| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Laboratoire Estrogènes et Reproduction (G.G., S.C., M.-L.K.), Equipe d’Accueil 2608 Unité sous contrat 2006 Institut National de la Recherche Agronomique, Université de Caen Basse-Normandie; and Département Génétique et Reproduction (N.R., S.C., M.-L.K.), Unité de Formation et de Recherche de médecine Centre Hospitalier Universitaire, F-14033 Caen, France
Address all correspondence and requests for reprints to: Marie-Laure Kottler, Département Génétique et Reproduction, UFR de médecine, F-14033 Caen, France. E-mail: kottler-ml{at}chu-caen.fr.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
ligand propyl pyrazole triol and the selective estrogen receptor-ß ligand diarylpropionitrile both mimic the E2 effects. By contrast, propyl pyrazole triol reduces LHß expression whereas diarylpropionitrile does not. In addition, using transient transfection assays in an LßT2 gonadotrope cell line, we provided evidence that GnRH agonist stimulated, in a dose-dependant manner, cyp19 promoters PII and PI.f and that E2 decreased the GnRH stimulation. In conclusion, our data demonstrate that GnRH is an important signal in the regulation of cyp19 in gonadotrope cells. Both common and specific intracellular factors were responsible for dissociated variations of LHß and cyp19 expression. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Estradiol (E2) is considered as the critical determinant of plasma gonadotropin levels in females by completing an endocrine feedback loop on the hypothalamus, affecting the pattern of GnRH release, and on the pituitary gland, suppressing the GnRH-stimulated LH secretion (3, 4, 5). Besides endocrine action, E2 is also reported to act in an auto- or paracrine manner to influence certain functions. In rodent brains, previous studies provide evidence of local estrogen production in the thalamus and hypothalamus of mouse (6) and in the hypothalamus and amygdala of adult rats (7). In the ovary, estrogens enhance FSH action on granulosa cells (8), inhibit androgen synthesis in thecal cells (9), and attenuate apoptosis (10). Finally, previous studies using in vitro and in vivo models have demonstrated that in situ aromatase is involved in local estrogen production in both bone (11) and breast cancer (12, 13, 14).
Several studies suggest a role of aromatase in the pituitary. Both E2 and testosterone significantly suppressed LHß mRNA and serum LH levels in wild-type castrates, whereas they had no significant effect on castrated estrogen receptor- knockout (ERKO) mice, suggesting that androgens must be metabolized to be effective (3). When the action of endogenous E2 was suppressed by an aromatase inhibitor in early and midpubertal boys, the LH amplitude and the GnRH-induced LH response increased, whereas LH pulse frequency did not change (15). Taken together, these data suggest that the normal regulation of the gonadotrope function requires the local aromatization of testosterone.
The intriguing presence of aromatase immunostaining in the pituitary (16) led us to question the exact location of its secretion and the mechanism of action of gonadal steroids. We report the results of in vivo and in vitro experiments designed to localize and study the expression of aromatase in the female rat pituitary. First, we characterized the aromatase transcripts expressed in the pituitary and carried out an immunolocalization of the P450arom protein. Second, we examined the total aromatase mRNA expression during the estrous cycle and assessed the effects of ovariectomy and steroid supplementation. We were also interested in establishing whether GnRH and E2 may alter cyp19 gene expression in vivo in female rats and in vitro using reporter gene assays.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Ovariectomy and vaginal smears
Ovaries were removed from female rats under anesthesia by injection of a ketamine-midazolam cocktail (ip, 100 mg/kg and 4 mg/kg, respectively) at random stages of the estrous cycle. All animals were housed individually after surgery. The estrous cycle and the estrogenic treatment were checked daily by examining vaginal smears between 0900 and 1000 h and measuring serum E2 (Table 1
).
|
Experiment 2.
The purpose of this experiment was to examine the effects of castration and E2 supplementation on pituitary aromatase mRNA. The treatment paradigm we used enabled measures to be taken at a time when the negative feedback of estrogens was operational. Ovariectomized (OVX) rats were injected (sc) every 2 d for 3 wk with 10 µg 17ß-estradiol (Sigma, L’Isle d’Abeau, France). Rats (n = 7 per group) were divided into two groups, OVX and OVX plus E2 supplementation, and were killed 3 wk later.
Experiment 3.
The purpose of this experiment was to determine the isoform of estrogen receptor (ER) used by estrogens to regulate the expression of the cyp19 gene. A selective ER agonist, propyl pyrazole triol (PPT), and an ERß agonist, diarylpropionitrile (DPN) were obtained from Tocris Cookson Ltd. (Avonmouth, UK). OVX rats (n = 5 per group) were randomly assigned to three groups: PPT, DPN, or PPT plus DPN. Three weeks after ovariectomy, rats were injected daily (sc) for 3 d with 1 mg PPT, 1 mg DPN, or 1 mg PPT plus DPN, as adapted from previous studies (17, 18).
Experiment 4.
The purpose of this experiment was to determine the effect of E2 at the pituitary or hypothalamic level using either a GnRH antagonist (Cetrorelix acetate, kindly provided by Serono, Boulogne, France) or a GnRH agonist (D-Trp6-GnRH, Triptorelin 3 mg, kindly provided by IPSEN Pharma Biotech, Signes, France). Animals (n = 7 per group) were randomly assigned to three groups: intact, OVX, and OVX plus E2 supplementation. Three weeks after ovariectomy, OVX and intact rats were injected daily (sc) for 5 d with 100 µg antagonist before being killed, as adapted from previous studies (19, 20). The long-acting GnRH agonist (75 µg) was administered once (im) on OVX and intact rats, as previously described (21), 3 wk before being killed.
Tissue collection
After each experiment, rats were weighed, anesthetized with ether, and killed by decapitation. Blood samples (3–5 ml) were obtained from each animal through a cardiac catheter. The blood was immediately centrifuged in dry tubes, and the serum was stored at –20 C until E2 evaluation. Pituitaries were dissected immediately after death, flash frozen in liquid nitrogen, and stored at –80 C for subsequent RNA extraction. For immunohistochemistry, pituitaries from male and female rats were carefully dissected and immediately fixed in 4% paraformaldehyde for 24 h and then rinsed with 1x PBS, dehydrated in ethanol, and embedded in paraffin. Tissue blocks were serially sectioned at 3 µm, and sections were mounted onto Super Frost Plus slides (Menzel-Glaser/CML, Nemours, France) before being air dried at 37 C. Sections were made in the rostrocaudal direction from all regions of the gland.
RT-PCR analysis
Total RNA was extracted using the Trizol reagent kit (Invitrogen, Life Technologies, Cergy Pontoise, France) and then retrotranscribed into a reaction mixture (40 µl) containing 100 U Moloney murine leukemia virus reverse transcriptase (Amersham Pharmacia Biotech, Orsay, France), 1x reaction buffer (Amersham), 0.2 mM deoxynucleotide triphosphates (Promega, Charbonnières, France), 50 pmol oligo(dT) primers (Invitrogen, Life technologies), 20 U ribonuclease inhibitor (Promega), and 2 µg total RNA. The reaction was carried out at 37 C for 60 min then at 95 C for 5 min. Primers for the gonadal-specific first exon II and for the brain-specific exon I.f were designed using Primer Express Software (version 2.0, Applied Biosystems) and synthesized by Life Technologies (Table 2). After PCR, amplified fragments were separated by 1.5% agarose gel electrophoresis and visualized by ethidium bromide staining.
|
Real-time PCR analysis.
Relative levels of total aromatase mRNA were examined by real-time PCR, using an ABI Prism 7000 Sequence Detector System (Applied Biosystems, Courtaboeuf, France) according to the manufacturer’s directions. The sequences of the primers selected for total aromatase analyses were located on exons IX-X. This allowed us to assess all aromatase variants, independently from the tissue-specific exon I inserted in the mRNA. Because our experimental manipulations are known to induce dramatic changes in the functioning of the gonadotropic axis, we also monitored LHß mRNA expression as a control. Total aromatase mRNA was quantified using the TaqMan Universal PCR Master Mix (Applied Biosystems) and a TaqMan specific probe (5'-CCATTTGGCTTTGGGCCC-3'), in a total volume of 30 µl reaction mixture whereas LHß mRNA was quantified using the SYBR Green Universal PCR Master Mix 2 x (Applied Biosystems), in a total volume of 20 µl. Samples were heated 10 min at 95 C, followed by 40 cycles of 15 sec at 95 C then 1min at 60 C or 64 C. For all samples, the quantification of the ß-actin gene was used as the endogenous control for the normalization of initial RNA levels. PCR were performed in duplicate and a reagent blank prepared using the RT blank was included with each plate to detect contamination by genomic DNA.
E2 production
Serum E2 levels were measured by using a RIA kit (ESTRADIOL, DiaSorin, Antony, France) validated for use in rats (Table
. 1). The limit of detection was 5 pg/ml of serum. All samples were run in duplicate in the same assay. The intra- and interassay coefficients of variation were 4 and 11%, respectively.
Western blot analysis
For microsomal protein extraction, pituitaries and testes were crushed and homogenized in phosphate buffer (0.1 M, pH 7.4; Sigma) containing a protease inhibitor cocktail tablet (Roche Applied Science, Meylan, France). Samples were first centrifuged at 4 C for 5 min at 8000 x g, and then the supernatants were ultracentrifuged at 4 C for 60 min at 150,000 x g. Protein concentration in the supernatant was measured with the Bradford micromethod. Finally, samples were frozen at –80 C until Western blotting.
The microsomal proteins (100 µg) were loaded into each lane and then separated on a 12% polyacrylamide denaturing gel, and proteins were blotted onto nitrocellulose membranes (Amersham Pharmacia Biotech) for 60 min at 15 V using a semidry transfer system (OWL, Illkirch, France). The membrane was blocked for 90 min at room temperature in PBS supplemented with 3% nonfat dried milk and then incubated for 2 h at room temperature with a rabbit polyclonal antibody directed against rat P450arom (kindly provided by Dr. Carretero, University of Salamanque, Spain) at 1/1000 dilution in PBS supplemented with 3% nonfat dried milk. After being washed in TBS (10 mM Tris, 15 mM NaCl, pH 7.4) for 10 min, the membrane was incubated for 90 min at room temperature with a rabbit Ig antibody at 1/4000 dilution in PBS supplemented with 3% nonfat dried milk. The antibody-protein complexes were revealed using the enhanced chemiluminescence visualization system (ECL; Amersham Pharmacia Biotech).
Fluorescence immunohistochemistry
Male and female pituitary sections were deparaffinized in toluene and rehydrated through a graded ethanol series, washed in H2O, then immersed in 100 mM citrate buffer (pH 6.0), and finally heated in a standard microwave twice for 5 min at 750 W and 5 min at 550 W. The container holding the sections was removed and allowed to cool for 20 min at room temperature. After a brief wash in PBS, nonspecific binding sites were blocked with 3% BSA in PBS for 60 min. We first tested immunostaining in adjacent sections with rabbit polyclonal antibody against rat P450arom at a 1/500 dilution, rabbit polyclonal antirat FSHß (NIDDK) at a 1/1000 dilution, rabbit polyclonal antirat prolactin (PRL) (22) at a 1/1000 dilution, or rabbit polyclonal antisynthetic human GH (NIDDK, no. IC-4, AFP-1613102481) at a 1/500 dilution. After overnight incubation at 4 C, polyclonal antibodies were revealed with a biotinylated secondary antibody directed against rabbit Ig associated with a streptavidin-fluorescein complex (RPN 1232; Amersham Pharmacia Biotech). Dual fluorescence labeling was then tested on the same section with the rabbit polyclonal antibody against rat P450arom plus mouse monoclonal antibody against rat LHß (kindly provided by Dr. Counis, University of Paris VI, France) at a 1/300 dilution. Texas Red-conjugated secondary antibodies directed against mouse Ig were subsequently used to reveal anti-LHß monoclonal antibody. After rinsing with 1x PBS, sections were coverslipped with Vectashield mounting medium (Vector Laboratories, Inc., Burlingame, CA). Exclusion of the primary antibody was run as negative control to confirm the specificity of the immunostaining, and rat testis sections were used as positive control for aromatase staining. Sections were viewed using fluorescence microscopy (Axioskop 2 plus; Carl Zeiss, Le Pecq, France), and digital images were collected using a Zeiss AxioCam digital camera and the AxioVision image analysis system (version 4.1; Carl Zeiss).
Reporter plasmids used in transfection studies
The construct containing region –1037 to + 94 of the rat aromatase gene promoter PII, inserted upstream of the Firefly luciferase gene into the pGL2-Basic vector (Promega), was generously provided by Dr. M. McPhaul and has been described previously (23). The 1029-bp fragment of the rat aromatase gene promoter PI.f inserted upstream of the Firefly luciferase gene into the pGL3-Basic vector (Promega) was isolated in our laboratory (unpublished data) and displays a nucleotide sequence homologous to that registered in the rat genome resource data bank of the National Center for Biotechnology Information.
Transfection experiments
LßT2 cells, generously provided by Dr. P. Mellon (La Jolla, CA), were maintained in monolayer cultures in high-glucose DMEM (Sigma) supplemented with 10% fetal bovine serum (Invitrogen, Life Technologies) and penicillin/streptomycin (Sigma) at 37 C in humidified 5% CO2. For transient transfection, cells were plated in 24-well plates at a density of 3 x 105 cells per well the day before transfection. On the day of transfection, Lipofectamine 2000 reagents (Invitrogen, Life Technologies; 2 µl/well) was used according to the manufacturer’s instructions to cotransfect cells with the pGL2-promoter construct (823 ng/well) or the pGL3-promoter (720 ng/well) and the Renilla luciferase pRL-TK expression vector (Promega; 15 ng/well) for 24 h. Transfected cells were washed once in DMEM with no phenol red (Sigma) and serum and then treated with either a GnRH agonist (Triptorelin acetate; IPSEN Pharma Biotech), E2 (Sigma), or a combination of Triptorelin plus E2. After 6 or 24 h, Firefly and Renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega). Cells were washed once with PBS (pH 7.4) and harvested by adding 100 µl/well lysis buffer. Firefly luciferase assays were performed by adding 100 µl luciferase assay reagent to 20 µl cell extract, and luminescence was measured with a 2-sec delay and a 10-sec measurement in a Mithras LB 940 microplate reader (Berthold Technologies, Bad Wildbad, Germany). One hundred microliters of Stop & Glo reagent were then added, and Renilla luciferase activity was determined under the same conditions. The Firefly luciferase data were corrected for transfection efficiency with Renilla luciferase activity.
Statistical analysis
All results are expressed as a ratio to a calibrator that was chosen to be the metestrous groups for in vivo studies or untreated cells for in vitro studies and then expressed as the mean values ± SEM. Statistical analysis was performed using one-way ANOVA, and differences between groups were determined by the Fisher protected least significant difference test (SigmaStat for Windows, version 3.1; SPSS Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
). After electrophoresis of PCR products, we observed two specific bands of 147 and 140 bp that fit the predicted sizes of the amplified sequences for brain and gonadal subtypes, respectively (Fig. 1A). No PCR product was obtained in control reactions in the absence of reverse transcriptase.
|
). The size of this protein is consistent with the predicted size.
Finally, immunohistochemistry was used to identify the cell type that synthesized aromatase (Fig. 2). Male pituitary sections were immunostained to adequately visualize the aromatase protein (16). Findings were then confirmed on female pituitary sections. Whatever the antibody used, the immunostaining was restricted to the anterior lobe, and no immunostaining was found in the pars intermedia or in the pars nervosa (only shown for P450arom and LH; Fig. 2, D1–D3). A positive labeling was detected in the cytoplasmic compartment but not in nuclei. Immunolabeling was undetectable when primary antibodies were omitted (Fig. 2, control –). A positive immunoreaction to P450arom was detected in Leydig cells within the interstitial tissue (Fig. 2, control +), thus confirming the specificity of the aromatase immunostaining. Examination of two consecutive sections indicated that P450arom and FSH were localized in the same cells (Fig. 2, A1 and A2). However, only certain cells were positive for the two proteins. Thanks to a specific mouse monoclonal antibody against LHß, we tested colocalization using two different immunolabeling methods. Results clearly showed that P450arom and LHß immunostaining colocalized within the same cells both in males (Fig. 2, D3 and E3) and females (Fig. 2, F3). P450arom was not found in the cells showing positive immunostaining for PRL and GH (Fig. 2, B1, B2, C1, and C2).
|
). Quantification was performed by amplifying the total aromatase mRNA with exons IX and X. The regularity of the estrous cycle was checked by measuring E2 levels and examining vaginal smears (Table 1). Figure 3 demonstrates that aromatase mRNA expression changed according to the estrous cycle stage. Aromatase mRNA levels were significantly higher during metestrus than during estrus and proestrus. We also observed that aromatase mRNA levels were lower in proestrus than in estrus, although not significantly.
|
). Figure 4A shows that ovariectomy resulted in a significant increase in the pituitary expression of total aromatase mRNA compared with metestrus, whereas E2 supplementation (Table 1) completely reversed the effects of ovariectomy. These variations paralleled those observed for LHß expression.
|
and ERß and ß isoforms) in E2 regulatory action on cyp19 gene expression, 3-wk OVX rats were injected over 3 d with the selective ER ligand, PPT, the potency-selective agonist of ERß, DPN, or the combined administration of PPT plus DPN. The potency of the treatment was checked by vaginal smears (Table 1). PPT administration and to a lesser extent DPN administration resulted in a significant decrease of aromatase mRNA compared with OVX values, at a level close to that observed in metestrus (Fig. 4B). The responses to the combined administration of PPT and DPN differed significantly from those caused by the administration of PPT alone (Fig. 4B). By contrast, PPT alone or in combination with DPN, but not DPN alone, reversed the effects of ovariectomy on LHß mRNA expression (Fig. 4B).
Dual regulation of the aromatase gene expression by estrogens at pituitary and hypothalamic levels
To highlight the E2 effect at the pituitary or hypothalamic level on aromatase mRNA expression, OVX and intact rats were treated with a GnRH antagonist, Cetrorelix (Fig. 4C) or a long-acting GnRH agonist, Triptorelin (Fig. 4D). The potency of different treatments was checked by measuring E2 levels and by examining vaginal smears (Table 1). In intact female rats (Table 1), daily administration of Cetrorelix for 5 d did not significantly suppress total aromatase mRNA levels compared with metestrus values (Fig. 4C). By contrast, Cetrorelix prevented the increase of total aromatase mRNA levels observed after ovariectomy (Fig. 4C). E2 administration (Table 1) amplified the decrease of cyp19 expression in a manner similar to that observed for LHß in OVX rats (Fig. 4C).
In intact female rats (Table 1), the single administration of Triptorelin, a long-acting GnRH agonist, did not significantly increase the total aromatase mRNA expression compared with metestrus values (Fig. 4D). Similarly, in OVX rats, this treatment did not modify total aromatase mRNA levels, although we observed the well-known desensitization of LHß, compared with OVX rats alone (Fig. 4D). Again, E2 (Table 1) reduced aromatase and LHß mRNA expression (Fig. 4D).
GnRH stimulates aromatase promoters PII and PI.f activity, and E2 reduces GnRH-dependent activities in LßT2 cells
To examine the direct effect of GnRH and E2 on aromatase expression, LßT2 cells were transfected with reporter genes containing aromatase promoters PII or PI.f. We found that GnRH agonist treatment at 10–9 or 10–7 M for 6 h stimulated promoter PII activity in a dose-dependent manner compared with the basal activity (7- and 18-fold, respectively; Fig. 5A). By contrast, the stimulation of promoter PI.f by GnRH is less important, with no significant difference between 10–9 and 10–7 M (1.4- and 1.5-fold, respectively; Fig. 5B). Interestingly, when cells were treated for 24 h with GnRH (10–7 M), the stimulation amplitude was lower for promoter PII (18- to 10-fold; Fig. 5, A and C) but was higher for promoter PI.f (1.5- to 2.4-fold; Fig. 5, B and D) compared with a 6-h treatment. Although treatment with E2 alone did not cause any significant response on promoter PII and PI.f activities, whatever the doses used (10–5 to 10–9 M, data not shown), the combined incubation of E2 (10–5 M) with Triptorelin (10–7 M) resulted in a negative effect on GnRH-stimulated promoter PII and PI.f activities at 6 h (data not shown) as well as at 24 h (Fig. 5, C and D).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
RT-PCR analysis indicates that aromatase is synthesized from two previously described transcripts in the rat pituitary, the gonadal-specific first exon II under the control of promoter PII and the brain-specific exon I.f under the control of promoter P1.f. In rodent brains, the brain subtype is the major transcript in the thalamic-hypothalamic areas of the mouse (6) and in the hypothalamus and amygdala of adult rats (7), although in both cases, low amounts of transcripts containing the gonadal subtype may also be present in these regions (6, 24). However, because brain and gonadal subtypes are found in low concentrations in the pituitary gland, we were unable to determine which was mainly expressed. Using immunolabeling with anti-LHß, -FSHß, and -P450arom antibodies, we observed that aromatase was expressed in cells that expressed LH and, to a lesser extent, FSH proteins, but not PRL or GH. We cannot, however, exclude the possibility that aromatase is present in other cell types such as ACTH and TSH cells.
Our analyses using real-time PCR showed for the first time that the level of P450arom mRNA significantly varied during the estrous cycle, its lowest level occurring in the morning of proestrus when maximal levels of estrogens were observed, just before the LH surge occurring in the afternoon. The effects of castration and steroid replacement recorded on aromatase mRNA levels concord with the histochemical data of Carretero et al. (16), who showed that aromatase labeling varied in intensity across the cycle. These results support a direct action of gonadal steroids on the pituitary to negatively control the expression of aromatase.
Numerous studies have demonstrated that E2 acts on the hypothalamic level to modify both GnRH pulse frequency and amplitude (3) and directly on gonadotropes to modify either the number of GnRH receptors (GnRH-R) (25, 26), intracellular responses to GnRH (26), or transcriptional activity of gonadotropin subunit genes (27). Therefore, the exact mechanism by which the steroid exerts its effect on cyp19 gene expression is difficult to ascertain. To separate the pituitary gland from endogenous GnRH secretion, we designed a protocol blocking GnRH-induced gonadotropin secretion. We found that administering a GnRH antagonist to OVX rats prevents the increase of cyp19 gene transcription. The main mechanism of action of GnRH antagonists was thought to be based on a competitive occupancy of GnRH-R and the counteraction of the stimulatory effect of endogenous GnRH (28). Thus, GnRH antagonists do not directly influence gene expression at the pituitary but do exert their suppressive effects by counteracting the up-regulation caused by GnRH. Accordingly, a significantly greater reduction in cyp19 as well as in LHß gene expressions by Cetrorelix was observed in OVX rats, which have higher GnRH concentration in the pituitary portal vessels than in controls. Thus, our experimental procedure clearly establishes that GnRH exerts a positive regulation of cyp19 gene expression in OVX rats. As an additional strategy to detect the direct effect of GnRH on cyp19 expression, specific aromatase promoters PII or PI.f driving reporter gene were introduced into LßT2 cells, a cell line that closely resembles a differentiated gonadotrope. As expected, GnRH markedly increased PII in a dose-dependent manner and, to a lesser extent, PI.f aromatase promoters.
Studies exploring cellular mechanisms involved in the LHß and cyp19 gene regulation have identified several common key factors such as the steroidogenic factor-1 (SF-1) and the cAMP response element-binding protein (CREB). In the gonads, promoter II activity of cyp19 is regulated by FSH and a cAMP-dependent signaling mechanism giving rise to an interaction between the gonadal promoter II with the transcription factors SF-1 and CREB, both in humans and in rats (29, 30, 31). SF-1, which is selectively expressed in the gonadotrope population (32), was also shown to be crucial for LHß, GnRH-R, and free -subunit (GSU) transcription activation (33, 34, 35, 36, 37). Taking these data into consideration, the presence of SF-1 in gonadotropes that is consequently linked to GnRH stimulation implies that GnRH may affect aromatase expression.
The variations of mRNA levels under administration of a long-acting GnRH agonist causing homologous desensitization provides evidence that the intracellular mechanisms responsible for GnRH activation of aromatase expression differ from those governing regulation and desensitization of LHß expression. Indeed, long-acting GnRH agonist depresses LHß in OVX rats using a postreceptor phenomenon (21), whereas it does not modify aromatase expression. Similarly, a dissociated regulation of transcriptional stimulation and mRNA stability was shown for the -subunit (38) and for nitric oxide synthase (39), illustrating the fact that elements under GnRH regulation in gonadotrope cells each respond to desensitization with distinct characteristics using specific mechanisms that remain to be characterized.
Estradiol levels in OVX rats receiving GnRH antagonist plus E2 were much higher than with any other treatment. Thus, we cannot conclude whether the decrease in aromatase mRNA levels is a direct pituitary effect or a result of the total inhibition of GnRH input. However, administration of E2 to long-acting GnRH-agonist-treated animals clearly demonstrated that E2 also acts directly on the pituitary by negatively controlling aromatase expression. When LßT2 cells were treated with E2, the basal activities of aromatase promoters PII and PI.f remained unchanged. However, E2 reduced the GnRH induction of promoters PII and PI.f activities according to our in vivo observations.
It is well established that estrogen-induced changes are mainly mediated via ER. Previous studies have demonstrated that the pituitary expresses both ER and ERß isoforms, with higher levels for ER than ERß (40, 41, 42) and a coexpression of both isoforms in the rat gonadotropes (41). In our study, activation of ER by the selective ligand PPT and, to a lesser extent, activation of ERß by the selective ligand DPN were able to mimic the effect of estrogen supplementation in OVX rats, thus suggesting the involvement of ER pathways in the regulation of the cyp19 gene. The responses of pituitary aromatase expression in the different experimental groups were closely paralleled by changes in pituitary LH mRNA levels after ovariectomy and E2 supplementation, except that ERß activation by the selective ligand DPN was unable to restore LHß mRNA levels. Concerning LH, this result is consistent with previous reports showing that estrogen-induced LHß regulation is heavily dependent upon the actions of ER, because only ERKO or ßERKO female mice exhibited elevated LHß gene expression, but not ßERKO female mice (43).
In our study, we found that aromatase and LH colocalized within the same gonadotrope cells. Therefore, the differences observed between the regulation of cyp19 and LHß genes are not related to a cell-specific effect but suggest the role of specific transcription factors. Our PCR analyses showed that the cyp19 gene is under the control of promoter II and promoter I.f. However, to date, no high-affinity ER binding sites have been identified in these promoters, despite the fact that cyp19 expression is negatively regulated by both potent ER and the ERß agonists. It is well known that ER can modulate the transcription from promoters that lack typical estrogen response elements, using alternative response elements to which ER are not bound or specific intracellular factors recruited by ER and/or ERß (44). For example, estrogens have been found to stimulate either the neurotensine or interleukin gene expression despite the lack of estrogen response element motifs in these promoters (45, 46). The regulation of the cyp19 gene by E2 also appeared to be cell type dependent, reinforcing the hypothesis that specific intracellular factors are implied. Indeed, E2 inhibited cyp19 gene expression in germ cells (47), whereas in Leydig cells, E2 enhanced it in a dose- and time-related manner (48). Thus, alternate mechanisms such as transcriptional interference via protein-protein interactions may be the molecular basis for the inhibitory functions of estrogens and could explain differences between the regulation of cyp19 and LHß gene expression.
It has been reported that the regulation of cyp19 gene expression in rat gonads mainly depends on SF-1 and CREB content. CREB contains several consensus phosphorylation sites for various kinases, in particular protein kinase A (49) and protein kinase C (50). In this model, the highest levels of phospho-CREB (pCREB) coincided with the maximal induction of endogenous cyp19 gene (30). It is known that ovariectomy increases pCREB in the pituitary, whereas E2 treatment dramatically decreases pCREB content via a mechanism linked to the GnRH signaling pathway (51). Thus, the increase of CREB phosphorylation in the gonadotrope cells could be responsible for GnRH positive regulation of cyp19 expression, whereas the decrease in CREB phosphorylation could be responsible for E2 negative regulation. However, the relevance of E2-induced change in pCREB has not yet been analyzed in the context of the initiation of the LH surge. Indeed, in the ewe, a combination of increased GnRH pulse frequency and estrogen leads to a pCREB response in gonadotrope cells (52).
The converging signaling of both pathways and concerted action of GnRH and E2 at the pituitary level are involved in the timing and initiation of the LH surge. At proestrus, the E2 circulating level is high, the number of gonadotrope cells that stained for ER/ERß increases (53), GnRH-R are up-regulated (54), and GnRH stimulates ER transactivation (55). The LH surge that is mainly dependent on the increase of GnRH input occurs in the afternoon of the proestrus, and we clearly demonstrated that GnRH enhanced aromatase expression. Thus, our results lead us to hypothesize that aromatase expression could be enhanced during the LH surge to amplify E2 signaling. Accordingly, Kazeto and Trant (56) have recently shown in catfish that the preovulatory induction of the CYP19A2 gene by E2 is similar to the pattern of gene expression for LHß in the pituitary. This may underlie some degree of redundancy within the control of the LH surge, a key component of reproductive hormone synthesis.
In conclusion, we have shown that P450arom is synthesized by gonadotrope cells from two different transcripts carrying the gonadal-specific first exon II and the brain-specific exon I.f. We report for the first time that cyp19 gene expression is positively regulated by GnRH in vivo in the rat pituitary gland and in vitro in LßT2 cells and negatively controlled by chronic exposure to E2 via ER. We also provide evidence for the involvement of both common and specific intracellular factors that could account for dissociated variations of LHß and cyp19 expression.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure statement: G.G., N.R., S.C., and M.-L.K. have nothing to declare.
First Published Online June 8, 2006
Abbreviations: CREB, cAMP response element-binding protein; DPN, diarylpropionitrile; E2, estradiol; ER, estrogen receptor; ERKO, ER knockout; GnRH-R, GnRH receptor; OVX, ovariectomized; P450arom, cytochrome P450 aromatase; pCREB, phospho-CREB; PPT, propyl pyrazole triol; SF-1, steroidogenic factor-1.
Received December 23, 2005.
Accepted for publication May 31, 2006.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
knockout mice. Endocrinology 139:4092–4101 and ERß, in estrogen target tissues in vivo through the use of an ER-selective ligand. Endocrinology 143:4172–4177 and ERß on progesterone receptor expression and action and on gonadotropin and prolactin secretion in the rat. Neuroendocrinology 79:247–258[CrossRef][Medline]T3–1 cells. Mol Cell Endocrinol 87:95–103[CrossRef][Medline]-subunit gene despite the absence of a high affinity binding site for estrogen receptor. Mol Endocrinol 5:725–733-subunit gene in pituitary gonadotropes. Mol Endocrinol 8:878–885-subunit messenger ribonucleic acid (mRNA) levels in T3 cells by increasing transcription and mRNA stability. Endocrinology 134:2475–2481 and ß in the rat pituitary gland. Endocrinology 139:3976–3983 but not ERß. Mol Endocrinol 17:1039–1053 and induction of Fos and phosphorylated cyclic adenosine monophosphate response element binding protein in pituitary gonadotropes: studies in hypothalamo-pituitary disconnected ewes. Endocrinology 146:1128–1137 and ß by gonadotropes during the estrous cycle. J Histochem Cytochem 49:665–666This article has been cited by other articles:
 |
|
 |
|
  D. Silandre, C. Delalande, P. Durand, and S. Carreau Three promoters PII, PI.f, and PI.tr direct the expression of aromatase (cyp19) gene in male rat germ cells J. Mol. Endocrinol., August 1, 2007; 39(2): 169 - 181. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Arreguin-Arevalo, T. L. Davis, and T. M. Nett Differential Modulation of Gonadotropin Secretion by Selective Estrogen Receptor 1 and Estrogen Receptor 2 Agonists in Ovariectomized Ewes Biol Reprod, August 1, 2007; 77(2): 320 - 328. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
