeng YJ, Strakova Z, Ives KL, Hellmich MR, Soloff MS 1999 Demonstration of functional oxytocin receptors in human breast Hs578T cells and their up-regulation through a protein kinase C-dependent pathway. Endocrinology 140:2258–2267
Jeng YJ, Lolait SJ, Soloff MS 1998 Induction of oxytocin receptor gene expression in rabbit amnion cells. Endocrinology 139:3449–3455
Bale TL, Dorsa DM 1998 NGF, cyclic AMP, and phorbol esters regulate oxytocin receptor gene transcription in SK-N-SH and MCF7 cells. Brain Res Mol Brain Res 53:130–137
Hoare S, Copland JA, Wood TG, Jeng YJ, Izban MG, Soloff MS 1999 Identification of a GABP / binding site involved in the induction of oxytocin receptor gene expression in human breast cells, potentiation by c-Fos/c-Jun. Endocrinology 140:2268–2279
Spencer TE, Mirando MA, Mayes JS, Watson GH, Ott TL, Bazer FW 1996 Effects of interferon- and progesterone on oestrogen-stimulated expression of receptors for oestrogen, progesterone and oxytocin in the endometrium of ovariectomized ewes. Reprod Fertil Dev 8:843–853
Smith CL 1998 Cross-talk between peptide growth factor and estrogen receptor signaling pathways. Biol Reprod 58:627–632
Stevenson KR, Gilmour RS, Wathes DC 1994 Localization of insulin-like growth factor-I (IGF-I) and -II messenger ribonucleic acid and type 1 IGF receptors in the ovine uterus during the estrous cycle and early pregnancy. Endocrinology 134:1655–1664
Saville B, Wormke M, Wang F, Nguyen T, Enmark E, Kuiper G, Gustafsson JA, Safe S 2000 Ligand-, cell-, and estrogen receptor subtype (/)-dependent activation at GC-rich (Sp1) promoter elements. J Biol Chem 275:5379–5387
Gao J, Mazella J, Seppala M, Tseng L 2001 Ligand activated hPR modulates the glycodelin promoter activity through the Sp1 sites in human endometrial adenocarcinoma cells. Mol Cell Endocrinol 176:97–102
Owen GI, Richer JK, Tung L, Takimoto G, Horwitz KB 1998 Progesterone regulates transcription of the p21(WAF1) cyclin-dependent kinase inhibitor gene through Sp1 and CBP/p300. J Biol Chem 273:10696–10701(JoAnn G. W. Fleming, Thomas E. Spencer, )