Synaptic regulation of paraventricular arginine vasopressin-containing neurons by neuropeptide Y-containing monoaminergic neurons in rats

Summary Synaptic regulation of arginine vasopressin (AVP)-containing neurons by neuropeptide Y (NPY)-containing monoaminergic neurons was demonstrated in the paraventricular nucleus of the rat hypothalamus. NPY and AVP were immunolabeled in the pre- and the post-embedding procedures, respectively, and monoaminergic fibers were marked by incorporating 5-hydroxydopamine (5-OHDA), a false neurotransmitter. The immunoreaction for NPY was expressed by diaminobenzidine (DAB) chromogen, and that for AVP by gold particles. The DAB chromogen was localized on the surface of the membrane structures, such as vesicles or mitochondria, and on the core of large cored vesicles. Gold particles were located on the core of the secretory granules within the AVP cell bodies and processes. The incorporated 5-OHDA was found as dense cores within small or large vesicular structures. From these data, three types of nerve terminals were discernible: NPY-containing monoaminergic, NPY-containing non-aminergic, and monoaminergic fibers. The AVP cell bodies appeared to have synaptic junctions formed by these nerve terminals as well as by the unlabeled nerve terminals which have small clear vesicles and large cored vesicles. These different types of nerve terminals were frequently observed in a closely apposed position on the same AVP cell bodies. The functional relationships of these three types of neuronal terminals are discussed.     

Reserpine-induced immunocytochemical change of neuropeptide Y in the hypothalamic arcuate nucleus

The effect of reserpine on neuropeptide Y immunoreactive (NPY-IR) neurons in the rat hypothalamic arcuate nucleus was examined by immunocytochemical techniques. Although only NPY-IR fibers and terminals were distributed in this nucleus in untreated and saline treated rats, single treatment of reserpine (10 mg/kg, i.p.) visualized abundant NPY-IR neuronal cell bodies: the increase began at 12 h of postinjection, reached its maximal level at 48 h, and returned to its normal level at 96 h. Pretreatment of nialamide, a monoamine oxidase inhibitor, prevented these acute reserpine-induced changes, suggesting reserpine acts on NPY neurons through monoaminergic mechanism. Chronic treatment of haloperidol (5 mg/kg, once daily for 5 days) a dopamine receptor antagonist, could induce the similar increase of NPY immunoreactivity. However, interruption of adrenergic and serotonergic neurotransmissions by chronic treatment of propranorol and methysergide, or chemical lesions of ascending noradrenergic and serotononergic pathways by 6-hydroxydopamine and 5,6-dihydroxytryptamine, could not induce any immunoreactive increase of NPY in arcuate neurons. These findings strongly suggest that reserpine-induced NPY increase occurs through dopaminergic afferents in hypothalamic arcuate neurons. Special issue dedicated to Dr. Kinya Kuriyama.     

Morphological correlates of serotonin-neuropeptide Y interactions in the rat suprachiasmatic nucleus: Combined radioautographic and immunocytochemical data

Summary The morphological substrate of putative serotonin (5-HT)/neuropeptide Y (NPY) interactions in thé suprachiasmatic nucleus (SCN) was investigated by combined radioautography and immunocytochemistry after intraventricular administration of (³H)5-HT in the rat. In the ventral portion of the SCN, the distribution of (³H)5-HT uptake sites overlapped closely the NPY-immunoreactive terminals. Previous investigations have shown that the dense 5-HT and NPY innervations of the SCN originate in different structures, i.e., the midbrain raphe nuclei and the ventral lateral geniculate nucleus, respectively. Accordingly, in the present study, destruction of 5-HT afferents by 5,7-dihydroxytryptamine was not found to induce any modification in NPY staining and, in ultrastructural immuno-radioautographic preparations, two distinct pools of axonal varicosities could be identified. Both 5-HT and NPY terminals established morphologically defined synaptic junctions, sometimes on the same neuronal target. Some cases of direct axo-axonic appositions between the two types of terminals were also encountered. These data constitute additional criteria for characterizing the cytological basis of the multiple transmitter interactions presumably involved in the function of the SCN as a central regulator of circadian biological rhythms.     

Synaptic relations of neurotensinergic neurons in the dorsal raphe nucleus

The ultrastructure and synaptic relations of neurotensinergic neurons in the rat dorsal raphe nucleus (DRN) were examined. The neurotensin-like immunoreactive (NT-LI) neurons in the DRN were fusiform or spherical. The NT-LI perikarya could only be detected in colchicine-treated animals whereas the immunoreactive axon terminals could only be found in the anirnals not treated with colchicine. Although many NT-LI dendrites received synapses from nonimmunoreactive axon terminals, the NT-LI perikarya received few synapses. NT-LI axon terminals also made synapses on nonimmunoreactive dendrites. Occasionally, synapses were found between the NT-LI axon terminals and NT-LI dendrites in the cases in which the animals were not treated with colchicine.     

Intragranular co-storage of neuropeptide Y and arginine vasopressin in the paraventricular magnocellular neurons of the rat hypothalamus

Summary Certain populations of arginine vasopressin (AVP) neurons in the magnocellular paraventricular nucleus became immunoreactive for neuropeptide Y (NPY) when rats were treated with colchicine or monosodium glutamate (MSG). The co-storage of these peptides was examined by empooying a post-embedding electron-microscopic immunohistochemistry technique using goldlabeled antibodies to the two peptides. In colchicinetreated rats, the neuronal perikarya contained numerous secretory granules showing co-storage of the two peptides. The cells of the MSG-treated rats were characterized by having well-developed Golgi bodies with the granular structures also co-storing the two peptides, although the secretory granules in the perikarya were rather fewer than in the colchicine-treated rats. It is concluded that the destruction of the arcuate nucleus by MSG-treatment may potentiate the synthesis of NPY in AVP neurons, the synthesis of which is latent in intact animals.     

Developmental changes in synaptic formation in the hypothalamic arcuate nucleus of female rats

Summary The hypothalamic arcuate nucleus (ARCN) of female rats at 5, 20, 45 and 90 days of age was examined ultrastructurally. Axodendritic and axosomatic synapses were counted in 18,000 m² area of the ARCN in each brain. Axodendritic and axosomatic synapses in the ARCN of day 5 rats were very small in number. Axon terminals contained small spherical vesicles (SSVs, 40–60 nm in diameter). Occasionally large granular vesicles (LGVs, 75–130 nm in diameter) were found to coexist with SSVs in the endings. Pre- and postsynaptic membranes were thin. The ARCN at this age exhibited a large extracellular space which decreased with advancing age. In day 20 rats, axodendritic and axosomatic synapses increased in number up to about one-half of those of day 45 or day 90 animals. Synaptic vesicles increased in number and mitochondria were frequently encountered in the axon terminals. Pre- and postsynaptic membranes became thicker than those of day 5 rats. Further increase in the number of axodendritic and axosomatic synapses in the ARCN of day 45 rats was observed, and there were no significant difference in the morphology and incidence of synapses between day 45 and day 90 rats. Synaptic vesicles were numerous and pre- and postsynaptic membranes were thick. In tissue incubated with 5-hydroxydopamine (5-OH-DA) before fixation, small granular vesicles (SGVs, about 50 nm in diameter) which were labeled with 5-OH-DA were detected in a certain number of endings in all material taken from each age group, but the incidence of synapses containing SGVs was usually low. From these results, it can be proposed that an increase in the number of synapses in the ARCN is correlated with functional maturation of the ARC neurons. Acknowledgements. The authors wish to thank Prof. T. Kojima, Nihon University, for valuable suggestions during the initial stage of this study. This study was supported by grants from the Ministry of Education of Japan     

Localization of neurophysin in the rat supraoptic nucleus

Summary Localization of neurophysin in neurons of the supraoptic nucleus was accomplished using an unlabeled-antibody, post-embedding, immunoperoxidase technique. Neurophysin was exclusively associated with neurosecretory granules within cell bodies of supraoptic neurons and their processes.Supported by U.S.P.H.S. Grant HD-08867     

Ultrastructural investigation of ACTH immunoreactivity in arcuate and supraoptic nuclei of the rat

Summary Fine structural localization of an ACTH-like substance was obtained in neurons of the rat arcuate nucleus using immuno-electron microscopy, whereas it could not be confirmed that ACTH-containing cell bodies are present in the supraoptic nucleus. The immunoreactive cells of the arcuate nucleus appeared to be more numerous than the unreactive neurons. Immunostaining was carried out before embedding in resin. Empty vesicles of irregular shape were found in dendrites of immunoreactive arcuate neurons, but their significance and nature remain enigmatic. The reaction product was distributed uniformly throughout the cytoplasm of the ACTH-positive cells, except that the mitochondria, rough endoplasmic reticulum and Golgi vesicles and cisternae were devoid of PAP molecules. This distribution differed from the localization reported in ACTH-secreting cells of the rat anterior pituitary, where the reaction product was found in the rough endoplasmic reticulum and Golgi complex as well as in secretory granules.     

Effects of ghrelin on gastric distention sensitive neurons in the arcuate nucleus of hypothalamus and gastric motility in diabetic rats

This study was performed to observe the effects of ghrelin on the activity of gastric distention (GD) sensitive neurons in the arcuate nucleus of hypothalamus (Arc) and on gastric motility in vivo in streptozocin (STZ) induced diabetes mellitus (DM) rats. Electrophysiological results showed that ghrelin could excite GD-excitatory (GD-E) neurons and inhibit GD-inhibitory (GD-I) neurons in the Arc. However, fewer GD-E neurons were excited by ghrelin and the excitatory effect of ghrelin on GD-E neurons was much weaker in DM rats. Gastric motility research in vivo showed that microinjection of ghrelin into the Arc could significantly promote gastric motility and it showed a dose-dependent manner. The effect of ghrelin promoting gastric motility in DM rats was weaker than that in normal rats. The effects induced by ghrelin could be blocked by growth hormone secretagogue receptor (GHSR) antagonist [d-Lys-3]-GHRP-6 or BIM28163. RIA and real-time PCR data showed that the levels of ghrelin in the plasma, stomach and ghrelin mRNA in the Arc increased at first but decreased later and the expression of GHSR-1a mRNA in the Arc maintained a low level in DM rats. The present findings indicate that ghrelin could regulate the activity of GD sensitive neurons and gastric motility via ghrelin receptors in the Arc. The reduced effects of promoting gastric motility induced by ghrelin could be connected with the decreased expression of ghrelin receptors in the Arc in diabetes. Our data provide new experimental evidence for the role of ghrelin in gastric motility disorder in diabetes.     

Synaptic relationships between proopiomelanocortin- and ghrelin-containing neurons in the rat arcuate nucleus

Both proopiomelanocortin (POMC) and ghrelin peptides are implicated in the feeding regulation. The synaptic relationships between POMC- and ghrelin-containing neurons in the hypothalamic arcuate nucleus were studied using double-immunostaining methods at the light and electron microscope levels. Many POMC-like immunoreactive axon terminals were found to be apposed to ghrelin-like immunoreactive neurons and also to make synapses with ghrelin-like immunoreactive neuronal perikarya and dendritic processes. Most of the synapses were symmetrical in shape. A small number of synapses made by ghrelin-like immunoreactive axon terminals on POMC-like immunoreactive neurons were also identified. Both the POMC- and ghrelin-like immunoreactive neurons were found to contain large dense granular vesicles. These data suggest that the POMC-producing neurons are modulated via synaptic communication with ghrelin-containing neurons. Moreover, ghrelin-containing neurons may also have a feedback effect on POMC-containing neurons through direct synaptic contacts.     

Synaptic interaction between ghrelin- and ghrelin-containing neurons in the rat hypothalamus

Synaptic relationships between ghrelin-like immunoreactive axon terminals and other neurons in the hypothalamic arcuate nucleus (ARC) were studied using immunostaining methods at the light and electron microscope levels. Many ghrelin-like immunoreactive axon terminals were found to be in apposition to ghrelin-like immunoreactive neurons at the light microscopic level. At the electron microscopic level, ghrelin-like immunoreactive axon terminals were found to make synapses on ghrelin-like immunoreactive cell bodies and dendrites in the ARC. While the axo-dendritic synapses between ghrelin- and ghrelin-like immunoreactive neurons were mostly the asymmetric type, the axo-somatic synapses were both asymmetric and symmetric type of synapses. Ghrelin at 10(-10) M increased cytosolic Ca(2+) concentration ([Ca(2+)](i)) in the neurons isolated from the ARC, some of which were immunocytochemically identified as ghrelin-positive. Ghrelin at 10(-10) M also increased [Ca(2+)](i) in 12% of ghrelin-like immunoreactive neurons in the ARC. These findings suggest that ghrelin serves as a transmitter and/or modulator that stimulates [Ca(2+)](i) signaling in ghrelin neurons of the ARC, which may participate in the orexigenic action of ghrelin. Our data suggests a possibility of existing a novel circuit implicating regulation of feeding and/or energy metabolism.     

Cyto- and synaptogenesis in the arcuate nucleus of the rat hypothalamus during fetal and early postnatal life

Summary Morphogenesis of the arcuate nucleus of the rat from the 15th fetal day to the 6th postnatal day was investigated light and electron microscopically. The arcuate neurons exhibit a gradual development after the 15th fetal day. All cytoplasmic constituents are present in these nerve cells already during the last days of gestation. Nevertheless, they are not fully differentiated at birth. The first synapse-like structures (presynapses) were observed in 17 day-old, the first synapses in 18 day-old fetuses. During the early postnatal period the number of presynapses decreases, but at the same time there is a gradual increase in the number of the relatively mature synapses. This process starts already during the last days of prenatal life. Although all structural elements of the arcuate nucleus of the adult rat appear to be present at birth, the extent of the neuropil area and the number of the presynapses indicate that the arcuate nucleus is still in a fairly undeveloped stage during the first postnatal days.     

Peripheral insulin administration attenuates the increase in neuropeptide Y concentrations in the hypothalamic arcuate nucleus of fasted rats

Fasting increases neuropeptide Y (NPY) concentrations in the arcuate nucleus (ARC), its site of synthesis, and in other regions of the rat hypothalamus. Neuropeptide Y is a potent central orexigenic agent and may therefore stimulate appetite during fasting. We tested the hypothesis that low plasma insulin levels stimulate ARC levels of NPY in fasted rats. Compared with freely fed controls (n = 8), rats fasted for 72 h (n = 8) showed significantly lower plasma insulin levels (28.9 ± 1.6 vs. 52.6 ± 5.7 pmol/l; p < 0.001) and higher ARC NPY concentrations (14.2 ± 1.8 vs. 8.4 ± 2.2 fmol/μg protein; p < 0.001). Fasted rats treated with subcutaneous insulin (5 U/kg/day; n = 10), which nearly normalized plasma insulin (46.6 ± 2.8 pmol/l), showed intermediate ARC NPY levels (11.2 ± 1.4 fmol/μg protein; p < 0.01 vs. controls and untreated fasted rats). Insulin administered peripherally, therefore, attenuates fasting-induced NPY increases in the ARC, supporting the hypothesis that hypoinsulinemia stimulates hypothalamic NPY.     

Ghrelin raises [Ca2+]i via AMPK in hypothalamic arcuate nucleus NPY neurons

Ghrelin, an orexigenic hormone, directly activates neuropeptide (NPY) neurons in the hypothalamic arcuate nucleus (ARC), and thereby stimulates food intake. The hypothalamic level of AMP-activated protein kinase (AMPK), an intracellular energy sensor, is activated by peripheral and central administration of ghrelin. We examined whether ghrelin regulates AMPK activity in NPY neurons of the ARC. Single neurons were isolated from the ARC and cytosolic Ca²⁺ concentration ([Ca²⁺]_i) was measured by fura-2 microfluorometry, followed by immunocytochemical identification of NPY, phospho-AMPK, and phospho-acetyl-CoA carboxylase (ACC). Ghrelin and AICAR, an AMPK activator, increased [Ca²⁺]_i in neurons isolated from the ARC. The ghrelin-responsive neurons highly overlapped with AICAR-responsive neurons. The neurons that responded to both ghrelin and AICAR were primarily NPY-immunoreactive neurons. Treatment with ghrelin increased phosphorylation of AMPK and ACC. An AMPK inhibitor, compound C, suppressed ghrelin-induced [Ca²⁺]_i increases. These results demonstrate that ghrelin increases [Ca²⁺]_i via AMPK-mediated signaling in the ARC NPY neurons.     

Study of neuropeptide Y-containing nerve fibers in the human penis

Summary Neuropeptide Y 1–36 (IR-NPY) immunoreactive nerve-fiber processes have been observed in tunicae of veins and arteries and in smooth muscles of the human penis taken at autopsy or during surgery by use of light-and electron-microscopic immunohistochemical techniques. Numerous IR-NPY nerve fibers were mostly concentrated in the inner part of the adventitia close to the media of the arterial and venous vessels and among the intracavernous smooth muscle cells. IR-NPY nerve fibers were less abundant in veins than in arteries. Positive somata were not observed in the penises. At the ultrastructural level, IR-NPY were localized exclusively in large, dense granules of nerve terminals by means of the postembedding immunogold technique. In the deep dorsal vein, IR-NPY nerve fibers were also located in the media formed by an outer circular and an inner longitudinal layer. In the intracavernous and dorsal arteries, they showed the highest density in the inner part of the adventitia. In the corpora cavernosa and in the corpus spongiosum, IR-NPY nerve processes were intermingled between the smooth-muscle fibers around the sinusoid spaces. IR-NPY nerve fibers were present in the cavernous nerves close to the central arteries. The urethra did not show any IR-NPY-positive nerve fibers. This peculiar distribution of IR-NPY nerve fibers suggested that they could participate in regulating arterial and venous blood flow and intracavernous smooth-muscle tone. NPY may therefore be of importance in some of the mechanisms of penile erection especially during detumescence.     

Localization of GRF-like immunoreactive neurons in the rat brain

The localization of human GRF1-44-immunoreactive neurons was studied in the rat brain. A dense accumulation of GRF-containing fibers was noted in the external layer of the median eminence. Cell bodies were observed in colchicine-treated rats. The most intensely fluorescent cluster of cells was contained in the arcuate nucleus. Other cells were seen on the base of the hypothalamus, within the median forebrain bundle, dorsal and ventral aspects of the ventromedial nucleus, zona incerta and dorsal part of the dorsomedial nucleus. These cells may influence the pulsatile release of pituitary growth hormone.     

AAV mediated expression of anti-sense neuropeptide Y cRNA in the arcuate nucleus of rats results in decreased weight gain and food intake

Neuropeptide Y (NPY) is the most potent stimulant of feeding when administered by intracerebroventricular injection. Despite this, there is conflicting evidence as to its importance in the regulation of daily food intake and energy balance. It has been suggested that whilst it is important in the response to starvation it has little role in the regulation of daily food intake. To investigate the role of NPY in the regulation of food intake, anti-sense cRNA to NPY was expressed in the arcuate nucleus of adult male rats. The anti-sense NPY (AS-NPY) construct was initially tested in vitro and there was a decrease of approximately 50% in NPY release from anti-sense treated cells compared to controls (16.3 +/- 2.0 fmol/L [AS-NPY] vs 37.3 +/- 7.7 fmol/L [control], mean +/- SEM p < 0.05). NPY release from hypothalamic explants from anti-sense injected animals was decreased by over 50% compared to those from controls at both 15 and 20 days after AAV injection (15 days 42% +/- 6.5% [AS-NPY] vs 100% +/- 36% [control], 20 days 41% +/- 6% [AS-NPY] vs 100% +/- 27% [control] mean+/-SEM, p < 0.05). In a study lasting for 50 days, weight gain was significantly lower in anti-sense injected animals from day 16 (day 16: 6.25 +/- 1.10 g [AS-NPY] vs 9.42 +/- 0.65 g [control] mean +/- SEM, p < 0.05) and remained so until the end of the study when they had gained approximately 40% less weight than controls (day 50: 52.0 +/- 9.6 g [AS-NPY] vs 82.0 +/- 6.3 g [control] mean +/- SEM, p < 0.01). Cumulative food intake was significantly lower in the anti-sense injected animals from day 23 (day 23: 225.8 +/- 1.9 g [AS-NPY] vs 250.6 +/- 8.7 g [control], mean +/- SEM, p < 0.05) and remained so until the end of the study (day 50: 834.5 +/- 14.8 g [AS-NPY] vs 926.0 +/- 31.7 g [control], mean +/- SEM, p < 0.05). Similarly mean daily food intake was also reduced in the anti-sense injected animals (days 7-14: 24.9 +/- 0.4 g/day [AS-NPY] vs 27.2 +/- 0.4 g/day [control], mean +/- SEM, p < 0.01). These data are supportive of a role for NPY in the regulation of daily food intake as well as in response to starvation.     

Ultrastructural characteristics of vasopressin-containing neurons in the paraventricular nucleus of the hypothalamus

Summary Vasopressin-containing neurons, identified by immunocytochemistry, are located predominantly in the posterior magnocellular division of the paraventricular nucleus of the rat hypothalamus. By electron microscopy, the immunoreaction product is seen within the cell bodies and neuronal processes. In the perikarya and dendritic processes, the immunoreactive material is associated primarily with neurosecretory granules. Axonal processes, identified by their content of microtubules and accumulation of neurosecretory granules, show the immunoreaction product in association with both of these organelles. Afferent axo-dendritic, axo-somatic and putative axo-axonic synapses with immunostained vasopressinergic neurons can be identified. The presynaptic profiles do not contain immunoreactive material. This study contributes to the ultrastructural characterization of vasopressinergic neurons in the paraventricular nucleus and of their afferent synaptic input.Supported by NIH Grants HD-12956 and 2SO7RR05403     

Serotonin-mediated increases in the extracellular levels of beta-endorphin in the arcuate nucleus and nucleus accumbens: a microdialysis study

Although the involvement of both endogenous opioid and serotonergic systems in modulation of pain and emotion was suggested, the neurochemical interaction between these systems in the brain has not previously been studied directly. Herein, the effects of the local application of serotonin (5-HT) and fluoxetine (a 5-HT reuptake inhibitor) on extracellular levels of beta-endorphin in the arcuate nucleus and nucleus accumbens were assessed in freely moving rats using in vivo microdialysis. The mean basal concentrations of beta-endorphin in dialysates obtained from the arcuate nucleus and nucleus accumbens were 259.9 and 143.3 pM, respectively. Specific lesion of the serotonergic system by 5,7-dihydroxytryptamine (5,7-DHT) caused a significant decrease in these dialysate beta-endorphin levels. When 5-HT (0.25-5 microM) was added to the perfusion solution, the levels of beta-endorphin in the dialysate from the arcuate nucleus increased (186-296% of baseline), in a concentration-dependent manner. In the nucleus accumbens, 0.5 and 2 microM 5-HT in the perfusion fluid did not affect the levels of beta-endorphin in the dialysate, whereas 5 and 10 microM 5-HT caused an increase of approximately 190% of baseline. When fluoxetine (250 microM) was present in the perfusing solution, the levels of beta-endorphin in the dialysates from the arcuate nucleus and nucleus accumbens increased two- to threefold. This effect was not obtained in the 5,7-DHT-lesioned rats. Thus, 5-HT, either endogenously or exogenously delivered, appears to facilitate the release of beta-endorphin in the arcuate nucleus and nucleus accumbens. This indication of an interaction between serotonergic and endorphinic systems may be relevant for assessing pain and mood disorder circuits and the mode of action of antidepressant drugs.