Delivery of pineal melatonin to the brain and SCN: role of canaliculi, cerebrospinal fluid, tanycytes and Virchow-Robin perivascular spaces.

Historically, the direct release of pineal melatonin into the capillary bed within the gland has been accepted as the primary route of secretion. Herein, we propose that the major route of melatonin delivery to the brain is after its direct release into the cerebrospinal fluid (CSF) of the third ventricle (3V). Melatonin concentrations in the CSF are not only much higher than in the blood, also, there is a rapid nocturnal rise at darkness onset and precipitous decline of melatonin levels at the time of lights on. Because melatonin is a potent free radical scavenger and antioxidant, we surmise that the elevated CSF levels are necessary to combat the massive free radical damage that the brain would normally endure because of its high utilization of oxygen, the parent molecule of many toxic oxygen metabolites, i.e., free radicals. Additionally, the precise rhythm of CSF melatonin provides the master circadian clock, the suprachiasmatic nucleus, with highly accurate chronobiotic information regarding the duration of the dark period. We predict that the discharge of melatonin directly into the 3V is aided by a number of epithalamic structures that have heretofore been overlooked; these include interpinealocyte canaliculi and evaginations of the posterodorsal 3V that directly abut the pineal. Moreover, the presence of tanycytes in the pineal recess and/or a discontinuous ependymal lining in the pineal recess allows melatonin ready access to the CSF. From the ventricles melatonin enters the brain by diffusion and by transport through tanycytes. Melatonin-rich CSF also circulates through the aqueduct and eventually into the subarachnoid space. From the subarachnoid space surrounding the brain, melatonin penetrates into the deepest portions of the neural tissue via the Virchow-Robin perivascular spaces from where it diffuses into the neural parenchyma. Because of the high level of pineal-derived melatonin in the CSF, all portions of the brain are better shielded from oxidative stress resulting from toxic oxygen derivatives.

Melatonin controls photoperiodic changes in tanycyte vimentin and neural cell adhesion molecule expression in the Djungarian hamster (Phodopus sungorus).

The Djungarian hamster displays photoperiodic variations in gonadal size synchronized to the seasons by the nightly secretion of the pineal hormone melatonin. In short photoperiod (SP), the gonads regress in size, and circulating sex steroids levels decline. Thus, the brain is subject to seasonal variations of both melatonin and sex steroids. Tanycytes are specialized glial cells located in the ependymal lining of the third ventricle. They send processes either to the meninges or to blood vessels of the medio-basal hypothalamus. Furthermore, they are known to locally modulate GnRH release in the median eminence and to display seasonal structural changes. Seasonal changes in tanycyte morphology might be mediated either through melatonin or sex steroids. Therefore, we analyzed the effects of photoperiod, melatonin, and sex steroids 1) on tanycyte vimentin expression by immunohistochemistry and 2) on the expression of the neural cell adhesion molecule (NCAM) and polysialic acid as markers of brain plasticity. Vimentin immunostaining was reduced in tanycyte cell bodies and processes in SP. Similarly, tanycytes and their processes contained lower amounts of NCAM in SP. These changes induced by SP exposure could not be restored to long photoperiod (LP) levels by testosterone supplementation. Likewise, castration in LP did not affect tanycyte vimentin or NCAM expression. By contrast, late afternoon melatonin injections mimicking a SP-like melatonin peak in LP hamsters reduced vimentin and NCAM expression. Thus, the seasonal changes in vimentin and NCAM expression in tanycytes are regulated by melatonin independently of seasonal sex steroid changes.

Kisspeptin neurons co-express met-enkephalin and galanin in the rostral periventricular region of the female mouse hypothalamus.

It is now well established that the kisspeptin neurons of the hypothalamus play a key role in regulating the activity of gonadotropin-releasing hormone (GnRH) neurons. The population of kisspeptin neurons residing in the rostral periventricular region of the third ventricle (RP3V), encompassing the anteroventral periventricular (AVPV) and periventricular preoptic nuclei (PVpo), are implicated in the generation of the preovulatory GnRH surge mechanism and puberty onset in female rodents. The present study examined whether these kisspeptin neurons may express other neuropeptides in the adult female mouse. Initially, the distribution of galanin, neurotensin, met-enkephalin (mENK), and cholecystokinin (CCK)-immunoreactive cells was determined within the RP3V of colchicine-treated mice. Subsequent experiments, using a new kisspeptin-10 antibody raised in sheep, examined the relationship of these neuropeptides to kisspeptin neurons. No evidence was found for expression of neurotensin or CCK by RP3V kisspeptin neurons, but subpopulations of kisspeptin neurons were observed to express galanin and mENK. Dual-labeled RP3V kisspeptin/galanin cells represented 7% of all kisspeptin and 21% of all galanin neurons whereas dual-labeled kisspeptin/mENK cells represented 28-38% of kisspeptin neurons and 58-68% of the mENK population, depending on location within the AVPV or PVpo. Kisspeptin neurons in the arcuate nucleus were also found to express galanin but not mENK. These observations indicate that, like the kisspeptin population of the arcuate nucleus, kisspeptin neurons in the RP3V also co-express a range of neuropeptides. This pattern of co-expression should greatly increase the dynamic range with which kisspeptin neurons can modulate the activity of their afferent neurons.

Negative correlation between neuropeptide Y profile in the cerebrospinal fluid and growth hormone pulses in the peripheral circulation in goats.

BACKGROUND/AIMS: Growth hormone (GH) is secreted in pulsatile fashion, but the involvement of neuropeptides in the generation of GH pulses is not fully understood. The present study was conducted to elucidate the relationship between GH pulses and neuropeptide levels in the cerebrospinal fluid (CSF) of the third ventricle in ovariectomized goats. METHODS: CSF and plasma samples were collected every 15 min. Levels of plasma GH and profiles of GH-releasing hormone (GHRH), somatostatin (SRIH) and neuropeptide Y (NPY) in the CSF were measured by radioimmunoassay. Pulse/trough characteristics and correlations were assessed by the ULTRA algorithm, cross-correlation analysis and approximate entropy test. RESULTS: The periodicity of GH pulses was 2.20 h. Although most GH pulses were associated with GHRH peaks, there was no correlation between GH and GHRH or GH and SRIH. NPY levels in the CSF fluctuated episodically at 2.03-hour intervals. GH pulse peaks occurred 0-30 min after NPY troughs, and there was a significant negative cross-correlation and negative synchronicity between GH and NPY profiles. In addition, intracerobroventricular infusion of NPY suppressed GH secretagogue (KP102)-induced GH release. CONCLUSION: The periodic decrease in NPY may be involved in the generation of GH pulsatility in goats.

Localization of angiotensin II (AT1)-receptor-immunoreactive fibres in the hypothalamus of rats: angiotensin II-sensitive tanycytes in the ependyma of the third ventricle?

Scanning the hypothalamus of rats for receptor binding sites of the octapeptide hormone angiotensin II (ANG II), we observed ANG II-sensitive fibres in the ventrolateral hypothalamus. The ANG II (AT(1))-receptor-immunoreactive processes originate from cells-probably tanycytes-embedded in the base and the ventrolateral walls of the third ventricle and reach into the retrochiasmatic area, the ventrolateral hypothalamus and the median eminence.

Transforming growth factor-alpha is expressed in astrocytes of the suprachiasmatic nucleus in hamster: role of glial cells in circadian clocks.

Transforming growth factor-alpha (TGF-alpha) is abundantly expressed in the suprachiasmatic nucleus of several rodent species. It was recently suggested to be a clock output signal regulating the activity/rest rhythm. In this study we further characterized the cellular identity of TGF-alpha-expressing cells in the suprachiasmatic nucleus of the Syrian hamster (Mesocricetus auratus). Using confocal laser scanning fluorescence imaging on brain sections immuno-histologically processed for TGF-alpha and GFAP double staining, we observed that in the suprachiasmatic nucleus, TGF-alpha staining is located mainly in GFAP-positive cells, indicating suprachiasmatic nucleus astrocytes produce TGF-alpha. Glial expression of TGF-alpha was also observed in the 3rd ventricle tanycytes of the retrochiasmatic area. In other brain regions where the TGF-alpha message is abundant, such as in the piriform cortex, we observed that TGF-alpha staining is mainly located in neurons. Our results provide the first evidence that glial cells are involved in the regulation of output from the suprachiasmatic nucleus circadian clock through a diffusible mechanism.

Differential distribution of the glutamate transporters GLT-1 and GLAST in tanycytes of the third ventricle.

The ventral one-third of the ventricular lining in the hypothalamus is formed by specialized ependymal cells called the tanycytes. These cells may serve a neuroendocrine transport function because of their structural specializations, which include apical microvili on the ventricular surface and long basal processes that terminate on blood vessels or on the glia limitans. Here, we describe the expression of mRNA and protein for the glutamate transporters GLT-1 and GLAST in unique tanycyte populations of the third ventricle in rat brain. Using nonisotopic in situ hybridization, we demonstrate GLAST mRNA labeling in tanycytes of the ventral floor and lateral walls in the tuberal and mammillary recess portions of the third ventricle. This GLAST mRNA labeling had a higher intensity than the labeling intensity observed in regular ependymal cells throughout the ventricular system. Furthermore, we have identified strong GLT-1 mRNA labeling in a population of tanycytes situated in the dorsolateral walls of caudal tuberal and mammillary recess portions. Immunocytochemical staining indicates that both GLT-1 and GLAST protein are expressed in the tanycyte populations as well. These data corroborate previous findings that third ventricle tanycytes are functionally heterogeneous. Furthermore, the GLT-1-expressing tanycytes represent a population of tanycytes that, to date, has not been recognized as functionally distinct. The strong GLAST expression by the ventral tanycytes in the hypophysiotropic area suggests a role of tanycyte-mediated glutamate transport in neuroendocrine activity. The functional role of GLT-1 in dorsal wall tanycytes remains to be explored.

Fos immunoreactivity in the lamina terminalis of adrenalectomized rats and effects of angiotension II type 1 receptor blockade or deoxycorticosterone.

Neural activity, as measured immunohistochemically by the presence of Fos protein, was determined in the lamina terminalis, a thin strip of tissue forming the anterior wall of the third brain ventricle, after adrenalectomy. Several weeks after surgery, the adrenalectomized rats were maintained with access to water and a low sodium diet for five days. In addition, hypertonic (0.5M) NaCl solution was available for the entire five-day period (sodium available) or only during the first four days (sodium unavailable). The number of neurons expressing Fos, determined at the end of the fifth day, was increased in the adrenalectomized rats with or without NaCl solution to drink. Fos activity in the median preoptic nucleus was increased only in adrenalectomized rats without access to NaCl solution. Treatment of adrenalectomized rats with the sodium-retaining mineralocorticoid hormone, deoxycorticosterone, at the end of the fourth day, decreased Fos expression in the subfornical organ and the organum vasculosum of the lamina terminalis when NaCl solution was available but not when the NaCl solution was unavailable. In the adrenalectomized rats with NaCl solution available, mineralocorticoid treatment decreased both urinary sodium excretion and daily sodium intake. Brain nuclei in the lamina terminalis also became activated in intact rats made sodium deplete by treatment with the diuretic, furosemide. Relative to sodium-deplete intact rats, however, sodium-deplete adrenalectomized rats had a greater number of neurons expressing Fos in the organum vasculosum. Treatment of sodium-deplete rats, adrenalectomized or intact, with the angiotensin II-type 1 receptor antagonist, ZD7155, decreased sodium intake and Fos expression in the subfornical organ but not in the organum vasculosum of the lamina terminalis or median preoptic nucleus. In conclusion, the results demonstrated that activation of the brain nuclei located in the lamina terminalis of adrenalectomized rats was primarily related to sodium deficit and not to the absence of the mineralocorticoid hormones, although the adrenal hormones may have a role in limiting the activation of organum vasculosum of the lamina terminalis during sodium depletion. Furthermore, the results obtained with the administration of the angiotensin receptor antagonist are consistent with the proposal that sodium appetite of the sodium-deplete rat, adrenalectomized or intact, is mediated by circulating angiotensin II acting in the subfornical organ.

Detailed localization of aquaporin-4 messenger RNA in the CNS: preferential expression in periventricular organs.

We have performed a detailed in situ hybridization study of the distribution of aquaporin-4 messenger RNA in the CNS. Contrary to expectation, we demonstrate that aquaporin-4 is ubiquitously expressed in the CNS. Strong hybridization labeling was detected in multiple olfactory areas, cortical cells, medial habenular nucleus, bed nucleus of the stria terminalis, tenia tecta, pial surface, pontine nucleus, hippocampal formation and multiple thalamic and hypothalamic areas. A low but significant hybridization signal was found, among others, in the choroid plexus of the lateral ventricles, ependymal cells, dorsal raphe and cerebellum. Overall, a preferential distribution of aquaporin-4 messenger RNA-expressing cells was evident in numerous periventricular organs. From the distribution study, the presence of aquaporin-4 messenger RNA-expressing cells in neuronal layers was evident in neuronal layers including the CA1 -CA3 hippocampal pyramidal cells, granular dentate cells and cortical cells. Further evidence of neuronal expression comes from the semicircular arrangement of aquaporin-4 messenger RNA-expressing cells in the bed nucleus of the stria terminalis and medial habenular nucleus exhibiting Nissl-stained morphological features typical of neurons. Combined glial fibrillary acidic protein immunohistochemistry and aquaporin-4 messenger RNA in situ hybridization demonstrated that aquaporin-4 messenger RNA is expressed by glial fibrillary acidic protein-lacking cells. We conclude that aquaporin-4 messenger RNA is present in a collection of structures typically involved in the regulation of water and sodium intake and that aquaporin-4 water channels could be the osmosensor mechanism responsible for detecting changes in cell volume by these cells.

Neuronal progenitor-like cells expressing polysialylated neural cell adhesion molecule are present on the ventricular surface of the adult rat brain and spinal cord.

In the adult rodent brain, it is now well established that neurons are continuously generated from proliferating neuronal progenitor cells located in the subventricular zone of the lateral ventricle (SVZ) and the dentate gyrus of the hippocampus. Recently, it has been shown that neurons can also be generated in vitro from various regions of the adult brain and spinal cord ventricular neuroaxis. As the highly polysialylated neural cell adhesion molecule (PSA-NCAM) has been shown to be specifically expressed by neuronal progenitor cells of the SVZ and the hippocampus, the present study was designed to determine whether cells expressing this molecule could be detected in the vicinity of the ventricular system of the adult rat brain and spinal cord. After double or triple immunostaining for different neuronal and glial markers, confocal microscopy was used to examine the surface of the ventricular neuroaxis in either 40- to 50-microm-thick transverse vibratome sections cut through different brain regions, or in 200- to 300-microm-thick tissue slices including the intact surface of the brain ventricles or of the spinal cord central canal. In untreated rats, PSA-NCAM, microtubule associated protein 2 (MAP2) and class III-beta-tubulin were found to be associated with a number of neuron-like cells located on the surface of the third and fourth ventricles and of the spinal cord central canal. The proliferation of the PSA-NCAM-immunoreactive (IR) neuron-like cells detected on the surface of the third and fourth ventricles was not affected by injection of epidermal growth factor (EGF) or basic fibroblast growth factor (bFGF) into these ventricles, but was stimulated by the combined injection of EGF + bFGF. These data indicate that cells exhibiting features of neuronal progenitors are present on the ependymal surface of the adult rat brain and spinal cord ventricular axis.

Negligible role of catecholaminergic receptors in the anteroventral third ventricular region in mediating vasopressin-releasing and cardiovascular actions of prostaglandin E2.

The aim of this study was to pursue the roles of the catecholamine receptors in the anteroventral third ventricular region (AV3V), a cerebral site engaged in various stress responses, in prostaglandin (PG) E2-evoked vasopressin (AVP) release and cardiovascular action. Experiments were conducted in conscious rats in which cerebral and vascular cannulae had been implanted chronically. Local infusion (0.5 microliter, 1 min) of dopamine (150 nmol), a D1-dopaminergic agonist SKF 38393 (17 nmol) and an alpha-adrenergic agonist phenylephrine (150 nmol), as well as PGE2 (7 nmol), into the AV3V enhanced plasma AVP 5 min later, without affecting plasma osmolality and electrolytes. In contrast to the increases in both arterial pressure and heart rate observed when PGE2 was applied, dopamine and SKF 38393 did not affect these variables, and phenylephrine elevated only arterial pressure. The AV3V infusion of a beta-agonist isoproterenol (100 nmol) did not change plasma AVP, although it decreased arterial pressure and increased heart rate. The increase in plasma AVP by dopamine was not blocked by the preinfusion of the D2-antagonist sulpiride (13 nmol) into the AV3V 10 min before, but was abolished by that of the D1-antagonist SCH-23390 (8 nmol). The effects of phenylephrine on both plasma AVP and the blood pressure were prevented by the preadministration of the alpha-antagonist phenoxybenzamine (13 nmol). However, the pretreatments with phenoxybenzamine, sulpiride or SCH 23390 did not inhibit the responses of AVP, arterial pressure and heart rate caused by PGE2. These antagonists were without significant effect on AVP and other variables when given alone. The infusion sites of PGE2 and the other drugs identified histologically included the AV3V structures such as the organum vasculosum laminae terminalis or its vicinity, median preoptic nucleus, medial preoptic nucleus and periventricular hypothalamic nucleus. Dopamine or phenylephrine administered into the cerebral ventricle at the same dose as used in the AV3V application did not exert a significant effect on plasma AVP, arterial pressure and heart rate. These results suggest that catecholamine receptors in the AV3V may not be involved in the AVP-secreting, tachycardiac and pressor responses evoked by topical action of PGE2 on this area, despite their ability to influence hormone release and cardiovascular function.

Comparative analysis of FMRFamide-like immunoreactivity in caiman (Caiman crocodilus) and turtle (Trachemys scripta elegans) brains.

The distribution of FMRFamide (FMRFa)-like peptides in caiman (Caiman crocodilus) and turtle (Trachemys scripta elegans) brains was studied by immunohistochemistry. In both species, distinct groups of FMRFa-like immunoreactive (ir) perikarya were present in the medial septal nucleus, accumbens nucleus, nucleus of the diagonal band of Broca, suprachiasmatic area, lateral hypothalamic area, and periventricular hypothalamic nucleus. A few FMRFa-ir neurons in the hypothalamic area were located in the neuroepithelial cell lining of the third ventricle. FMRFa-ir fibers were scattered in all major areas of the brain, from the olfactory bulbs to the rhombencephalon. They formed dense aggregates in the medial septal area, basal telencephalon, median eminence, and infundibulum, and adjacent to the fourth ventricle. The most obvious difference between the FMRFa-ir systems in caimans and turtles concerned the number of nuclei that contained neurons with this immunoreactivity. Eight such clusters were present in the caiman brain, whereas thirteen clusters were found in the turtle brain. The turtle also displayed scattered FMRFa-ir somata in the anterior olfactory nucleus, striatum, lateral septal nucleus, medial and lateral cortex, medial forebrain bundle, lateral preoptic area, and lateral geniculate nucleus. In the caiman brain, a few FMRFa-ir neurons were noted in the ventrolateral area of the pallial commissure and an even smaller number of ir neurons was found dispersed in the optic tracts. Neither formed nuclear aggregates. The results are compared with those described for other vertebrates.