Glucocorticoid regulation of proopiomelanocortin mRNA levels in rat arcuate nucleus

Food restriction-induced hyperactivity: addiction or adaptation to famine?

Increased physical activity is present in 30-80% of anorexia nervosa patients. To explain the paradox of low food intake and excessive exercise in humans and other animals, it has been proposed that increased physical activity along with food restriction activates brain reward circuits and is addictive. Alternatively, the fleeing-famine hypothesis postulates that refusal of known scarce energy-low food sources and hyperactivity facilitate migration towards new habitats that potentially contain new energy-rich foodstuffs. The use of rewarding compounds that differ in energy density, such as the energy-free sweetener saccharin and the energy rich sucrose makes it possible to critically test the reward-addiction and fleeing-famine hypotheses. The aims of the present work were to study if sucrose and/or saccharin could attenuate food restriction-induced hyperactivity, weight loss, increased plasma corticosterone, and activation of brain structures involved in neuroendocrine control, energy balance, physical activity, and reward signaling in rats. Its major findings are that access to sucrose, but not to saccharin, attenuated food restriction-induced running wheel activity, weight loss, rises in plasma corticosterone, and expression of the cellular activation marker c-Fos in the paraventricular and arcuate hypothalamus and in the nucleus accumbens. These findings suggest that the energy-richness and easy availability of sucrose interrupted a fleeing-famine-like hyperactivity response. Since corticosterone mediates food restriction-induced wheel running (Duclos et al., 2009), we propose that the attenuating effect of sucrose consumption on plasma corticosterone plays a role in reduced wheel running and weight loss by lowering activation of the nucleus accumbens and arcuate hypothalamus in these animals.

Effects of adrenalectomy on AGRP, POMC, NPY and CART gene expression in the basal hypothalamus of fed and fasted rats

Glucocorticoids regulate body energy balance through both peripheral and central mechanisms. In order to understand the central mechanisms that mediate these effects of glucocorticoids we studied the effects of adrenalectomy (ADX) and food deprivation on the expression of four neuropeptide genes (measured by S1 nuclease protection assay) in the medial basal hypothalamus (MBH), which are known to regulate energy balance: pro-opiomelanocortin (POMC), agouti-related peptide (AGRP), neuropeptide Y (NPY), and cocaine and amphetamine regulated transcript (CART). Adult male rats were ADX or sham operated (SHAM), and studied 1-2 weeks later. In the first study effects of ADX and corticosterone replacement on POMC and AGRP expression were determined. ADX decreased POMC and AGRP gene expression in the MBH by 27 and 38%, respectively, compared to SHAM rats. Corticosterone treatment increased the expression of POMC by 87% and AGRP by 45% in ADX rats. The second study was designed to determine if glucocorticoids are necessary for the fasting induced changes in POMC, AGRP, NPY and CART in the MBH. ADX caused a 20-30% decrease in the expression of all four neuropeptide genes in the MBH. As expected, fasting suppressed POMC and CART expression and increased AGRP and NPY expression. The fasting-induced increases in AGRP and NPY persisted after ADX but no further significant decreases in POMC or CART were noted after fasting in ADX rats. Plasma leptin and insulin declined significantly after ADX and increased with corticosterone replacement; both leptin and insulin declined further in fasted, ADX animals. In conclusion, ADX decreases both anorexigenic, POMC and CART, and orexigenic, AGRP and NPY, neuropeptide gene expression in the MBH. AGRP and NPY decrease after ADX despite the fall in plasma leptin and insulin concentrations which in other situations would increase these neuropeptides. Furthermore, glucocorticoids are not required for fasting-induced upregulation of AGRP and NPY expression.

Ablation of pituitary pro-opiomelanocortin (POMC) cells produces alterations in hypothalamic POMC mRNA levels and midbrain mu opioid receptor binding in a conditional transgenic mouse model

The hypothalamic-pituitary-adrenal (HPA) axis is regulated by stress-related excitatory inputs, and various inhibitory and negative-feedback controls by glucocorticoids and opioids, including pro-opiomelanocortin (POMC)-derived peptides. The role of POMC-derived peptides of pituitary origin in the modulation of brain POMC mRNA expression and opioid receptor binding was investigated using a line of transgenic mice that express a fusion gene composed of the pituitary expression-specific promoter region of the POMC gene driving the herpes simplex viral-1 thymidine kinase (TK). Male adult mice were treated with the antiherpes agent ganciclovir that selectively ablates cells expressing TK. Following treatment, POMC mRNA levels, measured by quantitative solution hybridization/RNase protection assays, were decreased by 48% in the pituitary of the TK+/+ mice, reflecting an expected loss of the pituitary corticotrope POMC cells. This treatment also significantly lowered pituitary beta-endorphin immunoreactivity content and plasma concentrations of corticosterone. In contrast, POMC mRNA levels were increased by 79% in the hypothalamus of the TK+/+ mice with pituitary POMC cell ablation. Binding of [(3)H]DAMGO to mu opioid receptors, as measured by quantitative autoradiography, was significantly reduced in several brain regions including the central grey, median raphe and superficial grey layer of the superior colliculus. These regions are innervated by hypothalamic POMC neurones. No significant differences in binding to either kappa or delta opioid receptors were found in the brain regions studied. These results suggest that POMC-derived peptides of pituitary origin may exert a tonic negative-feedback effect on hypothalamic POMC neurones. In turn, the downregulation of central mu opioid receptors in this model may be mediated through a mechanism related to hypothalamic POMC overexpression.

Melanocortin-3 receptor mRNA expression in pro-opiomelanocortin neurones of the rat arcuate nucleus

The melanocortins alpha- and gamma-melanocyte-stimulating hormones (alpha- and gamma-MSH) derive from the pro-opiomelanocortin (POMC) precursor. Melanocortins exert a wide range of biological activities in the brain through activation of at least three distinct melanocortin receptor (MC-R) subtypes. In order to determine whether POMC neurones can modulate their own activity, we looked for the possible expression of the MC3-R gene in POMC-positive cell bodies in the rat hypothalamus. In situ hybridization experiments revealed that the density of MC3-R mRNA is particularly high in the arcuate nucleus which contains the main population of POMC neurones in the brain. The occurrence of MC3-R mRNA in POMC-positive cell bodies was demonstrated using a double-labelling in situ hybridization technique. The proportion of POMC neurones expressing MC3-R mRNA was significantly higher in the most rostral (43.5%) than in the most posterior part of the arcuate nucleus (8.2%). These results indicate that melanocortins likely exert a direct regulatory feedback on POMC neurones through activation of MC3-R receptors. Our data also suggest that MC3-R may be involved in the neuroendocrine responses induced by centrally administered melanocortins.

The proopiomelanocortin system

POMC (31,000 MW) is localized to the pituitary, brain, skin, and other peripheral sites. The particular enzyme profile present within a cell dictates the nature of the hormonal ligand (melanocortin) synthesized and secreted: melanotropic peptides (alpha-MSH beta-lipotropin, lambda-MSH), corticotropin (ACTH), several endorphins (e.g., met-enkephalin). These POMC-derived peptides mediate their actions through typical seven-spanning membrane receptors (MCRs; MCR1, 2, 3, 4, and 5). A specific melanocortin acting on a specific MCR regulates a particular biological response; for example, alpha-MSH on MCR1 increases melanogenesis within melanocytes, ACTH on MCR2 increases cortisol production within adrenal zona fasciculata cells. Within the brain melanocortins regulate satiety (MCR4) and erectile activity (MCR?). MCRs have been localized by melanocortin macromolecular probes, for example, fluorescent to human epidermal melanocytes and also to keratinocytes, suggesting that systemic melanocortins or localized POMC products might regulate these integumental cellular elements in synchrony to enhance skin pigmentation and/or immunological responses. Superpotent, prolonged acting melanotropic peptides have been synthesized and their application in clinical medicine has been demonstrated. MCR antagonists have been used to discover and further delineate other roles of melanocortin ligands. For example, melanocortin-induced satiety can be antagonized by a melanocortin antagonist. Defects in melanocortin ligand biosynthesis, secretion, and melanocortin receptor function can lead to a diverse number of pathological states.

Influence of adrenal glands on the modulation of prolactin gene expression by the endogenous benzodiazepine ligand octadecaneuropeptide in the male rat pituitary gland

Recently, an 86-amino acid polypeptide with high affinity for diazepam binding sites, termed diazepam-binding inhibitor (DBI), has been found in the rat brain. DBI, as well as a peptide derived from DBI, the octadecaneuropeptide DBI[33-50] (ODN), interacts with the GABA(A) receptor complex. To investigate the role of these endogenous ligands for GABA(A) receptors on prolactin gene expression, we studied the effects of acute intracerebroventricular administration (4 h before sacrifice) of ODN on prolactin mRNA levels in the male rat. Because, in some neuropeptidergic systems, glucocorticoids play a role in the response to ODN, we also studied the influence of adrenal glands and the effect of dexamethasone administration in the response of prolactin gene expression to ODN. ODN injection produced an increase in prolactin mRNA levels. Adrenalectomy performed 5 days before sacrifice resulted in an increase in prolactin gene expression and also potentiated the stimulating effect of ODN. Because castration has been shown to decrease prolactin gene expression in the male rat, we used castrated and adrenalectomized animals to study the role of dexamethasone in the response of lactotrophs to ODN. In these steroid-deprived animals, dexamethasone treatment (for 4 days) decreased prolactin mRNA levels but did not modify the response to ODN. These data indicate that an endogenous neuropeptide interacting with the GABA(A) receptor complex can stimulate prolactin gene expression and suggest that the adrenal glands may produce factor(s) capable of decreasing prolactin mRNA. On the other hand, it does not appear that glucocorticoid hormones play a role in the effect of ODN on lactotroph activity.

Effect of morphine on proopiomelanocortin gene expression and peptide levels in the hypothalamus

Opiates have been reported to suppress POMC in the medial basal hypothalamus (MBH) but studies have been complicated by the fact that acutely, in the rat, opiates stimulate corticosterone and inhibit gonadal steroid release, which could both affect POMC in brain. We have therefore examined POMC gene expression and peptide levels in the MBH of castrated rats after 10 days of treatment with subcutaneous morphine or placebo pellets and after pellet removal. POMC mRNA was measured by solution hybridization assay and beta-endorphin (beta-EP) and alpha-MSH were measured by RIA. In castrated male rats, the mean POMC mRNA concentration in the MBH was 1.67 +/- 0.11 pg/microgram RNA in the control animals and decreased to 1.17 +/- 0.11 pg/microgram RNA in the morphine-treated animals (P < 0.01). Similarly in castrated, estradiol replaced female rats, the mean POMC mRNA level in the MBH was 1.36 +/- 0.19 pg/microgram RNA and decreased to 0.82 +/- 0.08 pg/microgram RNA after morphine treatment (P < 0.05). beta-EP levels were not significantly different in either study. When castrated male rats were similarly morphine pelleted and killed either on day 10 or 2 days later after pellet removal, the mean POMC mRNA level again fell from 1.83 +/- 0.21 in the controls to 1.28 +/- 0.20 pg/microgram RNA after 10 days of morphine; 2 days after pellet removal levels remained suppressed at 0.80 +/- 0.08 pg/microgram RNA (P < 0.01). In this study the concentrations of beta-EP and alpha-MSH were both noted to decline in the MBH after morphine treatment (P < 0.05). When the forms of beta-EP in the MBH were characterized by HPLC, a decrease in the concentration of beta-EP was again seen after morphine but no significant differences in the pattern of beta-EP processing or in the relative amounts of beta-EP1-31 compared to beta-EP1-27 and beta-EP1-26 were noted in morphine-treated animals. There was also no significant effect of 10(-6)-10(-4) M morphine on basal or KCl-stimulated release of beta-EP or gamma 3-MSH release from the perifused rat hypothalamus in vitro. We conclude that morphine suppresses POMC gene expression in the hypothalamus of chronically treated male and female rats. Persistent changes were also noted during morphine withdrawal. In some cases this was accompanied by a fall in beta-EP peptide content. These effects were seen in castrated animals with and without sex steroid replacement and are thus independent of the effects of morphine on the pituitary-gonadal axis. These results show that opiate drugs modify endogenous opioid systems in the brain and provide further support for the hypothesis that such changes may contribute to mechanisms of opiate dependence and withdrawal.

Evidence for short-loop feedback effects of ACTH on CRF and vasopressin expression in parvocellular neurosecretory neurons

Immuno- and hybridization histochemical methods were used to examine a possible role for adrenocorticotropic hormone (ACTH) in regulating the expression of corticotropin-releasing peptides in rat hypothalamus. Densitometric assessments of relative levels of mRNAs encoding corticotropin releasing factor (CRF), arginine vasopressin (AVP) and oxytocin (OT) in the parvocellular division of the paraventricular nucleus (PVH) were carried out in intact, adrenalectomized (ADX) and hypophysectomized (HYPOX) animals. Both surgeries resulted in comparable increases in relative levels of CRF and AVP transcripts in the parvocellular PVH; no effects on OT mRNA in this compartment were evident. In a second experiment, ACTH or saline vehicle were administered systemically via osmotic minipump for seven days to rats submitted to both HYPOX and ADX surgeries. Lower replacement doses of ACTH reduced the number of detectable AVP-immunoreactive (AVP-ir) cells in the parvocellular PVH to 53% of that seen in vehicle-treated HYPOX/ADX controls; the number of CRF-IR cells was not significantly affected. Higher doses of ACTH resulted in counts of AVP- and CRF-IR neurons that were reduced to 32% and 70%, respectively, of control values. Staining patterns for the two peptides in the external lamina of the median eminence generally followed the cell count data. Neither densitometric nor combined immunohistochemical (for CRF-ir) and hybridization histochemical (for AVP mRNA) assays revealed any marked effect of ACTH on AVP mRNA expression in the parvocellular PVH of HYPOX/ADX rats. The results indicate that ACTH is capable of inhibiting corticotropin-releasing peptide, but not mRNA, expression in hypophysiotropic neurons. The mechanisms underlying these effects remain to be fully clarified.

Arcuate nucleus neurons that project to the hypothalamic paraventricular nucleus: neuropeptidergic identity and consequences of adrenalectomy on mRNA levels in the rat

The possible role that the hypothalamic arcuate nucleus might play in mediating the increase in paraventricular nucleus corticotropin-releasing hormone mRNA levels following adrenalectomy was investigated in two series of experiments. In the first series in situ hybridization histochemistry was used to quantify levels of eight accurate nucleus neuropeptide and neurotransmitter mRNAs in neurons that potentially relay adrenal steroid feedback to the paraventricular nucleus. In the second series of experiments, arcuate neuropeptidergic projections to the hypothalamic paraventricular nucleus were characterized using retrograde tracing in combination with in situ hybridization histochemistry. Despite an increase in paraventricular nucleus corticotropin-releasing hormone (60%) and pituitary proopiomelanocortin mRNA levels (sixfold), arcuate mRNA levels for proopiomelanocortin, neuropeptide Y, somatostatin, galanin, dynorphin, tyrosine hydroxylase, glutamate decarboxylase, and the glucocorticoid receptor were unchanged 14 days following adrenalectomy. Neuropeptidergic characterization of arcuatoparaventricular projections was achieved by injection of the retrograde tracer fluorogold into the paraventricular nucleus; retrogradely labeled neurons were characterized with polyclonal antisera against fluorogold in combination with oligonucleotide probes directed against neuropeptide Y, proopiomelanocortin, or somatostatin. Out of these three arcuate neuropeptide Y mRNA was contained in 18% of the fluorogold-positive neurons in the arcuate, proopiomelanocortin mRNA was contained in 8%, and somatostatin mRNA was contained in 6%. Overall, the results from both experiments suggest that the arcuatoparaventricular neuropeptide Y, proopiomelanocortin, and somatostatin projections are not sensitive to a chronic (14 day) lack of adrenal steroids. These projections as well as the other arcuate neurotransmitter and neuropeptide systems appear not to contribute to the persistent elevations in paraventricular nucleus corticotropin-releasing hormone mRNA levels or pituitary proopiomelanocortin mRNA levels found in 14 day adrenalectomized rats.

Regulation of proopiomelanocortin gene expression by neuropeptide Y in the rat arcuate nucleus

In the arcuate nucleus which is richly innervated by both proopiomelanocortin (POMC) and neuropeptide Y (NPY) neurons, it has been shown that NPY fibers are in synaptic contact with POMC cell bodies. In order to determine whether NPY could influence POMC neuronal activity, we have studied the effects of NPY and some NPY analogs on POMC gene expression using quantitative in situ hybridization. The following peptides NPY, [Leu31,Pro34]-NPY (a Y1 receptor agonist), and NPY13-36 (a Y2 receptor agonist) were injected into the left lateral cerebral ventricle of adult male rats 4 h before being perfused for histological procedures. The intracerebroventricular injection of NPY and NPY13-36 induced a significant decrease in the number of grains overlying the labelled neurons. On the other hand, the Y1 receptor agonist [Leu31,Pro34]-NPY did not modify POMC mRNA levels. These data then strongly suggest that NPY negatively regulates the genetic expression of POMC neurons via the Y2 NPY receptor subtype.

Role of endogenous opioid peptides in central glucocorticoid receptor (GR)-induced decreases in circulating LH in the male rat

While it has been shown that intracerebral administration of exogenous glucocorticoids diminishes pituitary luteinizing hormone (LH) release, it is not known if these hormones act directly on gonadotropin-releasing hormone (GnRH)-synthesizing neurons, or if their central inhibitory effects are mediated by specific neural substrates. In the present study, we examined whether the opioid receptor antagonist, naltrexone (NALT), alters patterns of LH release elicited by either systemic or intracerebroventricular (i.c.v.) delivery of GR agonists. Subcutaneous (s.c.) injection of the GR agonist, RU362 (2.5 mg/kg), promoted a significant reduction in circulating LH levels; pretreatment by i.c.v. injection of 1.0 microgram NALT, however, attenuated this inhibitory hormonal response. It was also found that rats treated sequentially with NALT and RU362 exhibited significantly lower plasma LH levels compared to rats injected with NALT alone. In other experiments, intracranial delivery of the synthetic glucocorticoid, dexamethasone (DEX), into either the ventricular system or the hypothalamic ARC resulted in significantly decreased plasma LH concentrations; the central inhibitory effects of DEX on peripheral LH release were reversed, however, by i.c.v. pretreatment with NALT. In summary, the present studies show that opioid receptor blockade attenuates systemic, as well as intracerebral inhibitory effects of GR agonists on the GnRH-pituitary LH axis, suggesting that circulating glucocorticoids inhibit LH, in part, through central actions involving endogenous opioid receptors. The observed decline in peripheral plasma LH following intra-ARC injection of DEX suggests that local GR may be functional target sites for glucocorticoid effects on LH.(ABSTRACT TRUNCATED AT 250 WORDS)

Negative regulation of proopiomelanocortin gene expression by GABAA receptor activation in the rat arcuate nucleus

Activation of the GABAA-benzodiazepine receptor complex has previously been shown to inhibit the release proopiomelanocortin (POMC)-related peptides from the hypothalamus and to decrease mRNA levels in POMC neurons in the arcuate nucleus. To learn more about the precise role of the GABAergic system in POMC neuron regulation, we studied the effects of the administration (2 days) of the GABAA receptor agonist muscimol and the central-type barbiturate receptor agonist pentobarbital on POMC mRNA levels measured by in situ hybridization. Treatment with pentobarbital produced a 12% decrease in the hybridization signal. Similarly, muscimol treatment decreased the signal by 20%. The concomitant administration of the two GABAA receptor agonists resulted in a decrease (28%) of mRNA levels that was significantly more marked than that induced by pentobarbital or muscimol alone. The present results, together with previous data from our laboratory, indicate that different activators of the GABAA receptor complex, including barbiturates, can negatively regulate POMC neuronal activity in the rat arcuate nucleus.

Effect of acute stress on the expression of hypothalamic messenger ribonucleic acids encoding the endogenous opioid precursors preproenkephalin A and proopiomelanocortin

In addition to the release of adenohypophysial hormones adrenocorticotropin and beta-endorphin, most types of acute stress induce rapid release of prolactin (PRL) from the anterior pituitary lobe. Endogenous opioid peptides are believed to participate in the stress-induced PRL secretion via an action within the central nervous system. In the present study, we have investigated the effect of acute stress on anterior pituitary PRL secretion and the hypothalamic expression of messenger ribonucleic acids (mRNAs) encoding precursors of the endogenous opioids Met-enkephalin and beta-endorphin. Adult male rats were subjected to 1 h of restraint and the stress-induced rise in plasma PRL was measured both during and after termination of the stress paradigm. Using quantitative in situ hybridization histochemistry, it was observed that levels of proenkephalin A mRNA increased significantly within the medial parvocellular subset of the hypothalamic paraventricular nucleus, and within the arcuate nucleus levels of proopiomelanocortin (POMC) mRNA were slightly, but significantly, increased. The employed stress paradigm also induced an elevation of anterior pituitary levels of PRL and POMC mRNAs. The present data suggest that restraint stress activates gene expression of endogenous opioid systems in the hypothalamus and that the employed stress paradigm is of sufficient magnitude to stimulate both mRNA expression and release of PRL in the anterior pituitary lobe.

Opioid regulation of proopiomelanocortin (POMC) gene expression in the rat brain as studied by in situ hybridization

Proopiomelanocortin (POMC) is the precursor of the potent opioid peptide beta-endorphin as well as a number of other active peptides. On the basis of neuroanatomical data indicating the presence of contacts between POMC neurons in the rat arcuate nucleus, it has been proposed that POMC neurons could be autoregulated. In order to investigate the role of opiates in the regulation of POMC gene expression in the rat arcuate nucleus, we studied the effects of chronic administration of the opioid drug morphine and an opiate receptor antagonist naloxone on POMC mRNA levels as measured by in situ hybridization, 4-day treatment with naloxone (4 mg/kg/day) produced a 60% increase in the number of silver grains overlying POMC neurons. Conversely, morphine (40 mg/kg/day) also administered during 4 days decreased the hybridization signal by 30%. The concomitant administration of morphine and naloxone completely prevented the effect of morphine on POMC gene expression indicating that the inhibitory influence of morphine is likely to be mediated by opioid receptors. The data obtained clearly indicate that activation of opioid receptors decreased the biosynthetic activity of POMC neurons and that conversely opiate receptor blockade caused an increase in the activity of these neurons. They are consistent with the hypothesis of an autoregulation of the POMC neuronal system by endogenous opiate peptide(s).

Regulation of proopiomelanocortin gene expression in rat brain and pituitary as studied by in situ hybridization

The major sites of POMC gene expression are the intermediate and anterior lobes of the pituitary and the arcuate nucleus of the hypothalamus. We have investigated the regulation of POMC mRNA levels in the pituitary and hypothalamus by quantitative in situ hybridization using a 35S-labeled cDNA probe encoding or POMC. In the arcuate nucleus, where the POMC-producing neurons are concentrated, adrenalectomy induced a marked decrease in POMC mRNA levels, an effect that was completely reversed by dexamethasone administration. The stimulating effect of dexamethasone was much more striking in the most caudal regions of the nucleus. Since in the anterior pituitary, glucocorticoids exert an inhibitory action on POMC gene expression, it might be suggested that POMC is differentially regulated by glucocorticoids in the anterior pituitary and the hypothalamus. Ovariectomy induced an increase in mRNA levels in the most rostral region of the arcuate nucleus, an effect that was prevented by the concurrent administration of estradiol or dihydrotestosterone. The role of dopamine was investigated following the administration of the dopaminergic antagonist haloperidol and the D2 dopaminergic receptor agonist bromocriptine. In the arcuate nucleus, bromocriptine increased and haloperidol decreased the hybridization signal. Conversely, in the intermediate lobe of the pituitary, bromocriptine markedly depressed and haloperidol increased the levels of mRNA. These results indicate that the population of POMC neurons in the arcuate nucleus is heterogeneous. They also demonstrate that POMC gene expression is regulated by central and peripheral factors and that this regulation is different in the brain and the pituitary.

Role of dopamine in the regulation of proopiomelanocortin (POMC) mRNA levels in the arcuate nucleus and pituitary gland of the female rat as studied by in situ hybridization

The effects of the dopaminergic antagonist haloperidol (HAL) as well as the D2 dopamine receptor agonist bromocriptine (BRO) on proopiomelanocortin (POMC) mRNA levels in the female rat arcuate nucleus and pituitary were investigated by quantitative in situ hybridization. Since we had already shown that sex steroids could induce a decrease in POMC mRNA levels in the arcuate nucleus of castrated rats, the involvement of the dopaminergic system in the inhibitory effect of estradiol (E2) was also investigated. In situ hybridization was performed on paraformaldehyde-fixed cryostat sections through the arcuate nucleus and whole pituitary gland using a 35S-labelled cDNA probe encoding for POMC. In the arcuate nucleus of intact animals, a 14-day treatment with BRO increased by 54% the number of silver grains/unit of surface of labelled neurons while HAL decreased by 30% the value of this parameter. Hypophysectomy which induced a 20% decrease in the hybridization signal could not prevent the effects of BRO or HAL. Ovariectomy performed 14 days earlier increased by 20% the number of silver grains while a 14-day treatment of ovariectomized animals with E2 decreased the hybridization signal by 32%. On the other hand, the concomitant administration of HAL and E2 did not induce significant variations in POMC mRNA levels compared to those obtained following HAL administration, but slightly decreased the hybridization signal by 20% compared to that induced by E2 alone. In the intermediate lobe of the pituitary, BRO markedly depressed (30% of control values) and HAL increased by 50% the levels of POMC mRNA. The present data clearly demonstrate that POMC mRNA levels are differently regulated by dopamine in the intermediate lobe of the pituitary and the arcuate nucleus and that the effects of BRO and HAL on arcuate POMC mRNA are not mediated by the pituitary gland. They do not allow to draw any definite conclusion about the possible involvement of the dopaminergic system in the inhibitory role of E2 on POMC gene regulation.