See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/12568994
Stimulation of Gonadotropin-Releasing Hormone Surges by Estrogen. II. Role of Cyclic Adenosine 3′,5′-Monophosphate Article in Endocrinology · May 2000 DOI: 10.1210/en.141.4.1486 · Source: PubMed
3 authors, including: Jon E Levine Wisconsin National Primate Research Center 183 PUBLICATIONS 6,655 CITATIONS SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Energy and Reproduction View project
Sexual Differentiation and PCOS View project
All content following this page was uploaded by Jon E Levine on 06 June 2014.
The user has requested enhancement of the downloaded file.
0013-7227/00/$03.00/0 Endocrinology Copyright © 2000 by The Endocrine Society
Vol. 141, No. 4 Printed in U.S.A.
Stimulation of Gonadotropin-Releasing Hormone Surges by Estrogen. II. Role of Cyclic Adenosine 3ⴕ,5ⴕ-Monophosphate PATRICK E. CHAPPELL, JULIET LEE,
JON E. LEVINE
Department of Neurobiology and Physiology, Northwestern University, Evanston, Illinois 60208 ABSTRACT Release of GnRH surges in female rats is directed by a daily neural signal and occurs only after exposure of the hypothalamus to sustained, elevated estrogen (E2) levels in serum. We have proposed that preovulatory E2 couples the daily neural signal to the circuitry governing GnRH release by a two-step process, which includes stimulation of neuronal progesterone receptors (PRs) by E2 and subsequent activation of PRs by the daily neural signal. In the preceding report we documented that PR activation is obligatory for the stimulation of GnRH surges by E2. In these studies we assess the validity of a second essential feature of this model, that neural signals can activate PRs and thereby prompt the release of GnRH and LH surges. Our efforts specifically focused on the role of cAMP in mediating neural PR trans-activation leading to GnRH surges. To assess whether cAMP may function as a daily neural signal, cAMP levels were examined via a competitive binding assay in anteroventral periventricular nucleus (AVPV) homogenates obtained at 0900, 1200, 1500, 1800, and 2100 h on all days of the estrous cycle. A significant rise in cAMP concentrations was observed at 1500 h on all estrous cycle days. A similar rise at the same time was observed in AVPV tissues of ovariectomized (OVX) rats regardless of steroid treatment. No significant increase in cAMP levels was observed at any time point in homogenates of ventromedial nucleus or cerebral cortex. In a second experiment, female rats were OVX on the afternoon of diestrous day 2 and simultaneously administered 30 g estradiol benzoate or oil vehicle. On the following
VARIAN ESTROGEN (E2) permits the release of afternoon GnRH surges by coupling a daily neural signal to the neuronal circuitry controlling GnRH release (1). We have demonstrated that activation of progesterone receptors (PRs), particularly those expressed in the anteroventral periventricular nucleus (AVPV) of the hypothalamus and adjacent regions, is an obligatory component of this coupling process (2). The present experiments were conducted to determine how these E2-induced PRs in the AVPV may be activated as a part of the GnRH surge-generating mechanism. Our working model holds that a daily neural signal is conveyed from suprachiasmatic nucleus (SCN) neurons to AVPV neurons, where neurotransmitters mediating this signal bind membrane-bound receptors and activate associated intracellular signaling pathways; the intracellular signals thereafter prompt PR trans-activation, most likely in a ligandindependent manner (3, 4), and the transcriptional responses evoked by PR activation lead to the release of a GnRH surge. Received October 4, 1999. Address all correspondence and requests for reprints to: Jon E. Levine, Ph.D., Department of Neurobiology and Physiology, Northwestern University, Evanston, Illinois 60208. E-mail: [email protected]
day of presumptive proestrus, rats received intracerebroventricular infusions of the cAMP analog, 8-bromo-cAMP, or saline vehicle at 0900 h. Rats treated with 8-bromo-cAMP exhibited LH surges that were advanced by 3 h compared with those in saline-treated controls. This advance did not occur in 8-bromo-cAMP-treated rats not primed with E2, or in E2-treated rats given the antiprogestin RU486. In a third experiment, OVX, estradiol benzoate-primed rats received intracerebroventricular infusions of saline vehicle or the adenylyl cyclase inhibitor SQ22536; although saline-treated rats exhibited normal LH surges, no surges were observed in the rats receiving SQ22536. In additional SQ22536-treated animals, however, LH surge release was rescued and greatly augmented by a pharmacological dose of progesterone. These results demonstrate that 1) cAMP levels in the AVPV are significantly elevated at 1500 h on a daily basis; 2) cAMP elevations in the AVPV can prematurely evoke LH surges by a mechanism that requires PR activation; 3) inhibition of adenylyl cyclase activity in the AVPV blocks LH surges, an action that can be reversed by progesterone; and 4) cAMP generation leads to PR transactivation in the AVPV. Our observations thus provide support for the hypothesis that an increase in intracellular cAMP in the AVPV acts as a component of the daily neural signal required to initiate GnRH and subsequent LH surges, and that transmission of this signal is mediated by cAMP-induced PR trans-activation in the AVPV. (Endocrinology 141: 1486 –1492, 2000)
These experiments tested the specific hypothesis that intracellular cAMP mediates the daily neural signal that transactivates PRs in the AVPV, thereby prompting GnRH surges. Recent in vitro studies have demonstrated that cAMP can activate chicken, mouse, and human PR in the absence of progesterone (P) (5, 6) as well as potentiate agonist-mediated increases in transcriptional activity in a cell- and promoterspecific manner (7). Activation of either the A or B isoform of human PR can stimulate transcription in the presence of cAMP and in the absence of ligand (8). This second messenger has also been implicated in ligand-independent activation of PRs that appears to mediate GnRH self-priming (6) and the facilitation of sexual behavior by dopamine (DA) (9). Based on the foregoing evidence, we sought to determine whether cAMP manipulations in AVPV neurons result in alterations of LH surge release, and if any such effects are mediated by PR activation. Our model predicts that 1) cAMP levels in the AVPV are elevated just before GnRH and LH surges; 2) cAMP levels are similarly increased at the same time on all other days; 3) premature elevations of cAMP in the AVPV advance LH surges, whereas blockade of the endogenous cAMP signal prevents surges; 4) both normal and cAMP-advanced surges are blocked by PR antagonism; and
ROLE OF cAMP IN GnRH SURGES
5) P administration can rescue surge release after blockade of the cAMP signal. Experiments were thus designed to test each of these predictions, to thereby fully assess whether cAMP functions as a major component of the daily neural signal governing the release of E2-induced GnRH surges. Materials and Methods Animals All animal and surgical experimental procedures were used in accordance with protocols approved by the animal care and use committee of Northwestern University. Sprague Dawley female rats (Charles River Laboratories, Inc. Wilmington, MA), weighing 180 –300 g, were kept in a temperature-controlled environment under a 14-h light, 10-h dark (lights on, 0500 –1900 h) light cycle and fed laboratory chow ad libitum. Estrous cycles were monitored by examination of vaginal histology, and animals were used in experiments only after they had exhibited at least two 4-day estrous cycles.
Measurement of cAMP levels in brain tissues Proestrous, estrous, and metestrous rats were killed via decapitation at 0900, 1200, 1500h, 1800, and 2200 h. Additional groups of rats were ovariectomized (OVX) under methoxyfluorane anesthesia via bilateral abdominal incision. The latter rats were treated 5 days later with either estradiol benzoate (E2B; Sigma, St. Louis, MO) or oil vehicle, sc, and were killed by decapitation on the following day, at the same time points as the ovary-intact animals. Whole brains were removed and immediately fresh-frozen on dry ice. The brain tissues were blocked and mounted in a cryotome, and 40-m slices were removed until the caudal-most portion of the organum vasculosum of the lamina terminalis was exposed. At this point, small incisions were made with an ultrafine scalpel blade (Fisher Scientific, Pittsburgh, PA), defining a 1.0-mm wide region centered at the midline, and extending 2.5 mm from the ventral surface. A 1.0-mm thick tissue slice was then removed using the cryotome blade. Sections of equal dimensions were excised from the ventromedial hypothalamus (VMH) and the frontoparietal motor cortex to serve as controls. All sections were homogenized immediately in a Teflon-glass homogenizer on ice in a volume of 0.4 ml acid-ethanol solution. Fifty microliters of homogenate were removed, and protein content was determined by Bradford assay. Homogenates were centrifuged at 1000 ⫻ g for 20 min at 0 C. Supernatants were removed, placed into polypropylene tubes, and lyophilized in a Speed-Vac evaporator (SVC200H, Savant Instruments, Farmingdale, NY) for 8 h. Pellets were resuspended in 50 mm Tris-HCl assay buffer containing 4 mm EDTA and 100 m isobutylmethylxanthine (Sigma) to prevent cAMP degradation. cAMP measurements were determined via competition with 5.0 nm [8-3H]cAMP (Amersham Pharmacia Biotech, Piscataway, NJ) using a binding protein isolated from bovine adrenals. Standard curves were generated for each assay, protein-bound [8-3H]cAMP was placed in ScintiVerse (Fisher Scientific) fluid, and radioactivity was determined by a Beckman Coulter, Inc., liquid scintillation counter (Palo Alto, CA).
Surgical protocols and hormone treatments Rats were anesthetized with 10 mg/kg ketamine (Ketaset, Fort Dodge Laboratories, Fort Dodge, IA), ip, and 8 mg/kg xylazine (Gemini SA, Burns Veterinary Supply, Inc., Rockville Center, NY), im, and implanted with single barrel guide cannula (Plastics One, Roanoke, VA) stereotaxically directed to the rostral-most portion of the third ventricle (0.5 mm caudal to bregma; 8.0 mm ventral to the skull). During cannula insertion, the sagittal sinus was moved laterally using a blunted 27-gauge needle to minimize bleeding. An additional cohort of animals was implanted with cannulas directed 2 mm caudal to determine whether treatments distal to the AVPV could have similar effects. After resumption of estrous cyclicity, rats were anesthetized with methoxyfluoane (Metofane, Pittman-Moore, Inc., Washington Crossing, NJ), OVX at 0900 h on diestrous day 2 (day 1), given a sc injection of E2B (30 g), and fitted with indwelling atrial catheters (PE-50, Becton Dickinson and Co., Parsippany, NJ) inserted through the jugular vein and exteriorized at the nape of the neck. Stainless steel plugs were inserted into the free end of the catheters to occlude them until sampling on the following day.
Intracerebroventricular (icv) 8-bromo-cAMP administration On the day after E2B priming, rats received an icv infusion of 8-bromo-cAMP (Sigma, St. Louis, MO) or saline at 0900 h. The 8-Br-cAMP was dissolved in 0.9% saline to a final concentration of 10 mm, and 1.0 l of this solution was injected icv over 2 min via a 33-gauge injection cannula connected to a 10-l Hamilton syringe (Reno, NV). One cohort of 8-bromo-cAMP-treated rats received sc injections of 6 mg/kg RU486 (a gift from Roussel-UCLAF, Romainville, France) at 1000 h, and the rest received sesame oil vehicle. Blood samples (0.25 ml) were collected halfhourly from 0830 –1000 h and then hourly until 2100 h, and centrifuged at 4 C. Replacement of equal volumes of 0.9% saline followed the withdrawal of each sample. Plasma was removed and stored at ⫺20 C for LH RIA. Rats were killed the following day, brains were fresh-frozen on dry ice, and histochemical staining was performed for verification of cannula placement. Briefly, 40-m sections were sliced on a cryotome, thawmounted on glass slides, stained with 0.8% cresyl violet, dehydrated with increasing concentrations of ethanol, and coverslipped.
ICV SQ 22536 administration An adenylyl cyclase inhibitor, SQ 22536 [9-(tetrahydro-2⬘-furyl)adenine] (Calbiochem, San Diego, CA), was dissolved in 0.9% saline to a final concentration of 10 m and infused as described above at 1100 h on the day after E2B injection. One cohort of E2B-primed, SQ 22536treated animals was also given P (8 mg/kg in sesame oil; Sigma) or oil vehicle at 1000 h. Blood samples were withdrawn from atrial catheters hourly, beginning at 1200 h and ending at 2200 h. Blood samples were centrifuged, and plasma was removed and stored at ⫺20 C for later LH RIA, and cannula placement was assessed histologically as described above.
RIAs LH standard (RP-3) was generously provided by NIDDK. The sensitivity of the LH RIA was 40 pg/tube. The intraassay coefficient of variance for LH was 12.2%. LH and cAMP data are presented as the mean ⫾ sem. cAMP levels were normalized to the protein content of homogenates and presented as picomoles per mg protein.
Statistical analysis The mean cAMP content measured from AVPV homogenates at different sacrifice times was compared with levels at the 0900 h point via ANOVA, with P ⬍ 0.05 considered significant. Data obtained from treatment groups in the 8-Br-cAMP and SQ 22536 experiments were compared using a two-way ANOVA with repeated measures, followed by Bonferroni’s post-hoc tests, with P ⬍ 0.05 considered significant.
Results Measurements of tissue cAMP concentrations
In all intact and OVX rats, including those treated with E2B, increases in AVPV cAMP concentrations were observed at 1500 h; in proestrous rats, this time point corresponds to the hour before the initiation of LH surges in animals housed in our facility. cAMP concentrations in AVPV homogenates on proestrus are presented in Fig. 1A as picomoles per mg protein. ANOVA and post-hoc comparisons revealed a significant (P ⬍ 0.05) peak at the 1500 h point compared with the 0900 h point. Lowest levels of cAMP in the AVPV were observed at 1800 h. These patterns of intracellular cAMP accumulation, viz. significantly increased at 1500 h, were also evident in the data from rats killed on estrus (Fig. 1B) and metestrus (Fig. 1C) as well as in 5-day OVX rats (Fig. 1D) and 5-day OVX rats given E2B (Fig. 1E). Tissue blocks containing the AVPV were cut just caudal to the organum vasculosum of the lamina terminalis and extend 1 mm caudally. Although it is possible that these hypothalamic sections contain
ROLE OF cAMP IN GnRH SURGES
Endo • 2000 Vol 141 • No 4
FIG. 1. Temporal patterns of intracellular cAMP measured from AVPV homogenates taken from rats on proestrus (A), estrus (B), and metestrus (C) as well as from 5-day OVX rats (D) and 5-day OVX rats receiving E2B (E), at 0900, 1200, 1500, 1800, and 2200 h (n ⫽ 4 at each time point within each cycle stage and hormonal treatment group). ANOVA revealed that a significant peak (*, P ⬍ 0.05) in intracellular cAMP was observed in all three groups at the 1500 h point compared with the level at the 0900 h point.
cells from the rostral SCN, it is unlikely that the cAMP rhythm observed in this study represents oscillations in the SCN, as a previous study revealed a different pattern of cAMP accumulation in this region (10). Comparison of cAMP content in VMH (Fig. 2A) and motor cortex (Fig. 2B) among the same time points on proestrus and estrus revealed no such increase at 1500 h. An observable, yet not statistically significant, peak in VMH cAMP content was observed at 1200 h on estrus and may represent an increase in neuronal activity associated with mating behavior, although this is not clear from the present studies. cAMP levels in the motor cortex exhibited no significant differences among all time points. Effects of ICV 8-bromo-cAMP treatments on LH surges
Ovariectomized, E2B-treated rats given saline icv exhibited a significant elevation in plasma LH above baseline (P ⬍ 0.05), commencing at 1500 h and lasting until 2100 h. Similarly primed animals that received 8-bromo-cAMP icv instead of saline released LH surges of equal amplitude; however these surges commenced 3 h earlier. The plasma LH surges in rats treated with the cAMP analog also lasted until 2100 h (Fig. 3). In OVX rats that were not treated with E2B, no elevations in plasma LH were observed. In these unprimed animals, moreover, 8-bromo-cAMP infusions were entirely ineffective in altering LH secretion. In OVX, E2B-primed rats receiving saline infusions, pretreatment with RU486 blocked the initiation of LH surges (data not shown). The antiprogestin similarly blocked the
advanced LH surges in OVX, E2B-primed rats receiving 8-bromo-cAMP icv (Fig. 3). Effects of adenylyl cyclase inhibition on LH surges
Similar to the previous experiment, OVX, E2-treated rats given saline icv exhibited plasma LH surges beginning at 1600 h and remaining elevated until 2100 h. In contrast, OVX, E2-treated rats receiving infusions of the adenylyl cyclase inhibitor SQ22536 icv at the level of the AVPV exhibited no elevation in plasma LH (P ⬍ 0.05) compared with saline-treated controls (Fig. 4). Animals in which the inhibitor was infused 2 mm caudal of the AVPV, however, demonstrated no significant attenuation in LH surge levels compared with OVX, E2-primed controls infused with saline, suggesting that adenylyl cyclase inhibition only in the AVPV and rostral periventricular regions prevents GnRH surge generation. P treatment of OVX, E2-primed rats given SQ22536 icv, however, effectively restored LH surges. In these animals, LH surges were observed, which were significantly (P ⬍ 0.05) advanced and augmented compared with LH surges in E2-primed rats treated with saline icv (Fig. 4). Animals primed with E2B and P in the absence of SQ22536 treatment exhibited similar LH surge profiles as those given the inhibitor (data not shown), suggesting that P administration can initiate LH surge generation on proestrus in the presence or absence of adenylyl cyclase activity in the AVPV.
ROLE OF cAMP IN GnRH SURGES
FIG. 3. Effect of icv infusion of 8-bromo-cAMP (at 0900 h; as indicated by arrow) on plasma LH release in OVX, E2-primed rats. OVX, E2primed animals given either saline (open circles; n ⫽ 6) or 8-bromocAMP (solid circles; n ⫽ 6) exhibited LH surges compared with OVX animals not treated with E2 in both icv treatment groups (8-Br-cAMP, open triangles, n ⫽ 5; E2 alone, data not shown, n ⫽ 5) Elevations in plasma LH from 1200 –1400 h, however, were significantly higher (*, P ⬍ 0.05; **, P ⬍ 0.01) in 8-bromo-cAMP-treated rats than in OVX, E2-primed rats receiving saline icv. OVX, E2-primed rats treated with icv 8-bromo-cAMP as well as sc RU486 (n ⫽ 5) exhibited no significant increase in plasma LH release compared with unprimed controls.
FIG. 2. Temporal patterns of intracellular cAMP measured from VMH (A) and motor cortex (B) homogenates in proestrous and estrous rats at 0900, 1200, 1500, 1800, and 2200 h (n ⫽ 4 at each time point). No significant elevations above 0900 h levels were observed in VMH or motor cortical homogenates.
We have proposed a model that attempts to explain how the two major determinants of the GnRH surge, a daily neural signal and the positive feedback actions of E2, are integrated in hypothalamic neurons to produce surges that are appropriately timed and physiologically proportioned. The model holds that E2 couples the daily neural signal to the neuronal circuitry governing GnRH release by a two-step process: 1) E2 induces expression of neuronal transcription factors, specifically PRs in AVPV neurons; and 2) these factors are then activated at an appropriate time to regulate the transcriptional events that lead to GnRH surges. In a previous report we provided evidence for the validity of one important feature of this model: that E2-induced PRs are activated as an obligatory event in the release of GnRH surges. In these studies, we tested a second critical property of the model, viz. that PRs can be activated by neural signals, and that the result of this activation is release of GnRH and LH surges. These experiments focused on the specific hypothesis that cAMP is one important mediator of PR acti-
vation leading to GnRH surges. Our findings reveal that cAMP elevations, specifically those occurring in the AVPV and adjacent regions, fulfill several important criteria as neural signals for activation of PRs and consequent release of GnRH surges. One major prediction of our model is that cAMP levels in the AVPV are elevated on a daily basis, just before the initiation of the GnRH surge. Our measurements of GnRH release profiles in E2-treated rats previously showed that the initiation of GnRH surges occurs between 1500 –1600 h. Thus, the finding that cAMP levels are elevated at 1500 h in the AVPV, but not in control regions, directly supports the idea that this rise in cAMP represents a timed neural signal leading to the surge. That a similar elevation was seen in rats on all days of the cycle and in OVX rats regardless of E2 treatments additionally indicates that this signal is a daily one, probably stimulated by afferent impulses derived from the 24-h biological clock. Further studies are required, however, to determine in what cell type(s) this intracellular accumulation of cAMP occurs, to more definitively implicate this increase as playing a causative role in GnRH surge release. Our integrative model for the surge holds that neural signals for the surge stimulate GnRH surges via neural activation of PRs. Accordingly, premature delivery of a signal for the surge would be expected to advance the surge, whereas blockade of the endogenous neural signal should prevent its release. If cAMP in the AVPV comprises this daily neural signal for the surge, then by this reasoning it would be expected that cAMP elevations induced before the onset of an endogenous cAMP rise would temporally advance
ROLE OF cAMP IN GnRH SURGES
FIG. 4. Effect of icv infusion of the adenylyl cyclase inhibitor SQ22536 on plasma LH release measured from OVX, E2-primed rats. Animals infused with SQ22536 in the AVPV (open squares; n ⫽ 4) exhibited no increase in plasma LH throughout the sampling period (*, P ⬍ 0.05 compared with saline-treated group), in contrast to LH surges observed in E2-primed animals receiving saline icv in the AVPV (solid circles; n ⫽ 4). OVX, E2-primed rats infused with SQ22536 2 mm caudal to the AVPV (open triangles; n ⫽ 4) exhibited no significant attenuation in plasma LH compared with salinetreated controls. E2-primed rats infused with SQ22536 and receiving sc P (open circles; n ⫽ 4) displayed robust increases in plasma LH, in contrast to similarly treated animals given sc oil. In addition, P-treated rats exhibited elevated plasma LH (P ⬍ 0.05 compared with icv saline group) beginning 1 h earlier than E2-primed, oil-treated animals given saline icv.
GnRH surges, whereas blockade of cAMP production would prevent GnRH surges. The outcomes of our experiments fulfill both of these expectations; icv infusion of the cellpermeant cAMP analog, 8-bromo-cAMP, was found to advance the initiation of LH surges, whereas infusion of an adenylyl cyclase inhibitor completely blocked surge release. These observations additionally support the idea that cAMP elevations in the AVPV function as a manifestation of the daily neural signal governing the release of GnRH surges. It should be noted, however, that we could not directly measure GnRH release in experiments in which icv infusions were administered; instead, we made the assumption that the temporal characteristics of the LH surge faithfully represented the timing of the underlying GnRH surge in these animals. In the previous study, a close temporal association was demonstrated between the rising phases of GnRH and LH surges, and thus the presumption of such a relationship in these experiments would appear to be a reasonable one. The specific targeting of the icv pharmacological treatments to the AVPV and adjacent regions was also an important consideration in these experiments, as it was possible that significant concentrations of the administered drugs could have been transported to more caudal hypothalamic regions
Endo • 2000 Vol 141 • No 4
or to the pituitary gland via the portal vasculature. Infusions of SQ 22536 into more caudal ventricular positions, however, were ineffective in preventing the release of estrogen-primed LH surges, making it unlikely that any observed effects of pharmacological treatments resulted from actions at more caudal sites. An additional possibility is that 8-bromo-cAMP or SQ 22536 may have diffused more laterally and altered the activity of the cAMP/protein kinase A pathway in GnRH cell bodies themselves. This possibility seems unlikely considering that the response to 8-bromo-cAMP was seen only in E2-primed animals despite the fact that GnRH neurons in rats express few, if any, E2 receptors in female rats (11). Although a recent study demonstrates that isolated GnRH neurons express low levels of estrogen receptor-␣ (ER␣) and/or ER␤ primary transcript (12), it is as yet unclear whether this results in the generation of functional proteins. Similarly, the effects of adenylyl cyclase inhibition with SQ 22536 were probably not mediated by inhibition of adenylyl cyclase in GnRH neurons, as additional treatment with P restored LH surges in these animals; again, the probable absence of PRs in GnRH neurons makes this scenario unlikely. If cAMP represents the manifestation of a daily neural signal for the surge, then are its actions mediated by PR trans-activation, as proposed in our experimental model? Pretreatment of rats with the antiprogestin RU486 resulted in complete blockade of E2-induced LH surges regardless of whether they were temporally advanced by 8-bromo-cAMP. This action of RU486 was very likely a result of central PR antagonism, as similar treatments using this antiprogestin have been shown to decrease GnRH surge secretion on proestrus in rats (13) and to be ineffective in altering pituitary responses to exogenous GnRH in both rats and primates (13, 14). P administration, moreover, effectively restored LH surges in rats treated with the adenylyl cyclase inhibitor SQ 22536 icv. Thus, PR blockade prevents surge release induced by cAMP activation, whereas the effects of cAMP blockade can be reversed by PR activation. These results confirm the obligatory involvement of PRs in the E2-induced GnRH surge, and more importantly, they indicate that PR activation is a necessary signaling event that occurs downstream of cAMP accumulation. These findings are again consistent with the idea that neural signals for the GnRH and LH surges are mediated by cAMP generation, and that these signals trigger GnRH surges by evoking the trans-activation of PRs in the AVPV. It cannot be determined from these studies whether cAMP may activate PR in a ligand-dependent or a ligand-independent manner. It is possible that liganded PRs may be involved in this signaling event, as some P is produced (15, 16), along with cytochrome P450 (17, 18) and 3␤-hydroxysteroid dehydrogenase (19), in the brain. There is no evidence available, however, to support the idea that signal changes in P occur within neurons in response to neural signals, or that the exceedingly low levels of brain P (15) can activate PRs in a physiological manner. It is also not clear that any residual plasma P levels in OVX rats would provide sufficient concentrations of ligand for binding PR in these animals (20). By contrast, many studies, both in vitro and in vivo, have demonstrated the ability of cAMP to activate PR in the absence of P. cAMP has been shown to trans-activate PR in stably transfected CV-1 and COS-1 (7) cells as well as in
ROLE OF cAMP IN GnRH SURGES
enriched gonadotropes from primary pituitary cell culture (6). In vivo studies have demonstrated that DA agonists require unliganded PR to influence lordosis behavior in mice (9). As DA putatively acts only via its membrane-bound G protein-linked receptor, it is logical to assume that any trans-activational effects on PR by DA would be occurring via an increase in the neurotransmitter’s second messenger pathway activity. Other evidence suggests that phosphorylation of dopamine- and cAMPregulated phosphoprotein (DARPP-32) increases due to mating stimulation, and that this protein may play a role in transducing DA’s effects on PR activation in the VMH and other hypothalamic regions (21). Exactly how increased cAMP may activate unliganded PR remains unclear, although evidence indicates that this trans-activation requires activation of protein kinase A and subsequent phosphorylation of the receptor (22), which then would bind to DNA and increase transcriptional activation of a gene(s) as yet unknown. The availability and activation state of a multitude of transcriptional coactivators and coregulators could also provide a mechanism by which this cross-talk could be mediated, but as many of the characteristics and actions of these factors still remain undefined, further research will be required to determine their roles in this process. Previous studies have shown that an intact AVPV is required for generation of LH surges in intact and steroidprimed rodents (23–25), and ER activation in this region is critical for surge generation (26); ER activation also leads to PR messenger RNA and PR protein accumulation in the AVPV (27). Our studies have additionally provided evidence that cAMP-induced trans-activation of these PRs in the AVPV provides a neural signal for the surge. Which neurotransmitters may convey the daily neural signals to the AVPV, culminating in cAMP production, PR trans-activation, and release of GnRH surges? Candidate neurotransmitters include vasoactive intestinal peptide (28), vasopressin (29), and ␥-aminobutyric acid (30), as neurons expressing these transmitters have been identified among those located in the SCN and send projections to the AVPV (25, 31). Furthermore, these neurotransmitters have been found to be produced in the expected rhythmic fashion, and some (vasopressin and vasoactive intestinal peptide) are capable of activating cognate receptors that are positively coupled to adenylyl cyclase (32–34). It also remains to be determined how PR trans-activation may lead to GnRH surge generation. The 2- to 3-h latency between 8-bromo-cAMP administration and the stimulation of LH surges is consistent with the idea that the actions of PR in AVPV neurons are mediated by regulation of gene transcription. Much information has been obtained regarding the neurotransmitters produced by subpopulations of AVPV neurons. Earlier studies have shown that ER-immunoreactive cells in AVPV receive extensive input from SCN and also send afferents to GnRH perikarya (35, 36), and the neurotransmitters produced by these cells include ␥-aminobutyric acid, dynorphin, enkephalin, DA, and substance P, among others (25). Given these observations, it is possible that transactivated PRs in these neurons regulate the expression of genes encoding either these neurotransmitters or the enzymes involved in their synthesis. Alternatively, activated PRs may also regulate the expression of a variety of proteins that regulate cell excitability or secretion. We have recently
demonstrated, for example, that PR activation mediates increased hypothalamic neuropeptide Y1 receptor gene expression (37), which, in turn, results in heightened responsiveness of the GnRH secretory system to neuropeptide Y stimulation during the initiation of GnRH and LH surges. Convergence of two signaling pathways at steroid receptors has been proposed to be involved in other reproductive pathways, most notably the GnRH self-priming response by pituitary gonadotropes (6) and induction of sexual behavior by neurotransmitters in the VMH (38). Our results suggest that a similar mechanism may be required for timing and initiation of GnRH surges. Although these experiments focused primarily on E2-induced PRs and their activation by cAMP, there is evidence that E2 may exert additional actions mediated by E2’s ability to induce expression or activation of other transcriptional regulators. Both E2 and P treatment, for example, have been shown to rapidly phosphorylate cAMP response element-binding protein (CREB) in cells of various neuronal phenotypes in the AVPV (39), suggesting that an increase in phospho-CREB by steroids could play a part in regulating gene expression in this region. The relevance of rapid E2-induced alterations in phospho-CREB, however, has yet to be analyzed in the context of the initiation and timing of GnRH and LH surges; it is possible that E2’s effects on CREB are one of several amplification events that occur downstream from cAMP-induced PR activation, including E2 priming of pituitary gonadotropes (40). These studies provide evidence for the validity of a basic tenet of an integrative model for GnRH surges; a daily neural signal, comprised of a transient increase in cAMP in AVPV neurons, can evoke activation of E2-induced PRs and thereby stimulate the release of GnRH surges. It remains unknown whether this cross-talk mechanism figures as importantly in the control of GnRH and LH surges in other species, particularly those animals in which these events are not under the strict control of circadian timing signals. The prevalence of this type of integrative mechanism in the control of other neuroendocrine systems also remains to be determined. Cross-talk between membrane-bound and intracellular receptor-mediated signaling pathways has been established for a variety of cellular signaling pathways, such as growth factor stimulation of ER-dependent breast cancers (41). Future studies are likely to reveal that analogous cellular mechanisms mediate the permissive effects of a variety of other hormones on neurophysiological and behavioral processes. Acknowledgment The authors thank Venita DeAlmeida for her technical assistance and advice regarding the cAMP competitive binding assay.
References 1. Everett JW, Sawyer CH 1950 A 24-hour periodicity in the “LH-release apparatus” of female rats, disclosed by barbiturate sedation. Endocrinology 47:198 –218 2. Chappell PE, Levine JE 2000 Stimulation of gonadotropin-releasing hormone surges by estrogen. I. Role of hypothalamic progesterone receptors. Endocrinology 141:1477–1485 3. Bicknell RJ 1998 Sex-steroid actions on neurotransmission. Curr Opin Neurol 11:667– 671 4. Weigel NL, Zhang Y 1998 Ligand-independent activation of steroid hormone receptors. J Mol Med 76:469 – 479 5. Power RF, Mani SK, Codina J, Conneely OM, O’Malley BW 1991 Dopami-
6. 7. 8.
14. 15. 16. 17. 18. 19.
ROLE OF cAMP IN GnRH SURGES
nergic and ligand-independent activation of steroid hormone receptors. Science 254:1636 –1639 Turgeon JL, Waring DW 1994 Activation of the progesterone receptor by the gonadotropin-releasing hormone self-priming signaling pathway. Mol Endocrinol 8:860 – 869 Kazmi SM, Visconti V, Plante RK, Ishaque A, Lau C 1993 Differential regulation of human progesterone receptor A and B form-mediated trans-activation by phosphorylation. Endocrinology 133:1230 –1238 Kahmann S, Vaben L, Klein-Hitpass L 1998 Synergistic enhancement of PRB-mediated RU486 and R5020 agonist activities through cyclic adenosine 3⬘,5⬘-monophosphate represents a delayed primary response. Mol Endocrinol 12:278 –289 Mani SK, Allen JM, Lydon JP, Mulac-Jericevic B, Blaustein JD, DeMayo FJ, Conneely O, O’Malley BW 1996 Dopamine requires the unoccupied progesterone receptor to induce sexual behavior in mice [published erratum appears in Mol Endocrinol 1997 Apr;11(4):423]. Mol Endocrinol 10:1728 –1737 Murakami N, Takahashi K 1983 Circadian rhythm of adenosine-3⬘,5⬘-monophosphate content in suprachiasmatic nucleus (SCN) and ventromedial hypothalamus (VMH) in the rat. Brain Res 276:297–304 Herbison AE, Theodosis DT 1992 Localization of oestrogen receptors in preoptic neurons containing neurotensin but not tyrosine hydroxylase, cholecystokinin or luteinizing hormone-releasing hormone in the male and female rat. Neuroscience 50:283–298 Skynner MJ, Sim JA, Herbison AE 1999 Detection of estrogen receptor ␣ and ␤ messenger ribonucleic acids in adult gonadotropin-releasing hormone neurons. Endocrinology 140:5195–5201 Sanchez-Criado JE, Hernandez G, Bellido C, Gonzalez D, Tebar M, DiazCruz MA, Alonso R 1994 Preovulatory LHRH, LH and FSH secretion in cyclic rats treated with RU486: effects of exogenous LHRH and LHRH antagonist on LH and FSH secretion at early oestrus. J Endocrinol 141:7–14 Heikinheimo O, Gordon K, Lahteenmaki P, Williams RF, Hodgen GD 1995 Antiovulatory actions of RU 486:the pituitary is not the primary site of action in vivo. J Clin Endocrinol Metab 80:1859 –1868 Baulieu EE, Schumacher M, Koenig H, Jung-Testas I, Akwa Y 1996 Progesterone as a neurosteroid: actions within the nervous system. Cell Mol Neurobiol 16:143–154 Corpechot C, Collins BE, Carey MP, Tsouros A, Robel P, Fry JP 1997 Brain neurosteroids during the mouse oestrous cycle. Brain Res 766:276 –280 Le Goascogne C, Robel P, Gouezou M, Sananes N, Baulieu EE, Waterman M 1987 Neurosteroids: cytochrome P-450scc in rat brain. Science 237:1212–1215 Warner M, Gustafsson JA 1995 Cytochrome P450 in the brain: neuroendocrine functions. Front Neuroendocrinol 16:224 –236 Guennoun R, Fiddes RJ, Gouezou M, Lombes M, Baulieu EE 1995 A key enzyme in the biosynthesis of neurosteroids, 3␤-hydroxysteroid dehydrogenase/⌬5-⌬4-isomerase (3␤-HSD), is expressed in rat brain. Brain Res Mol Brain Res 30:287–300 Park OK, Ramirez VD 1987 Circulating blood progesterone is pulsatile throughout the rat oestrous cycle. Acta Endocrinol (Copenh) 116:121–128 Meredith JM, Moffatt CA, Auger AP, Snyder GL, Greengard P, Blaustein JD 1998 Mating-related stimulation induces phosphorylation of dopamine and cyclic AMP-regulated phosphoprotein-32 in progestin receptor-containing areas in the female rat brain. J Neurosci 18:10189 –10195 Denner LA, Schrader WT, O’Malley BW, Weigel NL 1990 Hormonal regulation of identification of chicken progesterone receptor phosphorylation sites. J Biol Chem 265:16548 –16555 Weigand SJ, Teresawa E 1982 Discrete lesions reveal functional heterogeneity of suprachiasmatic structures in the female rat. Neuroendocrinology 34:395– 404
View publication stats
Endo • 2000 Vol 141 • No 4
24. Nomura K, Takahashi S, Kawashima S 1988 Maintenance of estrous cycle in female rats with anterior or posterior deafferentation of the suprachiasmatic nucleus. Neuroendocrinology 47:444 – 452 25. Herbison AE 1998 Multimodal influence of estrogen upon gonadotropinreleasing hormone neurons. Endocr Rev 19:302–330 26. Petersen SL, Barraclough CA 1989 Suppression of spontaneous LH surges in estrogen-treated ovariectomized rats by microimplants of antiestrogens into the preoptic brain. Brain Res 484:279 –289 27. Simerly RB, Carr AM, Zee MC, Lorang D 1996 Ovarian steroid regulation of estrogen and progesterone receptor messenger ribonucleic acid in the anteroventral periventricular nucleus of the rat. J Neuroendocrinol 8:45–56 28. Van der Beek EM, Horvath TL, Wiegant VM, Van den Hurk R, Buijs RM 1997 Evidence for a direct neuronal pathway from the suprachiasmatic nucleus to the gonadotropin-releasing hormone system: combined tracing and light and electron microscopic immunocytochemical studies. J Comp Neurol 384:569 –579 29. Palm IF, van der Beek EM, Weigant VM, Buijs RM, Kalsbeek A 1999 Vasopressin induces a luteinizing hormone surge in ovariectomized, estradioltreated rats with lesions of the suprachiasmatic nucleus. Neuroscience 93:654 – 666 30. Herbison AE, Chapman C, Dyer RG 1991 Role of medial preoptic GABA neurones in regulating luteinizing hormone secretion in the ovariectomized rat. Exp Brain Res 87:345–352 31. Gu GB, Simerly RB 1997 Projections of the sexually dimorphic anteroventral periventricular nucleus in the female rat. J Comp Neurol 384:142–164 32. Rea MA 1990 VIP-stimulated cyclic AMP accumulation in the suprachiasmatic hypothalamus. Brain Res Bull 25:843– 847 33. Meeker RB, Michels KM, Hayward JN 1990 Vasopressin and oxytocin regulation of cyclic AMP accumulation in rat hypothalamo-neurohypophysial explants in vitro. Neurosci Lett 114:225–230 34. Vanecek J, Watanabe K 1998 Melatonin inhibits the increase of cyclic AMP in rat suprachiasmatic neurons induced by vasoactive intestinal peptide. Neurosci Lett 252:21–24 35. Langub Jr MC, Maley BE, Watson Jr RE 1994 Estrous cycle-associated axosomatic synaptic plasticity upon estrogen receptive neurons in the rat preoptic area. Brain Res 641:303–310 36. Watson Jr RE, Langub Jr MC, Engle MG, Maley BE 1995 Estrogen-receptive neurons in the anteroventral periventricular nucleus are synaptic targets of the suprachiasmatic nucleus and peri-suprachiasmatic region. Brain Res 689:254 –264 37. Xu M, Urban J, Levine JE, Hypothalamic NPY Y1 receptor mRNA levels vary in a cycle stage and steroid hormone-dependent manner. 28th Annual Meeting of the Society for Neuroscience, Los Angeles, CA, 1998, p 365 (Abstract 144.9) 38. Mani SK, Allen JM, Clark JH, Blaustein JD, O’Malley BW 1994 Convergent pathways for steroid hormone- and neurotransmitter-induced rat sexual behavior [published erratum appears in Science 1995 Jun 30;268(5219):1833]. Science 265:1246 –1249 39. Gu G, Rojo AA, Zee MC, Yu J, Simerly RB 1996 Hormonal regulation of CREB phosphorylation in the anteroventral periventricular nucleus. J Neurosci 16:3035–3044 40. Levine JE 1997 New concepts of the neuroendocrine regulation of gonadotropin surges in rats. Biol Reprod 56:293–302 41. Ignar-Trowbridge DM, Pimentel M, Parker MG, McLachlan JA, Korach KS 1996 Peptide growth factor cross-talk with the estrogen receptor requires the A/B domain and occurs independently of protein kinase C or estradiol. Endocrinology 137:1735–1744