Search

Mutations in the melanocortin-4 receptor (MC4-R) cause obesity in both mice and humans, and the receptor is presumed to have an important role in the regulation of energy homeostasis. The MC4-R is expressed in discrete sets of neurons in the central nervous system, and thus it has been technically difficult to study the regulation of expression and the signaling mechanisms of this receptor. We report here a neuronal cell line that exhibits endogenous functional expression for the MC4-R. Initially, RT-PCR analysis showed the presence of MC4-R RNA in the hypothalamic GT1-1 and GT1-7 cells. In addition, GT1-7 cells expressed melanocortin-3 receptor while the GT1-1 subclone specifically expressed predominantly the MC4-R RNA. High-affinity binding sites were demonstrated in the GT1-1 and GT1-7 cells for NDP-α melanocyte-stimulating hormone (MSH; Ki = 1.1 × 10–10 and 1.8 × 10–10M) and agouti-related protein (AGRP; Ki = 1.548 × 10–9 and 1.663–9M). α-MSH-stimulated cAMP production in GT1-1 cells with an EC50 of 2.2 × 10–8M, and cAMP production was inhibited in the presence of AGRP, an endogenous antagonist of the MC4-R. Stimulation of gonadotropin-releasing hormone (GnRH) secretion was achieved with 1 nM to 1µM concentrations of NDP-α-MSH while no GnRH secretion was observed when the GT1-1 cells were treated with AGRP. The data presented here show that GT1-1 cells specifically express a functional MC4-R that couples to GnRH release.

Transient exposure of rats to high doses of dexamethasone (DEX; 500 µg/day for 5 days) produced a host of symptoms that are indicative of hypothalamic-pituitary-adrenal (HPA) axis dysregulation, such as increased adrenocortical secretion over 24 h, blunted and prolonged secretory response to emotional stress, refractoriness of adrenocorticotropin in vitro release to stimulation with the secretagogues corticotropin-releasing hormone (CRH) and vasopressin, decreased levels of mRNA encoding type II corticosteroid receptors in the hippocampus and increased numbers of transcripts encoding CRH in the paraventricular nucleus. Daily administration of melatonin (MEL; 80 µg/kg) concomitantly with, and for 5 days after discontinuation of, glucocorticoid treatment ‘normalized’ most of the symptoms of impaired HPA regulation caused by the exposure to DEX. While none of the treatments used caused major shifts in circadian patterns of corticosterone secretion, MEL administration was associated with diminished overall corticosterone secretion and increased sensitivity to glucocorticoid feedback. Taken together, these findings indicate that chronic MEL treatment may protect several regulatory components of the HPA axis from glucocorticoid-induced deterioration.

A transgene expressing antisense RNA complementary to a fragment of the glucocorticoid receptor cDNA was incorporated into the mouse genome and resulted in a transgenic animal that has decreased glucocorticoid receptor function. The transgenic mice showed basal plasma ACTH and corticosterone levels similar to those of the normal control animals. We have further investigated changes in HPA axis regulation by use of different neuroendocrine challenge tests including a dexamethasone suppression test (DST). In comparison to normal mice, a tenfold higher dose of dexamethasone (i.e. 20 micrograms/100 g body weight) was required to suppress the basal corticosterone levels of transgenic mice. Dexamethasone (2 micrograms/100 g body weight) produced a long-lasting suppression of plasma ACTH and corticosterone levels in control mice, whereas in transgenic animals only a short-lasting decrease in ACTH levels was apparent. Corticotropin-releasing hormone (CRH) administration resulted in an enhanced response in plasma ACTH levels in transgenic mice, whereas the corticosterone response was markedly reduced. The discrepancy between ACTH and corresponding corticosterone secretions in transgenic mice could be attributed, in part, to a reduced sensitivity of the adrenal gland to stimulation by ACTH. Pituitaries of transgenic mice contained about 50% less proopiomelanocortin (POMC) mRNA than those of control animals. No significant differences were noted in the ACTH or protein contents of normal and transgenic mice pituitary glands although a slight increase in protein content of the transgenic mouse adrenal gland was apparent. In conclusion, transgenic mice with impaired GR function show major disturbances in HPA axis regulation which seem to be caused by the primary defect in conjunction with secondary modifications in, amongst others, pituitary CRH receptor system(s), sympathetic output and adrenal development. This mouse is therefore a useful model to study the consequences of life-long defective GR function and HPA axis regulation in general.

Serotonin (5-HT) stimulates aldosterone secretion in man through activation of 5-HT4 receptors coupled to adenylyl cyclase via a Gs regulatory protein. In adrenocortical cells, the levels of expression of the Gs protein and ACTH receptor are decreased when the cells are deprived of ACTH and angiotensin II (ANGII). In order to examine the possible influence of ACTH and ANGII on the responsiveness of human glomerulosa cells to 5-HT, we have investigated the effect of cisapride, a 5-HT4 receptor agonist, on plasma aldosterone in patients with suppressed plasma ACTH, i.e. patients with corticotropic insufficiency (CI), and in patients with suppressed renin-ANG II activity, i.e. patients with primary hyperaldosteronism (PH) including both aldosterone-producing adenoma and idiopathic hyperaldosteronism. After 2 h of recumbency, all patients received a single oral dose of 10 mg cisapride. In the CI group, cisapride induced a 5-fold increase in plasma aldosterone levels without any modification of plasma renin, potassium or cortisol levels. Combined administration of cisapride and ACTH caused an increase in plasma aldosterone similar to that produced by ACTH alone. In the PH group, cisapride was still able to cause a 3.6-fold increase in plasma aldosterone levels while renin remained suppressed throughout the study. Taken together, these data show that cisapride stimulates aldosterone secretion in CI and PH patients, indicating that prolonged suppression of plasma ACTH or renin-ANG II activity does not affect the sensitivity of glomerulosa cells to 5-HT. The present study also demonstrates that the stimulatory effects of 5-HT and ACTH on aldosterone secretion are not additive.

The potential roles of central and peripheral 5-HT3 receptors in the secretion of prolactin, adrenocorticotropic hormone (ACTH), corticosterone and renin was investigated. Male rats received the 5-HT

Histamine (HA) stimulates the release of adrenocorticotropic hormone (ACTH) and beta-endorphin (beta-END) via activation of central postsynaptic H1 or H2 receptors. The effect of HA is indirect and may involve the hypothalamic regulating factors corticotropin-releasing hormone (CRH), arginine vasopressin, or oxytocin (OT). We studied the effect of specific HA H1 or H2 receptor agonists on the concentration of CRH and OT in hypophyseal portal blood in urethane-anesthetized male rats. In addition we investigated the effect of the agonists on ACTH and beta-END immunoreactivity in peripheral plasma in conscious male rats pretreated with antiserum to CRH. Intracerebroventricular administration of the H1 receptor agonist 2-thiazolylethylamine (2-TEA) or the H2 receptor agonist 4-methylhistamine (4-MeHA) increased the CRH concentration in pituitary portal blood by 80-90% when compared to preinfusion levels (p < 0.05). Central infusion of saline had no effect. The level of OT in the pituitary portal blood was not affected by 2-TEA or 4-MeHA when compared to saline-treated rats. Intracerebroventricular infusion of 2-TEA or 4-MeHA increased the ACTH concentration in peripheral plasma 3- or 4-fold, respectively (p < 0.01). Pretreatment with a specific CRH antiserum (abCRH) inhibited the responses by 50 and 70%, respectively (p < 0.01). Intracerebroventricular administration of 2-TEA or 4-MeHA increased the beta-END immunoreactivity in peripheral plasma 3- or 2-fold, respectively (p < 0.01). These effects were inhibited by 80-90%, when rats were pretreated with abCRH (p < 0.01).(ABSTRACT TRUNCATED AT 250 WORDS)

Previous studies have shown that activation of the 5-HT1A receptor subtype enhances rat plasma ACTH concentration. Such receptors have been suggested to be located on CRH neuronal cell bodi

The cytokines interleukin-1 beta (IL-1 beta), interleukin-6 (IL-6) and tumor necrosis factor-alpha are known to be potent effectors of ACTH secretion. Some of the peripheral effects of IL-1 beta appear to be related to the secretion of IL-6 induced by IL-1 beta. Thus, we evaluated the effect of IL-6 on ACTH secretion and its interaction with IL-1 beta. Rats received recombinant human (rhIL-6) or murine (rmIL-6) IL-6 through indwelling jugular cannulae. rhIL-6 (200 ng or 2 micrograms/rat) produced peak plasma ACTH levels which were 3- to 4-fold greater than basal levels. rmIL-6 produced similar responses. Neither species of IL-6 affected plasma prolactin levels. Comparison of rhIL-1 beta (200 ng) to rhIL-6 (200, 100 or 50 ng) showed that IL-6 elevated ACTH in a dose-dependent manner and that IL-1 beta was significantly more effective. IL-1 beta was also administered concomitantly with or 10 min after IL-6. Delivered together, IL-1 beta (100, 30 or 10 ng) and IL-6 (100 ng) produced significantly higher ACTH levels than when given alone. This additivity was also evident when IL-6 was given 10 min prior to IL-1 beta. The coadministration of IL-6 (2 micrograms) with corticotropin-releasing factor (CRF, 1 micrograms/kg, b.w.) also had an additive effect on ACTH secretion (at 20 min: 300 +/- 40 pg/ml for CRF; 320 +/- 83 pg/ml for IL-6; and 540 +/- 44 pg/ml for CRF + IL-6), whereas a higher dose of CRF (10 micrograms/kg b.w.) yielded ACTH levels of 1,000 +/- 107 pg/ml at 20 min, with no further enhancement by IL-6. Incubation of pituitary cells with IL-6 alone (0.1, 1.0 or 3.0 nM) produced a slight but significant stimulation of ACTH secretion within 2 h in response to the higher doses of IL-6 only (p0.05), but did not modify the effect of CRF in vitro. To determine if the action of IL-6 was at a site(s) within the brain, IL-6 (30 or 100 ng/0.5 microliters) was injected into the third cerebroventricle of alert rats. 100 ng IL-6 elicited peak plasma ACTH levels (300 +/- 65 pg/ml) within 30 min; these were significantly higher than the buffer responses (90 +/- 25 pg/ml, p0.01), and lower than the responses to 30 ng IL-1 beta (530 +/- 50 pg/ml, p0.001). 30 ng IL-6 was ineffective.(ABSTRACT TRUNCATED AT 400 WORDS)

The suppression of plasma adrenocorticotropic hormone by very low levels of cortisol is reduced in pregnant adrenalectomized ewes, suggesting that pregnancy reduces the efficacy of the high-affinity corticosteroid receptor. This study was designed to determine the effects of pregnancy on the availability, immunoreactivity, and affinity of both corticosteroid receptors: the high-affinity mineralocorticoid receptor (MR) and the lower-affinity glucocorticoid receptor (GR). Availability was measured in the hypothalamus, pituitary, hippocampus and kidney using a saturation point radioligand binding assay. GR availability was significantly decreased in hippocampal cytosols obtained from pregnant ewes, but did not significantly change in other tissues. This finding is consistent with increased GR activation due to elevated circulating concentrations of cortisol. MR availability significantly increased from undetectable levels in hippocampal cytosols obtained from nonpregnant ewes to 2.8 ± 1.6 fmol/mg protein in pregnant ewes, suggesting a reduced MR activation in the hippocampus during pregnancy. MR availability tended to be greater in other tissues during pregnancy, but these differences were not significant. The amount of immunoreactive MR (iMR) and GR (iGR) protein was estimated by quantifying Western blots. iGR significantly increased in the pituitary, but did not significantly change in other tissues. In contrast, iMR was significantly increased during pregnancy in all tissues assayed, suggesting that an increased cytosolic MR protein amount contributes to the observed increase in MR availability. Since studies suggest that progesterone is a potent anticorticosteroid, we tested for evidence of endogenous inhibition of binding to MR and/or GR during pregnancy by determining MR and GR affinity in pituitary cytosols obtained from nonpregnant and pregnant ewes. Although there was a tendency towards a decreased affinity of the MR in pregnant ewes, there was no significant change in the KD of the pituitary MR or GR during pregnancy. We hypothesize that an alteration in activation and/or autoregulation of the MR during pregnancy, particularly in the hippocampus, may contribute to the observed changes in receptor availability and immunoreactivity and increase basal plasma cortisol levels during pregnancy.

Adrenocorticotropic hormone (ACTH) is the major regulator of adrenocortical steroidogenesis in mammals. By comparing the sensitivity to ACTH of isolated adrenocortical cells from two sources of the same strain (Sprague-Dawley, SD) of outbred rats, we have identified a source of rat with low sensitivity to ACTH in vitro. Cells isolated from Holtzman SD rats had a high sensitivity to ACTH (minimal effective concentration 50 pg/ml), whereas Taconic SD rats had a low sensitivity (minimal effective concentration 250 pg/ml; maximal steroidogenesis

Alpha-melanocyte-stimulating hormone (alpha-MSH) and adrenocorticotropic hormone (ACTH), peptides derived from the precursor proopiomelanocortin, share amino acid homology at the aminoterminus of ACTH, occur within the pituitary and the brain and are potent antipyretic compounds in cytokine-mediated fever. Because alpha-MSH and ACTH act within the hypothalamus to block leukocytic pyrogen- or cytokine-mediated fever, we hypothesized that these compounds might also be capable of blocking the action of interleukin-1 (IL-1) and interleukin-6 (IL-6) to stimulate corticotropin-releasing factor (CRF) release from the hypothalamus. Mediobasal hypothalami (MBH) were incubated in vitro. After 60 min preincubation in Krebs-Ringer bicarbonate buffer (KRB), MBH explants were incubated for 30 min with KRB alone or KRB containing IL-6 (10(-13) M), IL-1 (10(-16)-10(-10) M) and/or ACTH1-24 (10(-15)-10(-9) M) or alpha-MSH (10(-15)-10(-8) M); CRF release into the incubation medium was measured by RIA. None of the ACTH1-24 or alpha-MSH concentrations changed basal CRF release significantly. As we reported previously, IL-6 (10(-13) M) increased CRF release; this increase was suppressed, in a dose-dependent fashion, by alpha-MSH at concentrations of 10(-13)-10(-11) M, with the maximal inhibitory effect observed at 10(-13) M. ACTH1-24 also exerted a dose-dependent inhibitory effect on IL-6-stimulated CRF release but at even lower concentrations (10(-15)-10(-13) M) with the maximal inhibitory effect observed with the 10(-14) M concentration.(ABSTRACT TRUNCATED AT 250 WORDS)

The controversy about putative stimulatory and inhibitory functions of catecholamines in regulation of ACTH secretion has been recently shifted towards a consensus that during stress catecholamines stimulate corticotropin-releasing factor (CRF-41) containing neurons through alpha 1-adrenoreceptors, while inhibiting their own secretion acting on presynaptic alpha 2-receptors. In this study the effect of the alpha 2-agonist clonidine and the antagonist CH-38083 was studied on exogenous CRF-41/AVP-induced ACTH secretion in rats with/without paraventricular nucleus lesion. Clonidine (30 micrograms/kg) attenuated CRF-41/AVP (1 pmol/10 pmol)-induced ACTH secretion in sham-operated rats, but was ineffective in reducing CRF-41/AVP-induced ACTH secretion in rats with paraventricular nucleus lesion. In sham-operated rats, alpha 2-receptor antagonist CH-38083 slightly elevated the basal, and significantly potentiated the CRF-41/AVP-induced ACTH secretion, while it had no effect on the hypophyseotropic cocktail-induced ACTH response in paraventricular-lesioned rats. Neither the agonist nor the antagonist affected CRF-41/AVP-induced ACTH release from pituitary fragments in vitro. These results suggest that in response to activation of alpha 2-adrenoreceptors a corticotropin release-inhibiting substance is released from the paraventricular nucleus.

Histamine (HA) stimulates the release of the pro-opiomelanocortin (POMC)-derived peptides ACTH, beta-endorphin (beta-END), beta-lipotropin and alpha-melanocyte-stimulating hormone, and HA is involved in the mediation of the stress-induced release of these peptides. The effect of HA is indirect and may involve the hypothalamic regulating factors, corticotropin-releasing hormone (CRH) and/or arginine-vasopressin (AVP). We studied the effect of immunoneutralization with specific antisera against CRH or AVP on the response of ACTH and beta-END to HA, restraint stress, CRH, AVP or a posterior pituitary extract in male rats. Intracerebroventricular infusion of HA (34-540 nmol) increased the plasma levels of ACTH and beta-END immunoreactivity (beta-ENDir) in a dose-dependent manner. Pretreatment with antiserum to CRH or AVP prevented the ACTH response to 270 nmol HA and inhibited the beta-ENDir response by 30-60%. One to five minutes of restraint stress caused an increase in the plasma levels of ACTH and beta-ENDir. The increase was dependent on the duration of stress exposure. Pretreatment with CRH antiserum abolished the ACTH response to 5 min of restraint stress and inhibited the beta-ENDir response by 60%. Immunoneutralization with AVP antiserum had only half the inhibitory effect of that seen with CRH antiserum. CRH (100 pmol i.v.) increased the plasma levels of ACTH and beta-ENDir. This effect was abolished by pretreatment with CRH antiserum, whereas pretreatment with AVP antiserum prevented the CRH-induced ACTH release and inhibited the beta-ENDir response by 50%. AVP (24-800 pmol i.v.) stimulated ACTH and beta-ENDir in a dose-dependent manner. CRH and AVP antisera each prevented the effect of AVP (800 pmol) on ACTH secretion, whereas the beta-ENDir response to AVP was only inhibited by about 60% by the antisera. An extract of the posterior pituitary gland administered in a dose corresponding to 0.15 or 0.5 pituitary equivalents had no effect on ACTH secretion, while 1.0 pituitary equivalent increased the ACTH concentration in plasma. This effect was abolished by AVP antiserum. The posterior pituitary extract caused a dose-dependent rise in plasma beta-ENDir which might be due to an unavoidable contamination of the posterior pituitary extract by a small amount of beta-END from the intermediate lobe. Consistent with this view, AVP antiserum had no effect on the rise in the plasma concentration of beta-ENDir following administration of the posterior pituitary extract.(ABSTRACT TRUNCATED AT 400 WORDS)

beta-Endorphin and ACTH are secreted concomitantly in baseline conditions and in response to physiological and pharmacological stimuli. However, few and contradictory data are available on their feedback inhibition mechanisms. To investigate this aspect, the effects of exogenous ACTH1-24 and glucocorticoids on endogenous ACTH1-39 and beta-endorphin were tested in 18 patients with Addison's disease. Two main experimental protocols were employed: (1) 7 patients were given ACTH1-24 50 micrograms as an intravenous bolus followed by a 50-micrograms infusion in 90 min. Blood samples for beta-endorphin, ACTH and cortisol were obtained at 0, 15, 30, 45, 60, 90, 120 min. Six other patients were given oCRH 100 micrograms i.v. plus ACTH1-24, as described above. (2) In 5 other patients, hydrocortisone 37.5 mg was administered i.v. in 90 min. Blood samples for beta-endorphin, ACTH and cortisol were drawn at 0, 15, 30, 45, 60, 90, 120 min. One week later, the same patients were given oCRH 100 micrograms i.v. and hydrocortisone 37.5 mg, as described above. ACTH1-24 administration caused a significant (p less than 0.01) decrease in endogenous ACTH but not in beta-endorphin. oCRH injection significantly stimulated both ACTH and beta-endorphin. The response of both ACTH and beta-endorphin was inhibited by exogenous ACTH1-24. There was a potent inhibition by hydrocortisone on both basal and stimulated beta-endorphin, confirming that the feedback mechanism of glucocorticoids concomitantly inhibits ACTH and beta-endorphin. On the other hand, only CRH-stimulated but not basal secretion of beta-endorphin seems affected by ACTH ultrashort feedback, suggesting an intrapituitary regulation.

The possible implication of the amygdaloid central nucleus (ACE) of the rat in the control of ACTH secretion in response to immobilization stress was assessed. The ACTH secretion, in response to stress and/or bilateral lesions of the ACE, was correlated with the serotoninergic activity in specific hypothalamic and amygdaloid nuclei. Bilateral lesions of the ACE produced a striking decrease of plasma ACTH levels in response to stress. However, basal plasma ACTH levels measured between 7 and 11 a.m. were identical in both control and lesioned groups. Stress, applied to intact animals, did not modify the serotoninergic activity in any of the following areas: hypothalamic paraventricular (PVH), ventromedial (VMH) and dorsomedial (DMH) nuclei; the anterior hypothalamic area (AHA); the lateral part of the basal amygdaloid nucleus (ABL), the amygdaloid medial (AME) and cortical (ACO) nuclei. However, lesion of the ACE increased the serotoninergic activity in all these structures except for the VMH. Immobilization stress applied to lesioned animals decreased the serotoninergic activity to control levels in the PVH, AHA and DMH and decreased the activity to below control levels in the VMH. The serotoninergic activity remained at an increased level in the glucocorticoid receptor-rich areas of the amygdala, namely the AME, ACO and ABL nuclei. These results provide evidence for a stimulatory role of the central nucleus of the amygdala in the control of ACTH secretion. Moreover, they substantiate an implication of the amygdaloid complex in the control of the delayed negative feedback of glucocorticoids on ACTH secretion via interaction with the serotoninergic system.

To ascertain the physiological relevance of an autoregulation of adrenocorticotropin hormone(ACTH) secretion in man, we studied the effect of alsactide (beta-Ala1, Lys17-ACTH1-17-4-amino-N-butylamide), a synthetic ACTH analogue with potent steroidogenic activity but not recognized in the endogenous ACTH immunoassay, on plasma ACTH pattern in patients with Addison's disease. Three experimental models were employed as follows: (a) in 6 patients, whose steroid replacement therapy had been discontinued 36 h previously, we compared the effect of alsactide, administered at two dose levels (10 or 100 micrograms i.v. als bolus followed by the same dose infused over 2 h, and of placebo, on the plasma ACTH pattern; (b) the previous experiment was repeated in 4 patients in whom cortone replacement therapy was substituted for 3 days with dexamethasone, 0.5-1.5 mg daily p.o., so as to lower plasma ACTH levels to within the high normal range; (c) in 4 patients off therapy for 36 h, we evaluated the ACTH response to synthetic ovine corticotropin-releasing factor, 1 microgram/kg body weight injected intravenously, occurring during concomitant administration of alsactide, 100 micrograms i.v. as bolus plus 100 micrograms infused over 2 h, or placebo. Compared to placebo, alsactide did not significantly affect the pattern of ACTH under any of the experimental conditions investigated. Collectively, our findings, although they have to be interpreted with caution, do not support the idea that a self-regulation mechanism plays an important role in the control of ACTH secretion in man.

In the rat brain, the binding of corticosterone is mediated through two receptor types, the type I receptor and the type II receptor, which are presumed to be encoded by genes designated as MR and GR, respectively. We have studied the regulation of these receptors by glucocorticoids, utilizing a cytosol receptor binding assay. In addition, we have employed molecular probes for the GR and the MR to measure receptor mRNAs. The level of type II receptor binding is uniform across several brain regions, as is the expression of GR (type II) mRNA. In contrast, type I receptor binding is concentrated in the hippocampus, and the MR (type I) mRNA similarly shows a higher level of expression in hippocampus than in the other brain regions studied. Removal of endogenous glucocorticoids by adrenalectomy (ADX) induces an increase, and corticosterone administration results in a decrease, in the level of type I and type II binding in the hippocampus; however, no significant changes in the MR (type I) or GR (type II) mRNA levels are seen with these treatments. The diurnal variation of serum corticosterone in intact rats is correlated with a circadian regulation of type I receptor binding in the hippocampus, while MR (type I) mRNA expression is unaffected. Thus, the changes in type I and type II receptor binding capacity elicited by differing steroid conditions cannot be attributed to modulation of the steady state levels of MR (type I) or GR (type II) mRNA.

Early in the history of studies on the release of ACTH by stress there were indications that ACTH might be released by multiple factors. But the neurohumoral theory, as formulated by G. W. Harris, suggested that every hypophysial hormone had its unique hypothalamic controlling agent and a search for the unique ACTH-releasing hormone went on for about 20 years. This review reexamines the case for multiple releasers of ACTH.

This study reports on the kinetics of glucocorticoid-induced inhibition of ACTH release, together with steroid binding to specific pituitary receptors. It was shown that corticosterone (CORT) inhibited ACTH output provoked by either corticotropin-releasing factor (CRF) extracts or dbcAMP, in a manner which was both dose- and time-dependent. A close correlation appeared to exist between the degree of ACTH blockade and the percentage of filled steroid-binding sites. However, exposure of hypophyses to CORT for a critical period of time was a prerequisite for such a relationship to develop. Furthermore, it was found that dexamethasone (DEX) was more potent than CORT in inhibiting ACTH secretion and, in addition, bound to a greater extent to nuclei of pituitary cells. These data suggest the existence of a close correlation between occupancy of pituitary glucocorticoid receptors and modulation of ACTH secretion.