Previous studies from our laboratory, demonstrating that suppression of serum FSH by RU486 requires a high estrogen (E) background, suggested that E-inducible progesterone receptors play a role in the regulation of FSH secretion. We demonstrated further that the type II antiprogestin RU486 and the type I antiprogestin ZK98299 both suppressed the elevated serum FSH and FSHbeta messenger RNA levels similarly on the evening of proestrus, but had divergent effects on the morning of estrus, when only RU486, but not ZK98299, lowered the elevated serum FSH level (secondary FSH surge). In the present work we used primary anterior pituitary cell culture to examine whether RU486 caused direct, E-dependent suppression of basal and recombinant human activin A (activin)-induced FSH secretion in the gonadotrope and to compare this direct effect, if any, with that of ZK98299. Primary cell cultures were prepared from anterior pituitaries collected from cycling female rats either on metestrous or proestrous morning and cultured in DMEM, supplemented with charcoal-stripped serum without or with 10 nM estradiol (E2) for 96 h; exposure to test agents occurred during the last 48 h of culture. FSH released into the medium and intracellular FSH content were determined by RIA. In cells from the anterior pituitary of metestrous rats cultured in E2-free medium, neither antiprogestin (10 nM) affected FSH release; in contrast, when cells were cultured in medium to which E2 had been added, both antiprogestins caused profound suppression of both basal and activin (10 ng/ml)-stimulated FSH release. In cell cultures from proestrous rats, both antiprogestins caused a slight, but significant, suppression of basal FSH release even in the absence of added E2; activin-stimulated FSH release, however, was not affected. Upon exposure of the cells from proestrous rats to E2, the antiprogestins potently suppressed both basal and activin-stimulated FSH secretion. Because the foregoing incubations were performed in culture medium devoid of progesterone (P4), the actions of the antiprogestins on FSH secretion were independent of the natural ligand. Addition of P4 (10 nM) to the cell cultures stimulated basal and activin-induced FSH release more in the presence than in the absence of E2. The FSH response to P4 was completely blocked by both antiprogestins in both the absence and presence of E2. Finally, both RU486 and ZK98299 blocked the stimulatory effect of corticosterone (1 microM) on FSH secretion. The observed effects of P4 and antiprogestins were specific for FSH secretion; LH secretion was not similarly suppressed by either antiprogestin, but was, in fact, stimulated by ZK98299 in E2-treated cells. We conclude that 1) E2-inducible progesterone receptors interact with activin-mediated signal transduction to regulate FSH secretion, and 2) unlike on the morning of estrus in vivo, RU486 and ZK98299 affect FSH secretion similarly in the gonadotrope in vitro.

Previous in vivo studies from our laboratory indicated that administration of the antiprogestin RU486 on proestrus suppresses both the preovulatory gonadotropin surges and the secondary FSH surge, suggesting a role for the progesterone receptor (PR) in the generation of these surges. The present study was designed to test the effects of another antiprogestin, ZK98299, which has been reported to block the PR through a mechanism different from that of RU486, on gonadotropin secretion in vivo. RU486 and ZK98299 (2 and 6 mg/kg) were administered s.c. at 1230 h on proestrus; uterine intraluminal fluid content, serum gonadotropins, and gonadotropin subunit messenger RNAs (mRNAs) were determined at 1830 h on proestrus and at 0900 h on estrus. At 1830 h on proestrus, both RU486 and ZK98299 at both doses caused equal suppression of the preovulatory FSH surge and FSHbeta mRNA. Both antiprogestins also equally attenuated the preovulatory LH surge at this time, with the higher doses causing greater suppression. In contrast, at 0900 h on estrus, the antiprogestins affected serum FSH differentially; only RU486 suppressed the secondary FSH surge despite the fact that both drugs prevented the release of uterine intraluminal fluid, confirming blockade of progesterone action at the level of the uterus. Neither drug had a significant effect on FSHbeta mRNA at 0900 h on estrus. ZK98299 at the higher dose caused a small, but significant, increase in serum LH. In a subsequent experiment, we compared the effects of RU486 and ZK98299 (6 mg/kg, s.c.), administered at 1230 h on proestrus, on serum FSH raised above the natural secondary FSH surge on the morning of estrus by passive immunization with an antiserum to inhibin-alpha (anti-I) at 1700 h on proestrus. Consistent with the results of the first experiment, both antiprogestins blocked the release of uterine intraluminal fluid, but only RU486 lowered serum FSH in both the normal sheep serum-treated controls and anti-I-treated rats; in contrast, ZK98299 actually increased serum FSH in the normal sheep serum-treated control animals. ZK98299 also increased FSHbeta mRNA in the control group; RU486, on the other hand, reduced FSHbeta mRNA only in the anti-I group. The results demonstrate unequivocally that whereas the effects of the two antiprogestins on serum FSH and FSHbeta mRNA are similar on proestrus, they are divergent on estrus. The data suggest that the functional state of the PR/transcriptional activation complex in the gonadotrope on the morning of estrus is different from that on the evening of proestrus.

Recent evidence utilizing RU486 has implicated progesterone (P) and glucocorticoids, in addition to a drop in serum inhibin, in the development of the secondary FSH surge on the morning of estrus. To assess the role of these steroids, we treated proestrous female rats with the antiprogestin/antiglucocorticoid RU486 (6 mg/kg sc) at 1230 h, and with dexamethasone (dex; 8.4 or 16.2 mg/kg sc), or with the steroid biosynthesis inhibitor aminoglutethimide (AG; 150 mg/kg ip) at 1030 h, alone or in combination with RU486. The effects of these treatments on uterine ballooning and intraluminal fluid content (an index of P action), ovulation, and serum levels of P, corticosterone (B), FSH, LH, and inhibin-alpha at 1830 h proestrus and 0900 h estrus were examined. In accord with previous work from our laboratory, RU 486 caused uterine intraluminal fluid retention on the morning of estrus and significantly suppressed the preovulatory surges of both FSH and LH, and the secondary surge of FSH without affecting the fall in inhibin-alpha. Treatment with dex alone raised serum FSH at both 1830 h proestrus and 0900 h estrus, coincident with suppression of serum inhibin-alpha. When administered in combination with RU486, dex partially reversed the increased uterine intraluminal fluid retention at 0900 h estrus, but did not modify the inhibitory effect of RU486 on the primary gonadotropin surges or the secondary surge of FSH. AG alone significantly suppressed serum P, B, and gonadotropins (LH to a greater extent than FSH) at 1830 h proestrus and blocked ovulation and uterine intraluminal fluid release at 0900 h estrus; it did not, however, suppress the secondary FSH surge or prevent the fall in serum inhibin-alpha. When administered 2 h before RU486, AG did not prevent the RU486-induced inhibition of the primary gonadotropin surges or the secondary FSH surge. We conclude from these results that development of the secondary FSH surge does not require P or glucocorticoid action and that RU486 suppression of the secondary FSH surge does not involve blockade of binding of these steroids to their receptors. Our data are compatible with ligand-independent activation of the P receptor, susceptible to blockade by RU486, as the mechanism underlying the enhanced secretion of FSH from the gonadotrope on the morning of estrus.

We demonstrated previously that glucocorticoids differentially affect the levels of the two pituitary gonadotropins, LH and FSH, both in vivo and in vitro. In vivo, the effect of glucocorticoids is GnRH independent, indicating a direct action on the gonadotrope, and it leads to selective up-regulation of the pituitary content of FSH and FSH beta-subunit messenger RNA (mRNA). The objective of the present study was to confirm the direct action of corticosterone (B) on FSH beta-subunit mRNA in primary anterior pituitary cell culture and to assess whether the selective B-induced rise in FSH beta mRNA is mediated through altered stability of the FSH beta transcript. Anterior pituitary glands collected from randomly cycling female rats were dissociated with trypsin. Cells were incubated at 37 C for 48 h and subsequently exposed to vehicle or B (1.7 microM) for an additional 42 h. At the end of the incubation, media were sampled for FSH and LH, cells were lysed, and total RNA was isolated for Northern blot analysis. Exposure to B for 42 h caused direct and selective upregulation of FSH release, FSH content, and FSH beta mRNA; decreased alpha-subunit mRNA; and had no significant effect on LH release, LH content, or LH beta mRNA. To evaluate the mRNA stability of the three subunits, cells were exposed to the transcription blocker actinomycin D (act D; 5 micrograms/ml) for an additional 6 h. The combined 6-h treatment with B and act D slightly, but significantly, suppressed alpha-subunit mRNA and did not change LH beta mRNA, confirming a long half-life of the two gonadotropin subunit mRNAs. In contrast, FSH beta mRNA was significantly suppressed by act D to the same level in vehicle- and B-treated cells. The posttranscriptional decay rate was examined by sampling at 0, 1, 2, 3, and 6 h during the 6-h act D treatment period. Decay curves for FSH beta mRNA were parallel in vehicle- and B-treated cells, indicating that B did not alter FSH beta mRNA stability. We conclude that the selective B-induced rise in FSH beta mRNA is mediated at the level of transcription rather than mRNA stabilization.

We used passive immunization with an antiserum to the alpha-subunit of inhibin (anti-I) or acute ovariectomy to investigate the relationship between serum inhibin levels and FSH secretion in the presence of the progesterone/glucocorticoid antagonist RU486. We demonstrated previously that 1) anti-I administered at 1700 h causes serum FSH to rise on the morning of estrus, even in the presence of a GnRH antagonist, when the two treatments are delivered on proestrus; and that 2) RU486 given on proestrus (1230 h), a time when serum estradiol levels are high, not only blocks the natural secondary FSH surge, but also suppresses the anti-I-induced rise in serum FSH on the morning of estrus. We have now extended our studies of the relationship between inhibin and RU486 to investigate treatment with RU486 and anti-I on a different day of the cycle, estrus, when serum estradiol levels are low. When both RU486 and anti-I were given on estrus (1230 and 1700 h, respectively), RU486 failed to block the anti-I-induced rise in serum FSH on the next morning of metestrus, in contrast to the blockade seen with RU486 treatment on the day of proestrus. However, pretreatment with estradiol benzoate (50 microgram) on the evening of proestrus, before the RU486 and anti-I treatment on estrus, caused RU486 to suppress the effects of anti-I on serum FSH, as it does when given on proestrus. We then repeated the study, using ovariectomy on proestrus or estrus (1700 h) to raise serum FSH, and assessed the effects of RU486 treatment at proestrus and estrus and estradiol benzoate treatment on proestrus. Our results indicate that treatment with RU486 can block the postovariectomy rise in serum FSH only in the presence of high circulating estradiol levels. We conclude that the inhibitory action of RU486 on FSH secretion after a fall in serum inhibin depends on a precedent estradiol background, probably due to induction of progesterone receptors by estradiol.

The effects of corticosterone (B) and testosterone (T) on pituitary and serum bioactive and immunoreactive gonadotropins and on gonadotropin hormone subunit messenger RNA levels were compared in the absence of GnRH. Male rats were implanted with pellets of either cholesterol, B or T. At implantation, 2 and 4 days later half of each group received GnRH antagonist and animals were killed 5 days after implantation. As expected, GnRH antagonist lowered bioactive and immunoreactive serum FSH and LH, pituitary FSH, LHβ and FSHβ mRNA. B treatment alone lowered bioactive and immunoreactive serum FSH and immunoreactive serum LH. B reversed the antagonist effect on bioactive and immunoreactive pituitary FSH and FSHβ mRNA. T alone lowered bioactive and immunoreactive serum FSH and LH levels. T reversed the antagonist effect on bioactive and immunoreactive pituitary FSH. T lowered bioactive and immunoreactive pituitary LH and LHβ mRNA and partially reversed the antagonist effect on FSHβ mRNA. The data suggest that either B or T enhance FSH synthesis by acting directly at the gonadotrope, but that B does not affect LH variables to the same extent as T. The results suggest that in stressed animals, when T levels are reduced, B can substitute for T in sustaining FSH synthesis.

Experimental objectives were to determine: 1) if the native glucocorticoid, corticosterone (B), can selectively increase pituitary FSH and FSH beta messenger RNA (mRNA) in the presence or absence of a GnRH signal; and 2) if B affects the biological activity of the gonadotropins. Metestrous female rats were implanted with cholesterol or B. Each implant group received 100 micrograms GnRH antagonist or control injections every 48 h beginning at the time of implantation, and were killed 5 days later. B significantly increased bioactive serum FSH, with or without GnRH antagonist. GnRH antagonist decreased bioactive serum FSH. Immunoreactive serum FSH was not affected by any treatment. B did not affect bioactive serum LH, but GnRH antagonist significantly suppressed bioactive serum LH. Immunoreactive serum LH was significantly lowered by either B or GnRH antagonist. Neither bioactive nor immunoreactive pituitary FSH or LH content were affected by B, GnRH antagonist, or combined treatments, and no treatment affected alpha or LH beta mRNA. B significantly increased FSH beta mRNA specifically, in the presence or absence of GnRH antagonist. These results demonstrate that corticosterone can increase biological activity of secreted FSH and increase FSH beta mRNA in the absence of a GnRH signal, suggesting a direct effect on the anterior pituitary gland.

Previous work in this laboratory has shown that when the antiprogesterone RU486 is administered at 1230 h on proestrus, the primary surges of LH and FSH are significantly attenuated, and the secondary FSH surge is abolished. This suppression of the secondary FSH surge occurs in RU486-treated rats despite a drop in serum inhibin, which is presumably due to the partial primary LH surge. The experiments described investigated the interrelationships among the primary LH surge, the rising proestrous levels of progesterone, and the falling levels of inhibin on proestrus on the selective secretion of FSH during the night of proestrus and the morning of estrus. Exogenous LH given to rats treated first with RU486 cannot restore the secondary surge of FSH, as it does in GnRH antagonist-blocked rats, and progesterone alone cannot restore the secondary FSH surge in an animal in which the surges are blocked with GnRH antagonist. Antiserum to inhibin superimposed on GnRH antagonist causes an enhancement of FSH levels beyond that of the secondary FSH surge, but antiserum to inhibin superimposed on RU486 does not induce a similar elevation of FSH secretion. The failure of antiserum to inhibin to increase FSH in the face of RU486 suggests that either a small amount of progesterone, or perhaps another hormone blocked by RU486, is needed to facilitate FSH secretion in response to the drop in inhibin or RU486 is acting through nonprogesterone-mediated effects to block pituitary FSH secretion.

To determine if LH and FSH respond to cortisol exposure the same way in females as they do in males, metestrous females were implanted with cholesterol or cortisol (F) subcutaneously, and either ovariectomized or left intact 4 days later. Tail vein injections of 1000 ng of GnRH in saline, or saline alone, were given 4.5, 23.5, or 47.5 h after the time of ovariectomy. Animals were killed 30 min after the injections at 5, 24, and 48 h after surgery. F attenuated the postovariectomy increase in serum LH at 48 h. F also suppressed GnRH-stimulated LH release 24 and 48 h after surgery in ovariectomized animals and in intact animals at 48 h. Pituitary content of LH was increased moderately by F at 5 h. These effects of F are similar to those seen in males. In contrast to LH, F increased serum FSH in intact females and suppressed levels in ovariectomized animals at 24 and 48 h, while inducing a remarkable increase in pituitary FSH content at all three times. These divergent effects of F on serum FSH (suppression in gonadectomized and stimulation in intact groups) were not seen in males, and the increase in pituitary FSH as a result of exposure to F was much more profound and reliable in females than in males. To determine if the F-induced increase in pituitary FSH was dependent on endogenous secretion of GnRH, intact metestrous females were implanted with either cholesterol or F pellets. Each implant group received sc injections of 100 micrograms GnRH antagonist or control injections every 48 h beginning at the time of steroid implantation. Animals were killed 5 days after implantation. The antagonist suppressed both serum and pituitary LH. F also suppressed serum LH levels, but had no effect on pituitary content of LH. Neither the antagonist nor F affected serum FSH. F greatly increased pituitary content of FSH in the presence or absence of GnRH antagonist. These data suggest that 1) LH responds to F treatment in a similar way in females and males; 2) pituitary FSH content is more sensitive to the enhancing effect of F in females than in males; 3) the ability of F to increase pituitary FSH in females is not dependent on GnRH.

Cortisol treatment for 6 or 12 days had no effect on serum FSH in intact males and animals castrated for 1 or 7 days, but pituitary FSH was increased by the steroid in both intact and castrate groups. In contrast, cortisol inhibited serum LH in both intact and castrated animals while only increasing pituitary levels of LH in 7 day castrates. Cortisol also increased the mRNA for FSH beta without affecting alpha or LH beta mRNAs. These data suggest that the selective increase in pituitary content of FSH may be due to a selective increase in FSH beta mRNA following exposure to cortisol.

We have measured changes in circulating immunoreactive (ir-) inhibin in male and female rats using an RIA with an antiserum raised against porcine inhibin alpha (1-26)-Gly-Tyr. The same synthetic peptide was used for standards and for the preparation of tracer. Serum ir-inhibin levels were significantly higher in intact female than in intact male rats (p less than 0.001). Immunoreactive inhibin was significantly reduced in both sexes 24 h after bilateral gonadectomy (p less than 0.0001). Unilateral ovariectomy (ULO) of female rats on metestrus caused a transient decrease in serum inhibin 8 h after surgery, but levels were not significantly different from those of sham-operated controls at later times after surgery. Increases in serum FSH and LH were observed for 8-18 h after ULO. Serum ir-inhibin levels were also measured on the early morning of estrus during the secondary FSH surge. At this time, ir-inhibin levels were low, while FSH levels were high and LH levels were low. These results show that serum ir-inhibin levels in rats are decreased at times when serum FSH levels are high.

To verify the inhibitory effect of cortisol (F) on secretion of luteinizing hormone (LH) 24 h postorchidectomy, we implanted cholesterol (C) or F subcutaneously into male rats, and 4 days later orchidectomized or sham orchidectomized them under ether anesthesia. We injected gonadotropin-releasing hormone (GnRH) or saline into these rats 24 h postorchidectomy, collected blood 30 min later, and measured LH and follicle-stimulating hormone (FSH) in serum and pituitaries. F inhibited GnRH-induced secretion of LH without affecting secretion of FSH. We then implanted C, corticosterone (B), or F into rats, performed the same surgeries, and collected pituitaries 24 h after surgery for quantitation of receptors for GnRH. Neither F nor B affected the number of receptors for GnRH or their affinity for a GnRH analogue. Suppression of LH in serum occurred without decreased pituitary content of LH. In contrast, F increased pituitary content of FSH. Implantation of progesterone in a similar experiment did not affect circulating concentrations or pituitary contents of FSH or LH. These data suggest that glucocorticoids may inhibit responsiveness to GnRH by some mechanism distal to the receptor for GnRH that affects only LH.

Our previous work has suggested that glucocorticoid pretreatment suppresses the enhanced responsiveness to GnRH seen in serum LH 12 h after castration. By contrast, serum FSH continues to show the castration-induced hypersensitivity to GnRH. Our attempts to replicate this LH suppression in static pituitary culture in vitro were not successful. This suggested to us the possibility that corticoids in vivo might be preventing castration-induced increases in pituitary GnRH receptor levels. We tested this at 24 h post-castration and, in fact, corticoids did not suppress the increase in GnRH receptors. In addition to the aforementioned effects of corticoids, we have seen that cortisol reverses the castration-induced drop in pituitary FSH content. It does this for 7 days post-castration, even though it no longer has an effect in suppressing serum LH. Thus, our accumulated data reveal that glucocorticoids have a differential effect on LH and FSH synthesis and secretion. Further studies are needed to clarify the site(s) of action of glucocorticoids in gonadotropin secretion and synthesis. Glucocorticoids may well prove to be a key in unlocking the mystery of the mechanism of differential control of regulation of LH and FSH.

Castration leads to a rapid increase in LH and FSH secretion, and exogenous gonadotropin-releasing hormone (GnRH) elicits supernormal secretion of LH and FSH after orchidectomy. Superimposition of adrenalectomy or pretreatment with cortisol prevents the acute castration-induced LH and FSH response seen within 24 h. The gonadotropin response to exogenous GnRH is not suppressed by adrenalectomy, indicating that adrenalectomy acts by preventing hypothalamic GnRH secretion and not by diminishing pituitary responsiveness to GnRH. Cortisol treatment severely reduces the response of LH to injected GnRH after castration, suggesting that cortisol suppression is at the pituitary level. Surprisingly, FSH secretory response to GnRH after orchidectomy is not suppressed by cortisol treatment, indicating a striking point of separate regulation of the two gonadotropins.

Adrenalectomy performed at the same time as, or 12 h after, castration delays the postcastration rise in LH and FSH for at least 12 h. We tested three mechanisms previously advanced as possible mediators in this suppression: 1) blocking PRL in castrate-adrenalectomized males with bromoergocryptine did not restore the normal postcastration rise in serum LH and FSH, eliminating high PRL levels as a cause of this gonadotropin suppression; 2) exogenous ACTH given at the time of orchidectomy did not inhibit either gonadotropin, eliminating high peripheral ACTH as an agent of adrenalectomy-induced suppression of LH and FSH; and 3) intestinal traction performed at the same time as orchidectomy suppressed the secretion of LH and FSH to the same degree as adrenalectomy, ruling out the lack of any adrenal factor as a means by which adrenalectomy blocked gonadotropin secretion. Our data suggest that the adrenalectomy-induced suppression of LH is due to a neurally mediated stress response probably resulting in suppression of GnRH secretion. In other treatment groups, we implanted cortisol before surgery to test the effect of ACTH suppression on LH and FSH secretion in castrate-adrenalectomized animals. A striking divergence between LH and FSH was seen in response to cortisol treatment. Cortisol suppressed LH, but not FSH, in castrate animals, and restored postcastration FSH, but not LH, secretion 12 h after combined castration-adrenalectomy. This divergence between LH and FSH secretion suggests that the effect of adrenalectomy on the two gonadotropins might result from different mechanisms. It is also possible that the differential effect of cortisol on LH and FSH secretion is not relevant to the effect of adrenalectomy on the postcastration secretion of these gonadotropins. These data add to the evidence, however, that LH and FSH are regulated by different mechanisms under many experimental conditions, including stress and elevated corticoid levels.