THE GONADOTROPINS LH and FSH are regulated primarily

Transcription

1 /04/$15.00/0 Endocrinology 145(1):71 78 Printed in U.S.A. Copyright 2004 by The Endocrine Society doi: /en Regulation of Luteinizing Hormone- and Follicle- Stimulating Hormone (FSH)- Gene Transcription by Androgens: Testosterone Directly Stimulates FSH- Transcription Independent from Its Role on Follistatin Gene Expression LAURA L. BURGER, DANIEL J. HAISENLEDER, KEVIN W. AYLOR, ALAN C. DALKIN, KATHLEEN A PRENDERGAST, AND JOHN C. MARSHALL Division of Endocrinology, Department of Internal Medicine, and the Center for Research in Reproduction, University of Virginia, Charlottesville, Virginia The gonadotropin -subunit mrnas are differentially regulated by androgens. Testosterone (T) suppresses LH- and increases FSH-. We aimed to determine whether androgens regulate LH- and FSH- transcription [as measured by changes in primary transcript (PT)] and to determine whether androgens act directly on FSH- or via the intrapituitary activin/follistatin (FS) system. In castrate GnRH antagonisttreated rats, T increased FSH- PT between 3 and 48 h. In contrast, T suppressed LH- PT. The increases in FSH- mrna and PT were associated with reduced FS mrna. Activin B mrna was modestly suppressed. The increase in FSH- PT after T was androgen specific. Both T and dihydrotestosterone (DHT) increased FSH- PT 2-fold and decreased both FS and B mrna. Estradiol suppressed FSH- Abbreviations: AR, Androgen receptor; CAST, castrated (or castration); CT, competitive template; DHT, dihydrotestosterone; E 2, estradiol; FS, follistatin; GDX, gonadectomized; GnRH-A, GnRH antagonist; OVX, ovariectomized; PACAP, pituitary adenylate cyclase-activating polypeptide; PT, primary transcript; T, testosterone. Endocrinology is published monthly by The Endocrine Society ( the foremost professional society serving the endocrine community. PT 3-fold and had no effect on FS or B mrnas. LH- PT was suppressed by DHT. To determine whether T stimulation of FSH- PT reflected a decrease in pituitary FS, we gave androgen in the presence of exogenous FS in vitro. T and DHT increased FSH- PT 2- to 3-fold. FS alone decreased FSH- PT 40% but did not diminish the increase FSH- PT in response to T. T, DHT, and FS did not affect FS mrna, B mrna, or LH- PT. In conclusion, androgens acting directly on the pituitary increase FSH- and decrease LH- transcription. The increase in FSH- PT in response to T was androgen specific and occurs in the presence of excess FS, suggesting that T stimulates FSH- transcription independently of modulation of FS. (Endocrinology 145: 71 78, 2004) THE GONADOTROPINS LH and FSH are regulated primarily by the hypothalamic decapeptide GnRH. LH and FSH act on the gonads to regulate gametogenesis and steroid production, and the latter, in turn regulate gonadotropin subunit gene expression. LH and FSH are dimers composed of a common -subunit and a unique -subunit. The subunit genes are differentially regulated by GnRH pulse frequency, fast frequencies favoring and LH- and slow frequencies favoring FSH- (1), and also by gonadal steroids. Steroids act both at the hypothalamus to alter GnRH pulsatility and/or directly on the pituitary. Differential regulation of the subunit genes by steroids was first reported in gonadectomized (GDX) rats. Both estradiol (E 2 ) and testosterone (T) suppressed the post-gdx increases in pituitary and LH- mrnas. In contrast, the post-gdx increase in FSH- mrna was abolished by E 2, but T either had no effect or enhanced FSH- responses in a dose-dependent manner (2 5). Studies in castrated (CAST) and GnRH antagonisttreated (CAST GnRH-A) rats showed that the effects of T on LH- and FSH- were in part at the level of the pituitary. In both GnRH-deficient CAST rats (6 8) and cultured rat pituitary cells (4, 9, 10), T suppressed LH- mrna and increased FSH- mrna. The mechanism(s) by which androgens differentially regulate LH- and FSH- gene expression are not completely known. Gonadotropes contain androgen receptor (AR), and T increases translocation of the AR from the cytosol to the nucleus (11). To date, consensus androgen response elements have not been reported in the promoter of either the rat LH- or FSH- genes. Curtin et al. (12) have recently reported that AR is required for androgen suppression of rat LH- promoter activity in transgenic mouse pituitary cells. Studies using rat or ovine LH- promoter-reporter constructs have revealed that the AR suppresses LH- promoter activity through protein-protein interactions with other transcription factors (12, 13). The mechanism of T action on FSH- gene expression is less well understood, and the effects of androgens on the FSH- promoter are unknown. In an earlier study we proposed that T regulated FSH- mrna stability because T increased the half-disappearance time of FSH- mrna from 20 to 50 h in CAST GnRH-A rats but did not significantly increase FSH- mrna synthesis, as measured by nuclear runoff assays (7). 71

2 72 Endocrinology, January 2004, 145(1):71 78 Burger et al. Control of FSH- by Androgens and Follistatin In addition, androgens may regulate FSH- transcription indirectly via changes in the availability of intrapituitary activin. The activins are homo-/heterodimers of the inhibin -subunits ( Aor B). The activin B subunit is produced by the gonadotropes (14, 15), and activin increases both FSH- transcription (16) and FSH- promoter activity (17, 18). Activin activity is modulated by follistatin (FS), a glycoprotein produced in the pituitary by both the gonadotropes and folliculostellate cells (19), that binds to and bioneutralizes activin (20). Testosterone decreases both pituitary FS mrna expression (21 25) and secretion (26). Therefore, T may increase FSH- gene expression indirectly by decreasing FS and increasing activin availability. The aims of this report were to investigate androgen regulation of LH- and FSH- transcription in normal rat pituitary cells, by using sensitive quantitative RT-PCR assays to measure primary transcripts (PTs). Furthermore, we aimed to determine whether T actions on FSH- PTs were effected directly or involved modulation of pituitary FS and activin B. Animals Materials and Methods Adult ( g) male Sprague Dawley rats (Harlan Sprague Dawley, Inc., Indianapolis, IN) were used for all experiments. Rats were housed in a light (lights on h) and temperature (25 C) controlled room and allowed access to food and water ad libitum. All surgeries were performed under isoflurane (2.5% isoflurane, the balance being O 2 ; Iso-thesia, Vetus Animal Heath, Burns Veterinary Supply, Inc., Westbury, NY) anesthesia. At the completion of experiments rats were euthanized by decapitation under anesthesia. Trunk blood was collected for serum hormone measurements. Pituitaries were collected and snap frozen in liquid nitrogen, and stored at 70 C until RNA was extracted. The University of Virginia Animal Care and Use Committee approved the animal experimentation described within this report. Experiment 1: the effects of GnRH-A and T on FSH- transcription and FS gene expression after castration To determine the time course of changes in FSH- transcription after CAST and the roles of GnRH and T, male rats (n 5 9/group) were CAST and treated with 1) nothing (CAST), 2) GnRH-A LRF-147, or 3) GnRH-A T. Rats were killed 0, 8, or 24 h later. The water-soluble GnRH-A LRF-147 was synthesized by Dr. Jean Rivier (The Salk Institute, La Jolla, CA) and has been shown to abolish the post-cast rise in LH (7) and partially block the increase in pituitary FS mrna expression (22, 24). In this experiment, we treated rats with 200 g LRF-147 (in 0.5 ml 0.9% saline/0.1% BSA, sc) beginning 12 h before CAST and every 12 h thereafter to maximally suppress LH. Subsequently we found that pretreatment was unnecessary and 100 g LRF-147 prevented the post- CAST increase in serum LH. Testosterone was administered at the time of CAST via SILASTIC brand (Dow Corning, Midland, MI) implants (20 mm; two implants per rat) designed to achieve serum T levels of approximately 3.5 ng/ml (27). Pituitary FSH- and LH- PTs and mrnas, FS and activin B mrnas, and serum gonadotropins were measured. Experiment 2: time course of T action on pituitary FSH- transcription and FS mrna in GnRH-deficient CAST male rats To determine the time course of FSH- transcription after T, groups of male rats were CAST (n 4 7/group) and given the LRF-147 (100 g, sc) every 12 h. Four days after CAST, rats received T implants and were killed 0, 3, 8, 24, and 48 h later. Control groups were sham (blank implants) killed after 48 h and intact males. Pituitary FSH- and LH- PTs and mrnas, FS and activin B mrnas, and serum gonadotropins and T were measured. Experiment 3: to determine whether the effects of T on FSH- transcription are androgen specific As T can be aromatized to estradiol (E 2 ), we determined whether the effects of T on FSH- transcription were androgen specific. Male rats (n 5 6/group) were CAST and given LRF-147 (100 g, sc) every 12 h. Twenty-four hours after CAST, rats received implants containing nothing (sham controls), T, DHT, or E 2. DHT implants (20 mm; two implants per rat) were designed to achieve serum DHT levels of 250 pg/ml (5). E 2 implants [two per rat; 27-mm column of 1 mg/ml E 2 in sesame oil (Sigma Chemical Co., St. Louis, MO); SILASTIC brand tubing 1.6 mm inner diameter, 3.2 mm outer diameter] produced serum E 2 concentrations 4-fold greater than intact males. Rats were killed 8 h after steroid treatment. Intact male rats were included as controls. Pituitary FSH- and LH- PTs and mrnas, FS and activin B mrnas, and serum gonadotropins were measured. Experiment 4: to determine whether T inhibition of FS mrna is required to increase FSH- transcription To determine whether T increases FSH- transcription directly or by decreasing pituitary FS concentrations, we administered androgen in the presence of exogenous FS in vitro. Pituitaries from adult male Sprague Dawley rats ( g) were cultured as previously described (28), except that both the fetal calf serum (10%) and horse serum (5%) were charcoal stripped to remove steroids. After culturing cells for 24 h, coverslips were transferred to fresh medium containing T (4 ng/ml), DHT (250 ng/ml), recombinant human FS (30 ng/ml; R&D Systems, Minneapolis, MN) or T FS (3 or 30 ng/ml recombinant human FS). Control medium contained the appropriate vehicle (0.04% ethanol). After 24 h, cells were recovered and total RNA extracted. Measurement of serum hormones, RNA preparation, subunit mrnas, and subunit primary transcripts Serum LH and FSH were measured by RIA using the standards NIDDK RP-3 for LH and NIDDK RP-2 for FSH (National Hormone and Pituitary Program). The sensitivities for the LH and FSH assays are 0.09 ng/ml and 0.8 ng/ml, respectively. The coefficients of variation are 10.9% and 16.1% (intra- and interassay) for LH and 5.3% and 12.4% for the FSH assay. T, DHT, and E 2 were measure by RIA using kits available from Diagnostic Systems Laboratories (Webster, TX). The sensitivities of the assays are 0.1 ng/ml, 4 pg/ml, and 4.7 pg/ml, respectively. The coefficients of variation are 11.5% and 18.7% (intra- and interassay) for T and 6.2% and 14.8% for E 2. The intraassay coefficient of variation for DHT was 12.4%; all samples were measured within a single assay. Total pituitary RNA was extracted using the acid guanidinium method (29). Residual genomic DNA was removed by treatment with 1 U RNase-free DNase I/ g RNA (Roche Molecular Biochemicals, Indianapolis, IN) at 37 C for 1 h. RNA preparations were confirmed to be DNA-free by PCR in the absence of a preceding RT reaction. Subunit mrna concentrations were determined by dot blot hybridization assays using 4 g pituitary RNA per dot (30, 31) and a sense-strand RNA standard curve spotted on each nitrocellulose filter (32). Subunit PTs, FS mrna, and B mrna were measured by quantitative RT-PCR assays (22, 32, 33). Briefly, regions of intron/exon for PT assays (32) or portions of the mature mrnas for FS and B (22, 33) were amplified using specific oligonucleotide primers and a size-altered competitive template (CT) RNA specifically made for each gene. A four-point standard curve was generated by adding a fixed amount of pituitary RNA (5 400 ng/reaction) to a graduated amount (2, 10, 50, and 250 fg) of CT. The pituitary and CT RNA were reverse transcribed followed by 35 cycles of PCR in the presence of [ 32 P]dCTP. The PCR products were separated by electrophoresis in 3% agarose, the bands excised, and [ 32 P]dCTP incorporation determined by scintillation counting. Analysis All data were examined by ANOVA. Significant differences (P 0.05) were determined post hoc by Duncan s multiple range test. Before analyses, all measurements were transformed to the logarithmic scale to attain equal variation among treatments.

3 Burger et al. Control of FSH- by Androgens and Follistatin Endocrinology, January 2004, 145(1): Results Experiment 1: the effects of GnRH-A and T on FSH- transcription and FS gene expression after CAST The effects of CAST, GnRH-A, and T on -subunit mrnas and PTs and FS and B mrnas are shown in Fig. 1. Serum LH concentrations are shown in Table 1. Serum LH increased 2- and 8-fold by 8 h and 24 h post-cast, respectively, and was prevented by GnRH-A. LH- PT also increased 4- and 8-fold at 8 and 24 h after CAST, respectively (Fig. 1), and GnRH-A suppressed LH- PT to 60% of intact animals (0 h) at 24 h. T further decreased LH- PT to 30% of intact (0 h) values after 24 h. LH- mrna increased 2-fold 24 h after CAST, GnRH-A abolished the increase at 24 h, and T had no additional affect. FSH- PT was not increased by 24 h of CAST, unlike FSH- mrna, which increased 2-fold at 24 h. GnRH-A suppressed both FSH- PT and mrna (30 60% of intact rats). T returned both FSH- PT and mrna to intact levels or greater at 24 h. FS mrna concentrations increased 3- and 7-fold 8 24 h after CAST and was partially suppressed by GnRH-A ( % of intact values). The addition of T returned FS FIG. 1. The effects of GnRH-A (A) and T on pituitary -subunit PTs and mrnas and FS and B mrnas in CAST male rats. Rats (n 5 10/group) were CAST or treated with GnRH-A LRF-147 (200 g sc) every 12 h with or without T implants. Rats were killed 0, 8, and 24 h later. All data are presented as percentage ( SE) of 0-h controls. Bars with different letters are significantly different (P 0.05). mrna to intact (0 h) levels at 24 h. Neither CAST, GnRH-A, nor T had any effect on activin B mrna. Experiment 2: time course of T action on pituitary FSH- transcription and FS mrna in GnRH-deficient CAST male rats The time course of T action on -subunit mrnas and PTs and FS and B mrnas, in 4-d GnRH-deficient CAST rats are TABLE 1. Mean ( SE) serum LH after CAST with or without GnRH-A with or without T Group Hours after CAST Serum LH (ng/ml) CAST only a b c CAST GnRH-A d e CAST GnRH-A T d Means ( SE) with different superscripts are significantly different (P 0.05).

4 74 Endocrinology, January 2004, 145(1):71 78 Burger et al. Control of FSH- by Androgens and Follistatin shown in Fig. 2. Serum LH and FSH are shown in Table 2. The rats killed before steroid treatment (0 h T) and the 48-h CAST GnRH-A sham steroid groups were not different on any of the parameters measured; the two groups were combined (as 0 h) for analysis. GnRH-A prevented the post- CAST rise in both serum LH and FSH (Table 2). T had little effect on serum LH but significantly increased FSH. LH- mrna concentrations did not change significantly after T. In contrast, FSH- mrna increased after T, with maximal changes (2-fold) occurring at 8 h. Gonadotropin -subunit PTs showed similar changes; T reduced LH- PT (40% of controls at 8 h) and increased FSH- PT (2.5- to 3.5-fold) between 3 and 48 h. As in experiment 1, the rise in FSH- PT was associated with a decline in FS mrna. The post-cast rise in FS mrna was not fully suppressed by GnRH-A but was reduced to 30% of controls by T. Pituitary B mrna levels were elevated in CAST GnRH-A controls (vs. intact), but were unaffected by T. Experiment 3: to determine whether the effects of T on FSH- transcription are androgen specific The effects of androgens and estrogens on gonadotropin -subunit mrnas and PTs and FS mrna and B mrnas in CAST GnRH-A-treated rats are shown in Fig. 3. Serum LH and FSH were suppressed by GnRH-A. Steroids had no FIG. 2. The time course of T action on pituitary -subunit PT and mrnas and FS and B mrnas in GnRH-A-treated CAST male rats. Castrated rats (n 4 7/group) were treated with GnRH-A LRF-147 (100 g, sc) every 12 h. Four days after CAST, rats received T implants and were killed 0, 3, 8, 24, and 48 h later. All data are presented as percentage ( SE) of 0-h T controls. Intact rat ( SE) control values are represented by the shading. Points with different letters are significantly different (P 0.05). effect on LH, but T and DHT increased FSH 1.5- to 2-fold (vs. sham; data not shown). LH- PT was suppressed by GnRH-A (vs. intact) in all CAST groups and further suppressed by DHT (vs. sham control). GnRH-A prevented the post-cast increase in LH- mrna, and steroids had no additional effect. GnRH-A reduced FSH- PT and mrna, but T and DHT returned both FSH- PT and mrna to intact levels; E 2 further reduced FSH- PT (30% of sham controls). FS and to a lesser degree B mrnas were increased in sham controls vs. intact rats. Both T and DHT suppressed FS and B mrnas, whereas E 2 had no effect. TABLE 2. Mean ( SE) serum LH, FSH, and T in intact controls and after T administration to 4-d castrate GnRH-A-treated rats Serum concentration (ng/ml) Group Hours of T LH FSH Intact controls a a CAST GnRH-A T b b c c b,c c,d b,c d c d Means ( SE) with different superscripts are significantly different (P 0.05).

5 Burger et al. Control of FSH- by Androgens and Follistatin Endocrinology, January 2004, 145(1): FIG. 3. The differential effects of androgens and estrogen on pituitary -subunit PTs and mrnas and FS and B mrnas in GnRH-A-treated CAST male rats. Castrated rats (n 5 6/group) were given GnRH-A LRF-147 (100 g sc) every 12 h. Twentyfour hours after CAST, implants containing either nothing (sham controls), T, DHT, or E 2 were inserted sc, and rats were killed 8 h later. All data are presented as percentage ( SE) of sham controls. Bars with different letters are significantly different (P 0.05). Experiment 4: to determine whether T inhibition of FS mrna is required to increase FSH- transcription The effects of androgens with or without FS on gonadotropin -subunit PTs and FS mrna and B mrnas in cultured male rat pituitary cells are shown in Fig. 4. Due to limited RNA recovery, we could not measure gonadotropin -subunit mrnas in these samples. There were no differences between the groups treated with T and FS at 3 or 30 ng/ml for any of the parameters measured; for clarity, only the T FS (30 ng/ml) group is shown in Fig. 4. Media FSH was suppressed (60% of controls) by FS (30 ng/ml) but was unaffected by T (with or without FS) or DHT (data not shown). In contrast to the in vivo experiments, neither T nor DHT suppressed LH- PT in cultured rat pituitary cells. FS alone did not affect LH- PT, but LH- PT increased slightly in the FS T-treated groups (vs. controls). Both T and DHT increased FSH- PT vs. control. FS suppressed FSH- PT to 60% of controls, but the addition of T (FS T) resulted in marked (6-fold) elevation of FSH- PT (vs. FS only). Neither FS nor B mrnas were affected by androgens or FS. Discussion These studies augment data on the differential regulation of LH- and FSH- gene expression by androgens and are the first to demonstrate that androgens act directly to suppress LH- and increase FSH- transcription in normal gonadotropes. In addition to the effects on the -subunits, androgens suppress FS mrna that may contribute to, but is not required for, the stimulatory effects on FSH- gene expression. These data, together with the prior studies examining the role of GnRH, help to develop a more complete understanding of the differential changes in -subunit gene expression after CAST. FSH- and LH- transcription after CAST sharply contrast the changes that occur after ovariectomy (OVX). We previously reported that after OVX, FSH- PT increased within 12 h, and LH- PT did not increase until h later, coincident with enhanced GnRH secretion (34). In contrast, after CAST, LH- PT increased 4- to 8-fold at 8 and 24 h followed by a 2-fold increase in LH- mrna at 24 h. Although the increase in LH- mrna immediately post CAST lagged the change in PT, by d 7, we have previously reported that both LH- PT and mrna were increased 6-fold (32). FSH- PT did not change acutely after CAST, but FSH- mrna was increased 2-fold at 24 h. The mechanism(s) for the increase in FSH- mrna at 24 h without an increase in FSH- PT is unknown and may reflect enhanced mrna stability. In OVX females the acute increases in FSH- PT and

6 76 Endocrinology, January 2004, 145(1):71 78 Burger et al. Control of FSH- by Androgens and Follistatin FIG. 4. The effects of androgens and FS with or withoutton -subunit PTs and FS and B mrnas in male rat pituitary cells. Twenty-four hours after plating, in steroid-free media, cells were placed into fresh media containing no steroid (controls), T, DHT, FS (30 ng/ml), or T FS (30 ng/ml; n 4 wells/group). Cells were lysed and RNA recovered 24 h after steroid/fs treatment. All data are presented as percentage ( SE) of control. Bars with different letters are significantly different (P 0.05). mrna largely reflect the loss of circulating inhibin. In adult males inhibin plays little, if any, role in regulating FSH- expression (35 37). Hence the lack of changes in FSH- PT within 24 h of CAST may reflect a suboptimal GnRH input, increased pituitary FS, and/or the loss of circulating T. GnRH is a primary regulator of LH- and FSH- gene expression, and treatment with a GnRH-A suppresses the post-gdx increases in both LH- and FSH- PT (32, 34). In the present study, GnRH-A suppressed FSH- transcription below intact levels, abolished the post-cast rise in LH- PT, and partially suppressed the rise in FS mrna. GnRH differentially regulates FSH- and LH- gene expression via changes in GnRH pulse frequency, acting at the level of transcription. We have recently reported that GnRH pulses every 30 min in CAST T rats resulted in a rapid and sustained increase in LH- PT (34). In contrast, FSH- PT increased only transiently (2 4 h) after 30-min GnRH pulses, and the decline in FSH- PT coincided with an increase in FS mrna. Only slow-frequency (every 240 min) GnRH pulses resulted in a robust and sustained increase in FSH- PT. Because GnRH pulses in CAST rats are approximately every 30 min (38), GnRH input may not be optimal to stimulate FSH- PT acutely after CAST. GnRH pulses every 30 min also stimulate FS mrna (22, 33), and pituitary FS mrna increases rapidly after CAST (22, 24). In our male rat in vivo studies there appears to be an inverse relationship between FSH- PT and FS mrna levels; therefore the increase in FS mrna, and presumably FS protein, after CAST may restrain the acute increase in FSH- PT. However, the inverse relationship between FS mrna and FSH- PT is not maintained in long-term CAST rats. FS mrna continues to increase after CAST (21, 22), and we have reported that by d 7 after CAST FSH- PT is 3-fold greater than in intact rats (32), which indicates that FSH- transcription increases after CAST despite elevated pituitary FS gene expression. In these studies FS mrna increased acutely after CAST, was partially suppressed by GnRH-A, and was completely suppressed by GnRH-A T. The suppression of FS mrna by T has been shown previously both in vivo (21, 22, 24) and in vitro (23, 25). The administration of T to GnRH-deficient CAST rats also had differential actions on FSH- and LH- gene expression. T rapidly increased FSH- PT but suppressed LH- PT (by 40 70%). T did not suppress LH- mrna, which may reflect both its long half-life (44 65 h) (7, 39) and that T slightly increases the stability of the mrna (7). The effects of T on FSH- transcription were androgen specific. FSH- PT increased 2-fold (vs. control) after T and DHT and decreased 70% after E 2, which is in agreement with previous reports for the effects of T (or DHT) and E 2 on FSH- mrna both in vivo (3, 6) and in vitro (9). However, the data conflict with our earlier findings that T does not act transcriptionally to increase FSH- mrna (7). Paul et al. (7) reported that T increased FSH- mrna but did not increase FSH- mrna synthesis as measured by nuclear runoff assays. We believe that the measurement of primary transcripts by quantitative RT-PCR is a more sensitive method for assessing FSH- transcription in normal gonadotropes, although it remains possible that T also stabilizes PT as well as the mature FSH- mrna. In vivo, the increases in FSH- PT with androgen were associated with a decline in FS mrna and suggested that T increased FSH- transcription indirectly via changes in pituitary FS/activin activity. However, despite the presence of exogenous FS, treatment of rat pituitary cells with androgen significantly increased FSH- transcription. These data support the concept that T does not increase FSH- PT via modulating FS/activin but rather exerts a direct action on

7 Burger et al. Control of FSH- by Androgens and Follistatin Endocrinology, January 2004, 145(1): FSH- gene transcription. The mechanism by which androgens increase FSH- transcription remains unknown, and to date the regulation of the rat FSH- promoter by androgens has not been investigated. In contrast to the limited information on the mechanism of androgen regulation of the FSH- gene, the mechanism by which androgens suppress LH- transcription is better studied. Jorgensen et al. (13) found that androgen bound to AR suppressed the activity of a bovine LH- promoter-reporter construct, transfected into the L T2 gonadotrope cell line, by protein-protein interactions with SF-1 in the proximal portion of the promoter. Curtin et al. (12) reported that DHT (or T) suppressed GnRH-stimulated rat LH- promoter activity, in L T2 cells, by protein-protein interaction between AR and Sp1 (and possibly Egr-1) at the distal portion of the GnRHresponsive region of the promoter, but did not suppress basal LH- promoter activity. We also observed that neither T nor DHT reduced LH- PT in cultured pituitary cells, which is in contrast to the marked suppression of LH- PT by androgens in vivo, and possibly reflects a prolonged absence of GnRH stimuli. Alternatively, the conditions in culture do not mimic the environment in vivo; we cultured pituitary cells in steroid-free media, and it is also possible that other factors necessary for the regulation of LH- transcription were lost. Another difference between cultured pituitary cells and studies performed in vivo was the loss of androgen action on FS. In vivo, both T and DHT consistently suppressed FS mrna 50 70% vs. CAST GnRH-A rats, yet in culture neither steroid had any affect on FS mrna. This is consistent with a recent report by Bilezikjian et al. (26) in which T did not affect FS mrna in an immortalized rat folliculostellate cell line. However, this is in contrast to other studies from this lab in which T suppressed FS mrna expression in cultured male rat pituitary cells (23, 25). The differences between our results and Vale and co-workers are likely due to differences in cell culture parameters (we examined the effects of T on cells cultured for 48 h vs. 96 h) and/or T dose [our dose (4 ng/ml) is similar to T levels in intact male rats, whereas Vale and co-workers (23, 25) used a dose that was 3-fold greater]. Additionally, there are several possible explanations for the differences we observed in vivo vs. in vitro. First, as stated above, the conditions in cell culture and in vivo are not the same, and it is possible that in our attempts to ensure that the media was steroid free we removed some factor that is important for the regulation of FS. Second, FS expression may be regulated in part by hypothalamic input, other than GnRH, that is androgen sensitive. Winters and co-workers (40, 41) have reported that the hypothalamic peptide pituitary adenylate cyclase-activating polypeptide (PACAP) regulates FSH- gene expression through FS. PACAP decreases FSH- mrna and increases FS mrna in rat pituitary cells (26, 40, 41), and T prevents the post-cast increase in hypothalamic PACAP mrna in Xenopus laevis (42). Therefore, T may regulate pituitary FS mrna in vivo, in part, via hypothalamic secretion of PACAP. In addition to the lack of regulation of FS mrna in vitro by T, exogenous FS also had no effect on FS mrna. This contrasts with earlier reports where administration of FS suppressed its own mrna (23, 43). The differences between our data and the previous studies may be related to either FS dose [Dalkin et al. (43) used a dose that was 8-fold greater] and/or timing [Bilezikjian et al. (23) found that FS maximally suppressed FS mrna after 6 h]. Despite the lack of FS mrna suppression, FS was biologically active; both FSH secretion and FSH- transcription (FS only group) were suppressed to 60% of controls. In conclusion, we have shown that androgens differentially regulate FSH- and LH- transcription, increasing FSH- and suppressing LH- transcription. The increase in FSH- PT in response to T was androgen specific and in vivo was correlated to a decrease in pituitary FS mrna. However, the fact that FSH- PT increased in vitro in the presence of excess FS indicates that T stimulation of FSH- transcription occurs independently of modulation of FS. The divergent effects of T on gonadotropin subunit gene transcription provide an additional mechanism whereby differential regulation of LH and FSH can occur in males. Acknowledgments We thank the University of Virginia, Center for Research in Reproduction Ligand Preparation and Assay Core for conducting the RIAs. Received August 13, Accepted September 15, Address all correspondence and requests for reprints to: Laura L. Burger, University of Virginia, Department of Internal Medicine, P.O. Box , Charlottesville, Virginia lburger@virginia.edu. This work was supported by NIH Grants HD11489 and HD33039 (to J.C.M.), by postdoctoral fellowship F32 HD08572 (to L.L.B.), and by the Core Laboratories of Specialized Collaborative Centers Program for Research in Reproduction Center Grant U54 HD References 1. Haisenleder DJ, Dalkin AC, Marshall JC 1994 Regulation of gonadotropin gene expression. In: Knobil E, Neill JD, eds. The physiology of reproduction. Vol 1. 2nd ed. New York: Raven Press; Gharib SD, Wierman ME, Badger TM, Chin WW 1987 Sex steroid hormone regulation of follicle-stimulating hormone subunit messenger ribonucleic acid (mrna) levels in the rat. J Clin Invest 80: Wierman ME, Gharib SD, LaRovere JM, Badger TM, Chin WW 1988 Selective failure of androgens to regulate follicle stimulating hormone messenger ribonucleic acid levels in the male rat. Mol Endocrinol 2: Gharib SD, Leung PC, Carroll RS, Chin WW 1990 Androgens positively regulate follicle-stimulating hormone -subunit mrna levels in rat pituitary cells. Mol Endocrinol 4: Iliff-Sizemore SA, Ortolano GA, Haisenleder DJ, Dalkin AC, Krueger KA, Marshall JC 1990 Testosterone differentially modulates gonadotropin subunit messenger ribonucleic acid responses to gonadotropin-releasing hormone pulse amplitude. Endocrinology 127: Wierman ME, Wang C 1990 Androgen selectively stimulates follicle-stimulating hormone- mrna levels after gonadotropin-releasing hormone antagonist administration. Biol Reprod 42: Paul SJ, Ortolano GA, Haisenleder DJ, Stewart JM, Shupnik MA, Marshall JC 1990 Gonadotropin subunit messenger RNA concentrations after blockade of gonadotropin-releasing hormone action: testosterone selectively increases follicle-stimulating hormone -subunit messenger RNA by posttranscriptional mechanisms. Mol Endocrinol 4: Dalkin AC, Paul SJ, Haisenleder DJ, Ortolano GA, Yasin M, Marshall JC 1992 Gonadal steroids effect similar regulation of gonadotrophin subunit mrna expression in both male and female rats. J Endocrinol 132: Winters SJ, Ishizaka K, Kitahara S, Troen P, Attardi B 1992 Effects of testosterone on gonadotropin subunit messenger ribonucleic acids in the presence or absence of gonadotropin-releasing hormone. Endocrinology 130: Kawakami S, Winters SJ 1999 Regulation of luteinizing hormone secretion and subunit messenger ribonucleic acid expression by gonadal steroids in perifused pituitary cells from male monkeys and rats. Endocrinology 140: Okada Y, Fujii Y, Moore Jr JP, Winters SJ 2003 Androgen receptors in gonadotrophs in pituitary cultures from adult male monkeys and rats. Endocrinology 144: Curtin D, Jenkins S, Farmer N, Anderson AC, Haisenleder DJ, Rissman E, Wilson EM, Shupnik MA 2001 Androgen suppression of GnRH-stimulated

8 78 Endocrinology, January 2004, 145(1):71 78 Burger et al. Control of FSH- by Androgens and Follistatin rat LH- gene transcription occurs through Sp1 sites in the distal GnRHresponsive promoter region. Mol Endocrinol 15: Jorgensen JS, Nilson JH 2001 AR suppresses transcription of the LH- subunit by interacting with steroidogenic factor-1. Mol Endocrinol 15: Roberts V, Meunier H, Vaughan J, Rivier J, Rivier C, Vale W, Sawchenko P 1989 Production and regulation of inhibin subunits in pituitary gonadotropes. Endocrinology 124: Roberts VJ, Peto CA, Vale W, Sawchenko PE 1992 Inhibin/activin subunits are costored with FSH and LH in secretory granules of the rat anterior pituitary gland. Neuroendocrinology 56: Weiss J, Guendner MJ, Halvorson LM, Jameson JL 1995 Transcriptional activation of the follicle-stimulating hormone -subunit gene by activin. Endocrinology 136: Huang HJ, Sebastian J, Strahl BD, Wu JC, Miller WL 2001 Transcriptional regulation of the ovine follicle-stimulating hormone- gene by activin and gonadotropin-releasing hormone (GnRH): involvement of two proximal activator protein-1 sites for GnRH stimulation. Endocrinology 142: Pernasetti F, Vasilyev VV, Rosenberg SB, Bailey JS, Huang HJ, Miller WL, Mellon PL 2001 Cell-specific transcriptional regulation of follicle-stimulating hormone- by activin and gonadotropin-releasing hormone in the L T2 pituitary gonadotrope cell model. Endocrinology 142: Kaiser UB, Lee BL, Carroll RS, Unabia G, Chin WW, Childs GV 1992 Follistatin gene expression in the pituitary: localization in gonadotropes and folliculostellate cells in diestrous rats. Endocrinology 130: Shimonaka M, Inouye S, Shimasaki S, Ling N 1991 Follistatin binds to both activin and inhibin through the common subunit. Endocrinology 128: Kaiser UB, Chin WW 1993 Regulation of follistatin messenger ribonucleic acid levels in the rat pituitary. J Clin Invest 91: Kirk SE, Dalkin AC, Yasin M, Haisenleder DJ, Marshall JC 1994 Gonadotropin-releasing hormone pulse frequency regulates expression of pituitary follistatin messenger ribonucleic acid: a mechanism for differential gonadotrope function. Endocrinology 135: Bilezikjian LM, Corrigan AZ, Blount AL, Vale WW 1996 Pituitary follistatin and inhibin subunit messenger ribonucleic acid levels are differentially regulated by local and hormonal factors. Endocrinology 137: Dalkin AC, Haisenleder DJ, Gilrain JT, Aylor K, Yasin M, Marshall JC 1998 Regulation of pituitary follistatin and inhibin/activin subunit messenger ribonucleic acids (mrnas) in male and female rats: evidence for inhibin regulation of follistatin mrna in females. Endocrinology 139: Leal AM, Blount AL, Donaldson CJ, Bilezikjian LM, Vale WW 2003 Regulation of follicle-stimulating hormone secretion by the interactions of activin-a, dexamethasone and testosterone in anterior pituitary cell cultures of male rats. Neuroendocrinology 77: Bilezikjian LM, Leal AM, Blount AL, Corrigan AZ, Turnbull AV, Vale WW 2003 Rat anterior pituitary folliculostellate cells are targets of interleukin-1 and a major source of intrapituitary follistatin. Endocrinology 144: Steiner RA, Bremner WJ, Clifton DK 1982 Regulation of luteinizing hormone pulse frequency and amplitude by testosterone in the adult male rat. Endocrinology 111: Haisenleder DJ, Yasin M, Marshall JC 1992 Enhanced effectiveness of pulsatile 3,5 -cyclic adenosine monophosphate in stimulating prolactin and -subunit gene expression. Endocrinology 131: Chomczynski P, Sacchi N 1987 Single step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162: Papavasiliou SS, Zmeili S, Herbon L, Duncan-Weldon J, Marshall JC, Landefeld TD 1986 and luteinizing hormone messenger ribonucleic acid (RNA) of male and female rats after castration: quantitation using an optimized RNA dot blot hybridization assay. Endocrinology 119: Ortolano GA, Haisenleder DJ, Dalkin AC, Iliff-Sizemore SA, Landefeld TD, Maurer RA, Marshall JC 1988 Follicle-stimulating hormone -subunit messenger ribonucleic acid concentrations during the rat estrous cycle. Endocrinology 123: Dalkin AC, Burger LL, Aylor KW, Haisenleder DJ, Workman LJ, Cho S, Marshall JC 2001 Regulation of gonadotropin subunit gene transcription by gonadotropin-releasing hormone: measurement of primary transcript ribonucleic acids by quantitative reverse transcription-polymerase chain reaction assays. Endocrinology 142: Burger LL, Dalkin AC, Aylor KW, Haisenleder DJ, Marshall JC 2002 GnRH pulse frequency modulation of gonadotropin subunit gene transcription in normal gonadotropes: assessment by primary transcript assay provides evidence for roles of GnRH and follistatin. Endocrinology 143: Burger LL, Dalkin AC, Aylor KW, Workman LJ, Haisenleder DJ, Marshall JC 2001 Regulation of gonadotropin subunit transcription after ovariectomy in the rat: measurement of subunit primary transcripts reveals differential roles of GnRH and inhibin. Endocrinology 142: Rivier C, Cajander S, Vaughan J, Hsueh AJ, Vale W 1988 Age-dependent changes in physiological action, content, and immunostaining of inhibin in male rats. Endocrinology 123: Dalkin AC, Knight CD, Shupnik MA, Haisenleder DJ, Aloi J, Kirk SE, Yasin M, Marshall JC 1993 Ovariectomy and inhibin immunoneutralization acutely increase follicle-stimulating hormone- messenger ribonucleic acid concentrations: evidence for a nontranscriptional mechanism. Endocrinology 132: Winters SJ, Pohl CR, Adedoyin A, Marshall GR 1996 Effects of continuous inhibin administration on gonadotropin secretion and subunit gene expression in immature and adult male rats. Biol Reprod 55: Urbanski HF, Pickle RL, Ramirez VD 1988 Simultaneous measurement of gonadotropin-releasing hormone, luteinizing hormone, and follicle-stimulating hormone in the orchidectomized rat. Endocrinology 123: Bouamoud N, Lerrant Y, Ribot G, Counis R 1992 Differential stability of mrnas coding for and gonadotropin subunits in cultured rat pituitary cells. Mol Cell Endocrinol 88: Winters SJ, Dalkin AC, Tsujii T 1997 Evidence that pituitary adenylate cyclase activating polypeptide suppresses follicle-stimulating hormone- messenger ribonucleic acid levels by stimulating follistatin gene transcription. Endocrinology 138: Fujii Y, Okada Y, Moore Jr JP, Dalkin AC, Winters SJ 2002 Evidence that PACAP and GnRH down-regulate follicle-stimulating hormone- mrna levels by stimulating follistatin gene expression: effects on folliculostellate cells, gonadotrophs and L T2 gonadotroph cells. Mol Cell Endocrinol 192: Hu Z, Lelievre V, Tam J, Cheng JW, Fuenzalida G, Zhou X, Waschek JA 2000 Molecular cloning of growth hormone-releasing hormone/pituitary adenylyl cyclase-activating polypeptide in the frog Xenopus laevis: brain distribution and regulation after castration. Endocrinology 141: Dalkin AC, Haisenleder DJ, Yasin M, Gilrain JT, Marshall JC 1996 Pituitary activin receptor subtypes and follistatin gene expression in female rats: differential regulation by activin and follistatin. Endocrinology 137: Endocrinology is published monthly by The Endocrine Society ( the foremost professional society serving the endocrine community.