Barbara Attardi

Glucocorticoid Repression of the Mouse Gonadotropin-releasing Hormone Gene Is Mediated by Promoter Elements That Are Recognized by Heteromeric Complexes Containing Glucocorticoid Receptor

We have identified two regions of the mouse gonadotropin-releasing hormone (GnRH) promoter, one between −237 and −201 (distal element) and the other between −184 and −150 (proximal element), which are required for glucocorticoid repression in transiently transfected GT1-7 cells. These sequences show no similarity to known positive or negative glucocorticoid response elements (nGREs) and do not function when placed upstream of heterologous viral promoters. The glucocorticoid receptor (GR) does not bind directly to the distal or proximal promoter elements but may participate in glucocorticoid repression of GnRH gene transcription by virtue of its association within multiprotein complexes at these nGREs. Electrophoretic mobility shift assays with GT1-7 nuclear extract demonstrate the presence of GR-containing protein complexes on GnRH nGREs. One protein that co-occupies the distal nGRE in vitro along with GR is the POU domain transcription factor Oct-1. Thus, the tethering of GR to the GnRH distal nGRE, by virtue of a direct or indirect association with DNA-bound Oct-1, could play a role in hormone-dependent transcriptional repression of the GnRH gene. In contrast, Oct-1 does not appear to be a component of the GR-containing protein complex that is bound to the proximal nGRE.

Progesterone blockade of a luteinizing hormone surge blocks luteinizing hormone-releasing hormone Fos activation and activation of its preoptic area afferents

Progesterone is capable of facilitating or blocking the luteinizing hormone (LH) surge, depending on the timing of its administration. However, the precise targets of progesterones actions are unknown. Since recent studies described the presence of a periventricular preoptic area (pePOA) neuron population afferent to LH-releasing hormone (LHRH) neurons that is co-activated to express c-Fos with LHRH neurons at the time of the LH surge, the present study was designed to determine if the pePOA neurons contain progesterone receptors (PRs) and whether progesterone inhibition is manifested by a failure of LHRH and pePOA neurons to become activated at the time of an LH surge. For progesterone facilitation, a group of immature rats each received a silastic capsule (1.57 mm i.d., 3.18 mm o.d., 1.5 cm long) containing estradiol-17beta (E2) in peanut oil (150 microg/ml) at 09.00 h on postnatal day 28 followed 24 h later by a progesterone implant (crystalline, 1.57 mm i.d., 3.18 mm o.d., 1.5 cm long). For progesterone inhibition, a second group of rats received the estrogen capsule and a progesterone capsule (3.35 mm i.d., 4.65 mm o.d., 3.0 cm long) together at 09.00 h on day 28, and 24 h later received only a blank capsule. On the afternoon of postnatal day 29, all animals were anesthetized and perfused for localization of c-Fos and LHRH, PRs alone, or c-Fos and PRs. The present studies determined that following a progesterone-inhibition paradigm, along with blockade of the LH surge, both activation of LHRH and pePOA neurons was low or absent. Staining of PRs in progesterone-facilitated and progesterone-inhibited rats indicated that the pePOA neurons contained PRs in similar patterns. Double labeling of c-Fos and PRs in progesterone-facilitated rats indicated that nearly all the c-Fos-positive neurons of the pePOA (80 +/- 4.2%) co-expressed PRs; in progesterone-inhibited rats, only 32 +/- 12% of few c-Fos-positive neurons also contained PRs. In no instance were LHRH neurons found to contain PRs. Taken together, these data suggest that both progesterone facilitation and inhibition likely involve direct actions of progesterone on the pePOA neurons, and are consistent with a role for the pePOA neurons in transducing steroid effects on LHRH neurons.

Glucocorticoid repression of gonadotropin-releasing hormone gene expression and secretion in morphologically distinct subpopulations of GT1-7 cells

Two morphologically distinct subpopulations of GT1-7 cells have been characterized and examined for their responsiveness to glucocorticoids. Type I cells have a neuronal phenotype, extending many lengthy processes, and express neuronal, but not glial, markers. Type II cells show weaker or negative immunostaining for neuronal markers and exhibit fewer processes. The effect of glucocorticoids on gonadotropin-releasing hormone (GnRH) secretion and gene expression was compared in type I and type II GT1-7 cells. For secretion studies, cells were attached to Cytodex beads and perifused with control medium or medium containing dexamethasone (dex). The high level of GnRH secreted by type I cells was slightly enhanced in the presence of dex, whereas dex rapidly and profoundly decreased the already low level of GnRH secreted by type II cells. Immunocytochemistry for GnRH showed dark reaction product in the cell bodies and processes of type I cells and little or no immunoreactivity in type II cells. Both the endogenous mouse GnRH mRNA and the transcriptional activity of a mouse GnRH promoter luciferase reporter gene plasmid were suppressed to a greater extent in type II cells than in type I. In electrophoretic mobility shift assays, there was no difference between type I and type II nuclear extracts in the pattern of protein-DNA complexes formed on two previously identified negative glucocorticoid response elements located at -237 to -201 and -184 to -150 bp of the mouse promoter. Both cell types contained glucocorticoid receptors (GR) by Western blot analysis. Cytosols from type I or type II cells were incubated with [3H]dex to obtain GR binding parameters. Binding data were consistent with a one-site model for dex binding in each case. Small differences in Kd (1.7 nM, type I; 3.1 nM, type II) or Bmax (approximately 3600 sites/cell, type I; approximately 1800 sites/cell, type II) were not likely to account for the differential sensitivity to dex treatment. In conclusion, nuclear alterations in type II cells leading to greater transcriptional susceptibility to dex, coupled with low GnRH storage levels, may be reflected in exquisite sensitivity of GnRH secretion to glucocorticoid repression. This represents the first example of a steroid hormone acting directly on GnRH-producing cells to alter GnRH secretion.

Regulation of α-subunit mRNA transcripts by pituitary adenylate cyclase-activating polypeptide (PACAP) in pituitary cell cultures and α T3-1 cells

Pituitary adenylate cyclase activating polypeptide (PACAP) increases glycoprotein hormone α-subunit mRNA levels suggesting a role for PACAP in maintaining the high levels of α-subunit protein characteristic of the pituitary. The present study used primary pituitary cell cultures and the α T3-1 pituitary cell line to investigate how PACAP affects α-subunit mRNA transcripts. Stimulation of cultured pituitary cells with 10 nM PACAP38, 10 nM GnRH, or the combination, for 24 h increased α-subunit mRNA levels 1.5-fold, whereas GnRH more effectively (P < 0.01) stimulated α-subunit protein release than did PACAP38 (3.2- vs. 2.0-fold). α-Subunit mRNA levels in α T3-1 cells were also increased by PACAP38 and by GnRH to maximum values at 12 h (P < 0.05), and α-subunit protein secretion rose proportionately and in parallel with α-subunit MRNA levels. PACAP38 was a 100-fold more potent stimulator of α-subunit mRNA than was VIP, and a VIP-antagonist failed to block the stimulatory effect of PACAP38, suggesting an effect via type PACAP 1 receptors. Type I receptor mRNA transcripts were identified by Northern analysis in α T3-1 cells. Depletion of PKC activity by PMA failed to block the stimulatory effect of PACAP38, but prevented GnRH from increasing α-subunit mRNA levels and α-subunit secretion. PACAP38, like 8Br-cAMP and forskolin, stimulated (P < 0.05) luciferase (LUC) activity in α T3-1 cells transfected with a plasmid containing the first 846 or 180 base pairs of the 5′-flanking region of the human α-subunit gene linked upstream to a LUC reporter gene. Finally, experiments using the transcription inhibitor DRB reveal that PACAP does not appreciably change α-subunit mRNA half-life. These findings are consistent with the proposal that PACAP contributes to the high levels of α-subunit protein characteristic of the pituitary by activating Type I receptors and stimulating α-subunit gene transcription in part by the cAMP/PKA pathway.

Facilitation or Inhibition of the Estradiol‐Induced Gonadotropin Surge in the Immature Rat by Progesterone: Regulation of GnRH and LH Messenger RNAs and Activation of GnRH Neurons

We have developed and extensively characterized immature female rat models to demonstrate inhibition or facilitation of the estradiol (E2)‐induced gonadotropin surge by progesterone (P). We show here that the surge of free α‐subunit is regulated similarly by P in these models. To investigate the possibility that P alters the biosynthesis of GnRH and/or LH, we measured levels of LH subunit mRNAs by Northern blot hybridization and GnRH mRNA by a solution hybridization‐RNase protection assay. In the P inhibition model, α‐subunit mRNA was significantly decreased when P was administered together with E2 for 32 or 48 h, and LHβ, at 29 h. In the facilitation model, neither α‐subunit nor LHβ mRNA increased with premature and enhanced release of LH and free α‐subunit. Levels of GnRH mRNA in E2‐treated rats were significantly higher on the afternoon of the LH surge than on that or the following morning. There was no effect of P on GnRH mRNA levels, however, before, during, or after the LH surge in either paradigm. The time course of activation of GnRH neurons in P‐facilitated rats was determined by double‐label immunocytochemistry for GnRH and cFos. When serum LH concentrations were basal there was no expression of cFos in GnRH neurons. LH secretion in P‐facilitated rats was initiated at ≈14.00 h and remained elevated until at least 19.00 h. During this time 63–78% of GnRH neurons were cFos positive. Both serum LH concentrations and the percentage of cFos‐activated GnRH neurons were significantly lower in control rats treated with E2 alone than in those treated also with P. In conclusion: 1) suppression of LH and free α‐subunit secretion by P can be accounted for at least partly by suppression of α‐subunit mRNA levels; 2) P facilitation is not associated with changes in LH subunit or GnRH mRNA levels; 3) the large proportion of cFos‐positive GnRH neurons in P‐facilitated rats closely parallels increases in serum LH concentrations but is not accompanied by changes in GnRH mRNA levels. It is likely, therefore, that P acts in the facilitation model to trigger release of pre‐existing GnRH stores by altering synthesis or activity of neuro‐transmitters/neuropeptides involved in GnRH regulation and/or release of LH stores by altering, for example, pituitary responsiveness to GnRH (including self‐priming) and components of the LH secretory apparatus. Similar possibilities may also obtain for the blockade of the gonadotropin surge in the inhibition model.

17β-Estradiol treatment retards excitotoxic delayed degeneration in substantia nigra reticulata neurons

Estrogen treatment offers neuro-protection in animal experiments in which excitotoxic mechanisms destroy neurons. In a model of delayed neuronal degeneration that depends on excitotoxicity, we tested whether females had an altered susceptibility, and whether physiologic doses of estrogen administered after the brain insult would protect susceptible neurons. Females were ovariectomized, exposed to striatal-pallidal ibotenic acid injury that caused delayed degeneration of substantia nigra neurons, and treated with 17 beta -estradiol (30 microg, subcutaneously every other day, beginning 2 days after the striatal injury) or vehicle. At 6 and 8 days post lesion, the 17beta-estradiol treatment group maintained over 87 and 70% of control nigral neuron number, respectively. Physiologic levels of estrogen delivered days after the excitotoxic stress completely protected neurons in the substantia nigra reticulata 6 days post lesion and slowed degeneration 8 days post lesion.

Multiple forms of nuclear estrogen receptor in the hypothalamus and pituitary after in vitro exchange with [3H]estradiol or [3H]hydroxytamoxifen.

Abstract We have examined the sedimentation properties of hypothalamic and pituitary estrogen receptors (ER) translocated to the nucleus by in vivo estradiol (E2) or nafoxidine treatments in the immature female rat. Nuclear ER were extracted with 0.4 M KC1 and incubated in vitro with saturating concentrations of [3H]E2under conditions which allowed exchange with endogenous unlabeled ligands (23°C for 1–16 h). Sucrose gradient centrifugation of pituitary extracts from rats treated for 1, 24 or 48 h with E2 yielded two components, sedimenting at 3 and 4–5S; in hypothalamic extracts the 3S form and variable amounts of a heavier shoulder were observed. After nafoxidine treatment the 4–5S species was present in greater proportions than after E2. Both species interacted with monoclonal antibodies prepared against cytoplasmic ER from MCF-7 cells to yield more rapidly sedimenting complexes (7–8S). The 4–5S receptor was preferentially labeled in nuclear extracts subjected to exchange with [3H]E2 for l h in vitro. In addition, nuclear ER from rats treated with [3H]E2for l h in vivo sedimented at about 5S. We also investigated the direct binding of [3H]4-hydroxytamoxifen to nuclear ER. A single 4.2–4.3S component was present in both the hypothalamus and pituitary; this peak was displaced to about 7.5S after incubation with antibodies to MCF-7 ER. In the same nuclear preparations the 3S receptor predominated after exchange with [3H]E2;however, a 1000-fold excess of unlabeled nafoxidine or hydroxytamoxifen abolished the 3S peak. These results suggest that 4–5S nuclear ER may give rise to the 3S form through degradation by endogenous proteases and that binding of antiestrogens to the larger species may interfere with this conversion.

Neuroendocrine Regulation of Sexual Behavior

Beginning with the pioneering work of Frank Beach and his colleagues, together with W.C. Young and his colleagues, the effects of sex-steroid hormones on the reproductive behaviors of experimental animals have comprised one of the most reliable sets of empirical phenomena in endocrinology as well as in neurobiology. Estrogenic compounds are known to foster female-typical sex behaviors in genetic females. In a wide variety of cases, administration of progestins following estrogenic priming greatly amplifies the behavioral response. In genetic males, circulating testosterone fosters courtship and copulatory behaviors using at least three routes: 1. Actions in the chemical form of testosterone itself. 2. Actions following reduction to dihydrotestosterone. 3. Actions following aromatization to estrogens.

Effects of Sulfhydryl Reagents on Uterine Nuclear Estrogen Receptors Labeled by in Vitro Exchange with [3H]Estradiol or [3H]4-Hydroxytamoxifen

We have attempted to convert 4 S uterine nuclear estrogen receptors obtained after in vitro labeling with [3H]antiestrogens to 3 S, the form observed after in vitro exchange with [3H]estradiol, in order to examine the possible relationship between these forms. Treatment of nuclear extracts labeled with the high affinity antiestrogen, [3H]4-hydroxytamoxifen, with a variety of nucleases, phosphatases, or proteases either had no effect on the 4 S antiestrogen-receptor complex or led to loss of ligand binding. The sulfhydryl reducing agents, cysteine or reduced glutathione, on the other hand, brought about conversion of 4 S estrogen receptors to components sedimenting at about 3 S. Conversely, when oxidized glutathione was included in all buffers used for preparation and labeling of nuclear estrogen receptors with [3H]estradiol, more rapidly sedimenting (approximately 4.6 S) forms of estrogen-receptor complex predominated. Cysteine still effected the 4 S to 3 S conversion when nuclear estrogen receptors, partially purified by sucrose gradient centrifugation, were used as substrate, suggesting a direct action of the sulfhydryl reagents on receptor molecules. From these results we propose that nuclear estrogen and antiestrogen-receptor complexes may differ in conformation such that the former may be more sensitive to the action of an endogenous reducing agent which contributes to formation of 3 S [3H]estradiol-receptor complexes.

Multiple forms of nuclear estrogen receptor in the immature rat uterus after in vitro exchange with [3H]estradiol or [3H]antiestrogens

Abstract We have compared the properties of salt-extracted uterine nuclear estrogen receptors (ER) labeled with [3H]estradiol (E2) or [3H]antiestrogens after in vivo exposure of immature rats to these compounds or after in vitro exchange. As expected, nuclear ER labeled during a l h in vivo treatment with [3H]E2 migrated predominantly at 5S, although this peak was skewed towards the lighter side. [3H]4-Hydroxytamoxifen ([3H]OH tamoxifen, the active metabolite of tamoxifen)-nuclear ER complexes showed comparable sedimentation rates (4.6 ± 0.05S) after injection of the [3H]antiestrogen. On the other hand, when nuclear extracts from rats treated with unlabeled E2 in vivo were subjected to exchange with [3H]E2 or [3H]antiestrogens for 16 h at 23°C, a marked difference. in sedimentation profiles was observed. While the [3H]E2 -nuclear ER complex sedimented primarily as a 3S species with variable amounts of a heavier (4.0–4.5S) shoulder, complexes between nuclear ER and [3H]tamoxifen, [3H]OH tamoxifen or [3H]CI628M formed sharp, symmetrical peaks at 4S. Both the 3S and 4S components represented forms of ER as they were eliminated in the presence of excess unlabeled DES, and they were displaced to 7–8S after reaction with specific anti-ER antibodies. The 3S peak was also abolished by addition of excess nonradioactive nafoxidine or OH tamoxifen during exchange with [3H]E2. These results suggest that more rapidly sedimenting forms of uterine nuclear ER (4–5S) may be converted to a 3S species by the action of endogenous proteases and that association of these large forms with antiestrogens may stabilize them in a conformation less or differentially susceptible to cleavage.