Regulation by osmotic stimuli of galanin-R1 receptor expression in magnocellular neurones of the paraventricular and supraoptic nuclei of the rat.

Neurones of the supraoptic nucleus (SON) and the magnocellular and parvocellular divisions of the paraventricular nucleus (PVN) express galanin and [125I]galanin binding sites. Although the precise role(s) of galanin in these different cell populations is still unknown, it has been shown to regulate the electrophysiological, neurochemical and secretory activity of magnocellular neurones. In light of the well-described effects of hyperosmotic stimuli, such as salt-loading on magnocellular neurone activity and galanin synthesis and release, and the recent identification of multiple galanin receptors in brain, this study assessed the possible regulation of galanin receptor subtype expression in the PVN/SON of salt-loaded, dehydrated and food-deprived rats. Gal-R1 mRNA was abundant in the SON (and magnocellular PVN) of control rats and levels were increased in these same cells after 4 days of salt-loading (2% NaCl solution as drinking water) or water deprivation. The density of specific [125I]galanin(1-29) binding and the intensity of Gal-R1-like immunostaining were also increased in the characteristically enlarged, magnocellular neurones of the PVN and SON after these treatments. Gal-R2 mRNA was detected in the parvocellular PVN, but levels were not altered by the hyperosmotic stimuli. In contrast, food deprivation (4 days), which has been shown to reduce levels of several neurochemical markers in magnocellular neurones, produced a significant reduction in Gal-R1 (and galanin) mRNA levels in the SON, but no consistent change in neurone size, [125I]galanin binding levels, or Gal-R1 immunostaining. Along with previous findings from this and other laboratories, these data suggest that the expression of galanin and Gal-R1 receptors is regulated in parallel with functional and morphological changes in hypothalamic magnocellular neurones. Furthermore, Gal-R1 immunoreactivity was primarily detected in somatodendritic areas and thus galanin may influence the activity of these cells, particularly vasopressin synthesis/release, via autocrine or paracrine activation of Gal-R1 receptors, especially during long-lasting stimulation.

Galanin-like peptide mRNA in neural lobe of rat pituitary. Increased expression after osmotic stimulation suggests a role for galanin-like peptide in neuron-glial interactions and/or neurosecretion.

Galanin-like peptide (GALP) was recently identified in the porcine hypothalamus, pituitary gland and gut, and has reported selectivity for the GalR2, c.f. the GalR1 receptor. GALP cDNAs have been cloned from pig, rat and human, and GALP mRNA expression is restricted to the arcuate nucleus in normal rat brain. This study examined the regional and cellular distribution of GALP mRNA in the rat pituitary gland, and subsequently determined the effect of osmotic stimulation on GALP transcript levels. GALP mRNA was not detected in the anterior or intermediate lobes, but moderate levels of GALP mRNA were present in the neural (posterior) lobe, in presumed pituicytes, of normal male and female rats. Osmotic stimulation by dehydration or salt loading produced a time-dependent increase in GALP mRNA levels in the neural lobe. Thus, dehydration for 4 days increased GALP mRNA 40-fold, while salt loading for 4, 7 or 10 days increased GALP levels 14-, 21- and 25-fold, respectively (p < or = 0.001). Levels of vasopressin (VP) mRNA in the neural lobe were also increased by these treatments, consistent with previous reports. Galanin (GAL) and GalR2 receptor mRNAs were not detected in the neural lobe, under normal or osmotic stimulation conditions. In addition, GALP mRNA levels in the arcuate nucleus were not altered in dehydrated or salt-loaded rats; and GALP mRNA was not detected in magnocellular neurons of the supraoptic or paraventricular nucleus, despite the characteristic up-regulation of VP and GAL mRNA in these cells. In view of the established anatomy and function of VP/oxytocin neurons in the hypothalamo-neurohypophysial system and the role played by pituicytes in their regulation, the likely synthesis/release of GALP by these specialized astrocytes strongly suggests a role for this novel peptide in regulation of pituicyte morphology/function and/or neurohormone release.

Distribution, regulation and role of hypothalamic galanin systems: renewed interest in a pleiotropic peptide family.

1. Galanin peptide and galanin receptor-binding sites are known to be widely distributed within the central nervous system, particularly in the hypothalamus in the preoptic area, the paraventricular (PVN) and supraoptic (SON) nuclei and the arcuate nucleus/median eminence. 2. The present brief review focuses on some recent studies of the regional and cellular localization of mRNA encoding galanin and two galanin receptor subtypes (GalR1 and GalR2) in the hypothalamus, regulation of galanin and/or galanin receptor expression in various nuclei by physiological stimuli, electrophysiological effects of galanin on hypothalamic neurons and the isolation and cloning of galanin-like peptide (GALP), a putative endogenous ligand for GalR2. 3. In situ hybridization studies in rat brain have demonstrated an abundance of GalR1 mRNA in SON, magnocellular (m) and parvocellular (p) PVN and dorsomedial, ventromedial and arcuate nuclei. In contrast, GalR2 mRNA is enriched in pPVN, but not mPVN, and is not detected in SON. In addition, GalR2 mRNA is present in the dorsomedial nucleus and is enriched in the arcuate nucleus compared with GalR1 transcripts, with numerous labelled cells in all subdivisions. 4. Neurons of the SON and PVN contain vasopressin and/or oxytocin, along with several other peptides, and the production and release of these hormones and peptides are modulated by various physiological stimuli. In relation to galanin systems, GalR1 and galanin expression is increased in magnocellular neurons by salt loading and is downregulated by lactation, consistent with an increased inhibition by galanin of vasopressin release following osmotic stimulation and a decreased inhibition of oxytocin release during lactation. 5. Powerful inhibitory effects of galanin on the electrical (and secretory) activity of magnocellular neurons and complex presynaptic actions of galanin on the synaptic release of glutamate in the arcuate nucleus in vitro suggest an active role for multiple galanin receptor subtypes in the regulation of these hypothalamic systems in vivo. 6. The recent isolation of a peptide from porcine hypothalamus (GALP-1-60) that is structurally related to galanin and appears to be selective for GalR2 over GalR1 and the subsequent cloning of GALP cDNA from pig, rat and humans should allow studies to help reveal the physiological role played by galanin receptor subtypes (especially GalR2) and their multiple ligands in the hypothalamus and other brain areas.

Galanin-like peptide (GALP) mRNA expression is restricted to arcuate nucleus of hypothalamus in adult male rat brain.

Galanin-like peptide (GALP) is a novel 60-amino acid neuropeptide, isolated from porcine hypothalamus and subsequently identified in rats and humans, which has reported selectivity for the Gal-R2 galanin receptor [Ohtaki T et al: J Biol Chem 1999; 274: 37041-37045]. In the current study, the regional and cellular distribution of GALP mRNA in rat brain has been investigated by in situ hybridization of [(35)S]-labelled oligonucleotide probes. In a thorough screening of adult male rat brain, GALP mRNA expression was detected only throughout the rostrocaudal extent of the arcuate nucleus (ARC) with the most abundant hybridization signal in the posterior, periventricular zones. GALP mRNA-positive neurons were mostly localized in the ventromedial division of the ARC, with many closely adjacent to the wall of the third ventricle. Smaller numbers of labelled neurons were also found in ventrolateral areas. The distribution of GALP mRNA was somewhat complementary to that for galanin (GAL) mRNA in the ARC, but contrasted with the broad distribution of this transcript throughout the hypothalamus. GAL mRNA was also distributed along the rostrocaudal extent of the ARC, but was most abundant in the anterior to middle levels and in ventrolateral regions. Interestingly, somatostatin mRNA expression appeared to overlap the distribution of GALP mRNA in posterior, ventromedial regions of the ARC. Thus, in adult rat brain GALP mRNA expression was restricted to a discrete subpopulation of neurons in the ARC, with a unique localization pattern unlike GAL or many other known peptide- and transmitter-containing cells in this region. GALP could, however, be co-expressed in sub-populations of other neuronal phenotypes (e.g. somatostatin neurons) or within cells that express Gal-R2 receptors. In view of the established anatomy and function of the ARC and the restricted localization of GALP mRNA, this novel peptide is likely to play a role in regulation of anterior pituitary hormone secretion, or in regulation of other hypothalamic peptide and transmitter systems.

Galanin-R1 and -R2 receptor mRNA expression during the development of rat brain suggests differential subtype involvement in synaptic transmission and plasticity.

The present study employed 35S-labelled oligonucleotides and in situ hybridization to examine the distribution in the developing rat brain of mRNA encoding two galanin receptor subtypes, i.e. Gal-R1 and Gal-R2. Gal-R1 and/or Gal-R2 mRNA was detected at embryonic day (E) 20 and from postnatal day (P) 0-70. Gal-R1 mRNA was highly expressed in olfactory regions, ventral hippocampal CA fields, dorsomedial thalamic areas and many hypothalamic nuclei at all ages studied. In adult brain, Gal-R2 mRNA was most abundant in the dentate gyrus, anterior and posterior hypothalamus, raphe and spinal trigeminal nuclei, and in the dorsal motor nucleus of the vagus. At P0-P7, Gal-R2 mRNA was more widely distributed and abundant than at other ages, with highest levels of expression detected throughout the neocortex and thalamus. Thus, Gal-R2 transcripts had a more restricted distribution than Gal-R1 and were differentially abundant at different ages, while the distribution and relative abundance of Gal-R1 mRNA did not alter substantially during postnatal development. In general, Gal-R1 and -R2 mRNAs were localized in regions previously shown to contain [125I]-galanin binding sites and galanin-positive terminals in adult brain. Galanin-immunostaining was assessed in postnatal brain to determine whether peptide innervation correlated with observed transient receptor expression, but was not particularly enriched in Gal-R2 mRNA-positive areas of P4 or P7 brain. These results, together with earlier findings [e.g. Burazin, T. C. D. & Gundlach, A. L. (1998) J. Neurochem., 71, 879-882], suggest that Gal-R1 receptors have a broad role in normal synaptic transmission, while Gal-R2 receptors, in addition to a similar role in particular pathways, may be involved in processes prominent during the establishment and maturation of synaptic connections in developing brain and during neural damage and repair in the mature nervous system.

Kainate-induced apoptosis correlates with c-Jun activation in cultured cerebellar granule cells.

We have investigated the involvement of c-Jun in cell death induced by exposure of primary cultures of murine cerebellar granule cells to the glutamate receptor agonist kainate (KA) and evaluated its possible use as a marker for apoptosis. Using cerebellar granule cell neurones from postnatal day 7 mice, we found that 1 hr exposure to KA (1-1000 microM) induced a concentration-dependent neuronal cell death with characteristic apoptotic morphology, including cell shrinkage, neurite blebbing and DNA fragmentation. In addition KA-induced a concentration-dependent expression of c-Jun mRNA and protein as determined by in situ hybridization and immunocytochemistry respectively. DNA fragmentation was detected using terminal transferase-mediated nick-end (TUNEL) labelling and agarose gel electrophoresis. KA-induced cell death was significantly attenuated by the non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 50 microM), which shifted the concentration-response curve significantly rightward. The number of apoptotic cell bodies, determined by TUNEL, was also reduced by CNQX (50 microM), with only 15-20% of neurones staining positive after exposure to 1mM KA. In addition, the number of positively stained cells for c-Jun protein and mRNA was substantially reduced by CNQX (50 microM) as determined by random and representative cell counts. These results show for the first time that KA induced apoptotic neuronal death in cultured murine cerebellar granule cells involves the induction of c-Jun mRNA and protein, suggesting the involvement of this immediate early gene in excitotoxic receptor-mediated apoptosis and its potential use as a marker for apoptotic cell death.

Neurotoxin domoic acid produces cytotoxicity via kainate- and AMPA-sensitive receptors in cultured cortical neurones.

Domoic acid, a naturally occurring kainoid, has been responsible for several outbreaks of fatal poisoning after shellfish ingestion, and we examined its neurotoxic mechanism in cultured murine cortical neurones. Using observations of neuronal viability and morphology, exposure to domoic acid for 24 h was found to induce substantial concentration-dependent neuronal cell death. Domoic acid-mediated neuronal death was attenuated by the non-N-methyl-D-aspartate receptor antagonist 6-cyano-7-nitro-quinoxaline-2,3-dione and the alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) receptor-selective antagonist LY293558 ((3S,4aR,6R,8aR)-6-[2-(1H-tetrazol-5-yl)-ethyl]-1,2,3, 4,4a,5,6,7,8,8a-decahydroisoquinoline-3-carboxylic acid), but unaffected by NS-102 (5-nitro-6,7,8,9-tetrahydrobenzo[g]indole-2, 3-dione-3-oxime)--a low-affinity kainate receptor antagonist. Domoic acid was equipotent with (S)-AMPA (EC50 values 3.8 and 3.4 microM respectively); however, (S)-AMPA induced only 50% cell death compared to > 80% cell death induced by domoic acid. Kainate also killed > 80% of cortical neurones; however, domoic acid was about 19 times more potent than kainate (EC50 75 microM). We show the potent neurotoxicity of domoic acid for the first time in a pure neuronal model and indicate that domoic acid acts via high-affinity AMPA- and kainate-sensitive glutamate receptors to produce excitotoxic cell death.

Apoptosis induced via AMPA-selective glutamate receptors in cultured murine cortical neurons.

We have investigated the mechanisms of cell death induced by long-term exposure to the glutamate receptor agonist (S)-alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate [(S)-AMPA]. Using primary cultures of pure neurons (95%) grown in serum-free conditions, we found that 24-h exposure to (S)-AMPA (0.01-1,000 microM) induced concentration-dependent neuronal cell death (EC50 = 3 +/- 0.5 microM) with cellular changes including neurite blebbing, chromatin condensation, and DNA fragmentation, indicative of apoptosis. (S)-AMPA induced a delayed cell death with DNA fragmentation occurring in approximately 50% of cells at concentrations between 100 and 300 microM detected using terminal transferase-mediated dUTP nick end-labeling (TUNEL) and agarose gel electrophoresis. Apoptotic chromatin condensation was detected using 4,6-diamidino-2-phenylindole, a fluorescent DNA binding dye. Cell death induced by (S)-AMPA was attenuated by the AMPA receptor-selective antagonist LY293558 (10 microM) and the non-NMDA receptor antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX; 50 microM), yielding EC50 values of 73 +/- 5 and 265 +/- 8 microM, respectively, and was unaffected by the NMDA receptor antagonist MK-801 (10 microM). The number of apoptotic nuclei induced by 300 microM (S)-AMPA (57%) was also reduced substantially by the antagonists LY293558 and CNQX, with only 20% and 18% of neurons, respectively, staining TUNEL-positive at 24 h. In addition, cycloheximide (0.5 microg/ml) also inhibited (S)-AMPA-induced DNA fragmentation and cell death. Our results show that long-term exposure to AMPA can induce substantial neuronal death involving apoptosis in cultured cortical neurons, suggesting a wide involvement of AMPA-sensitive glutamate receptors in excitotoxic injury and neurodegenerative pathologies.

Cholecystokinin-GABA interactions in rodent cortex: analyses of cholecystokinin effects on K(+)- and L-glutamate-induced release of [3H]GABA from rat cortical slices and cultured mouse cortical neurones.

Neurones of the cerebral cortex immunoreactive for the neuropeptide, cholecystokinin (CCK), also invariably contain GABA. Hence CCK is believed to modulate some aspect of GABAergic synaptic activity. The present study therefore investigated the effects of CCK on basal, K(+)- and L-glutamate-induced release of [3H]GABA from slices of rat neocortex and cultured murine neocortical neurones. Rat neocortical prisms loaded with [3H]GABA (10 nM) were superfused with Krebs-Henseleit buffer and stimulated twice (S1 and S2, 2 min) with K+ (30 mM). Release associated with each stimulus was measured and expressed relative to basal release (R1 and R2). The effects of non-selective and CCKB selective agonists, CCK-8S and CCK-4, respectively, on basal and K(+)-induced release of [3H]GABA were subsequently assessed by alternately including the peptides in S2 and comparing R2/R1 and S2/S1 ratios to control experiments. Contrary to previous findings, CCK-8S (30 nM-1 microM) and CCK-4 (0.3 nM-1 microM) failed to influence basal or K(+)-induced release. In similar experiments, murine cortical neurones superfused with HEPES balanced salt buffer, released exogenous [3H]GABA upon stimulation (1 min) with either K+ (55 mM) or L-glutamate (30 microM). However, CCK-8S, CCK-4 (both 300 nM-1 microM) and the CCKB selective antagonist, L365,260 (1 microM), failed to influence basal, K(+)- or L-glutamate-induced release of [3H]GABA from these neurones when included in S2. These data therefore do not support the postulate that CCK acting via CCKA or CCKB receptors modulates release of GABA under the present experimental conditions. GABA-CCK interactions were not specifically studied because only L-glutamate (30 microM) significantly elevated release of CCK-like immunoreactivity (115% above basal) in murine cortical neurones: basal release of CCK was estimated to be 7 and 11 pM from neurones and slices, respectively. Further studies employing more rigorous stimulation and perhaps examining endogenous GABA release are necessary to fully investigate the co-release of CCK and GABA.

Effectors of the mammalian plasma membrane NADH-oxidoreductase system. Short-chain ubiquinone analogues as potent stimulators.

In the presence of effectors variations in the two recognized activities of the plasma membrane NADH-oxidoreductase system were studied in separate, specific in vitro assays. We report here that ubiquinone analogues that contain a short, less hydrophobic side chain than coenzyme Q-10 dramatically stimulate the NADH-oxidase activity of isolated rat liver plasma membranes whereas they show no effect on the reductase activity of isolated membranes. If measured in assays of the NADH:ferricyanide reductase of living cultured cells these compounds have only a limited effect; the oxidase activity of whole cells is not measurable in our hands. We have furthermore identified selective inhibitors of both enzyme activities. In particular, the NADH-oxidase activity can be significantly inhibited by structural analogues of ubiquinone, such as capsaicin and resiniferatoxin. The NADH:ferricyanide reductase, on the other hand, is particularly sensitive to pCMBS, indicating the presence of a sulfhydryl group of groups at its active site. The identification of these specific effectors of the different enzyme activities of the PMOR yields further insights into the function of this system.

(S)-5-fluorowillardiine-mediated neurotoxicity in cultured murine cortical neurones occurs via AMPA and kainate receptors.

We have examined the neurotoxic effects of kainate, (S)-alpha-amino-3-hydroxy-5-methylisoxazole-4-propionate (AMPA) and the novel AMPA-receptor preferring agonist (S)-5-fluorowillardiine in murine cultured cortical neurones. Kainate induced > 90% cell death (EC50 65 microM) and (S)-AMPA only about 50% cell death (EC50 3.1 microM), both in a monophasic dose-dependent manner. (S)-5-Fluorowillardiine also killed > 90% of neurones, however, in a biphasic dose-dependent manner (EC50 0.70 and 170 microM). Additionally, the neurotoxic effects of (S)-AMPA and (S)-5-fluorowillardiine (high-affinity component) were attenuated by the AMPA receptor antagonists LY293558 ((3,S,4aR, 6R,8aR)-6[2h91 H-tetrazol-5-yl)ethyl]-1,2,3,4,4a,5,6,7,8,8a-decahydroisoquinol ine- 3-carboxylic acid). A component of kainate and (S)-5-fluorowillardiine (low-affinity component) neurotoxicity was blocked by the low-affinity kainate receptor antagonist NS-102 (5-nitro-6,7,8,9-tetrahydrobenzo[g]indole-2,3-dione-3-oxime). We have shown that both kainate and (S)-AMPA can effect substantial cell death in cortical neurones and that the novel agonist (S)-5-fluorowillardiine exerts its excitotoxicity through both AMPA- and kainate-preferring receptors.

Apoptosis induced by inhibitors of the plasma membrane NADH-oxidase involves Bcl-2 and calcineurin.

Activation of the plasma membrane NADH-oxidoreductase (PMOR) system by addition of growth factors or extracellular electron acceptors stimulates cellular proliferation. We now show that the vanilloids capsaicin, dihydrocapsaicin, and resiniferatoxin are inhibitors of the NADH-oxidase activity of the PMOR system and that both these and two previously identified PMOR inhibitors (chloroquine and retinoic acid) induce apoptosis in human B-cell and mouse myeloid cell lines. At the optimal concentration, PMOR inhibitors can induce between 50 and 70% of apoptosis in mouse myeloid and human B-cell lines within 8-12 h, provided these cell lines do not express Bcl-2. The immunosuppressants cyclosporin A and fujimycin (tacrolimus) inhibit PMOR inhibitor-induced apoptosis. By using combinations of these immunosuppressants and excess amounts of their nonimmunosuppressive analogues, we demonstrate that in human B-cell lines the Bcl-2-sensitive apoptotic pathway triggered by PMOR inhibitors involves signaling through the protein phosphatase calcineurin. We suggest that the PMOR system is a redox sensor that can, depending on the ambient redox environment and the availability of growth factors, regulate plasma membrane calcium fluxes and signal for apoptosis through calcineurin. Bcl-2, a protein that is thought to inhibit apoptosis by regulating reactive oxygen species and calcium fluxes in the cell, inhibits this apoptotic pathway.

Up-regulation of the plasma membrane oxidoreductase as a prerequisite for the viability of human Namalwa rho 0 cells.

We have studied aspects of the regulatory interrelationship between the plasma membrane oxidoreductase (PMOR) system and the mitochondrial respiratory capacity of human Namalwa cells. Although the role of mitochondria in the maintenance of cellular redox and energetic states is well established, the PMOR system in comparison is a poorly characterized enzyme system whose functions, particularly in relation to cellular metabolism, have not been clearly elucidated. Therefore we compared the PMOR and mitochondrial respiratory activities of human Namalwa cells during the induction by ethidium bromide treatment of rho 0 cells, which lack a functional mitochondrial respiratory system. The plasma membrane NADH-ferricyanide reductase activity of the PMOR system was found to increase in a stepwise manner concomitant with a decline in cellular mitochondrial respiratory activity. Addition of p-chloromercuriphenylsulfonic acid to the culture medium, at a concentration totally inhibiting the plasma membrane NADH-ferricyanide reductase in vitro, leads to cell death of rho 0 but not of rho + cells. Thus, the up-regulation of a functional PMOR system is a necessary phenomenon in maintaining the viability of mammalian rho 0 cells.