Alternative role for prolactin-releasing peptide in the regulation of food intake.

Prolactin-releasing peptide (PrRP) is a peptide ligand for the human orphan G-protein-coupled receptor hGR3/GPR10 and causes the secretion of prolactin from anterior pituitary cells. However, the lack of immunoreactive staining for PrRP in the external layer of the median eminence seems to rule out this peptide as a classical hypophysiotropic hormone and, furthermore, PrRP is less effective than another inducer of prolactin secretion, thyrotropin-releasing hormone, both in vitro and in vivo. Here we show a reduction in the expression of PrRP mRNA during lactation and fasting and an acute effect of PrRP on food intake and body weight, supporting the hypothesis of an alternative role for the peptide.

KCC3 and KCC4 expression in rat adult forebrain.

Potassium chloride ion cotransporters (KCCs) are part of a family of transporters classically described as being involved in cell volume regulation. Recently, KCC2 has been shown to have a role in the development of the inhibitory actions of amine transmitters, whereas KCC3 also plays a fundamental role in the development and function of the central and peripheral nervous system. We have re-assessed the expression of each of the known KCCs in the rat forebrain using RT-PCR and in situ hybridisation histochemistry. As well as confirming the widespread expression of KCC1 and KCC2 throughout the brain, we now show a more restricted expression of KCC3a in the hippocampus, choroid plexus and piriform cortex, as well as KCC4 in the choroid plexus and the suprachiasmatic nucleus of the hypothalamus. The expression of KCC4 in the latter and KCC2 in the lateral hypothalamic and ventromedial hypothalamic nuclei suggests that these cotransporters may have selective roles in neuroendocrine or homeostatic functions. Finally, we demonstrate the existence of a truncated splice variation of KCC3a in the rat that appears to be expressed exclusively in neurons (as is KCC2), whereas the native form of KCC3a and KCC4 appears to be expressed in glial cells.

Intravenous peptide YY3-36 and Y2 receptor antagonism in the rat: effects on feeding behaviour.

Systemic injection of peptide YY3-36 reduces food intake in rodents and humans, although some groups have reported a lack of response. PYY3-36 is thought to act via the Y2 receptor to presynaptically inhibit the release of neuropeptide Y and GABA from hypothalamic arcuate neurones. Due to the controversy surrounding its action in rodents, we tested the peptide intravenously on feeding behaviour in rats and attempted to block its actions with the Y2 receptor antagonist BIIE0246. PYY3-36 significantly decreased food intake during the first hour in male Sprague-Dawley rats fasted overnight and then re-fed. BIIE0246 had no effect alone on re-feeding, but completely blocked the action of PYY3-36. In a second experiment of similar design, the behavioural satiety sequence (BSS) was studied. Normal rats eat, drink, explore and groom before entering rest. PYY3-36 significantly reduced food eaten maintaining the normal BSS, although shifting it to the left as expected for a natural satiety factor. The latency to rest occurred earlier for animals given PYY3-36 alone and PYY3-36 tended to increase the total time in rest compared with controls. These behavioural effects of PYY3-36 were blocked by BIIE0246, and BIIE0246 alone did not have an effect on the BSS. These results support the role of PYY3-36 as a natural satiety factor acting through Y2 receptors.

Hypothalamic STAT proteins: regulation of somatostatin neurones by growth hormone via STAT5b.

Signal transducers and activators of transcription (STATs) are a family of transcription factors linked to class I cytokine receptors. In the present study, we investigated whether their distribution in the hypothalamus reflects the feedback regulation by growth hormone and what role they might play in the functioning of target neurones. We demonstrate that each of the seven known STATs has a distinct distribution in the hypothalamus. Notably, the STAT5 proteins, that are important in growth hormone (GH) and prolactin signalling in peripheral tissues, were expressed in somatostatin neurones of the periventricular nucleus and dopamine neurones of the arcuate nucleus. Because somatostatin neurones are regulated by feedback from circulating GH, we investigated the importance of STAT5 in these neurones. We demonstrate that STAT5b protein expression, similar to somatostatin mRNA, is sexually dimorphic in the periventricular nucleus of rats and mice. Furthermore, chronic infusion of male dwarf rats with GH increased the expression of STAT5b, while a single injection of GH into similar rats induced the phosphorylation of STAT5 proteins. The cellular abundance of somatostatin mRNA in STAT5b-deficient mice was significantly reduced in the periventricular nucleus, effectively reducing the sexually dimorphic expression. These results are consistent with the hypothesis that STAT5 proteins are involved in the feedback regulation of somatostatin neurones by GH, and that these neurones may respond to patterned GH secretion to reinforce sexual dimorphism in the GH axis.

Induction of c-fos messenger ribonucleic acid in neuropeptide Y and growth hormone (GH)-releasing factor neurons in the rat arcuate nucleus following systemic injection of the GH secretagogue, GH-releasing peptide-6.

In this study we investigated the neurochemical identity of the arcuate cells activated following GH-releasing peptide-6 (GHRP-6) injection by comparing, on consecutive sections, the distribution c-fos messenger RNA (mRNA) with that of mRNAs for peptides synthesized in arcuate cells, including neuropeptide Y (NPY), GH-releasing factor (GRF), tyrosine hydroxylase, POMC, and somatostatin. Rats bearing chronically implanted jugular catheters were injected with either 50 micrograms GHRP-6 or vehicle. Thirty minutes later they were terminally anesthetized and perfused with fixative. Paraffin-embedded sections of 7 microns thickness were processed using in situ hybridization for either c-fos mRNA or mRNAs for the neurochemical markers. In GHRP-6-treated rats the mean (+/-SEM) number of cells expressing c-fos mRNA in the arcuate nucleus (23 +/- 2 cells/section per rat; n = 5) was significantly higher than for vehicle-treated controls (2 +/- 1 cells/section per rat; n = 5; P < 0.001, Mann-Whitney U test). Superimposed camera lucida maps indicated that, in GHRP-6-injected rats, neurochemically identifiable cells expressing c-fos mRNA also express NPY mRNA (51 +/- 4%), GRF mRNA (23 +/- 1%) tyrosine hydroxylase mRNA (11 +/- 3%), POMC mRNA (11 +/- 2%), or somatostatin mRNA (4 +/- 1%). Thus, the majority of cells expressing c-fos mRNA following GHRP-6 injection are NPY and GRF-containing cells.

Induction of uterine activity with oxytocin in late pregnant rats replicates the expression of c-fos in neuroendocrine and brain stem neurons as seen during parturition.

We investigated whether delivery could be induced with pulsatile oxytocin and whether such treatment activated neurons in the supraoptic nucleus (SON) and putative afferent neurons in the nucleus tractus solitarii and ventrolateral medulla, as seen during spontaneous parturition. Rats were implanted with a jugular venous cannula 1 day before the expected term and on the next morning were given a pulse of oxytocin or saline every 10 min for 4 h. Pulses of 10 and 20 mU oxytocin induced delivery in 77% (14 of 18) of rats, whereas none of the control animals (0 of 9) gave birth during the treatment. Lower doses of oxytocin were not effective at this time in inducing delivery. Animals were killed either before (prepartum groups) or during (parturient groups) delivery, and the brains were processed for immunocytochemistry. Oxytocin treatment induced Fos expression in SON and brain stem neurons in both parturient rats and rats in which parturition was not induced. Fos expression in all sites was significantly higher than that in control prepartum rats, but was similar in extent and distribution to that in spontaneous parturient rats. In the brain stem, a substantial proportion of Fos-immunoreactive cells contained tyrosine hydroxylase-like immunoreactivity, and the number of these cells was increased in response to oxytocin treatment. As only very few Fos-immunoreactive nuclei in either the SON or the nucleus tractus solitarii were observed in virgin rats injected with oxytocin, we suggest that intermittent oxytocin injections in late pregnant rats induce strong uterine activity, which can stimulate magnocellular and putative afferent neurons even before the expulsion of pups.

Morphological plasticity that occurs in the neurohypophysis following activation of the magnocellular neurosecretory system can be mimicked in vitro by beta-adrenergic stimulation.

The osmotic stress to rats of replacing drinking water with 2% NaCl for four days (salt loading) led to dramatic ultrastructural morphological changes in the neural lobe of the pituitary, including a decrease in the ratio of glial to neurosecretory terminal coverage of the pericapillary basal lamina. The decrease in the proportion of the basal lamina covered by pituicytes, the specialized astrocytic glial cells of the neural lobe, was due to a decrease in the number of pituicyte processes reaching the pericapillary spaces. The concomitant increase in the proportion of neuronal coverage was due to the combination of an increase in the length of individual nerve terminals and a change in the number of terminals. Similar structural changes to those seen in vivo were produced by incubation of the isolated neural lobe with the beta-adrenergic agonist isoprenaline. A decrease in the ratio of glial to neuronal coverage of the basal lamina was achieved within 1 h and could be blocked by inclusion of the beta-adrenergic antagonist propranolol. It is proposed that the morphological plasticity is explained by the active movement of pituicyte cytoplasm.

Induction of members of the Fos/Jun family of immediate-early genes in identified hypothalamic neurons: in vivo evidence for differential regulation.

In situ hybridization was used to measure the expression of members of the Fos/Jun family of immediate-early genes in hypothalamic neurons in vivo following defined stimuli that utilize different afferent pathways. Only c-jun messenger RNA was expressed in the hypothalamic supraoptic and paraventricular nuclei of control animals. Intravenous infusions of sodium chloride solutions of different tonicity produced a range of plasma osmolalities within physiological limits. While the induction of c-fos and jun B messenger RNAs followed the stimulus intensity, the expression of c-jun was repressed at low levels of stimulation. A higher level of osmotic stimulation was able to co-induce c-jun with the c-fos, jun B and fos B genes, suggesting that other signalling pathways may then be activated. Parturition or systemic administration of cholecystokinin, that activate supraoptic and paraventricular neurons via ascending afferent pathways from the brainstem, both induced c-fos, but not the other genes, in the magnocellular nuclei. Use of double in situ hybridization confirmed that, unlike with osmotic stimulation, induction of c-fos only occurred in oxytocin neurons. These two stimuli did not cause a concomitant repression of c-jun messenger RNA expression in magnocellular oxytocin neurons. These patterns of induction provide evidence for the differential regulation of members of this family of genes in a physiological context.

Effects of neurotensin on discharge rates of rat suprachiasmatic nucleus neurons in vitro.

The neuropeptide neurotensin and two classes of its receptors, the neurotensin receptor-1 and 2, are present in the suprachiasmatic nucleus of the mammalian hypothalamus. The suprachiasmatic nucleus houses the mammalian central circadian pacemaker, but the effects of neurotensin on cellular activity in this circadian pacemaker are unknown. In this study, we examined the effects of neurotensin on the spontaneous discharge rate of rat SCN cells in an in vitro slice preparation. Neurotensin (1-10 microM) increased cell firing rate in approximately 50% of cells tested, while approximately 10% of suprachiasmatic cells tested showed a decrease in firing rate in response to neurotensin. These effects of neurotensin were not altered by the GABA receptor antagonist bicuculline (20 microM) or the glutamate receptor antagonists, D-aminophosphopentanoic acid (50 microM) and 6-cyano-7-nitroquinoxaline-2,3-dione (20 microM). The neurotensin receptor selective antagonists SR48692 and SR142948a (10 microM) failed to antagonise neurotensin responses in the majority of cells examined. Compounds that function as agonists selective for the neurotensin-receptor subtypes 1 and 2, JMV-510 and JMV-431 respectively, elicited neurotensin-like responses in approximately 90% of cells tested. Six out of seven cells tested responded to both JMV-510 and JMV-431. Neuropeptide Y (100nM) treatment of suprachiasmatic nucleus slices was found to elicit profound suppression of neuronal firing rate. Co-application of neurotensin with neuropeptide Y significantly (P<0.05) reduced the duration of the response, as compared to that elicited with neuropeptide Y alone. Together, these results demonstrate for the first time the actions of neurotensin in the suprachiasmatic nucleus and raise the possibility that this neuropeptide may play a role in modulating circadian pacemaker function.

Up-regulation of nitric oxide synthase messenger RNA in an integrated forebrain circuit involved in oxytocin secretion.

The hypothalamo-neurohypophysial system contains high levels of neuronal nitric oxide synthase and this increases further during times of neurohormone demand, such as that following osmotic stimulation. Using double in situ hybridization, we demonstrate here an increase in the expression of nitric oxide synthase messenger RNA by oxytocin neurons, but not vasopressin neurons, of the supraoptic nucleus at the time of lactation, when oxytocin is in demand due to another neuroendocrine stimulus, the milk-ejection reflex. In addition, using immunocytochemical retrograde tracing, we show that neurons of the subfornical organ, median preoptic nucleus and organum vasculosum of the lamina terminalis, which project to the supraoptic nucleus, contain nitric oxide synthase. These three structures of the lamina terminalis, together with the hypothalamo-neurohypophysial system, make up the forebrain osmoresponsive circuit that controls osmotically-stimulated release of oxytocin in the rat. The expression of nitric oxide synthase messenger RNA in the lamina terminalis was also shown to increase during lactation. The increases in nitric oxide synthase messenger RNA were not apparent during pregnancy. These results provide evidence for an integrated nitric oxide synthase-containing neural network involved in the regulation of the hypothalamo-neurohypophysial axis. The expression of nitric oxide synthase messenger RNA increases in this circuit during lactation and correlates with a reduction in the sensitivity of the circuit to osmotic stimuli also present in lactation but not pregnancy. As nitric oxide is believed to attenuate neurohormone release, it seems that the increased nitric oxide synthase messenger RNA expression detected here during lactation at a time of high oxytocin demand may be involved in reducing the sensitivity of the whole forebrain circuit to osmotic stimuli.

Involvement of the noradrenergic afferents from the nucleus tractus solitarii to the supraoptic nucleus in oxytocin release after peripheral cholecystokinin octapeptide in the rat.

Activation of abdominal vagal afferents by peripheral injection of cholecystokinin octapeptide induces oxytocin release into the circulation. To test the hypothesis that cholecystokinin increases oxytocin release via activation of noradrenergic afferents from the brainstem, we injected rats with 5-amino-2,4-dihydroxy-alpha-methylphenylethylamine, a selective neurotoxin to noradrenergic fibres, into a lateral cerebral ventricle. The neurotoxin treatment reduced the noradrenaline content in the hypothalamus by 75% and reduced the oxytocin secretion in response to cholecystokinin by over 90%. In separate experiments, the neurotoxin was injected unilaterally in the vicinity of the supraoptic nucleus to test whether direct noradrenergic afferents to the supraoptic nucleus are involved in the response to cholecystokinin. The injection reduced the immunoreactivity for dopamine beta-hydroxylase in the supraoptic nucleus and significantly decreased the number of the supraoptic neurons expressing Fos-like protein after cholecystokinin but not after hypertonic saline. In further experiments, rhodamine-conjugated latex microspheres were injected into the supraoptic nucleus to retrogradely label afferent neurons, and the brains were processed with double-immunohistochemistry for tyrosine hydroxylase and Fos-like protein. In the C2/A2 but not the C1/A1 region of the brainstem, cholecystokinin increased the expression of Fos-like protein in the population of retrogradely-labelled catecholaminergic cells. In the C2/A2 region, the majority of retrogradely labelled cells expressing Fos-like protein after cholecystokinin were catecholaminergic. We conclude that noradrenergic afferents from the A2 but not from the A1 region of the brainstem to the hypothalamus mediate, at least in part, oxytocin release following cholecystokinin.

Involvement of cholecystokinin receptor types in pathways controlling oxytocin secretion.

1. Intravenous administration of cholecystokinin (CCK) results in a transient activation of oxytocin neurones in the rat, and hence to oxytocin secretion: this activation is followed by expression of c-fos mRNA and of Fos-like immunoreactivity (Fos-LI) in magnocellular oxytocin neurones. Fos-like immunoreactivity is also induced in the regions of the brainstem that are thought to relay information from the periphery to the hypothalamus. 2. Administration of the selective CCKA receptor antagonist MK-329, but not the CCKB receptor antagonist L-365,260, prior to CCK injection, prevented oxytocin release as measured by radioimmunoassay and oxytocin neuronal activation as measured by electrophysiology and by the lack of induction of c-fos mRNA. 3. MK-329 abolished the release of adrenocorticotrophic hormone (ACTH) following injection of CCK. 4. MK-329 prevented the expression of Fos-LI in the hypothalamic magnocellular nuclei and in the area postrema and dorsal vagal complex of the brainstem. 5. L-365,260 had no effect on the expression of Fos-LI in the brainstem, but attenuated that seen in the hypothalamic magnocellular nuclei. 6. We conclude that CCK acts on CCKA receptors, either in the area postrema or on peripheral endings of the vagus nerve, to cause the release of hypothalamic oxytocin and ACTH. Information may be carried to the hypothalamus in part by CCK acting at CCKB receptors.

Expression of inducible cAMP early repressor (ICER) in hypothalamic magnocellular neurons.

Cyclic AMP-responsive genes are regulated both positively and negatively by a number of constitutively expressed nuclear proteins. These proteins bind to cAMP-responsive DNA elements in their target genes and they are activated by protein kinase A-mediated phosphorylation. The cAMP response element modulator gene encodes for several constitutively expressed products. However, a second intronic promoter within the gene is inducible and produces another negatively acting transcription factor, inducible cAMP early repressor (ICER). ICER shows a diurnal pattern of expression in the pineal gland, but to date it has not been noted elsewhere in the brain. Here we show expression of ICER mRNA in hypothalamic magnocellular neurons following osmotic stimulation over a time course consistent with a modulatory effect on the expression of other immediate-early genes, such as c-fos. However, since ICER was not present in magnocellular neurons during parturition, its presence is not a prerequisite for the transient expression of c-fos.

Fos-like immunoreactivity in the brainstem of the rat following peripheral administration of cholecystokinin.

Ninety min after intraperitoneal (ip) injection of Cholecystokinin (CCK, 50 µg/kg body wt) Fos-like protein was expressed in cells throughout the nucleus of the tractus solitarii (NTS) and area postrema, and also in scattered cells in the lateral reticular area of the brainstem. Using dual fluorescent immunocytochemistry for Fos-like protein and tyrosine hydroxylase (TH), catecholaminergic cells in the A2 region of the NTS and the A1 region of the lateral reticular area were shown to be activated.

Is there such a thing as a healthy appetite?

Not surprisingly, since we have a need to live, grow and reproduce, evolution has equipped us with a powerful drive to seek food and devour it. If that is not enough, our hedonistic emotions have made eating and drinking a pleasurable experience. So, what tells us to stop eating? Have we evolved systems to do so effectively? The ever increasing problems of excess weight and obesity would suggest not.

Centrally administered galanin-like peptide modifies food intake in the rat: a comparison with galanin.

Galanin-like peptide (GALP) is a recently identified neuropeptide that shares sequence homology with the orexigenic neuropeptide, galanin. In contrast to galanin, GALP is reported to bind preferentially to the galanin receptor 2 subtype (GalR2) compared to GalR1. The aim of this study was to determine the effect of GALP on feeding, body weight and core body temperature after central administration in rats compared to the effects of galanin. Intracerebroventricular (i.c.v.) injection of GALP (1 micro g-10 micro g) significantly stimulated feeding at 1 h in both satiated and fasted Sprague-Dawley rats. However, 24 h after GALP injection, body weight gain was significantly reduced and food intake was also usually decreased. In addition, i.c.v. GALP caused a dose-related increase in core body temperature, which lasted until 6-8 h after injection, and was reduced by peripheral administration of the cyclooxygenase inhibitor, flurbiprofen (1 mg/kg). Similar to GALP, i.c.v. injection of galanin (5 micro g) significantly increased feeding at 1 h in satiated rats. However, there was no difference in food intake and body weight at 24 h, and galanin only caused a transient rise in body temperature. Thus, similar to galanin, GALP has an acute orexigenic effect on feeding. However, GALP also has an anorectic action, which is apparent at a later time. Therefore, GALP has complex opposing actions on energy homeostasis.

The effect of pinealectomy on osmotically stimulated vasopressin and oxytocin release and Fos protein production within the hypothalamus of the rat.

Osmotically stimulated vasopressin and oxytocin release were measured in pinealectomized and sham operated male rats infused with hypertonic sodium chloride. Neuronal activation in the hypothalamic regions associated with oxytocin and vasopressin release was investigated by quantitative assessment of Fos protein production. The osmotically stimulated release of both vasopressin and oxytocin was significantly lower in pinealectomized animals as compared to sham operated controls. The slope of regression lines between plasma osmolality and hormone concentrations in the sham animals showed a 1.0 +/- 0.1 pmol per mosm/kg rise in vasopressin and 2.0 +/- 0.4 pmol per mosm/kg rise in oxytocin whilst in the pinealectomized animals these values were significantly lower at 0.4 +/- 0.1 pmol vasopressin per mosm/kg and 0.8 +/- 0.2pmol oxytocin per mosm/kg. The osmotic thresholds for hormone release were unaffected by pinealectomy. Fos production was also significantly lower in the supraoptic nucleus and organ vasculosum of the lamina terminalis in the pinealectomized rat at 62 +/- 20 and 59 +/- 9 Fos immunoreactive cells/section as compared to corresponding values of 202 +/- 31 and 123 +/- 20 Fos immunoreactive cells/section in the shams. These observations suggest that reduced hormone release in the pinealectomized animal is due to lowered responsiveness of central osmoregulatory mechanisms and that melatonin may therefore influence the activation of the magnocellular system.

Neuromedin U neurones in the rat nucleus of the tractus solitarius are catecholaminergic and respond to peripheral cholecystokinin.

Centrally administered neuromedin U (NMU) has profound effects on food intake and energy expenditure. In the rat, central expression of NMU mRNA is confined to the brainstem and the hypothalamus/pituitary, while mRNA for the receptor NMU2R is expressed in the hypothalamus and hippocampus, as well as in the lining of the ventricular system, but not in the brainstem. We demonstrate that a subpopulation of catecholaminergic neurones in the brainstem nucleus of the tractus solitarius contain NMU and are activated by the gut-derived peptide, cholecystokinin. This is consistent with NMU neurones having an anorectic action, probably via their interaction with other neurones in the paraventricular hypothalamus.

The maintenance of normal parturition in the rat requires neurohypophysial oxytocin.

The neuropeptide oxytocin has long been known as a potent contractor of the uterus. However, it has remained difficult to attribute a definite role for neurohypophysial oxytocin in either the initiation or continuation of labour. Most recently, Lefebvre and colleagues have suggested that oxytocin produced in the uterus, rather than in the hypothalamus, may be more important in parturition since at term the uterus of the rat contains 70-fold more mRNA for oxytocin than the hypothalamus, and this disappears at about the time of parturition. Despite the high levels of mRNA the uterus contains only nanogram quantities of immunoreactive oxytocin per gram wet weight at term, compared to microgram quantities present in the pituitary. Here we show that activation of the neurohypophysial oxytocin system occurs, as reflected by expression of immunoreactivity for Fos in the hypothalamic supraoptic nucleus, and that this activation is indeed critical for normal parturition, since its inhibition results in a significant prolongation of parturition. In addition, we present evidence that pulsatile delivery of oxytocin into the circulation is important for the efficient progress of parturition, indicating that a major role of the neuronal circuits regulating oxytocin secretion for parturition, as is already known for suckling, is to produce an appropriately patterned hormonal output for efficient biological action.

Fos expression within regions of the preoptic area, hypothalamus and brainstem during pregnancy and parturition.

Vaginocervical stimulation, that occurs during mating or with the birth of pups, is believed to induce specific sexual and maternal behaviours in the rat as well as stimulating a number of neuroendocrine responses including the secretion of oxytocin, prolactin and luteinizing hormone. Since the medial preoptic area has been implicated in the induction of maternal behaviour, the expression of the immediate-early gene product Fos was compared between non-pregnant, late pregnant and parturient rats. Although no difference was detected in the number of Fos-positive neuronal profiles in the preoptic area of non-pregnant and late-pregnant rats, a large increase was observed in the medial preoptic nucleus and the anteroventral periventricular region, as well as in the hypothalamic supraoptic nucleus, of parturient rats. Double labelling for Fos and tyrosine hydroxylase immunoreactivity in the brainstem of parturient rats showed the activation of catecholaminergic neurons in both the nucleus of the tractus solitarius and in the ventrolateral medulla that may form part of the afferent pathway from the uterus and cervix to the preoptic area and hypothalamus.