| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ARTICLES |
a Institut National de la Recherche Agronomique/Unité de Recherche Associée CNRS 1291, Station de Physiologie de la Reproduction des Mammifères Domestiques, 37380 Nouzilly, France
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
anterior pituitary, estradiol, FSH, LH
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Estradiol plays a key role in the control of the estrous cycle by exerting negative or positive feedback on both the hypothalamus and the pituitary gland. At the hypothalamic level, E2 reduces both the amplitude and frequency of the GnRH pulses in the portal blood during the negative feedback [7]. In contrast, during the positive feedback, the high preovulatory levels of circulating E2 induce a GnRH surge [7–9]. At the pituitary level, the negative feedback of E2 is characterized by an inhibitory effect on 
, LHß, and FSHß mRNA levels [10–12], persisting even during the preovulatory gonadotropin surge. Moreover, E2 increases the number of GnRH receptors in the pituitary during the negative [13] as well as the positive feedback [14–17], thereby enhancing pituitary responsiveness to GnRH during the surge.
Whether the pituitary actions of E2 involve specific subpopulations of gonadotrope cells is not known. Therefore, in this study, we determined changes in the proportion of the different subpopulations of gonadotrope cells in response to E2. Because modifications in these proportions may reflect changes in gene expression, protein storage, and/or protein release, the relationship between steady-state concentrations of gonadotropin subunit mRNAs, gonadotrope immunoreactive cell proportions, and gonadotropin release was examined. Both negative and positive effects of E2 were first examined in ovariectomized (OVX) ewes. Second, we focused on the direct pituitary effect by treating OVX ewes, which were deprived of endogenous active GnRH by passive immunization, with E2. Because immunization against GnRH induces a depletion in gonadotropin storage [18, 19], these ewes were maintained under constant hourly pulses of GnRH analogue to ensure gonadotrope function.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Production of GnRH-antibodies followed the method described by Molter-Gérard et al. [6]. Briefly, the conjugate made of synthetic GnRH (Bale Biochem, Bale, Switzerland) linked to BSA (Sigma, L'Isle-d'Abeau, France) was dissolved in phosphate buffer and emulsified with Freund complete adjuvant (one-third conjugate and two-thirds adjuvant). Rams were immunized against the GnRH-BSA conjugate (n = 8), or against the carrier BSA alone (n = 2) according to the method described by Caraty et al. [20]. The antigen (1 mg of GnRH per 4 mg of BSA) was injected intradermally (at 
30 sites) at weekly intervals for 4 wk, and intradermal booster injections were performed over 6 mo at 2-mo intervals. Antibody titer was determined by incubating serial dilutions of serum with 30 000 cpm of 125I-GnRH (specific activity, 300 µCi/µg). Radiolabeled GnRH bound to the antibody was isolated by ethanol precipitation. The titer was defined as being the dilution of antiserum that bound 50% of the radiolabeled GnRH. The sera of the rams in which the specific GnRH-antibody titer exceeded 1/50 000 (n = 2) were pooled, sterilized by 0.2-µm filtration, and stored at -20°C.
Experimental Design
The experiment was conducted between November 1996 and January 1997. Forty mature, Ile de France OVX ewes, which were ovariectomized at least 1 wk before, were used. Twenty of the OVX ewes were subjected to GnRH immunization (OVX-IMG ewes) by two injections of 50 ml of GnRH antiserum produced as described earlier and performed 7 and 10 days after ovariectomy. Furthermore, these ewes were given hourly pulses of the GnRH analogue (GnRH-A) DesGly 10GnRH Ethylamide (Sigma). The GnRH-A treatment (100 ng/pulse in 2 ml of physiological saline) was initiated 6 h after the last anti-GnRH injection and continued for 48 h until slaughtering. A complete absence of binding by this GnRH analogue to the antiserum was checked according to the method described by Caraty et al. [20]. These ewes were designed as OVX-IMG ewes. Efficiency of the GnRH antiserum was demonstrated in a previous work [6], in which no endogenous LH pulse was detected at least 8 days after the first antiserum injection. The 20 remaining ewes constituted the OVX set. Both sets of ewes (OVX and OVX-IMG) were given 50 µg of E2 (25 µg i.m. in oil and 25 µg i.v. in physiological saline) at 6, 12, 16, and 24 h before slaughtering (n = 4 for each time point). This treatment has been shown previously to induce a gonadotropin surge within 12–18 h of injection [7]. One control group in each set of ewes (n = 4) received no E2. Ewes were killed by decapitation 12 days after ovariectomy.
One day before the experiment, an indwelling catheter was installed in the jugular vein of each animal to facilitate the injections and the collection of blood samples. Blood samples were collected throughout the experiment every 15 min for 6 h before E2 injection and every hour after E2 injection until slaughtering. Moreover, for the OVX-IMG ewes, blood samples were additionally collected every 15 min for 4 h before and 4 h after each anti-GnRH injection, as well as before and after the beginning of the hourly GnRH-A pulses. The samples were heparinized and centrifuged (2000 x g for 20 min at 4°C), and the plasma was separated and stored at -20°C until assayed.
Anterior pituitaries were immediately collected after slaughtering and hemisected midsagittaly. One half was fixed during 72 h in Bouin Holland fixative containing 10% HgCl2 for immunohistochemistry, and the other half was embedded in OCT compound (Tissue-Tek, Miles Laboratories, Elkhart, IN) and stored at -70°C for mRNA studies.
Radioimmunoassay
Assays for LH were performed in duplicate 100-µl aliquots of plasma using the RIA method as described by Pelletier et al. [21] and modified by Montgomery et al. [22]. The results were expressed as nanograms of LH CY1051 (equivalent to 2.5 NIH-LH-S1). The minimum detectable concentration for LH was 0.1 ng/ml. The intra- and interassay coefficients of variation for three plasma pools averaged 8.5% and 9.5%, respectively. The concentrations of plasma FSH were measured using the reagents supplied by NIADDK (Bethesda, MD), and the results were expressed as nanograms of ovine FSH (oFSH) 19-SIAFP RP2. The minimum detectable concentration for FSH was 0.2 ng/ml. Mean intra- and interassay coefficients of variation were less than 9.5% and 9.8%, respectively. The cross-reaction with ovine LH (oLH) was 0.6%.
Immunohistochemistry and Quantitative Microscopical Analysis
Double labeling was used to determine the percentage of LH and FSH cells (mono- and bihormonal) among the total pituitary cells and was performed as described elsewhere [3]. Briefly, pituitary sections (7 µm) were incubated with rabbit anti-oLHß and horse anti-oFSHß. These antibodies were revealed with goat antirabbit immunoglobulins conjugated to fluorescein isothiocyanate and goat antihorse immunoglobulins conjugated to lissamine rhodamine (Jackson Immunoresearch, West Grove, PA), respectively. Appropriate controls were performed as described elsewhere [3]. Light microscope images from the stained sections were acquired through a 40x objective and a tri-CCD camera (Lhesa, Cergy-Pontoise, France). The analysis was performed using the Visilog 5.4.1 Image Analyzer (Noesis, Velizy, France). The proportion of monohormonal cells was assessed by the ratio of single-stained area to total area, and the proportion of bihormonal cells was assessed by the ratio of double-stained area to total area. A preliminary experiment led us to select two consecutive sections after the first 20 sections. Twenty sections later, two more consecutive sections were selected. These four pituitary sections of each animal were treated and analyzed in two different immunohistochemistry assays. On each section, 20 fields selected according a grid covering the entire section were analyzed.
Dot-Blot Analysis
Amounts of LHß and FSHß mRNAs were quantified in triplicate by dot-blot analysis. Total RNA was isolated from frozen pituitary sections (20 sections of 20 µm each) using the method described by Chomczynski and Sacchi [23]. The amount and purity of the mRNA were determined using absorption spectrometry to measure optical densities at 260 and 280 nm. A sample of 3 µg of total RNA from each pituitary was electrophoresed through a 1.5% agarose gel containing ethidium bromide to check the quality of the preparation. Probes were prepared from cDNAs for LHß and FSHß cloned in pUC18 and kindly provided by R. Counis (CNRS, Paris, France). The probe encoding for LHß was a 443-base pair (bp) fragment, and the probe encoding for FSHß was a 277-bp fragment. The probe encoding for actine was a 520-bp fragment obtained by reverse transcription-polymerase chain reaction with specific primers provided by V. Piketty (INRA, Nouzilly, France). The probes were radiolabeled with 32P by random priming using the multiprime DNA labeling system (Amersham, Buckinghamshire, UK). The RNAs (3 µg for FSHß, 1 µg for LHß) were denatured for 5 min at 65°C in three volumes of 65% formamide, 8% formaldehyde, and 1.3x MOPS and applied in 4-µl aliquots to Hybond N+ membranes (Amersham) prewetted in 10x saline-sodium citrate (SSC). Prehybridization was performed in 6x SSC containing 50% formamide, 100-µg/ml denatured herring sperm, 5x Denhardt's, and 0.5% SDS during 2 h at 42°C. Hybridization was performed overnight in the same buffer without Denhardt's and with 5 million dpm/ml of the radiolabeled LHß, FSHß, or actine probes. The blots were then washed for 30 min in, respectively, 2x SSC/0.1% SDS at room temperature, 2x SSC/0.1% SDS at 42°C, 0.2x SSC/0.1% SDS at 42°C, and 0.1x SSC/0.1% SDS at 65°C. Blots were then exposed to phosphor screen during 1–2 days. A phosphor-imaging system (Molecular Dynamics, Sunnyville, CA) was used to visualize and to quantitate spots of 32P cDNA-RNA heteroduplexes. Relative abundance of LHß or FSHß mRNA was normalized to relative amounts of actine.
Data Analysis
The mean plasma LH or FSH values obtained through the 2 h preceding slaughter of the ewes not treated with E2 or those killed at 6, 12, 16, or 24 h after E2 injection were evaluated independently on OVX or OVX-IMG ewes by one-way ANOVA. The Newman-Keuls test was used to determine which means were different. Moreover, the mean plasma LH and FSH values obtained through the 2 h preceding slaughter of the ewes not treated with E2 were compared with the values obtained through the 2 h preceding E2 injection of the treated ewes by ANOVA, followed by the Newman-Keuls test. The results showed that LH and FSH values of the non-E2-treated ewes were not different than the preinjection values of the animals receiving E2 (data not shown).
Statistical comparison of the percentage of mono- and bihormonal cells was performed independently for OVX and OVX-IMG ewes using one-way ANOVA, followed by Newman-Keuls test for individual comparisons. Because the immunohistological studies were performed separately for the OVX and OVX-IMG ewes, no comparison of the gonadotropin storage pattern between these two sets was done. Statistical comparison of mRNA expression was performed for OVX and OVX-IMG ewes using two-way ANOVA, followed by Newman-Keuls test for individual comparisons. All data were expressed as mean ± SEM. Differences were considered to be significant for P values of 0.05 or less.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
As described elsewhere [6], passive immunization against GnRH suppressed the pulsatility of LH immediately after the GnRH-antiserum injection and decreased the mean FSH levels within 3 days (Fig. 1). In contrast, injection of the BSA-antiserum had no effect on LH pulsatility or on FSH plasma levels (data not shown). Pulsatile replacement with the GnRH-A restored the LH pulses, with each GnRH-A injection producing an LH pulse, but it failed to restore the FSH levels that were observed before immunization (Fig. 1).
|
Gonadotropin Release after E2 Treatment
A biphasic effect on mean LH and FSH plasma levels in OVX and OVX-IMG ewes was observed after E2 treatment (Fig. 2, inserts). In the OVX ewes, injection of E2 6 h before slaughter induced a decrease in LH plasma levels compared with non-E2-treated ewes (P < 0.01). In contrast, an increase in LH plasma concentrations was observed in ewes killed 12 h after injection (P < 0.01 vs. nontreated ewes). This increase was enhanced 16 h after injection (P < 0.0001 vs. nontreated ewes). Similarly, injection of E2 6 h before slaughter induced a decrease in FSH plasma levels (P < 0.05). These low concentrations were maintained 12 h after E2 injection (P < 0.01 vs. nontreated ewes) and increased 16 h after E2 treatment to reach the levels observed in non-E2-treated ewes. The OVX-IMG ewes showed a similar biphasic pattern of gonadotropin release (Fig. 2, inserts).
|
Gonadotropin Storage Pattern
Analysis of dual immunostaining in OVX and OVX-IMG ewes showed that the gonadotrope cell population was mainly constituted by monohormonal LH and bihormonal LH/FSH cells (Fig. 3). The proportion of monohormonal FSH cells never exceeded 0.5% of the total pituitary cells.
|
Figure 2 shows the percentage of the gonadotrope cells (monohormonal LH + bihormonal LH/FSH cells) among the pituitary cells in the different experimental groups. This percentage was lower in the E2-treated ewes than in the non-E2-treated ewes, both in the OVX set (P < 0.001), and the OVX-IMG set (P < 0.001). In this last set, the decrease was more pronounced 16 h after E2 injection (P < 0.001 vs. other groups). The proportion of monohormonal LH cells was constant throughout the experiment, except at 16 h after E2 injection, when it was enhanced in the OVX set (P < 0.001 vs. other groups) and in the OVX-IMG set (P < 0.001 vs. other groups). In the OVX ewes, the proportion of bihormonal cells was lower in the E2-treated ewes compared with the nontreated ewes (P < 0.001), with a more pronounced decrease 16 h after E2 injection. A slight increase was observed 12 h after E2 injection compared with 6 h (P < 0.05 vs. nontreated ewes). In the OVX-IMG ewes, the pattern of the proportion of bihormonal cells was similar to that in the OVX ewes, except between 6 and 12 h after E2 treatment, when no difference was observed.
LHß and FSHß mRNA Expression
Relative amounts of LHß and FSHß mRNAs (± SEM) for the different experimental groups are shown in Figure 4. Amounts of mRNA for actine did not differ before or after treatment with E2 in either OVX or OVX-IMG ewes (data not shown). Relative levels of LHß and FSHß mRNAs were lower in the treated OVX-IMG group than in the OVX group (P < 0.05), independent of E2 treatment. In OVX as well as in OVX-IMG ewes, injection of E2 did not alter the relative levels of LHß mRNA. In contrast, injection of E2 decreased FSHß mRNA expression in OVX ewes (P < 0.003) and in OVX-IMG ewes. In OVX ewes, FSHß mRNA relative levels were lowered 6 or 12 h after E2 injection (P < 0.05), and the decrease was enhanced 16 h after injection compared with nontreated ewes (P < 0.01). Twenty-four hours after injection, FSHß mRNA levels were not different from those observed in nontreated ewes. In OVX-IMG ewes, FSHß mRNA levels became undetectable in E2-treated animals.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The injection of E2 rapidly (within 6 h) led to a decrease in gonadotropin release and in the percentage of gonadotrope cells. This latter fits with the decrease in LH and FSH pituitary contents that has been previously observed [24]. Both the reductions in FSH release and in the percentage of FSH immunoreactive cells associated with a decrease in FSHß mRNA level indicate reduced FSH synthesis. In contrast, E2 treatment did not affect the level of LHß mRNA in either OVX or OVX-IMG ewes, raising the possibility of a regulation at a post-transcriptional level. These results extend previous data obtained in OVX, hypothalamopituitary-disconnected ewes [25]. However, in OVX ewes, a decrease in the LHß mRNA amounts after E2 injection was reported [25]. These divergent results may be caused by the delay after ovariectomy. More interesting, the negative effect of E2 is accompanied by a decrease in the percentage of bihormonal cells, whereas the percentage of monohormonal LH cells is not affected. This strongly suggests that E2 acts on bihormonal cells. However, it cannot be excluded that the monohormonal LH cells originally present may disappear, to be replaced by bihormonal cells that have lost their FSH.
Moreover, similar changes in the percentages of gonadotrope cell subpopulations were observed both in OVX and OVX-IMG ewes, which indicates that E2 acts, at least in part, directly at the pituitary level. These data are consistent with the effects of E2 on gonadotropin release reported in previous studies of ovine species [26] and with the presence of E2 receptors (ER) in the pituitary [27]. Whether a differential distribution of ER exist in the gonadotrope subpopulations is not known. Recently, Mitchner et al. [28] demonstrated in rat that only 25% of LH-containing cells expressed the ER mRNA and, to a lesser extent, ERß. Alternatively, E2 may act indirectly on gonadotrope cells via another cell type, such as somatotropes or lactotropes [28].
The negative effect of E2 was followed by the induction of a surge in LH release, as expected [19, 20], as well as by a significant increase in FSH plasma concentrations. Besides the enhanced LH and FSH release 12 or 16 h after E2 injection, the percentages of LH-containing cells as well as FSH-containing cells were maintained at lower levels compared with the non-E2-treated ewes. Furthermore, the LHß mRNA levels were not modified, whereas the decrease in FSHß mRNA levels initiated 6 h after E2 injection was maintained and even magnified when FSH release was maximal. Hence, development of the LH or FSH surges is not related to a recruitment of new cells in the gonadotrope pool or to an enhancement in gene expression [29]. More probable is that the gonadotropin surge is related to changes from nonreleasable pools to releasable pools.
The stimulatory effect of E2 on LH release, occurring in both OVX and OVX-IMG ewes, may result, at least in part, from the ability of E2 to enhance responsiveness to GnRH by increasing the number of GnRH receptors in the pituitary [13–17, 30–32].
Initiation of the gonadotropin surge 12 h after E2 injection did not modify the percentage of monohormonal LH cells in either OVX or OVX-IMG ewes, but the percentage of bihormonal cells showed a small and transitory increase in OVX animals compared with the preceding stage. This transitory rise could suggest that E2 can enhance the storage of LH and FSH in only a percentage of bihormonal cells via its stimulatory effect on GnRH release. These cells might participate in the sustained gonadotropin release. The lack of increase in the percentage of bihormonal cells in OVX-IMG ewes 12 h after E2 treatment may relate to the end of the inhibitory effect of E2 observed 6 h after treatment, which is less pronounced than in OVX ewes, or reflect the slower rise in gonadotropin release in this experimental set.
When plasma concentrations of LH and FSH were maximal (16 h after E2 injection), a pronounced decrease in the percentage of FSH-containing cells (bihormonal) occurred, whereas the percentage of LH-containing cells (monohormonal LH + bihormonal) remained constant in OVX ewes and decreased only in OVX-IMG ewes. During the surge, only a small portion of the pituitary store of LH might be released, whereas almost all the gonadotrope content of FSH is released. However, even if the percentage of LH-containing cells remained constant in this study within OVX ewes, the staining intensity in LH cells was lower after the surge peak (data not shown), suggesting a reduction in LH pituitary content.
The decrease in the proportion of bihormonal cells observed at this time coincided with an increase in the percentage of monohormonal LH cells. Inversely, 24 h after E2 injection, the percentage of monohormonal LH cells was decreased, and this accompanied an increased percentage of bihormonal cells. The opposite changes of these subpopulations led us to postulate that at least part of the gonadotrope cell population can be either monohormonal LH or bihormonal, depending on the physiological conditions. The switch from one subpopulation to another seems to be driven by the inhibition or stimulation of FSH synthesis and/or release in these cells.
Taken together, the data presented herein show that E2 induces rearrangement in gonadotrope cell subpopulations, at least in the presence of GnRH. This effect may result from the ability of E2 to increase the number of GnRH receptors on the gonadotrope cells. In ovine pituitary cell culture, E2 plays a minor role in absence of GnRH stimulation in rearranging the types of gonadotropes, but it was shown to increase the number of GnRH-responsive cells [32]. The negative effect of E2 seems to target mainly the bihormonal cells and occurs, at least in part, directly at the pituitary level. Moreover, the gonadotropin surge induced by E2 would cause a switch from bihormonal cells to monohormonal LH cells by the loss of FSH. The combination of immunohistochemistry and in situ hybridization would provide further information about the physiological relevance of these switches.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 Supported by a grant from the Région Centre. 
2 Correspondence: Catherine Taragnat, Institut National de la Recherche Agronomique, Station de Physiologie de la Reproduction des Mammifères Domestiques, 37380 Nouzilly, France. FAX: 33 247427743;taragnat{at}tours.inra.fr
Accepted: May 17, 2000.
Received: August 3, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M-O Faure, L Nicol, S Fabre, J Fontaine, N Mohoric, A McNeilly, and C Taragnat BMP-4 inhibits follicle-stimulating hormone secretion in ewe pituitary J. Endocrinol., July 1, 2005; 186(1): 109 - 121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Xu, X.-Z. Xu, C. S. Nunemaker, and S. M. Moenter Dose-Dependent Switch in Response of Gonadotropin-Releasing Hormone (GnRH) Neurons to GnRH Mediated through the Type I GnRH Receptor Endocrinology, February 1, 2004; 145(2): 728 - 735. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Meeran, H. F. Urbanski, S. J. Gregory, J. Townsend, and D. J. Tortonese Developmental Changes in the Hormonal Identity of Gonadotroph Cells in the Rhesus Monkey Pituitary Gland J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2934 - 2942. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) and FSH (
) levels from a representative OVX-IMG ewe before and after GnRH-antiserum injection and during administration of hourly pulses of GnRH analogue (shadowed area)
