Estradiol is concentrated in tyrosine hydroxylase-containing neurons of the hypothalamus.

Localization of [3H]estradiol in tyrosine hydroxylase-containing neurons of rat brain was shown by a combined technique of autoradiography and immunohistochemistry. [3H]Estradiol was concentrated in the nuclei of tyrosine hydroxylase-containing neurons in the nucleus arcuatus, nucleus periventricularis hypothalami, and the zona incerta. These results suggest that estradiol acts directly on dopamine-producing neurons of the tuberoinfundibular system and incertohypothalamic system.

Estrogen decreases rat hypothalamic proopiomelanocortin messenger ribonucleic acid levels.

Ovariectomy and estrogen (E) or E plus progesterone treatment has previously been shown to alter both hypothalamic content and portal plasma levels of beta-endorphin. To determine if these changes were accompanied by changes in beta-endorphin synthesis, we used a RNA dot blot method to quantify proopiomelanocortin (POMC) mRNA levels in the arcuate nucleus-median eminence region of rats. Animals were bilaterally ovariectomized, implanted with Silastic capsules containing E or oil, and killed 1 or 3 days after implantation. Total nucleic acid was isolated from dissections of the arcuate-median eminence by proteinase-K/sodium dodecyl sulfate/phenol extraction, and POMC mRNA was quantified by dot blot analysis. Although 1 day of E treatment had no effect on hypothalamic POMC mRNA levels, 3 days of E treatment caused a significant reduction of approximately 40% of POMC mRNA levels relative to oil controls in two replicate experiments. These results suggest that the decreases in hypothalamic POMC peptide levels after E administration reported previously may be due to a decrease in POMC peptide biosynthesis resulting from a decrease in hypothalamic POMC mRNA.

Testosterone regulation of proopiomelanocortin messenger ribonucleic acid in the arcuate nucleus of the male rat.

GnRH regulates the secretion of LH and FSH, which stimulate the secretion of testicular hormones. Acting in a reciprocal fashion, these hormones, including testosterone and inhibin, exert a negative feedback effect on GnRH and gonadotropin secretion. Endogenous opioid peptides (EOPs) have been implicated to play a role in steroid-mediated regulation of gonadotropin secretion. In this context, certain steroid hormones (e.g. testosterone) increase EOP activity and ultimately inhibit GnRH secretion; however, the cellular mechanism by which this occurs is unknown. beta-Endorphin is one of these EOPs, and it is derived from a larger precursor molecule, POMC. We tested the hypothesis that testicular hormones and testosterone, in particular, stimulate POMC gene expression in the arcuate nucleus of the male rat brain. First, we compared POMC mRNA levels between intact and castrated male rats. Adult male rats were killed 4 days (n = 4) and 21 days (n = 5) after castration. Intact animals (sham-operated; n = 6) were used as controls. Using in situ hybridization and a computerized image analysis system, we measured the POMC mRNA content in individual cells of the arcuate nucleus. POMC mRNA signal was significantly lower (P less than 0.0003) in both 4-day (126 +/- 2 grains/cell) and 21-day (117 +/- 5 grains/cell) castrates than in controls (142 +/- 2 grains/cell). In a second experiment we tested whether testosterone would reverse the castration-induced loss of POMC message. Again, we castrated animals and immediately implanted them with either empty (sham; n = 6) or testosterone-containing Silastic implants (n = 5) of a size that would deliver physiological levels of testosterone (3.6 +/- 1.5 ng/ml). We observed that testosterone-treated animals had significantly higher levels of POMC mRNA signal (121.8 +/- 3.8 grains/cell) than sham-treated castrates (111.4 +/- 3.6 grains/cell; P less than 0.03) and that the testosterone-treated castrates had POMC mRNA signal levels indistinguishable from those of intact controls (122.0 +/- 1.1 grains/cell). These observations lend credence to the theory that one mechanism by which testosterone may regulate GnRH secretion is by increasing the synthesis of POMC in the arcuate nucleus.

Pubertal changes in gonadotropin-releasing hormone and proopiomelanocortin gene expression in the brain of the male rat.

Pubertal development in mammals is in part attributable to a brain-dependent process, whereby increased pulsatile GnRH secretion leads to the awakening of the entire reproductive system. However, the brain mechanisms controlling this event are unknown. The apparent increase in GnRH secretion at puberty could reflect an autonomous change in the activity of GnRH neurons themselves or in the afferent networks leading to GnRH neurons. If there were a significant increase in the secretion of GnRH with puberty onset, we hypothesized that there would be a commensurate increase in the biosynthetic capacity of GnRH neurons to meet the increasing demand. We tested this hypothesis by comparing the level of cellular prepro-GnRH mRNA (GnRH mRNA) observed between prepubertal (25-day-old; n = 5) and adult (75-day-old; n = 4) male rats by in situ hybridization. We detected no significant change with puberty in GnRH mRNA signal levels in any of the anatomical areas examined, which included the vertical limb of the diagonal band of Broca, medial septum, lateral preoptic area, and medial preoptic area. Given the variance of our analytical technique, we determined that there was a greater than 90% probability that we would have detected a 20% increase in GnRH mRNA had there been one. Endogenous opioid peptides have been implicated in timing the onset of puberty in the rat, with the argument being that a loss in opioid tone could effect a disinhibition of GnRH secretion. One opioid peptide, beta-endorphin, is among several peptides cleaved from the precursor POMC. We hypothesized that with puberty, POMC neurons in the arcuate nucleus would have an attenuated capacity to produce beta-endorphin. We tested this hypothesis by comparing cellular pre-POMC mRNA (POMC mRNA) levels in the arcuate nuclei of prepubertal (n = 6) and adult (n = 7) male rats with in situ hybridization. We observed an increase in POMC mRNA levels with puberty; prepubertal rats had relative POMC mRNA signal levels of 119 +/- 10 grains/cell, while adult rats contained 167 +/- 12 grains/cell (P less than 0.02). This increase in cellular POMC mRNA was confined to the rostral portion of the arcuate nucleus. We conclude that the GnRH gene is fully expressed well before the time of normal puberty onset and that the increase in POMC mRNA that occurs with the onset of puberty may be important for the development of pulsatile GnRH secretion.

Adrenocorticotropic hormone (ACTH) in pituitary colloid and extrapituitary tissues.

The level of ACTH in bovine pituitary intraglandular colloid (IGC/ACTH), the holocrine secretion of the marginal half of the intermediate lobe (IL) cells, was found to correlate with the concentration of this peptide in blood plasma (BP/ACTH), cerebrospinal fluid (CSF/ACTH), the paraventricular nuclei (PVN/ACTH) and the arcuate nuclei (AN/ACTH). The ACTH content in all sites was measured by radioimmunoassay (RIA). Although the IL is virtually avascular, the intraglandular lumen housing the IGC communicates nonetheless with the venous cavernous sinuses as well as the subarachnoid cerebral spinal fluid (CSF). Therefore, we content that the IGC serves as a transport medium whereby IL materials gain access to these extrapituitary sites.

Topographical and ontogenetic study of the neurons producing growth hormone-releasing factor in human hypothalamus.

Neurons producing growth hormone-releasing factor have been characterized and analyzed by immunohistochemistry in the hypothalami of human fetuses, neonates, infants and adults, using two antibodies against human pancreatic GRF (hpGRF). One of the antibodies recognized both the hpGRF(1-40)OH and hpGRF(1-44)NH2 in the mid portion (between the 28th and 39th amino acid), the other one specifically recognized the C-terminal end of hpGRF(1-44)NH2. These two antibodies stain a single neuronal system with cell bodies mainly located in the infundibular (arcuate) nucleus, and in the ventromedial and lateralis tuber nuclei. These neurons project to the median eminence where they give numerous endings in contact with portal vessels. These neurons are distinct from those containing LH-RH, somatostatin, CRF or pro-opiocortin. In fetuses, neurons immunoreactive with hpGRF antibodies are first detected at the 29th week. They display a neuroblastic aspect which persists after birth. Immunoreactive fibers are detectable in the median eminence after the 31st week. These results demonstrate that the infundibular nucleus plays a major role in control of GH secretion in man and that secretion of GRF appears late during fetal life; this suggests that the first stages of differentiation and development of GH producing cells in the human fetus do not depend on hypothalamic GRF secretion.

Immunocytochemical localization of progesterone receptor in the guinea pig central nervous system.

The distribution of neurons containing progesterone receptors in the brain of guinea pigs ovariectomized and primed by estradiol was investigated immunohistochemically using monoclonal antibodies to the progesterone receptor. We found that the picric acid paraformaldehyde perfusion provided satisfactory conditions of fixation for visualizing the progesterone receptor in sections of frozen tissue. Among the different techniques of immunocytochemical detection used, the indirect antibody peroxidase-antiperoxidase method gave the best results. The immunostained neurons were mainly located in two specific regions of the hypothalamus: the preoptic area and the mediobasal hypothalamus. Within the preoptic region, the majority of immunoreactive cells were present in the nucleus preopticus periventricularis particularly at the level of the pars suprachiasmatica. Within the mediobasal hypothalamus, immunostained neurons were found in the nucleus periventricularis, arcuatus, ventromedialis and premamillaris. Differences in the intensity of immunoreactivity appeared from one region to another. A marked cellular heterogeneity was observed: in each neuroanatomical structure, labeled cells alternated with non-labeled cells. The receptor, even in absence of progesterone, was localized in the nucleus. The nucleolus was not stained and only neurons were labeled. There was no progesterone receptor immunoreactivity in other regions of the brain, especially in the amygdala, hippocampus and cortex where biochemical studies have shown the presence of a non-estrogen regulated protein binding the progestin [3H]R 5020. Control experiments with antibody pretreated with purified progesterone receptor or with mouse receptor-unrelated monoclonal antibody did not show fluorescent or immunoreactive cells.

Light and electron microscopic immunocytochemistry of beta-endorphin/beta-LPH-like immunoreactive neurons in the arcuate nucleus and surrounding areas of the rat hypothalamus.

beta-Endorphin/beta-LPH-like immunoreactive neurons in the hypothalamic arcuate nucleus and its surrounding areas were visualized by light and electron microscopic immunocytochemistry. Immunoreactive processes were found in the vicinity of the pia mater, in the lateral part of the external layer of the median eminence and near the lateral wall of the third ventricle. Neuronal perikarya contained immunoreactive dense granules as well as developed cell organellae. They received neuronal inputs from other neurons through axoplasmic and axodendritic synapses. Immunoreactive neuronal processes containing dense granules and mitochondria were found as preterminal elements on non-immunoreactive neuronal soma and dendrites. Immunoreactive processes also make intimate contact with capillaries in the arcuate nucleus near the median eminence.

Effects of sulpiride treatment on the levels of immunoreactive beta-endorphin in rat hypothalamic nuclei.

The effects of sulpiride, a specific dopamine receptor antagonist, on the levels of immunoreactive beta-endorphin (ir-beta-END) in discrete hypothalamic nuclei of rats were investigated. Single injections of sulpiride significantly decreased the levels of ir-beta-END in the nucleus arcuatus, nucleus paraventricularis and median eminence. Repeated injections of sulpiride significantly increased the levels of ir-beta-END in the nucleus arcuatus, nucleus periventricularis and median eminence. These findings suggest that beta-endorphinergic neurons are under dopaminergic control.

Ultrastructural morphometric analysis of somatostatin-like immunoreactive neurones in the rat central nervous system after labelling with colloidal gold.

Nerve terminals of the rat median eminence, arcuate nucleus and spinal cord were examined in the electron microscope after post-embedding, colloidal gold labelling of immunoreactivity to somatostatin. Strong immunostaining was thus obtained together with adequate morphological preservation. Reactive boutons showed clusters of gold particles essentially confined to dense-cored vesicles. In the median eminence, the positive varicosities made up more than half of all terminals and averaged 735 nm in diameter. Those in the arcuate nucleus and spinal cord were much less numerous and generally smaller (575 nm). The labelled vesicles had mean external diameters of 109, 95 and 79 nm in the median eminence, arcuate nucleus and spinal cord, respectively. Calculations of the likely amounts of somatostatin within the vesicles of the median eminence and arcuate nucleus yielded values of 0.7 and 1.4 mM, corresponding to 190 and 230 molecules of the peptide, respectively. These data support a neurotransmitter or modulator role for somatostatin in the central nervous system.

Adrenaline concentration and turnover in the arcuate nucleus and median eminence during the critical period in the rat.

Previous work has shown a relatively high turnover of adrenaline in the mediobasal hypothalamus during the critical period (15.00-17.00 h) of the proestrous rat. We now report that this high level of adrenergic activity can be detected in the median eminence (turnover rate 1.62 +/- 0.36 pg/micrograms protein/h) rather than the arcuate nucleus (turnover rate 0.18 +/- 0.32 pg/micrograms protein/h). In addition the median eminence was isolated as medial and lateral components and determination of catecholamine concentrations revealed a greater proportion of adrenaline (A) (59%) in the lateral median eminence whereas a larger proportion of dopamine (60%) was found in medial median eminence.

Origin of serotonin innervation of the arcuate and ventromedial hypothalamic region.

Combined fluorescence serotonin immunohistochemistry and retrograde transport labelling with Fast blue and Fluoro-gold were used to identify serotonin-immunoreactive neurons in the midbrain and pons which project to the region of the arcuate and ventrome-dial hypothalamic nuclei. Approximately 90% of doubly labelled neurons were located in the 3 major mesencephalic serotonin-containing cell groups: dorsal raphe (38%), median raphe (21%) and medial lemniscus group (29%). Within these groups, there were numerous non-retrogradely labelled serotonin-immunoreactive neurons as well as numerous non-serotonin-immunoreactive retrogradely labelled neurons. No doubly labelled neurons were observed caudal to raphe pontis although non-serotonin-immunoreactive neurons were retrogradely labelled in the more caudal raphe nuclei.

Serotonin innervation of neuropeptide Y-containing neurons in the rat arcuate nucleus.

Structural non-synaptic appositions between serotonin (5-HT) nerve endings and neuropeptide Y (NPY)-containing neurons were demonstrated in the rat arcuate nucleus by means of a combined radioautographic and immunocytochemical detection of [3H]5-HT uptake sites and NPY-immunoreactivity. Such cellular relationships are proposed to constitute a morphological substrate for putative 5-HT/NPY interactions in neuroendocrine hypothalamus.

Lesions of the hypothalamic arcuate nucleus modify discrete brain and pituitary pools of dynorphin in addition to beta-endorphin in the rat.

Radiofrequency destruction of the hypothalamic arcuate nucleus, the exclusive central location of beta-endorphin (beta-EP)-synthesizing neurones, greatly depressed levels of immunoreactive (ir)-beta-EP in the hypothalamus. However, the content of ir-dynorphin (DYN) therein was also decreased, in parallel with that of ir-beta-EP. Levels of ir-Met-enkephalin (ME) were, in contrast, unaffected. In the midbrain and septum, the content of ir-beta-EP was greatly depleted; in the former but not latter tissue, levels of ir-DYN were also diminished whilst levels of ir-ME were not affected in either of these structures. Further, in the neurointermediate pituitary, the content of ir-beta-EP was depressed while that of ir-DYN was not changed, whereas, in the anterior pituitary, levels of ir-beta-EP were not altered in contrast to those of ir-DYN which were decreased. Destruction of the arcuate nucleus is not, therefore, restricted to selective effects upon central beta-EP neurones: this finding is of importance for studies of the role of central beta-EP via lesioning techniques. The data suggest an interaction of the arcuate nucleus and, possibly, beta-EP with particular brain and pituitary pools of DYN.

Demonstration of gastrin/cholecystokinin-like immunoreactivity in the arcuate nucleus, median eminence and pituitary of the guinea pig.

In the brains of colchicine-treated guinea pigs, antibodies directed against the C-terminal sequence of cholecystokinin octapeptide (CCK-8) and cross-reacting with gastrin, visualized a dense population of cell bodies in the arcuate nucleus, periportal nerve endings in the external layer of the median eminence and a few fibers in the neurohypophysis. In the adenohypophysis, at least the major part of corticomelanotrophs were labelled. Immunostainings are compared to those reported in the rat.

Decreased tuberoinfundibular and pituitary dopamine and dihydroxyphenylacetic acid concentrations in rats with estrogen-induced pituitary tumors.

The majority of pituitary adenomas are prolactin (PRL)-secreting, but it is still uncertain whether their pathogenesis results from a central nervous system (CNS) disturbance or autonomous lactotroph growth and function. We have measured dopamine (DA) and dihydroxyphenylacetic acid (DOPAC) concentrations in rats bearing estradiol-induced PRL-secreting pituitary tumors. Female rats, injected at 3-week intervals with 2 mg estradiol valerate (EV), had increased plasma prolactin concentrations, up to 3 micrograms/ml, at 31 weeks. Inversely, there was a reduction of DA and DOPAC in the median eminence and arcuate nucleus as well as a decreased DOPAC/DA ratio in the arcuate nucleus. DA-containing nuclei in the other parts of the brain were not affected. Anterior pituitary weight increased while its DA concentration decreased during estradiol treatment. In the neurohypophysis, DA concentrations were unchanged while DOPAC and the ratio DOPAC/DA decreased following the estrogen treatment. Our data suggest that rats with primary estrogen-induced PRL-secreting tumors have a defect in the CNS-DA neurotransmission.

Lateral hypothalamic innervation of the cerebral cortex: immunoreactive staining for a peptide resembling but immunochemically distinct from pituitary/arcuate alpha-melanocyte stimulating hormone.

The combination of retrograde transport of fluorescent dyes and indirect immunofluorescence has been used to study the putative neurotransmitter specificity of the tuberal lateral hypothalamic projection to the cerebral cortex. Injections of either fast blue or diamidino yellow dye into the cerebral cortex or hippocampus retrogradely labeled large, multipolar neurons scattered through the lateral hypothalamic area and zona incerta at the level of the ventromedial nucleus of the hypothalamus. Approximately 80% of these neurons stained immunohistochemically with an antiserum against alpha-melanocyte stimulating hormone (alpha-MSH). A second population of smaller, predominantly bipolar alpha-MSH-like immunoreactive neurons was seen in the arcuate nucleus and retrochiasmatic area, but none of these projected to the cerebral cortex. Immunohistochemical staining for ACTH (18-24), another proopiomelanocortin series peptide, or with an antiserum against alpha-MSH (4-10) demonstrated only the second of these cell groups. Our results indicate that the tuberal lateral hypothalamic projection to the cerebral cortex contains a substance similar but not identical to alpha-MSH, and that this material is probably not derived from the same proopiomelanocortin precursor as true alpha-MSH.

Differential alterations in opioid analgesia following neonatal monosodium glutamate treatment.

Neonatal administration of monosodium glutamate (MSG) produces necrosis of circumventricular structures, including perikarya in the medial-basal hypothalamus that contain beta-endorphin (BEND) and met-enkephalin. Since neonatal MSG treatment alters morphine analgesia, the present study examined neonatal MSG effects upon opioid analgesia observed following either BEND or d-ala d-leu enkephalin (DADL). Rats treated with either MSG or vehicle over the first ten post-natal days, were surgically prepared with a lateral ventricle cannula at 100 days of age. Respective groups received central injections of either BEND (0, 0.1, 0.5 or 1.0 microgram) or DADL (0, 4, 20 or 40 micrograms), and jump thresholds were assessed 15, 30, 45 and 60 min thereafter. Following testing, selected MSG-treated and control animals were prepared for BEND immunocytochemistry. While the magnitude, duration and sensitivity of BEND analgesia on the jump test failed to differ between groups, MSG-treated rats displayed a 10-fold leftward shift in sensitivity and a 200-300% increase in the magnitude of DADL analgesia. Immunocytochemical analysis indicated that MSG treatment depleted perikarya in the medial-basal hypothalamus, periventricular thalamic fibers and periaqueductal gray terminal fields that contained BEND. The differential effects of MSG treatment upon opiate and opioid analgesia are discussed in terms of possible alterations in opiate receptor subpopulations.

Prenatal stress impairs maternal behavior in a conflict situation and reduces hippocampal benzodiazepine receptors.

Maternal behavior (pup retrieval) was assessed in prenatally stressed rats during control and conflict situations (having to pass through an airstream) when their pups were 4-5 days old. There was no difference in pup retrieval between experimental and control rats under normal conditions but only 52% of the former retrieved their pups during the conflict situation, compared with 96% of the controls. Catecholamine (CA) levels in the arcuate nucleus (Arc.n.) and noradrenaline in the medial preoptic nucleus (POM) were not altered in prenatally stressed females, but their dopamine levels in the POM tended to be lower (p less than 0.1). The number of benzodiazepine (BZ) receptors in the hippocampi of prenatally stressed females was significantly lower than in controls. We conclude from these results that random prenatal noise and light stress increases the vulnerability to stressful situations in the female offspring during adulthood, which may be accompanied by altered CA function in the hypothalamus and BZ binding in the hippocampus.

Endogenous opioid peptides and the control of the menstrual cycle.

This paper reviews recent experimental evidence which supports a role for endogenous opioid peptides in the control of gonadotropin function. In primates, cell bodies containing endogenous opioid peptides have been located within the hypothalamus in areas rich in gonadotropin-releasing hormone (GnRH) and dopamine. The release of beta-endorphin from these hypothalamic neurons is influenced by gonadal steroids, maximal release being observed when both estradiol and progesterone are present. beta-Endorphin has been shown to decrease LH secretion, and naloxone, an opiate antagonist, reverses this action. The LH-releasing activity of naloxone parallels variations in hypothalamic beta-endorphin secretory activity, so that maximal effects are seen during the luteal phase of the cycle. Present evidence indicates that opiates exert their effect on LH via a hypothalamic site. It is concluded that increased opioid inhibition of the GnRH-LH axis is responsible for the decline in LH pulse frequency during the luteal phase. The studies provide evidence for a chemical basis rationalizing relationships between reproductive function and stress, and have further implication on other forms of amenorrhea.