Tamoxifen but Not Other Selective Estrogen Receptor Modulators ...

Central Effects of Estrogens and Reproductive Neuroendocrinology Neuroendocrinology 2002;76:203–213 DOI: 10.1159/000065952

Received: April 22, 2002 Accepted after revision: August 1, 2002

Tamoxifen but Not Other Selective Estrogen Receptor Modulators Antagonizes Estrogen Actions on Luteinizing Hormone Secretion while Inducing Gonadotropin-Releasing Hormone Self-Priming in the Rat José E. Sańchez-Criado a Pedro Guelmes b Carmina Bellido a Mirian Gońzalez c Guadalberto Hernańdez b Rafaela Aguilar a José C. Garrido-Gracia a Aixa R. Bello c Rafael Alonso b Departments of a Cell Biology, Physiology and Immunology, University of Co´rdoba, and b Physiology and of Cell Biology, University of La Laguna, Spain

c Section

Abstract Selective estrogen receptor modulators (SERMs) are compounds which may function as agonists or antagonists depending upon the target tissue. This study compares the actions of different SERMs on luteinizing hormone (LH) secretion, and on gonadotropin-releasing hormone (GnRH) self-priming in the rat. To do this, 4-day cyclic rats were injected twice, on day 2 (metestrus) and day 3 of the estrous cycle, with one of the following SERMs: 0.25 mg ICI 182,780, 3 mg tamoxifen (TX), LY139481-HCl or LY117018-HCl, or 0.5 mg RU58668. Control rats were given subcutaneous injections of 0.2 ml oil. On the morning of day 4 (proestrus in controls), rats from each group were either injected intraperitoneally with pentobarbital (40 mg/kg) for in vivo study or decapitated and their pituitaries collected for incubation (in vitro study). Additionally, pituitaries taken on each day of the estrous cycle from control rats as well

ABC

© 2002 S. Karger AG, Basel 0028–3835/02/0764–0203$18.50/0

Fax + 41 61 306 12 34 E-Mail [email protected] www.karger.com

Accessible online at: www.karger.com/journals/nen

as on day 4 from SERM-treated rats were processed for immunohistochemical determination of the estrogen receptor-· (ER·) gonadotrope. The plasma concentration or accumulation of LH in the medium was determined after 1 h (basal secretion). Thereafter, an intravenous bolus of GnRH (50 ng/0.5 ml/100 g BW) or 10 –8 M GnRH was injected or added to the medium, respectively. After 1 h of GnRH exposure, blood or medium were taken, and another challenge of GnRH was made. At the end of the 3rd h of the experiment, blood or medium samples were taken again and the LH plasma concentration or accumulation in the medium were determined. All SERM treatments reduced uterus weight and decreased basal and stimulated LH secretion. Also, on day 4, rats treated with any SERM other than TX showed vaginal smears infiltrated by leukocytes and a reduction in GnRH self-priming. TX-treated rats exhibited cornified vaginal smears and an estrogenic effect on GnRH self-priming. Moreover, 15-min exposure to two consecutive GnRH (10 –8 M) challenges 1 h apart in incubated pituitaries with estradiol (E2, 10 –8 M), TX (10 –7 M), E2 + TX, or medium alone form ovariectomized rats injected for 3 days with estradiol benzoate (25 Ìg), TX (3 mg), estradiol benzoate + TX, or 0.2 ml oil, respectively, showed that TX increased

Jose E. Sanchez-Criado Seccioń de Fisiologıá, Facultad de Medicina Avda. Menendez Pidal s/n E–14004 Co´rdoba (Spain) Tel. +34 957 218283, Fax +34 957 218288, E-Mail [email protected]

Downloaded by: Dr.Alonso Rafael - 31020 198.143.45.65 - 6/19/2015 12:42:50 PM

Key Words Selective estrogen receptor modulators W Tamoxifen W Raloxifene W Gonadotropins W Gonadotropin-releasing hormone W Gonadal steroids W Gonadal steroid receptors

Copyright © 2002 S. Karger AG, Basel

Introduction

The preovulatory surge of luteinizing hormone (LH) is the consequence of the combined action of the hypothalamic decapeptide gonadotropin-releasing hormone (GnRH) and ovarian estradiol (E2) on the pituitary [1]. Thus, E2 is a key regulator of the synthesis and secretion of pituitary reproductive hormones and, consequently, of the maintenance of reproductive processes [2, 3]. E2 triggers the preovulatory GnRH surge in proestrus [1], sensitizes the pituitary to GnRH [4], and potentiates GnRH self-priming [5]. Also, E2 induces expression of pituitary progesterone receptors (PRs) [6, 7], and, in turn, the preovulatory LH-dependent proestrous progesterone secreted by preovulatory follicles, acting on these receptors, enhances the magnitude of LH release [8, 9]. Selective estrogen receptor modulators (SERMs), previously known as anti-estrogens, are synthetic molecules with mixed agonist/antagonist characteristics depending on the target tissue, the animal species and the response measured [10, 11]. The estrogen-dependent GnRH-LH system governs all reproductive functions in females [2]. Antagonizing estrogen actions by the administration of different SERMs to the female rat varies in its effect on GnRH-LH-dependent reproductive processes [12, 13]. For example, raloxifene decreases LH secretion in neonatal female rats mimicking the action of neonatal estrogenization [14], and the benzothiophene raloxifene derivative LY117018-HCl acts as an anti-estrogen, decreasing the GnRH activation of estrogen-dependent PRs in the pituitary [15]. On the other hand, the triphenylethylene tamoxifen (TX) increases LH accumulation in a 4-hour incubation of proestrous pituitary in the same way as estrogen [16]. Without disregarding the possible phenom-

204

Neuroendocrinology 2002;76:203–213

enon of pituitary desensitization under continuous exposure to GnRH [17], accumulated LH in incubation medium during prolonged exposure to GnRH may be the result of a complex secretory process involving basal, GnRH-stimulated LH secretion and GnRH self-priming. The purpose of these experiments was to compare the effects of four different SERMs (tamoxifen (TX), raloxifene (RX), LY117018-HCl (LY) and RU58668 (RU)) with those of ‘pure’ anti-estrogen ICI 182,780 (ICI) on basal and GnRH-stimulated secretion and on GnRH selfpriming in pentobarbital-injected rats and in incubated rat pituitaries. Since we found that TX, in contrast to the other SERMs used, inhibited basal and GnRH-stimulated LH secretion while inducing GnRH self-priming, and increased estrogen receptor-· (ER·) immunoreactivity in the cytosol of gonadotropes whilst blocking the nucleocytoplasmatic shuttling, we compared in vitro the effects of TX on LH secretion and on GnRH self-priming with those of E2 under pulsatile exposure of pituitary glands to GnRH. Materials and Methods Adult cyclic female Wistar rats weighing 220 B 20 g were used. Rats were housed (5 per cage) under a 14:10-hour light-dark cycle (lights on at 05.00 h) and 22 B 2 ° C room temperature, with ad libitum access to rat chow and tap water. Vaginal smears were taken daily and only rats showing consistent 4-day cycles were used. The day of vaginal estrus (day of ovulation) was arbitrarily defined as day 1 of the cycle. All experimental protocols were approved by the Ethical Committee of the Universities of Co´rdoba and La Laguna, and the experiments performed in accordance with the rules of laboratory animal care and international law on animal experimentation. Drugs and Treatments The different SERMs used in this study were: pure anti-estrogen ICI (Astra-Zeneca) [18], TX (Sigma Chemical Co., St. Louis, Mo.), LY139481-HCl (RX) and LY (Ely Lilly & Co., Indianapolis, Ind.), a pyrrolidine analog of RX [19], and RU (Roussel-Uclaf, Romainville, France), a new class of ER antagonist devoid of partial agonist activity [20]. Rats were injected subcutaneously at 09.00 h on days 2 and 3 with either 0.25 mg ICI, 3 mg TX, 3 mg RX, 3 mg LY, or 0.5 mg RU. Controls received 0.2 ml olive oil. In vivo Experiments At 11.00 h on day 4, heparinized (1,000 IU/rat) rats, which had had a right atrial cannula inserted through the external jugular vein on day 3 [21], were given an intraperitoneal injection of pentobarbital (PB; 40 mg/kg) to block the endogenous surge of GnRH and gonadotropins. One hour after PB injection about 700 Ìl of blood was obtained (1st h) and 50 ng/0.5 ml saline/100 g BW intravenous bolus of GnRH (LHRH, Peninsula Laboratories Inc., Merseyside, UK) was given. One hour later, another sample of blood was removed (2nd h) and a second bolus of GnRH was given. Blood was

Sańchez-Criado et al. Downloaded by: Dr.Alonso Rafael - 31020 198.143.45.65 - 6/19/2015 12:42:50 PM

GnRH self-priming, as did E2, whereas it reduced the E2sensitizing effect on GnRH-stimulated LH secretion and cancelled the E2-dependent GnRH self-priming. All SERMs prevented the physiological nucleocytoplasmic shuttling of ER· exhibited during proestrus in control rats, and TX, in addition, induced a significantly larger number of gonadotropes displaying strong cytosolic immunosignals corresponding to ER· than the rest of the experimental groups. Overall, data from this study indicated that, in contrast to the general antagonistic effect of the tested SERMs, TX seemed to display both selective agonist and antagonist activity at the gonadotrope level and on GnRH self-priming of LH secretion respectively.

In vitro Experiments At 10.00 h on day 4, rats were decapitated and trunk blood collected. Blood was allowed to clot and then centrifuged at 4 ° C; serum was stored at –20 ° C until quantified for LH and compared with plasma LH concentrations of the in vivo experiment (1st h) to determine the effect of PB on LH secretion. In addition, wet and dry uterus weights were recorded and compared with those from a group of rats on day 2. Anterior pituitaries were dissected out for the incubation experiments. Incubation of pituitaries was carried out as described previously [15, 16, 22]. Briefly, halves of pituitaries were incubated at 37 ° C with constant shaking (60 cycles/min) in an atmosphere of 95% O25% CO2. Each vial contained 1 ml of Dulbecco’s modified Eagle’s medium (DMEM), without L-glutamine and phenol red, containing glucose (4.5 g/l) and bovine serum albumin (BSA, 0.1%, w/v, pH 7.4). In addition, each vial contained the corresponding SERM: ICI (10 –8 M ), TX (10 –7 M ), RX (10 –7 M ), LY (10 –7 M ), RU (10 –8 M ) or, in the case of control pituitaries, medium alone or 10 –8 M E2 (Sigma Chemical Co.). At the end of the preincubation period (60 min), pituitaries were incubated for 3 h: the 1st without GnRH (basal release) and the 2nd and 3rd h with GnRH (10 –8 M ). Medium was removed at the end of each 3-hour period to determine LH accumulation in the medium. In a second in vitro experiment pituitaries from 2-week ovariectomized (OVX) rats injected subcutaneously over 3 days with 25 Ìg E2 benzoate (EB), TX (3 mg), EB + TX, or 0.2 ml olive oil were incubated during 3 h with 10 –8 M E2, TX (10 –7 M ), E2 + TX, or medium alone, respectively. GnRH (10 –8 M ) was added to the incubation medium for 15 min at the beginning of the 2nd and 3rd h of incubation. During the 1st hour the pituitaries were incubated without GnRH. Medium was aspirated every 15 min for quantification of LH concentration by RIA. Radioimmunoassay of LH Concentrations of LH in plasma and in the incubation medium were measured in duplicate by radioimmunoassay using a doubleantibody method with kits supplied by NIH (Bethesda, Md.) and a previously described microassay method [9, 22]. Rat LH-I-9 was labelled with 125I by the chloramine T method. The intra-assay coefficient of variation was 7% and assay sensitivity was 7.5 pg/tube. The LH concentration (mean B SEM of 5 or 8 observations for in vivo and in vitro study, respectively) was expressed as ng/ml or ng/hemipituitary of the reference preparation LH-rat-RP-3. GnRH Self-Priming Self-priming was evaluated as the percentage increase in LH plasma concentration or accumulation in the medium during the 3rd h of the experiment (2nd GnRH exposure) with respect to the 2nd h (1st GnRH exposure). In the third experiment GnRH self-priming was similarly evaluated, but the peak of LH accumulation in the medium was taken into account (15 min after the beginning of GnRH exposure).

the estrous cycle. At 14.00 h rats were injected with pentobarbital (40 mg/kg) and then fixed by intracardiac perfusion of sodium phosphate buffer (pH 7.4, 0.1 M ) containing 0.4% paraformaldehyde. After decapitation of the animals, the pituitaries were dissected out, postfixed in the same fixative for 2 h, and immersed overnight in sodium veronal buffer (pH 7.4, 0.1 M ) with 20% saccharose. The pituitaries were embedded in TissueTek and frozen in isopentane cooled with liquid nitrogen. Horizontal cryostat sections (8–10 Ìm) were collected on gelatin-coated slides, allowed to dry, and rehydrated in sodium veronal buffer, which was also used for further incubations and washes. Antiserum to ER· (Upstate, Buckingham, UK) was raised in rabbits against a KLH-conjugated synthetic peptide representing the last 15 amino acids of the rat ER·, and its immunological characteristics have been reported elsewhere [23]. Antiserum raised against human FSH (Chemicon International, Temecula, Calif.) was also used as described previously [24]. The indirect immunohistochemical procedure was carried out by incubating sections overnight at room temperature with this antiserum (1:500) and after rinsing with biotinylated goat antirabbit F(ab) fragments and peroxidase-labeled streptavidin complex (Biosys, Paris, France). Peroxidase activity was revealed in Tris-HCl buffer (pH 7.6, 0.05 M ) containing 0.04% 4-chloro-1-naphthol (Sigma-Aldrich, Madrid, Spain) and 0.01% hydrogen peroxide in Tris-HCl buffer. The specificity of immunostaining was assessed by replacing the specific antiserum with normal serum, omitting one step of the reaction, or following preabsortion of the antiserum with the corresponding antigen. The efficiency of the elution procedure was checked by substituting normal serum (1:50, 60 min) for the specific antiserum of the second reaction. To demonstrate the coexistence, an elution-restaining procedure was used [25] which allowed us to carry out a second immunohistochemical sequence using antibodies against FSH (1:800). Peroxidase activity was revealed using 3-amino-9ethylcarbazol and 0.01% hydrogen peroxidase in sodium acetate buffer (pH 5, 0.05 M ). For quantitative analysis, anterior pituitary sections were examined by light microscopy (Leitz Laborlux S, Germany). At least three fields from the same section were digitalized through a camera (Analysis Imagen System Leica Q-Win, Barcelona), and the number of immunoreactive cells for the ER· antigen per unit area were counted (1,000 Ìm2). Data are expressed as the percentage of cells that were positive for ER· within a defined area. Statistical Analysis All data are presented as means B SEM. Statistical analysis was performed with two-way ANOVA followed by Newman-Keuls multiple range test. Student’s t test was also used when appropriate. A p level of 0.05 was considered significant.

Results

Immunohistochemical Detection of Pituitary ER· For immunohistochemical studies, groups of 5–7 rats were treated with the different SERMs as described above and sacrificed on day 4. Controls were oil-injected rats decapitated on each day of

SERM Effects on Vaginal Cytology and Uterine Weight On day 4 of the estrous cycle, oil-injected rats typically exhibited nucleated epithelial cells in vaginal smears. All SERM-treated rats, except those injected with TX, predominantly showed vaginal smears infiltrated by leuko-

Tamoxifen Induces GnRH Self-Priming


205 Downloaded by: Dr.Alonso Rafael - 31020 198.143.45.65 - 6/19/2015 12:42:50 PM

taken at the end of the 3rd h of the experiment. After centrifugation of blood, plasma was stored at –20 ° C until assayed for LH.

cytes. The vaginal smears of TX-treated rats were similar to those of controls with variable amounts of estrous cornified cells. Wet and dry uterine weights decreased in all SERM-treated rats when compared to control rats in proestrus (p ! 0.01) but not when compared to control rats in metestrus (fig. 1). All OVX rats injected with EB and/or TX, but not OVX oil-injected rats, had exclusively cornified cells in the vaginal smears. Furthermore, while EB treatment induced intense uterine ballooning, significantly reduced by TX, injections of TX alone did not induce ballooning of the uterus.

206


Fig. 2. Effects of SERMs on basal (a) and GnRH-stimulated LH secretion (b, c) and on GnRH self-priming (d) in pentobarbital-

injected rats ()). Rats were treated with SERMs as described in the legend to figure 1. At 11.00 h on day 4 rats were injected intraperitoneally with 40 mg/kg pentobarbital and 1 h later blood was taken (a; )) and injected intravenously with 50 ng/100 g BW of GnRH. At 13.00 h another blood sample was taken (b) and a second bolus of GnRH was given. At 14.00 h blood was drained to determine the LH plasma concentration (c). GnRH self-priming (d) is expressed as the percentage increase in LH concentration during the 2nd h of exposure to GnRH with respect to the first. a The hatched bars represent basal serum LH concentrations from rats of the in vitro experiment (fig. 1). Data are the mean B SEM of 5–6 rats. Two-way ANOVA and Newman-Keuls multiple range test. * p ! 0.01, and # p ! 0.05 vs. controls in proestrus.

SERM Effects on in vivo Basal and GnRH-Induced LH Secretion The serum LH concentration before the first GnRH bolus injection (1st h) was significantly (p ! 0.01) reduced in all SERM-treated groups (fig. 2a, hatched bars). Pento-


Fig. 1. Wet (a) and dry (b) uterine weights on metestrus (M) and on day 4 of the estrous cycle (proestrus; P) in oil-injected rats, and on day 4 in rats injected on days 2 (metestrus) and 3 (diestrus) with 0.25 mg ICI182,780 (ICI), 3 mg tamoxifen (TX), 3 mg raloxifene (RX), 3 mg LY117018-HCl (LY), or 0.5 mg RU58668 (RU). Data are the mean B SEM of 8 rats. Two-way ANOVA and NewmanKeuls multiple range test. * p ! 0.01 vs. controls in proestrus; # p ! 0.05 vs. controls in metestrus.

suppressive effect of LY was slightly less evident than that of the other SERMs (fig. 2b). SERM Effects on in vitro Basal and GnRH-Induced LH Secretion As in in vivo experiments, LH released into the medium in the absence of GnRH (basal release) by incubated pituitaries from rats injected with ICI, TX, RX, LY or RU, was significantly (p ! 0.05) reduced with respect to control pituitaries incubated with medium alone (fig. 3a). Also, all SERMs reduced GnRH-stimulated LH secretion during both the first and the second GnRH challenge in the same way as ICI (fig. 3b, c). As in in vivo conditions, the suppressive effect of LY during the first GnRH challenge was less evident (fig. 3b). E2 added to the medium of control pituitaries tended to increase the GnRH response. SERM Effects on GnRH Self-Priming The LH concentration during the second GnRH challenge (3rd h of the experiments) were, in all experimental groups, significantly greater than during the second hour (1st GnRH challenge) thereby showing GnRH self-priming. However, all SERMs, with the exception of TX which mimics the action of E2 in in vitro conditions (fig. 3c) and augmentated GnRH self-priming in in vivo conditions (fig. 2c), significantly reduced GnRH self-priming (fig. 2c, 3c).

pituitaries. Rats were treated as described in the legend to figure 1, and their pituitaries incubated for 3 h with ICI (10 –8 M ), TX (10 –7 M ), RX (10 –7 M ), LY (10 –7 M ) or RU (10 –8 M ), respectively. Hemipituitaries from oil-injected rats were incubated either with 10 –8 M estradiol (E2) or with medium alone (M). During the 2nd and 3rd hours of incubation, 10 –8 M of GnRH was added to the incubation medium. Medium was aspirated to determine the LH accumulation at the end of each 3-hour period. Data are the means B SEM of 7–8 hemipituitaries. Two-way ANOVA and Newman-Keuls multiple range test. * p ! 0.01 and # p 0.05 vs. control pituitaries incubated with medium alone. See legend of figure 2 for additional details.

barbital injection reduced the plasma LH concentration in all experimental groups when compared with the serum LH levels of rats decapitated at 11.00 h (in vitro experiment). All SERMs reduced GnRH-stimulated LH secretion during both the first and second GnRH challenge, 2nd and 3rd h, respectively, in the same way as ICI (fig. 2b, c). However, during the first GnRH challenge, the


E2 and/or TX Effects on Basal, GnRH-Stimulated LH Secretion and GnRH Self-Priming in Pituitaries from Ovariectomized Rats Table 1 shows peak LH secretion by hemipituitaries from OVX rats injected with oil, EB, TX, or EB + TX in response to two GnRH administrations in the presence of medium alone, E2, TX, or E2 + TX, respectively. All treatments induced significant increases in LH secretion in response to both the first and the second GnRH pulses compared to basal LH secretion. Treatment with E2 enhanced basal and stimulated LH secretion and induced GnRH self-priming in comparison with pituitaries from rats injected with oil and incubated with medium alone. Simultaneous treatment with TX reduced both basal and GnRH-stimulated LH secretion, and abolished GnRH self-priming (table 1). TX alone had no effect on either basal or stimulated LH secretion; however, it stimulated GnRH self-priming. The complete pattern of LH secretion from pituitaries in the presence of these in vivo + in vitro treatments is shown in figure 4.



Fig. 3. Effects of SERMs on basal (a) and GnRH-stimulated LH secretion (b, c) and on GnRH self-priming (d) in incubated rat hemi-

Table 1. LH response (ng LH/hemipituitary) of incubated pituitary glands from ovariectomized rats treated with

25 Ìg estradiol benzoate (EB) and/or 3 mg tamoxifen (TX) over 3 days to two 15-min GnRH (10 –8 M ) pulses 1 h apart and incubated with 10 –8 M estradiol (E2) and/or 10 –7 M TX Treatment

Basal LH

in vivo

in vitro

Oil EB TX EB + TX

Medium E2 TX E2 + TX

14.2B2.0 25.6B3.3b 15.6B3.4 19.8B3.6

Peak LH response to 1st GnRH pulse

2nd GnRH pulse

29.4B4.7a 98.3B3.4a, b 25.6B3.1a 54.2B5.5a, c

29.0B2.3a 127.0B10.3a, b 43.1B5.0a 60.6B3.4a, c

GnRH self-priming % 98.6B8.1 133.9B15.5b 161.7B20.1d 118B16.6

Values are mean B SEM of 7–8 hemipituitaries (ANOVA and Newman-Keuls multiple range test). GnRH selfpriming = peak response to 2nd GnRH pulse ! 100/peak response to 1st GnRH pulse. a GnRH-induced LH secretion significantly (p ! 0.01) different from basal LH secretion. b Estradiol treatment enhanced basal (p ! 0.01) and GnRH-stimulated LH secretion (p ! 0.01) and induced GnRH self-priming (p ! 001). c TX antagonized estradiol priming on basal and GnRH-stimulated LH secretion and annulled GnRH self-priming. d TX alone augmented GnRH self-priming (p ! 0.01).

SERM Effects on the Expression of ER· in the Pituitary In control rats, ER·-like immunoreactivity (ER·-IR) was mainly localized in cytosol during all stages of the estrous cycle, except during proestrus when the immunosignal associated with ER· was exclusively found within the nucleus in all cells observed (fig. 5a, b, 6). Pretreat-

Fig. 4. LH secretion from ovariectomized rat pituitaries treated for 3 days with oil (0.2 ml), 25 Ìg estradiol benzoate (EB), 3 mg tamoxifen (TX) or EB + TX and incubated for 3 h with medium alone ([), 10 –8 M estradiol (P), 10 –7 M TX ($), or estradiol + TX ()), respectively, in response to two consecutive 15-min GnRH challenges (10 –8 M ) at the beginning of the 2nd and 3rd h of incubation. Data are the means B SEM of 7–8 hemipituitaries. * p ! 0.01 vs. peak LH levels during the first GnRH pulse (Student’s t test). Fig. 5. Distribution of ER· in the anterior pituitary of female rats. a Sections of the anterior lobe from untreated rats at proestrus (day 4) showing the exclusive nuclear localization of ER·-IR at this stage. !250. b Sections from untreated rats at diestrus (day 3) showing anterior pituitary cells with cytoplasmic ER·-IR. !300. c Representative sections on day 4 from rats treated with all SERMs used, except TX, showing very few cells that display cytoplasmic ER·-IR. !250. d Representative sections from proestrous rats treated with TX showing a large number of cells with cytoplasmic ER·-IR. In contrast to control rats in proestrus, no nuclear staining was observed in pituitaries from animals treated with any SERMs. e, f Higher magnifications (!720) of anterior pituitary sections from rats treated with SERMs showing the same cells with cytoplasmic staining for ER·-IR (e) and FSH (f).

208



4




5

209

Discussion

ER·-IR. Pituitary sections from rats at estrus (E), metestrus (M), diestrus (D) and proestrus (P) or, on day 4, in rats treated with SERMs immunostained for ER· as described in Materials and Methods. No nuclear staining was observed in control rats in M or on day 4 in rats pretreated with SERMs. Similarly, no cytosolic staining was observed in controls in P. See legend to figure 1 for details on SERMs treatment.

ment with the different SERMs caused complete abolition of nuclear ER·-IR in anterior pituitary cells on day 4, with only a few cells showing cytosolic immunostaining for ER·. Interestingly, in TX-treated rats the number of ER·-IR cells was significantly higher than that of rats treated with the other SERMs (fig. 5c, d, 6). Following elution of the first reaction performed with anti-ER· and restaining with FSH antiserum, it was found that the majority of ER·-IR cells were FSH-positive (fig. 5e, f).

210



Fig. 6. Percentage of gonadotropes expressing cytosolic and nuclear

Estrogen is essential for all aspects of reproductive function in females. In the rat, rising estrogen levels during diestrus are critical for the preovulatory release of GnRH and LH in proestrus and, consequently, the release of eggs during the early hours of estrus [2]. Both basal and GnRH-stimulated LH secretion are strictly dependent on estrogen action on the pituitary [26]. All SERMs used in these experiments reduced both basal and stimulated LH secretion, as did the ‘pure’ anti-estrogen ICI. This indicated that SERMs functioned mainly as estrogen antagonists blocking estrogen priming actions in the pituitary. Small differences found in these experiments between in vivo and in vitro actions of SERMs on basal LH secretion may be attributable to the fact that the effects of antagonists in vitro are related to their affinities for the receptors while, in vivo, their metabolic clearance rates may, in addition, influence their biological activity [27]. Furthermore, the action of estrogen on the pituitary is also responsible for GnRH self-priming [1]. During continuous stimulation of the pituitary gland with GnRH the phenomenon of self-priming is shown by a two-phase LH response: a phase of low pituitary responsiveness (unprimed response) followed by a second phase of increased LH release (primed state response) [28, 29]. The response to the first GnRH challenge demonstrates the secretory effect and the LH response to the second the priming GnRH action [30]. Although in the present study GnRH self-priming was seen in all experimental groups, ICI, as well as all SERMs except TX, reduced GnRH priming. Thus, TX seemed to act as an estrogen agonist on GnRH self-priming and as an estrogen antagonist on GnRHstimulated LH secretion. During continuous exposure of pituitary glands to GnRH the phenomenon of pituitary desensitization to GnRH cannot be overlooked [17, 31]. The GnRH self-priming effect is better recognized if the interval between two consecutive GnRH pulses, the physiological manner in which GnRH is released [32], is separated by 60 min [1]. Results from the last in vitro experiment of this study indicated that TX antagonized the sensitizing action of estrogen, reduced basal and GnRHstimulated LH secretion, and annulled GnRH priming. When given alone to OVX rats, TX did not affect basal and GnRH-stimulated LH secretion, but induced GnRH priming. The suppressive activity of SERMs on LH secretion observed in these experiments coincides with that described in postmenopausal women but not in women of reproductive age [33]. Clinical studies have demonstrated that TX functions as an estrogen agonist in bone, in the

cardiovascular system and in the uterus [12, 34], while manifesting antagonist activity in the breast [13, 35]. In rat reproductive tissues, in contrast with other SERMs, TX displayed agonistic activity at the vaginal level, inducing in intact rats on day 4 (day of proestrus in controls) vaginal smears typically seen in estrus, and in ovariectomized rats the same effects as EB. Therefore, the SERM TX, which acted as an antagonist at the uterus level prompting a reduction in uterus weight in intact rats and antagonization of estrogen action on the uterus in OVX rats, presents tissue selectivity (vagina vs. uterus). TX is generally considered to be uterotropic, especially in women [36]. It is possible, however, that the 2-day treatment used in the rat antagonizes the effects of estrogen on the uterus, while its agonistic effect (endometrial hyperplasia and cancer) in women could be a consequence of longterm treatment. In addition to the antagonistic action of SERMs on LH secretion, we found that TX stimulates prolactin secretion from the lactotrope [16], as described for estrogen [37], while RX acts as an antagonist on both pituitary cell types [16]. In this regard, TX displayed cell selectivity in the pituitary (lactotrope vs. gonadotrope). Until recently, it was thought that all biological actions of estrogen and anti-estrogen were manifested through a single ER [38]. However, the discovery of a second ER [39–42], ERß, suggested that the differential distribution of ER· and ERß in different tissues [13, 40] or cells [43], together with different affinity of ligands for each of these receptors, could clarify SERM selectivity. This possibility could explain, at least in part, the antagonist action of TX at the uterine level and its partial agonist activity at the vaginal level and even the reported agonistic activity of TX on prolactin secretion by the lactotrope and the antagonist action on LH secretion by the gonadotrope reported in this and other studies [16]. We found that subcellular localization of ER· in gonadotropes was estrous cycledependent and strongly affected by treatment with SERMs. Control pituitaries showed cytosolic immunoreactivity signals associated with ER· in estrus, metestrus and diestrus, while in proestrus ER· immunoreactivity was accumulated exclusively within the nucleus. This may indicate dynamic nucleocytoplasmic shuttling of ER· in response to the rising estrogen levels [44]. In fact, treatment with any SERMs dramatically altered this picture, since no nuclear signal for ER· was detected on day 4 in SERM-treated rats. Interestingly, the ER· immunosignal in the cytosol of gonadotropes from TX-treated rats was significantly higher than that of the other SERMtreated groups. Therefore, TX, like the other SERMs

tested, interrupted the expected nucleocytoplasmic shuttling of ER· during proestrus. However, it apparently maintained high levels of ER· in the cytosolic compartment. Intriguingly, TX also seemed to have cellular function specificity within the gonadotrope: antagonist on LH secretion and agonist on GnRH self-priming. It has been reported that the agonist/antagonist effects of TX vary according to the parameter in question and depend upon promoter and cell context [11]. In cultured human breast cancer cells TX acts as agonist in terms of induction of PR [27], while it is an anti-estrogen in terms of their proliferation rate [45]. In addition, some breast tumors that initially recognize TX as an antagonist switch and respond to this drug as if it were an agonist [43]. The distribution of ER· and ERß in the gonadotrope cannot solely explain the dual action of tamoxifen on the same pituitary cell population (agonist on GnRH self-priming and antagonist on LH secretion) and raises the possibility that factors within the cell are important for TX selectivity [10]. It has been suggested that certain differences in the agonist/ antagonist properties of SERMs may depend upon their differential recruitment of intracellular coregulator complexes. Thus, in several mammary and endometrial cell lines TX, but not RX, is able to induce the recruitment of specific coactivators, such as steroid receptor coactivator protein 1, which appear to be related to its agonistic activity [46]. However, while the recruitment of particular coactivators may justify the differential effects of estrogenic compounds on gene expression activation in different cell types, it does not explain the different effects of a particular compound within the same cell. Intracellular messenger systems involved in LH secretion are sufficiently known: acute and chronic LH secretory responses to GnRH involve intracellular Ca2+ increase and activation of protein kinase C [47, 48]. GnRH self-priming, which is a protein synthesis-dependent phenomenon [1, 49, 50], requires cAMP-protein kinase A system interaction with PR [50–53] in the form of a ligand-independent activation of steroid receptor [54]. The anti-progesterone at the receptor RU486 [55], in the absence of the cognate ligand, suppresses in vivo and in vitro basal and GnRHstimulated LH secretion at the pituitary level in the rat in an estrous cycle stage-dependent manner [15, 22]. Basal and GnRH-stimulated LH secretion elicited through this ligand-independent activation of estrogen-induced PR [6, 7, 56] is blocked by RX or its derivatives [15] but not by TX [22] which in human breast cancer cells induces PR [27]. Furthermore, it has been reported that TX reduced estrogen induction of PR in the pituitary, while inducing




211

PR in the absence of estrogen [57]. The relationship between TX effects on GnRH self-priming and its ability to induce PR in the gonadotrope should be investigated in greater depth. In summary, the triphenylethylene TX, which was the first compound identified with noticeable SERM activity, seems to display a wider spectrum as a SERM on the rat productive system than any other compound tested. Although the mechanism is not known, the surprising finding of this study was evidence indicating that TX acted as estrogen antagonist on LH secretion and as estrogen agonist on GnRH self-priming.

Acknowledgments This study was subsidized by grants (1FD97-1065-CO3-01, -03; PM98-0167; PB97-1472-CO3-01) from the European Commission and DGESIC, Spain. The authors thank Drs. F. Marin (Eli Lilly & Co., Madrid, Spain), P. van de Velde (Roussel Uclaf, Romainville, France) and M.L. Garcıá de Paredes (Astra Zeneca, Madrid, Spain) for generously providing the SERMs, LY139481-HCl, LY117018HCl, RU58668 and ICI182789, respectively; and are grateful to the National Hormone and Pituitary Program (Baltimore, Md., USA) for the LH radioimmunoassay kit.

References

212

12 Cosman F, Lindsay R: Selective estrogen receptor modulators: Clinical spectrum. Endocr Rev 1999;20:418–434. 13 Jordan VC, Morrow M: Tamoxifen, raloxifene, and the prevention of breast cancer. Endocr Rev 1999;20:253–278. 14 Pinilla L, Gonza´lez LC, Gaytań F, Tena-Sempere M, Aguilar E: Oestrogenic effects of neonatal administration of raloxifene on hypothalamic-pituitary-gonadal axis in male and female rats. Reproduction 2001;121:915–924. 15 Aguilar R, Bellido C, Gonza´lez D, GarridoGracia JC, Sańchez-Criado JE: The in vitro inhibitory action of antiprogestin RU486 on LH and FSH secretion in the absence of progesterone in rats is estrogen-dependent. Pituitary 2000;3:153–158. 16 Gonza´lez D, Bellido C, Aguilar R, GarridoGracia JC, Hernandez G, Alonso R, SańchezCriado JE: Luteinizing hormone secretion elicited in a ligand-independent activation of progesterone receptor manner at pituitary level in the rat: Differential effects of two selective estrogen receptor modulators. Neurosci Lett 2000;289:111–114. 17 De Koning J, Van Dieten JAM, van Rees GP: Refractoriness of the pituitary gland after continuous exposure to LHRH. J Endocrinol 1978; 79:311–318. 18 Howell KB, Osborne CK, Morris C, Wakeling AE: ICI182780 (Faslodex): Development of a novel ‘pure’ antiestrogen. Cancer 2000;89: 817–825. 19 Black LJ, Goode RL: Uterine bioassay of tamoxifen, trioxifene and a new estrogen antagonist (LY117018-HCl) in rats and mice. Life Sci 1988;26:1453–1454. 20 Van de Velde P, Nique F, Bouchoux F, Bremaud J, Hameau MC, Lucas D, Moratille C, Viet S, Philibert D, Teusch G: RU 58,668, a new pure antiestrogen inducing a regression of human mammary carcinoma implanted in nude mice. J Steroid Biochem Mol Biol 1994; 48:187–196.


21 Harms PG, Ojeda SR: A rapid and simple procedure for chronic cannulation of the rat jugular vein. J Appl Physiol 1974;3:391–392. 22 Bellido C, Gonzalez D, Aguilar R, SańchezCriado JE: Antiprogestins RU486 and ZK299 suppress basal and LHRH-stimulated FSH and LH secretion at pituitary level in the rat in an oestrous cycle stage-dependent manner. J Endocrinol 1999;163:79–85. 23 Schreihofer DA, Resnick EM, Soh AY, Shupnik MA: Transcriptional regulation by a naturally occurring truncated rat estrogen receptor (ER), truncated ER product-1 (TERP-1). Mol Endocrinol 1999;13:320–329. 24 Bello AR, Hernandez G, Gonzalez M, Rweyes R, Marrero A, Sańchez-Criado JE, Tramu G, Alonso R: Immunoreactive neurotensin in gonadotrophs and thyrotrophs is regulated by sex steroid hormones in the female rat. J Neuroendocrinol 1999;11:786–794. 25 Tramu G, Pillez A, Léonardelli J: An efficient method of antibody elution for the successive or simultaneous location of two antigens by immunocytochemistry. J Histochem 1978;26: 322–324. 26 Meijs-Roelofs HMA, Uilenbroek JThJ, de Greef WJ, de Jong FH, Kramer P: Gonadotropin and steroid levels around the time of first ovulation in the rat. J Endocrinol 1975;67: 275–282. 27 Baulieu EE, Mester J: Steroid hormone receptors; in DeGroot LJ (ed): Endocrinology. Philadelphia, Saunders, 1989. 28 De Koning J, van Dieten JAM, van Rees GP: LHRH-dependent synthesis of protein necessary for LH release from rat pituitary glands in vitro. Mol Cell Endocrinol 1976;5:151–160. 29 Fink G, Chiappa SA, Aiyer MS: Priming effect of luteinizing hormone releasing factor elicited by preoptic stimulation and by intravenous infusion and multiple injections of the synthetic decapeptide. J Endocrinol 1976;69:359– 372.


1 Fink G: Gonadotropin secretion and its control; in Knobil E, Neill J (eds): The Physiology of Reproduction. New York, Raven Press, 1988. 2 Freeman ME: The ovarian cycle of the rat; in Knobil E, Neill J (eds): The Physiology of Reproduction. New York, Raven Press, 1988. 3 Brann DW, Mahesh VB: Regulation of gonadotropin secretion by steroid hormones. Front Neuroendocrinol 1991;12:165–207. 4 Aiyer MS, Fink G: The role of sex steroid hormones in modulating the responsiveness of the anterior pituitary gland to luteinizing hormone releasing factor in the female rat. J Endocrinol 1974;62:553–572. 5 Aiyer MS, Chiappa SA, Fink G: A priming effect of luteinizing hormone releasing factor on the anterior pituitary gland in the female rat. J Endocrinol 1974;62:573–588. 6 Turgeon JL, Van Patten SM, Shyamala G, Waring DW: Steroid regulation of progesterone receptor expression in cultured rat gonadotropes. Endocrinology 1999;140:2318–2325. 7 Szabo M, Kilen SM, Nho SJ, Schwartz NB: Progesterone receptor A and B messenger ribonucleic acid levels in the anterior pituitary of rats are regulated by estrogen. Biol Reprod 2000;62:95–102. 8 Rao IM, Mahesh VB: Role of progesterone in the modulation of the preovulatory surge of gonadotropins and ovulation in the pregnant mares’ serum gonadotropin-primed immature rat and the adult rat. Biol Reprod 1986;35: 1154–1161. 9 Sańchez-Criado JE, Bellido C, Galiot F, Lo´pez FJ, Gaytań F: A possible dual action of the anovulatory action of antiprogesterone RU486 in the rat. Biol Reprod 1990;42:877–886. 10 McDonnell DP: The molecular pharmacology of SERMs. Trends Endocrinol Metab 1999;10: 301–311. 11 Jordan CV: Biochemical pharmacology of antiestrogen action. Pharmacol Rev 1984;21: 247–271.


40 Kuiper GGJ, Carsson B, Grandien K, Enmark E, Häggablad J, Nilsson S, Gustafsson JA: Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors · and ß. Endocrinology 1997;138: 863–870. 41 Wilson ME, Richard H, Price J, Handa RJ: Estrogen receptor ß messenger ribonucleic acid expression in the pituitary gland. Endocrinology 1998;139:5151–5156. 42 Gustafsson J-A: Estrogen receptor ß – A new dimension in estrogen mechanism of action. J Endocrinol 1999;163:379–383. 43 Gottardis MM; Jordan VC: Development of tamoxifen-stimulated growth of MCF-7 tumors in athymic mice after long-term antiestrogen administration. Cancer Res 1988;48: 5183–5187. 44 Dauvois S, White R, Parker MG: The antiestrogen ICI 182780 disrupts estrogen receptor nucleocytoplasmic shuttling. J Cell Sci 1993; 106:1377–1388. 45 Lippman M, Bolan G, Huff R: The effects of oestrogen and antiestrogen on hormone-responsive breast cancer in long-term tissue culture. Cancer Res 1976;36:4595–4601. 46 Shang Y, Brown M: Molecular determinants for the tissue specificity of SERMs. Science 2002;295:2465–2468. 47 Stojilkovic SS, Catt KJ: Novel aspects of GnRH-induced intracellular signaling and secretion in pituitary gonadotrophs. J Neuroendocrinol 1995;7:739–757. 48 Billiard J, Koh D-S, Babcock DF, Hille B: Protein kinase C as a signal for exocytosis. Proc Natl Acad Sci USA 1997;99:12192–12197. 49 De Koning J, van Dieten JAM, van Rees GP: LH-RH-dependent synthesis of protein necessary for LH release from rat pituitary glands in vitro. Mol Cell Endocrinol 1976;5:151–160.

50 Turgeon JL, Waring DW: The timing of progesterone-induced ribonucleic acid and protein synthesis for augmentation of luteinizing hormone secretion. Endocrinology 1991;129: 3234–3239. 51 Turgeon JL, Waring DW: Activation of the progesterone receptor by the gonadotropinreleasing hormone self-priming signaling pathway. Mol Endocrinol 1994;8:860–869. 52 Waring DW, Turgeon JL: A pathway for luteinizing hormone-releasing hormone self-potentiation: Cross-talk with the progesterone receptor. Endocrinology 1992;130:3275–3282. 53 Chappell PE, Schneider JS, Kim P, Xu M, Lydon JP, O’Malley BW, Levine JE: Absence of gonadotropin surges and gonadotropin-releasing hormone self-priming in ovariectomized (OVX), estrogen (E2)-treated, progesterone receptor knockout (PRKO) mice. Endocrinology 1999;140:3653–3658. 54 Cenni B, Picard D: Ligand-independent activation of steroid receptors: New roles for old players. Trends Endocrinol Metab 1999;10: 41–46. 55 Baulieu EE: Contragestion and other clinical applications of RU486, an antiprogestagen at the receptor. Science 1989;245:1351–1357. 56 Scott REM, Wu-Peng X-S, Pfaff DW: Regulation and expression of progesterone receptor mRNA isoforms A and B in the male and female rat hypothalamus and pituitary following oestrogen treatment. J Neuroendocrinol 2002;14:175–183. 57 Roy EJ, MacLusky NJ, McEwen BS: Antiestrogen inhibits the induction of progestin receptors by estradiol in the hypothalamuspreoptic area and pituitary. Endocrinology 1979;104:1333–1336.



30 De Koning J, Tijssen AMI, van Rees GP: The involvement of ovarian factors in the maintaining the pituitary glands of female rats in a state of low responsiveness to LHRH. J Endocrinol 1987;112:265–273. 31 Sandow J: Gonadotropic and antigonadotropic actions of LH-RH analogues; in Muller EE, MacLeod RM (eds): Neuroendocrine Perspectives. Amsterdam, Elsevier Biomedical Press, 1982, vol 1. 32 Knobil E: The neuroendocrine control of the menstrual cycle. Recent Prog Horm Res 1980; 36:53–88. 33 Plouffe L Jr, Siddhanti S: The effect of selective estrogen receptor modulators on parameters of the hypothalamic-pituitary-gonadal-axis. Ann NY Acad Sci 2001;949:251–258. 34 Rutqvist LH, Johansson H, Signomklao T, Johansson U, Fornander T, Wilking N: Adjuvant tamoxifen therapy for early stage breast cancer and second primary malignancies. Stockholm Breast Cancer Study Group. J Natl Cancer Inst 1995;87:645–651. 35 Jordan VC, Murphy CS: Tamoxifen, raloxifen, and the prevention of breast cancer. Endocr Rev 1999;20:235–278. 36 White INH: The tamoxifen dilemma. Carcinogenesis 1999;20:1153–1160. 37 Shull JD, Gorski J: Estrogen regulates the transcription of the rat prolactin gene in vivo through at least two independent mechanisms. Endocrinology 1985;116:456–462. 38 Koike S, Sakai M, Muramatsu M: Molecular cloning and characterization of rat estrogen receptor cDNA. Nucleic Acid Res 1987;15: 2499–2513. 39 Kuiper GGJ, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson J-A: Cloning of a novel estrogen receptor espressed in rat prostate and ovary. Proc Natl Acad Sci USA 1996;93:5925– 5930.