Origin of adenohypophysial lobes and cells from Rathke's pouch in Swiss albino mice. Proliferation and expression of Pitx 2 and Calbindin D28K in corticotropic and somatotropic cell differentiation.

A developmental study of the adenohypophysis of the mouse was carried out in response to several as yet unanswered questions about its organogenesis and differentiation. The main objectives were to establish the origin of adenohypophysial lobes and cells from the Rathke's pouch (RP) and elucidate the mechanisms of development and functional differentiation of the gland. Using diverse techniques, the morphological development, proliferation and differentiation were studied in order to delimit different proliferative regions in the RP, and provide a satisfactory explanation for the distribution of each cell type in the adult gland. Combining the proliferation and differentiation studies, corticotropic and somatotropic cells appear to mainly originate from undifferentiated precursors located within each of these proliferative regions. The involvement of transcription factor Pitx 2 and calcium-binding protein Calbindin D 28K in the differentiation of corticotropic and somatotropic cells is further clarified.

Immunohistochemical localization of hormones and peptides in the human pituitary cells in a case of hypercortisolism by ACTH secreting microadenoma.

This study assesses the action of hypercortisolism on the hormone and peptide periadenoma region of removed ACTH-producing microadenoma. Our findings show that cortisol excess affects both ACTH and GH production, with no immunoreaction for these hormones. The remaining pituitary hormones (TSH, FSH and PRL) and POMC-derived peptides (betaEnd, alphaMSH and betaMSH) were not modified. Likewise, we observed pituitary immunoreactive cells for Neurotensin (NT), Intestinal vasoactive peptide (VIP), Substance P (SP) and Angiotensin-II (Ang-II). The colocalization demonstrated that NT was expressed in thyrotrope and gonadotrope cells, VIP in gonadotrope cells and SP in corticotrope cells. The results about Ang-II were inconclusive. On the other hand, immunoreaction for the NPY and Gal peptides were not present. In the adenomatous cells, the peptide NT is present in ACTH cells as well as SP. These results suggest a peptide regulation of pituitary cells in the pathological state that can differ between normal and tumoural cells of the same pituitary.

An androgen-dependent sexual dimorphism visible at puberty in the rat hypothalamus.

Morphological studies in rodents have well documented the masculinization of the perinatal brain by estradiol derived from aromatized testosterone, and the resulting irreversible quantitative sex-differences generated in cell numbers or expression of chemical phenotypes. Here, using immunohistochemistry, we explored how this applies to the postnatal development and masculinization of the neurokinin B (NKB)-containing system of the arcuate nucleus/median eminence complex (ARC/ME). In adult rats, NKB-immunoreactive neurons exhibit an unusual, qualitative sexual dimorphism of their ventral axonal projections: to the neuropil in females, to capillary vessels in males. In adults, there was no sex-difference in the numbers of NKB-immunoreactive perikarya or capillary vessels in the ARC/ME, suggesting that this sexual dimorphism cannot be explained by the existence of supernumerary structures. At birth (day 0) the NKB system was immature in both sexes, and while its adult features emerged progressively until puberty in females, they did not develop before puberty (day 40) in males, revealing a sexual dimorphism only late postnatally. When males were orchidectomized at day 30, the masculine distribution of NKB-immunoreactive axons expected at day 40 was not seen, while it was apparent after chronic treatment with testosterone or dihydrotestosterone, suggesting a testicular masculinizing action via androgen receptors at puberty. Moreover in these prepubertal-orchidectomized males, the distribution of NKB-immunoreactive axons was surprisingly feminized by chronic estradiol alone, suggesting that NKB neurons are not irreversibly programmed before puberty. Last, in adult females, the distribution of NKB-immunoreactive axons was feminine 30 days after ovariectomy, and it was masculinized after concurrent chronic dihydrotestosterone, suggesting that NKB neurons remain responsive to androgens late in reproductive life. Thus, the sexual differentiation of the hypothalamus proceeds well beyond the perinatal period and includes the epigenetic action of non-aromatizable androgens upon subsets of neurons that have retained bipotent features.

Immunocytochemical detection of cholecystokinin and corticotrophin-releasing hormone neuropeptides in the hypothalamic paraventricular nucleus of the jerboa (Jaculus orientalis): modulation by immobilisation stress.

The hypothalamic response to an environmental stress implicates the corticotrophin-releasing hormone (CRH) neuroendocrine system of the hypothalamic parvicellular paraventricular nucleus (PVN) in addition to other neuropeptides coexpressed within CRH neurones and controlling the hypothalamo-pituitary-adrenal (HPA) axis activity as well. Such neuropeptides are vasopressin, neurotensin and cholecystokinin (CCK). It has previously been demonstrated that the majority of the CRH neuronal population coexpresses CCK after a peripheral stress in rats. In the present study, we explored such neuroendocrine plasticity in the jerboa in captivity as another animal model. In particular, we studied CCK and CRH expression within the hypothalamic PVN by immunocytochemistry in control versus acute immobilisation stress-submitted jerboas. The results show that CCK- and CRH-immunoreactive neuronal systems are located in the hypothalamic parvicellular PVN. The number of CCK-immunoreactive neurones within the PVN was significantly increased (138% increase) in stressed animals compared to controls. Similarly, the number of CRH-containing neurones was higher in stressed jerboas (128%) compared to controls. These results suggest that the neurogenic stress caused by immobilisation stimulates CCK as well as CRH expression in jerboas, which correlates well with previous data obtained in rats using other stressors. The data obtained also suggest that, in addition to CRH, CCK is another neuropeptide involved in the response to stress in jerboa, acting by controlling HPA axis activity. Because CCK is involved in the phenotypical plasticity of CRH-containing neurones in response to an environmental stress, we also explored their coexpression by double immunocytochemistry within the PVN and the median eminence (i.e. the site of CRH and CCK corelease in the rat) following jerboa immobilisation. The results show that CCK is not coexpressed within CRH neurones in either control or stressed jerboa, suggesting differences between jerboas and rats in the neuroendocrine regulatory mechanisms of the stress response involving CRH and CCK. The adaptative physiological mechanisms to environmental conditions might vary from one mammal species to another.

Sexual dimorphism in the organization of the rat hypothalamic infundibular area.

The hypothalamic infundibular area is located outside the blood-brain barrier and includes, the ventromedial arcuate nucleus (vmARC) sensing circulating substances, and the median eminence (ME) where neurohormones are released into the hypothalamo-hypophysial vasculature. This integrated functional unit, pivotal in endocrine control, adjusts neuroendocrine output to feedback information. Despite a differing physiology in males and females, this functional unit has not appeared differently organized between sexes. Using immunocytochemistry, we describe here for the first time in adult rats, a conspicuous sex-difference in its axonal wiring by intrinsic glutamatergic neurons containing the neuropeptides neurokinin B (NKB) and dynorphin. In the male, NKB neurons send axons to capillary vessels of the vmARC and of the ME (only where gonadotropin-releasing hormone (GnRH) axons terminate). Electron microscopy revealed that NKB axons target the barrier of tanycytes around fenestrated capillary vessels (in addition to GnRH axons), suggesting a control of regional bidirectional permeability. In the female, NKB neurons send axons to the neuropile of the vmARC, suggesting a direct control of its sensor neurons. The other projections of NKB neurons, studied by surgical isolation of the ARC-ME complex and confocal microscopy, are not sexually dimorphic and target both integrative and neuroendocrine centers controlling reproduction and metabolism, suggesting a broad influence over endocrine function. These observations demonstrate that the mechanisms subserving hypothalamic permeability and sensitivity to feedback information are sexually dimorphic, making the infundibular area a privileged site of generation of the male-to-female differences in the adult pattern of pulsatile hormonal secretions.

Vasopressin-containing neurons of the hypothalamic parvocellular paraventricular nucleus of the jerboa: plasticity related to immobilization stress.

The corticotropin-releasing hormone (CRH) neurons of the hypothalamic parvocellular paraventricular nucleus (PVN) have a high potential for phenotypical plasticity, allowing them to rapidly modify their neuroendocrine output, depending upon the type of stressors. Indeed, these neurons coexpress other neuropeptides, such as cholecystokinin (CCK), vasopressin (VP), and neurotensin, subserving an eventual complementary function to CRH in the regulation of the pituitary. Unlike in rats, our previous data showed that in jerboas, CCK is not coexpressed within CRH neurons in control as well as stressed animals. The present study explored an eventual VP participation in the phenotypic plasticity of CRH neurons in the jerboa. We analyzed the VP expression within the PVN by immunocytochemistry in male jerboas submitted to acute stress. Our results showed that, contrary to CRH and CCK, no significant change concerned the number of VP-immunoreactive neurons following a 30-min immobilization. The VP/CRH coexpression within PVN and median eminence was investigated by double immunocytochemistry. In control as well as stressed animals, the CRH-immunopositive neurons coexpressed VP within cell bodies and terminals. No significant difference in the number of VP/CRH double-labeled cells was found between both groups. However, such coexpression was quantitatively more important into the posterior PVN as compared with the anterior PVN. This suggests an eventual autocrine/paracrine or endocrine role for jerboa parvocellular VP which is not correlated with acute immobilization stress. VP-immunoreactive neurons also coexpressed CCK within PVN and median eminence of control and stressed jerboas. Such coexpression was more important into the anterior PVN as compared with the posterior PVN. These results showed the occurrence of at least two VP neuronal populations within the jerboa PVN. In addition, the VP expression did not depend upon acute immobilization stress. These data highlight differences in the neuroendocrine regulatory mechanisms of the stress response involving CRH/CCK or VP. They also underline that adaptative physiological mechanisms to stress might vary from one mammal species to another.

Mapping of tachykinins in the cat spinal cord.

Using an indirect immunoperoxidase technique, the location of cell bodies and fibers containing substance P, neurokinin A and neurokinin B was studied in the cat spinal cord. The former two neuropeptides showed a widespread distribution throughout the whole spinal cord, whereas the distribution of neurokinin B was more restricted. Neurokinin A-immunoreactive structures showed a more widespread distribution and a higher density than the immunoreactive structures observed to contain substance P. In the cat spinal cord, we observed cell bodies containing neurokinin A, but no cell bodies containing neurokinin B or substance P were found. These cell bodies were located in laminae V (sacral 1 and 2 levels), VI (sacral 1 and 3), VII (lumbar 7, sacral 1 and 3, caudal 1) and X (sacral 1). Laminae I and II showed the highest density of immunoreactive fibers for each of the three tachykinins studied, being in general lamina IV who showed the lowest number of immunoreactive fibers containing substance P, neurokinin A or B. The anatomical distribution of the three tachykinins studied in the cat spinal cord indicates that the neuropeptides could be involved in the neurotransmission and/or in the neuromodulation of nociceptive information, as well as in autonomic and affective responses to pain. Moreover, the involvement of substance P, neurokinin A or B in other functions unrelated to the transmission of pain is also possible (autonomic and motor functions). The distribution of the neuropeptides studied in the cat is compared with the location of the same neuropeptides in the spinal cord of other species. The possible origin of the tachykinergic fibers in the cat spinal cord is also discussed.

Oestradiol effect on galanin-immunoreactive neurones in the diencephalon of the ewe.

Galanin is a neuropeptide involved in the regulation of numerous functions such as reproduction. In female rats, this peptide stimulates gonadotropin-releasing hormone (GnRH)/luteinizing hormone release and its synthesis is stimulated by oestradiol. It could therefore be an intermediary between the oestrogenic signal from the ovaries and the GnRH neurones (e.g. during the time course leading to the preovulatory GnRH surge). However, although the involvement of galanin is well-known in rodents, it is poorly understood in ewes. Using immunohistochemistry with a specific antigalanin antiserum, we detected the peptide in neurones of two groups of ovariectomized ewes treated for 6 h with subcutaneous implants, either with oestradiol (experimental group) or empty (control group). The galanin-immunoreactive neurones were counted in three areas, the preoptic area, the bed nucleus of the stria terminalis and the infundibular nucleus, using a computerized image analysis system. There was no change in the mean number of galanin-immunoreactive (GAL-ir) neurones in the infundibular nucleus (37 +/- 12 neurones/section in treated animals and 31 +/- 11 in controls) or in the bed nucleus of the stria terminalis (22 +/- 5 neurones/section in treated animals and 16 +/- 4 in controls), but the number of GAL-ir neurones was higher in the preoptic area in treated than in control ewes (35 +/- 4 versus 14 +/- 10, P < 0.001). To determine whether the neurones of the preoptic area were directly sensitive to oestradiol, we performed double immunohistochemical labelling for oestradiol receptor alpha and galanin. More than 50% of the GAL-ir neurones contained the oestradiol receptor alpha and therefore could be directly regulated by oestradiol. These results indicate that oestradiol might act directly on a GAL-ir neuronal population situated in the preoptic area, without any effect on the GAL-ir neurones of the infundibular nucleus or the bed nucleus of the stria terminalis. Because a 6-h oestradiol treatment can induce a preovulatory GnRH surge in ewes, the GAL-ir neuronal population of the preoptic area might be one of the neuronal systems by which oestradiol activates the GnRH neurones. However, although the morphological relationships between galanin and GnRH neurones have been described in rodents, they remain to be demonstrated in the ewe.

An immunocytochemical mapping of methionine-enkephalin-Arg(6)-Gly(7)-Leu(8) in the human brainstem.

Using an indirect immunoperoxidase technique, we studied the distribution of immunoreactive fibers and cell bodies containing methionine-enkephalin-Arg(6)-Gly(7)-Leu(8) in the adult human brainstem. Immunoreactive cell bodies were found in the reticular formation of the medulla oblongata (in which we observed the highest density of immunoreactive cell bodies) and the pons, the solitary nucleus, the hypoglossal nucleus, the medial and spinal vestibular nuclei, the lateral cuneate nucleus, the nucleus prepositus, the central gray of the pons and mesencephalon, the central and pericentral nuclei of the inferior colliculus, the superior colliculus, ventral to the superior olive and in the midline region of the pons and mesencephalon. The highest density of immunoreactive fibers containing methionine-enkephalin-Arg(6)-Gly(7)-Leu(8) was found in the spinal trigeminal nucleus, the central gray and the reticular formation of the medulla oblongata, pons and mesencephalon, the solitary nucleus, the spinal vestibular nucleus, the dorsal accessory olivary nucleus, the raphe obscurus, the substantia nigra and in the interpeduncular nucleus. The widespread distribution of immunoreactive structures containing methionine-enkephalin-Arg(6)-Gly(7)-Leu(8) in the human brainstem indicates that this neuropeptide might be involved in several physiological mechanisms, acting as a neurotransmitter and/or neuromodulator.

Distribution of ACTH immunoreactivity in the diencephalon and the brainstem of the dog.

The present work describes for the first time the anatomical distribution of adrenocorticotropic hormone (ACTH) in the diencephalon and the brainstem of the dog by means of the indirect immunoperoxidase technique. The distribution found in this species agrees well with the pattern found in other mammals and particularly confirms much of the findings reported in the cat. An exception to that concordance is the presence of ACTH perikarya in the nucleus of the solitary tract of the dog, a population that has been described in the rat but not in the cat, and in the ventral mesencephalon. This last population spread across the ventral tegmental area from the raphe to the cerebral peduncle and appeared to be a specific feature of the canine brain. On the other hand, we can not see ACTH fibers in the substantia nigra of the dog which could be a characteristic of the domestic carnivores, opposite to rodents, since these fibers appeared to be also lacking in the cat. Nevertheless, the widespread distribution of ACTH fibers in the brain of the dog included many other nuclei containing monoaminergic neurons which supported a possible role for ACTH in the regulation of these neurotransmitter systems.

Developmental expression of neurotensin in thyrotropes and gonadotropes of male and female rats.

Besides its potential roles as a central neuromodulator or a hypothalamic neurohormone, neurotensin (NT) may also have endocrine function in the anterior pituitary of mammals. We previously found that NT immunoreactivity is present in the secretory granules of gonadotropes and thyrotropes in both male and female rats, where its levels of expression are under the control of sex steroids. In this work, using immunocytochemistry and in situ hybridization, we have studied the postnatal development of NT-like immunoreactivity (NTir) and the mRNA encoding NT (mRNA-NT) in specific anterior pituitary cells of both male and female rats. NT expression starts after birth and displays an identical pattern in both sexes until sexual maturity, with mRNA-NT being detected from day 2 of postnatal life in thyrotropes localized in the central portion of the anterior lobe. This pattern of expression develops progressively throughout the 2nd and 3rd weeks in both sexes. By the beginning of the 3rd week, mRNA-NT can also be detected in gonadotropes localized in the periphery of the gland coinciding with a rise in serum estradiol concentrations in both sexes, and by day 21, mRNA-NT is extensively present in both the periphery and the central region. NTir is observed from days 5-6 in thyrotropes predominantly localized in the central portion of the anterior lobe, and by day 21, NTir is also detected in gonadotropes localized in the periphery of the gland. This pattern remains similar in both sexes until the time of puberty, when female rats start displaying plastic changes in NT expression according to the stage of the estrous cycle. These findings indicate that NT expression in the rat anterior pituitary is cell specific, and develops from birth to adulthood under the control of sex steroid hormones. In addition, preliminary data showing the presence of NT receptors in rat pituitary cells support the hypothesis of a paracrine or an autocrine role for this peptide within the pituitary.

An immunocytochemical mapping of somatostatin in the cat auditory cortex.

Using an indirect immunoperoxidase technique, the localization of somatostatin-28 (1-12)-like immunoreactive fibers and cell bodies in the auditory cortex of the cat (anterior, primary, secondary, temporal, ventral, ventroposterior, posterior and dorsoposterior auditory fields) was studied. In general, the distribution of SOM-ir structures is widespread in the auditory cortex of the feline. A high density of immunoreactive fibers as well as a low density of cell bodies containing somatostatin were observed in all the layers of the eight above-mentioned auditory fields. These data indicate that somatostatin-28 (1-12) could act as a neurotransmitter and/or a neuromodulator in the auditory cortex of the cat. The origin of the SOM-ir fibers in the auditory cortex of the cat, as well as the issue of whether the cell bodies containing somatostatin-28 (1-12) are local or projecting neurons is discussed.

Influence of monoamines on differentiating gonadotropin-releasing hormone neurones in foetal mice.

This study evaluated the influence of monoamines, serotonin (5-hydroxytryptamine, 5-HT) and noradrenaline, on differentiating gonadotropin-releasing hormone (GnRH)-producing neurones in foetal mice. The differentiation and migration of GnRH neurones were compared in Tg8 mice (the knocked-out gene encoding monoamine oxidase A) with increased levels of 5-HT and noradrenaline and in C3H mice with normal metabolism of monoamines in C3H mice. To achieve this, immunocytochemistry for GnRH combined with quantitative and semiquantitative image analysis were employed. GnRH neurones in foetuses at the 18th embryonic day were detected in the forebrain along the trajectory of their migration from the olfactory bulbs to the hypothalamic retrochiasmatic region. The total number of GnRH neurones in the forebrain in knockout mice was significantly lower compared to C3H mice, suggesting an inhibiting influence of monoamines on the proliferation of precursor cells. The fraction of GnRH neurones in the caudal part of the trajectory of their migration in Tg8 mice exceeded significantly those in C3H foetuses, whereas there was a reverse in the rostral part of the trajectory. These data suggest that an excess of 5-HT and noradrenaline served to accelerate the GnRH neurone migration in Tg8 mice. Moreover, an excess of 5-HT and noradrenaline provided a minor effect on the area and optical density of GnRH neurones (i.e. on GnRH neurone differentiation). Thus, an excess of 5-HT and noradrenaline appears to inhibit the proliferation of the precursor cells of GnRH neurones and stimulates the GnRH neurone migration to the place of their final location in the septo-preoptic region.

Prostaglandins mediate lipopolysaccharide effects upon cholecystokinin and neurotensin phenotypes in neuroendocrine corticotropin-releasing hormone neurons: in situ hybridization evidence in the rat.

Intraperitoneal injection of the endotoxin lipopolysaccharide produces an inflammation accompanied by immune system activation and secretion of cytokines that stimulate the hypothalamo-pituitary-adrenal (HPA) axis to release the anti-inflammatory corticosterone. Upstream in HPA axis are neuroendocrine corticotropin-releasing hormone neurons in the paraventricular nucleus whose multipeptidergic phenotype changes during inflammation: coexisting corticotropin-releasing hormone and cholecystokinin mRNAs are up-regulated whereas neurotensin mRNA expression is induced de novo. These changes may be mediated by prostaglandins released from perivascular and microglial cells in response to circulating cytokines. We examined by quantitative in situ hybridization histochemistry whether blockade of prostaglandin synthesis by indomethacin alters phenotypic expression in paraventricular nucleus neurons after lipopolysaccharide. Because indomethacin also elevated circulating corticosterone, animals were adrenalectomized and corticosterone replaced. Results showed that i.p. indomethacin administration suppressed lipopolysaccharide effects in a phenotype non-specific manner: one injection was sufficient to prevent both the increase in corticotropin-releasing hormone and cholecystokinin mRNAs expression and the induction of neurotensin mRNA expression. Therefore, neuroendocrine corticotropin-releasing hormone neurons with different peptidergic phenotypes appear to respond as a whole in the acute phase response to systemic infection.

Sensitivity of galanin- and melanin-concentrating hormone-containing neurones to nutritional status: an immunohistochemical study in the ovariectomized ewe.

The sensitivities of galanin and melanin-concentrating hormone (MCH) neuronal systems to nutrition are poorly understood in sheep compared to rodents. The aim of this study was to describe the changes in the numbers of galanin and MCH neurones in ovariectomized ewes submitted to different nutritional levels. In the first experiment, ewes were fed ad libitum or food deprived for 24 h. In the second experiment, two groups of ewes were fed at maintenance level (group 100) or undernourished (group 40) for 167 days, after which one-half of each group was killed or refed ad libitum (group 100R and 40R) for 4 days. The MCH neuronal population located in the lateral hypothalamic area was not affected by these nutritional changes. Long-term undernutrition enhanced the number of galanin neurones located in the infundibular nucleus and the dorsal hypothalamic area (DHA), refeeding resulted in an increase of neurones in the DHA and preoptic area, but short-term starvation had no effect on any galanin subpopulations. Our data suggest that the sensitivity of MCH neuronal populations to nutrition in sheep differs from that of rodents. Various populations of galanin-containing neurones differ in sensitivity in ewes subjected to long undernutrition and refeeding but not to short starvation.

Influence of serotonin on the development and migration of gonadotropin-releasing hormone neurones in rat foetuses.

This study used a pharmacological approach to evaluate the consequences of the metabolic perturbations of neurotransmitters on brain development. Pregnant rats received p-chlorophenylalanine (pCPA), an inhibitor of serotonin (5-hydroxytryptamine, 5-HT) synthesis, or saline (control) from the 11th day of gestation once or daily up to the 15th, 17th and 20th day, followed by processing of the forebrain and/or nasal cranium of foetal males and females for high-performance liquid chromatography of monoamines, radioimmunoassay of gonadotropin-releasing hormone (GnRH) and quantitative and semiquantitative immunocytochemistry for GnRH. The pCPA treatment resulted in a 50-70% depletion of 5-HT in the nasal crania and forebrains at any studied age. Radioimmunoassay showed no change in GnRH content in 5-HT deficient foetuses at E16 compared to controls, being higher in both cases in the rostral forebrain than in the hypothalamus. In controls at E21, the GnRH content in the hypothalamus exceeded that in the rostral forebrain, whereas in the 5-HT deficient group the opposite was found. These data suggest that 5-HT provided a stimulating effect on GnRH neurone migration, and this was confirmed by quantification of GnRH-immunoreactive neurones in the forebrain along the trajectory of their migration. At E18 and E21, the fractions of GnRH neurones in the rostral part of the trajectory in pCPA-treated foetuses were greater than those in control foetuses but the opposite was true for the caudal part of the trajectory. Moreover, 5-HT appeared to control the proliferation of the precursor cells of GnRH neurones and their differentiation, as derived from the observations of the increased number of GnRH neurones in the forebrain of foetuses of both sexes, as well as the region-specific decreased neuronal size and content of GnRH in 5-HT-deficient females. Thus, 5-HT appears to contribute to the regulation of the origin, differentiation and migration of GnRH neurones.

Spatiotemporal analysis of signal transducer and activator of transcription 3 activation in rat brain astrocytes and pituitary following peripheral immune challenge.

The host response to peripheral inflammation induces fever and behavioural depression that are supposed to be centrally mediated by cytokines. Several proinflammatory cytokines activate 'signal transducer and activator of transcription' 3 (STAT3) via gp130-like receptor signaling. In order to determine which cells in the rat brain and pituitary are activated during bacterial inflammation, we investigated in a spatiotemporal manner the activation of STAT3 in these organs following peripheral lipopolysaccharide (LPS) challenge. Under basal conditions, STAT3 immunoreactivity was observed in neurones and some glial cells throughout the brain. Two hours after the administration of LPS, nuclear localisation of STAT3 (hallmark of activation) was observed in zones at the interface between brain and blood or cerebrospinal fluid such as pituitary, ependymal layer, meninges, glia limitans, circumventricular organs and surrounding nervous parenchyma. Four hours after LPS, the nuclear activation of STAT3 propagated to cells located inside the parenchyma (cortex, hypothalamus, corpus callosum and hippocampus among others) and declined 8 h after treatment. Double labelling of STAT3 and glial fibrillary acidic protein identified activated cells in the parenchyma as astrocytes. These data show that STAT3 is activated in the pituitary and in brain astrocytes after a peripheral LPS challenge as demonstrated by immunohistochemistry. Astrocytes may therefore play a key role in the brain response to peripheral inflammation.

Localisation of GABA(A) receptor epsilon-subunit in cholinergic and aminergic neurones and evidence for co-distribution with the theta-subunit in rat brain.

In situ hybridisation and immunohistochemical methodologies suggest the existence of a large diversity of GABA(A) receptor subtypes in the brain. These are hetero-oligomeric proteins modulated by a number of clinically important drugs, depending on their subunit composition. We recently cloned and localised the rat GABA(A) receptor epsilon-subunit by in situ hybridisation and immunohistochemical procedures. Here, in a dual-labelling immunohistochemical study in the rat brain, we used our affinity-purified antiserum to epsilon with antisera to markers of cholinergic, catecholaminergic, and serotonergic neurones. As far as cholinergic systems were concerned, epsilon-immunoreactivity was expressed in all forebrain cell-groups, as well as in the caudal lateral pontine tegmentum and dorsal motor nucleus of the vagus nerve. As far as dopaminergic systems were concerned, epsilon-immunoreactivity was found to be expressed in a great number of hypothalamic cell-groups (A15, A14 and A12) and in the substantia nigra pars compacta. The noradrenergic, and to a lesser extent, adrenergic cell-groups were all epsilon-immunoreactive. Also, epsilon-immunoreactivity was detected in all serotonergic cell-groups. We also revealed by in situ hybridisation in a monkey brain that epsilon mRNA was expressed in the locus coeruleus, as previously observed in rats. Finally, by using in situ hybridisation in rat brains, we compared the distribution of the mRNA of epsilon with that of the recently cloned theta-subunit of the GABA(A) receptor. Both subunits showed strikingly overlapping expression patterns throughout the brain, especially in the septum, preoptic areas, various hypothalamic nuclei, amygdala, and thalamus, as well as the aforementioned monoaminergic cell-groups. No theta-mRNA signals were detected in cholinergic cell-groups. Taken together with previously published evidence of the presence of the alpha3-subunit in monoamine- or acetylcholine-containing systems, our data suggest the existence of novel GABA(A) receptors comprising alpha3/epsilon in cholinergic and alpha3/theta/epsilon in monoaminergic cell-groups.

Long-term undernutrition followed by short-term refeeding effects on the corticotropin-releasing hormone containing neurones in the paraventricular nucleus: an immunohistochemical study in sheep.

The effect of nutritional level on the immunoreactivity of corticotropin-releasing hormone (CRH) in neurones of the hypothalamic paraventricular nucleus was described in sheep, a ruminant, whose feeding strategy differs from that of monogastric species. Two groups of ewes were underfed (40%), or fed at maintenance (100%) for 167 days, after which one-half of each group was killed or ad libitum refed (at least 150% of maintenance) for 4 days before killing. The presence of CRH in the paraventricular nucleus was examined by immunohistochemistry. The number of CRH immunoreactive neurones was increased in underfed ewes, but without modification of the plasma concentration of cortisol, indicating that the rise of CRH was not released in the portal blood nor linked to the pituitary-adrenal axis. Refeeding did not modify significantly the number of CRH immunoreactive neurones in the nucleus although these neurones were increased, only in refed ewes that were previously underfed. These data differ from those for rats and mice where CRH expression is decreased or not modified by underfeeding which could reflect different effects of undernutrition on CRH immunoreactive neurones in monogastric compared to ruminants species.

Distribution of alpha-neoendorphin immunoreactivity in the diencephalon and the brainstem of the dog.

Alpha-neoendorphin (alpha-NE) is an opiate decapeptide derived from the prodynorphin protein. Its anatomical distribution in the brain of mammals other than the rat, particularly in carnivores, is less well known than for other opiate peptides. In the present work, we have charted the distribution of alpha-NE immunoreactive fibers and perikarya in the diencephalon and the brainstem of the dog. The highest densities of labeled fibers were found in the substantia nigra and in patches within the nucleus of the solitary tract. Moderate densities appeared in the arcuate nucleus (Ar), median eminence, entopeduncular nucleus, ventral tegmental area, retrorubral area, periaqueductal central gray, interpeduncular nucleus and lateral parabrachial nucleus. Groups of numerous labeled perikarya were localized in the magnocellular hypothalamic nuclei, Ar and in the central superior and incertus nuclei in the metencephalon. Moreover, less densely packed fibers and cells appeared widely distributed throughout many nuclei in the region studied. These results are discussed with regard to the pattern described in other species. In addition, the present results were compared with the distribution of met-enkephalin immunoreactivity in the diencephalon and the brainstem of the dog that we have recently described. Although the distributions of these two peptides overlap in many areas, the existence of numerous differences suggest that they form separate opiate systems in the dog.