Neurite Outgrowth Stimulated by Oxytocin Is Modulated by Inhibition of the Calcium Voltage-Gated Channels.

Neuropeptide oxytocin contributes to the regulation of the neuron differentiation and cell morphology. However, the precise mechanisms are not yet fully understood. Oxytocin receptor function and its coupling to calcium entry are obvious objects of interest in relation to the neuron morphology. Postsynaptic scaffolding proteins including SHANK proteins interact with other synaptic molecules and change dendritic morphology. SH-SY5Y neuroblastoma cell line represents a useful neurobiological in vitro model to study the short-term oxytocin effects on neurite outgrowth and underlying mechanisms. In the present study, we show that oxytocin induces an increase in the intracellular calcium in SH-SY5Y cells. Specificity of the calcium influx was verified by blockade of the oxytocin receptors with oxytocin receptor antagonist L-371,257. Neurite outgrowth stimulated by oxytocin was inhibited by specific voltage-gated calcium channel blockers. The exposure of SH-SY5Y cells to oxytocin resulted in a significant increase in the gene expression of SHANK1 and SHANK3 proteins. Overall, the present data indicate that oxytocin may contribute to the regulation of scaffolding proteins expression known to be associated with clusters of calcium channels at the cell membrane. It appears that oxytocin stimulated neurite outgrowth is, at least, in part dependent on the voltage-gated calcium channels.

Oxytocin Modulates Expression of Neuron and Glial Markers in the Rat Hippocampus.

Neuropeptides including oxytocin belong to the group of factors that may play a role in the control of neuronal cell survival, proliferation and differentiation. The aim of the present study was to investigate potential contribution of oxytocin to neuronal differentiation by measuring gene and protein expression of specific neuron and glial markers in the brain. Neonatal and adult oxytocin administration was used to reveal developmental and/or acute effects of oxytocin in Wistar rats. Gene and protein expression of neuron-specific enolase (NSE) in the hippocampus was increased in 21-day and 2-month old rats in response to neonatal oxytocin administration. Neonatal oxytocin treatment induced a significant increase of gene and protein expression of the marker of astrocytes - glial fibrillary acid protein (GFAP). Oxytocin treatment resulted in a decrease of oligodendrocyte marker mRNA - 2',3'-cyclic nucleotide 3'-phosphodiesterase (CNPase) - in 21-day and 2-month old rats, while no change of CD68 mRNA, marker of microglia, was observed. Central oxytocin administration in adult rats induced a significant increase of gene expression of NSE and CNPase. The present study provides the first data revealing the effect of oxytocin on the expression of neuron and glial markers in the brain. It may be suggested that the oxytocin system is involved in the regulation of development of neuronal precursor cells in the brain.

Hyperosmotic environment blunts effectivity of ischemic preconditioning against ischemia-reperfusion injury and improves ischemic tolerance in non-preconditioned isolated rat hearts.

Several studies have shown that diabetes mellitus modulates heart resistance to ischemia and abrogates effectivity of cardioprotective interventions, such as ischemic preconditioning (IP). The aim of this study was to evaluate whether the effect of hyperglycemic conditions on the severity of ischemia-reperfusion (I/R) injury in preconditioned and non-preconditioned hearts (controls, C) is related to changes in osmotic activity of glucose. Experiments were performed in isolated rat hearts perfused according to Langendorff exposed to 30-min coronary occlusion/120-min reperfusion. IP was induced by two cycles of 5-min coronary occlusion/5-min reperfusion, prior to the long-term I/R. Hyperosmotic (HO) state induced by an addition of mannitol (11 mmol/l) to a standard Krebs-Henseleit perfusion medium significantly decreased the size of infarction and also suppressed a release of heart fatty acid binding protein (h-FABP - biomarker of cell injury) from the non-IP hearts nearly to 50 %, in comparison with normoosmotic (NO) mannitol-free perfusion. However, IP in HO conditions significantly increased the size of infarction and tended to elevate the release of h-FABP to the effluent from the heart. The results indicate that HO environment plays a cardioprotective role in the ischemic myocardium. On the other hand, increased osmolarity, similar to that in the hyperglycemic conditions, may play a pivotal role in a failure of IP to induce cardioprotection in the diabetic myocardium.

Oxytocin Increases Neurite Length and Expression of Cytoskeletal Proteins Associated with Neuronal Growth.

Neuropeptide oxytocin acts as a growth and differentiation factor; however, its effects on neurite growth are poorly understood. The aims of the present study were (1) to evaluate time effects of oxytocin on expression of nestin and MAP2; (2) to measure the effect of oxytocin on gene expression of ß-actin, vimentin, cofilin, and drebrin; and (3) to measure changes in neurite length and number in response to oxytocin/oxytocin receptor antagonist L-371,257. Exposure of SH-SY5Y cells to 1 µM oxytocin resulted in a significant increase in gene expression and protein levels of nestin after 12, 24, and 48 h. Oxytocin treatment induced no changes in gene expression of MAP2; however, a decrease of protein levels was observed in all time intervals. Gene expression of ß-actin, vimentin, and drebrin increased in response to oxytocin. Oxytocin induced significant elongation of neurites after 12, 24, and 48 h. No change in neurite length was observed in the presence of the combination of retinoic acid and oxytocin receptor antagonist L-371,257. Oxytocin treatment for 12 h increased the number of neurites. Overall, the present data suggest that oxytocin contributes to the regulation of expression of cytoskeletal proteins associated with growth of neuronal cones and induces neurite elongation mediated by oxytocin receptors at least in certain types of neuronal cells.

Prolactin increases expression of cytoskeletal proteins in SK-N-SH cells.

Although many studies have demonstrated the role of prolactin in the central nervous system, there is a considerable lack of known effects of prolactin on the parameters of neurogenesis and neuronal differentiation. The aim of the present study was to test whether prolactin changes gene expression and protein levels of nestin and microtubule-associated protein 2 (MAP2) in neuroblastoma (SK-N-SH) and glioblastoma (U-87MG) cells. Nestin and MAP2 represent cytoskeletal proteins associated with neuronal differentiation and they contribute to radial growth of the axons, dendrites and glial processes. SK-N-SH and U-87MG cells were exposed to prolactin (10 nM) for 48 h. Total mRNA was extracted. After reverse transcription, qPCR with specific primers for nestin and MAP2 was performed. The levels of proteins were measured by the In-Cell Western assay. Mitochondrial activity test was used to evaluate the viability of cells under the influence of prolactin. Incubation with 10 nM prolactin did not change the viability, either in SK-N-SH or in U-87MG cells. Prolactin significantly increased the gene expression and protein levels of both nestin and MAP2 in SK-N-SH cells, while no significant changes were observed in U-87MG cells. The presented data suggest that prolactin is linked to the regulation of cytoskeletal proteins in the neuronal type of cells and might be important for their differentiation.

Effect of cellular cholesterol changes on insulin secretion by tumor cell lines.

UNLABELLED: Glucose and cell swelling induce insulin secretion by alternative signaling pathways. Swelling-induced secretion is in most systems independent of calcium and various mediators of glucose stimulation. Comparison of two insulinoma tumor cell lines revealed surprising difference; INS-1E cells in contrast to INS-1 cells and isolated rat pancreatic islets do not respond to hypotonicity in the presence of calcium. To delineate the role of cholesterol the effect of its extraction or addition on the insulin secretion in response to glucose and cell swelling was compared. INS-1E cells have significantly higher cholesterol content than INS-1 cells (58.5 ± 2.9 and 46.3 ± 2.5 mg chol/mg prot respectively). After cholesterol desorption by 1.0, 5.0 and 10.0 mM of carboxymethyl-ß-cyclodextrin, methyl-ß-cyclodextrin, or 2-hydroxypropyl-ß- cyclodextrin the response to hypotonicity in INS-1E cells emerged. On the contrary, supplementation of INS-1 cells with cholesterol inhibited their response to cell swelling. Cyclodextrin pretreatment inhibited glucose-induced insulin secretion from INS-1 cells while INS-1E cells were more resistant to their effect. CONCLUSION: Cellular cholesterol content substantially affects secretory process; both high and low levels could be inhibitory. Absence of swelling-induced insulin secretion in INS-1E cells despite adequate response to glucose is related to their high cholesterol content. Optimal cholesterol concentration is different for either type of stimulation; swelling-induced mechanism is more sensitive to higher cholesterol content. The difference is likely to reflect involvement of sequential type exocytosis after cell swelling. Sensitivity of secretory processes suggests that either hypercholesterolemia or excessive effort to decrease plasma cholesterol in patients could have adverse effect on insulin secretion.

Different secretory response of pancreatic islets and insulin secreting cell lines INS-1 and INS-1E to osmotic stimuli.

Objective of this study was to characterize osmotically-induced insulin secretion in two tumor cell lines. We compared response of freshly isolated rat pancreatic islets and INS-1 and INS-1E tumor cell lines to high glucose, 30 % hypotonic medium and 20 % hypertonic medium. In Ca(2+)-containing medium glucose induced insulin release in all three cell types. Hypotonicity induced insulin secretion from islets and INS-1 cells but not from INS-1E cells, in which secretion was inhibited despite similar increase in cell volume in both cell types. GdCl(3) (100 micromol/l) did not affect insulin response from INS-1E cells to hypotonic challenge. Hypertonic medium inhibited glucose-induced insulin secretion from islets but not from tumor cells. Noradrenaline (1 micromol/l) inhibited glucose-induced but not swelling-induced insulin secretion from INS-1 cells. Surprisingly, perifusion with Ca(2+)-depleted medium showed distinct secretory response of INS-1E cells to hypotonicity while that of INS-1 cells was partially inhibited. Functioning glucose-induced insulin secretion is not sufficient prerequisite for hypotonicity-induced response in INS-1E cells suggesting that swelling-induced exocytosis is not essential step in the mechanism mediating glucose-induced insulin secretion. Both cell lines are resistant to inhibitory effect of hyperosmolarity on glucose-induced insulin secretion. Response of INS-1E cells to hypotonicity is inhibited by the presence of Ca(2+) in medium.

Thyrotropin-releasing hormone in rat heart: effect of swelling, angiotensin II and renin gene.

AIM: This study was performed to examine thyrotropin-releasing hormone (TRH) secretion and regulation in rat heart. METHODS: Expression of prepro-TRH gene in left atrium and left ventricle was studied by RT-PCR. TRH secretion from slices of left auricle and left ventricle in response to cell swelling (induced by hypotonic medium or ethanol in isosmotic medium), angiotensin II and losartan and their combinations was studied. RESULTS: RT-PCR revealed two times higher prepro-TRH expression in left auricle than left ventricle. In transgenic rats with extra copy of mouse renin gene a marked increase of prepro-TRH expression in the heart was noted but the relative difference between left atrium and left ventricle persisted. The swelling stimulated TRH release from both left auricle and left ventricle and this stimulation could not be inhibited by bumetanide. Angiotensin II (10 nmol L-1) added into medium significantly decreased basal secretion of TRH. The inhibiting effect of Angiotensin II was prevented by 1 micromol L-1 losartan, an angiotensin II AT1 receptor blocker. When angiotensin II and hypotonicity were applied simultaneously, swelling-induced secretion persisted. CONCLUSION: TRH secretion from heart slices has attributes of regulated secretion--depending on the stimulus it could be either stimulated or inhibited. Renin positively affects prepro-TRH expression in the heart. Angiotensin II inhibits TRH secretion from heart tissue by a mechanism involving AT1 receptors. Swelling-induced TRH secretion overrides inhibitory effect of angiotensin II. Swelling could be a useful tool when natural or pharmacological secretagogue is unknown. Peptides and proteins released by swelling could be mediators of local and remote ischaemic preconditioning protecting from subsequent ischaemia.

The effect of swelling on TRH and oxytocin secretion from hypothalamic structures.

1. Cell swelling induces exocytosis of material stored in secretory vesicles resulting in a secretory burst of peptidic hormones or enzymes from various types of cells including endocrine cells and neurons. We have previously shown that swelling-induced exocytosis possesses limited selectivity; hypotonic medium evokes TRH but not oxytocin release from hypothalamic paraventricular nucleus (PVN) and neurohypophysis (NH). 2. It is the aim of this study to ascertain whether the swelling-induced oxytocin secretion could be unmasked by the inhibition of specific osmotic response using Ca(2+)-free medium and GdCl(3), an inhibitor of stretch activated channels. 3. Oxytocin release from the PVN was stimulated by the hypotonic medium only in the presence of 50 or 100 microM GdCl(3.) Oxytocin release from supraoptic nucleus (SON) was also stimulated by the Ca(2+)-free hypotonic medium in the presence of GdCl(3). Oxytocin secretion from the NH was not stimulated even in the presence of GdCl(3), both in Ca(2+) containing and Ca(2+)-free medium. TRH response to swelling-inducing stimulus was not affected by the presence of GdCl(3). 4. An intranuclear oxytocin secretion to hyposmotic stimulation within the PVN and the SON could be unmasked by the inhibiting specific response by GdCl(3). At these conditions general secretory response to swelling-inducing stimuli emerged. Secretion of oxytocin from the NH was not affected by any of these treatments. 5. Peptides and proteins released after cell swelling can play an important role in the pathophysiology of ischemia and could be mediators of local or remote preconditioning. Disruption of mechanosensitive gating in magnocellular neurosecretory cells could result in an inadequate secretory response (e.g. stimulation instead of inhibition and vice versa) of hormones engaged in water and salt metabolism regulation.

Effect of neonatal streptozotocin and thyrotropin-releasing hormone treatments on insulin secretion in adult rats.

Neonatal STZ (nSTZ) treatment results in damage of pancreatic B-cells and in parallel depletion of insulin and TRH in the rat pancreas. The injury of B-cells is followed by spontaneous regeneration but dysregulation of the insulin response to glucose persists for the rest of life. Similar disturbance in insulin secretion was observed in mice with targeted TRH gene disruption. The aim of present study was to determine the role of the absence of pancreatic TRH during the perinatal period in the nSTZ model of impaired insulin secretion. Neonatal rats were injected with STZ (90 microg/g BW i.p.) and the effect of exogenous TRH (10 ng/g BW/day s.c. during the first week of life) on in vitro functions of pancreatic islets was studied at the age 12-14 weeks. RT-PCR was used for determination of prepro-TRH mRNA in isolated islets. Plasma was assayed for glucose and insulin, and isolated islets were used for determination of insulin release in vitro. The expression of prepro-TRH mRNA was only partially reduced in the islets of adult nSTZ rats when compared to controls. nSTZ rats had normal levels of plasma glucose and insulin but the islets of nSTZ rats failed to response by increased insulin secretion to stimulation with 16.7 mmol/l glucose or 50 mmol/l KCl. Perinatal TRH treatment enhanced basal insulin secretion in vitro in nSTZ animals of both sexes and partially restored the insulin response to glucose stimulation in nSTZ females.

Effect of immobilization on in vitro thyrotropin-releasing hormone release from brain septum in wild-type and corticotropin-releasing hormone knock-out mice.

There is considerable evidence linking alcohol consumption, sedation, and thyrotropin-releasing hormone (TRH) in the brain septum. We have shown that ethanol in clinically relevant concentrations can in vitro induce TRH release from the septum by a mechanism involving neuronal swelling. Corticotropin-releasing hormone-deficient (CRH-KO) mice serve as an interesting model to help us understand the role of CRH in the regulation of different neuroendocrine systems. The aim of this study was to compare TRH release activity in the brain septum at basal and stress conditions in CRH-KO mice and their wild-type (WT) littermates. Experimental mice were decapitated immediately or 3 h after single (2 h) or repeated (seven times for 2 h daily) immobilization stress. The brain septum was immediately cut out and incubated to measure basal-, ethanol-, and hyposmosis-stimulated TRH release in vitro. Ethanol in isosmotic medium or hyposmotic medium stimulated TRH release from mice septal explants from WT and CRH-KO mice. The response was disturbed immediately after immobilization and recovered 3 h later. Our results show that immobilization stress transiently affects the TRH system in brain septum. Inborn absence of CRH does not affect septal TRH and its response to ethanol before and 3 h after immobilization.

Cell swelling induced secretion of TRH by posterior pituitary, hypothalamic paraventricular nucleus and pancreatic islets: effect of L-canavanine.

The aims of this study were to test if ethanol induces thyrotropin-releasing hormone (TRH) secretion in vitro from the posterior pituitary and hypothalamic explants by a mechanism involving cell swelling, and to characterize the pathway of stimulated secretion. Ethanol, at a concentration of 80 mM, stimulated the release of TRH from the posterior pituitary, the hypothalamic paraventricular nucleus, the median eminence, and the brain septum, when administered only in isosmolar but not in hyperosmolar medium. This indicates the involvement of a cell swelling-inducing mechanism. L-canavanine in a concentration of 3 mM, increased the basal and hyposmosis-induced TRH secretion from the posterior pituitary and the paraventricular nucleus, and both basal and ethanol-induced TRH secretion from isolated pancreatic islets. This indicates the presence of both constitutive and regulatory secretory pathways. Our results suggest that cell swelling induces exocytosis from clathrin coated granules.

The osmotic component of ethanol and urea action is critical for their immediate stimulation of thyrotropin-releasing hormone (TRH) release from rat brain septum.

There is considerable evidence linking alcohol consumption and sedation and TRH in the brain septum. Moreover, innate septal TRH concentration is inversely related to the degree of ethanol preference. Recently we demonstrated in rats that four-week ethanol drinking increased the septal TRH content by 50 %. We had shown previously that ethanol induces neuronal swelling, which is known to evoke the secretion of hormones, peptides and amino acids from various types of cells. We have therefore explored the effect of hyposmotic medium and of 80 and 160 mM ethanol and 80 mM urea (both permeant molecules) in isosmotic and hyperosmotic (preventing cell swelling) media on the in vitro release of TRH by the rat septum. Lowering medium osmolarity resulted in a hyposmolarity-related increase in TRH secretion. Both ethanol and urea stimulated TRH release only in isosmolar solution. Our data indicate that ethanol in clinically relevant concentrations can induce TRH release from the septum by a mechanism involving neuronal swelling.

Body mass, plasma leptin, glucose, insulin and C-peptide in offspring of diabetic and non-diabetic mothers.

OBJECTIVE: The aim was to investigate the relationship between body mass index (BMI), plasma leptin, glucose, insulin and C-peptide levels in the offspring of diabetic mothers (DM) and non-diabetic healthy mothers (HM). DESIGN: Seventy-two offspring (37 girls and 35 boys, age 4-20 years) of DM were investigated in a prospective study. Those 14-16 years old (Tanner stage II-IV) were compared with age-matched offspring of HM (33 girls and 33 boys). RESULTS: BMI strongly correlated with plasma leptin concentration in the offspring of both DM and HM children. There were higher BMI and plasma leptin and glucose levels in DM than in HM children. There was no difference in plasma insulin or C-peptide levels between HM and age-matched DM children. There was a highly significant positive correlation between plasma leptin and C-peptide in boys of DM. CONCLUSIONS: The higher plasma leptin found in the offspring of DM reflects their higher BMI. A moderately high but still normal glycemia might be a preclinical sign of insulin resistance or other disturbance of glucoregulation.

Regulation of hormone secretion by acute cell volume changes: Ca(2+)-independent hormone secretion.

Exocytosis of intravesicular material should help a cell meet a relative extracellular hyposmotic challenge by expanding the plasmalemma through fusion with vesicular membrane. Cell swelling evokes an immediate secretory burst of hormones stored in secretory vesicles with dynamics indistinguishable from those induced by specific secretagogues. Hormone secretion induced by cell swelling is not associated with a rise in cAMP, IP(3), or prostaglandins, and it is not depressed by inhibition of stretch mechano-receptors or aquaporin channels. In contrast to most types of regulated secretion, that induced by cell swelling in normal cells does not require a rise in intracellular Ca(2+) through opening L-type Ca(2+) channels. However, such Ca(2+) influx is essential for cell-swelling induced secretion in tumor-derived pituitary cells. Cell swelling induces universal secretion of exocytotic material. The response of cells specialized in osmoregulation is, however, different. Possible physiological significance: Consistent stimulation of secretion occurs with a 4% hyposmolar challenge. It is likely that fluctuations in osmotic pressure with resultant cell volume changes have a significant regulatory role in hormone secretion. Released hormones could also play an important role in the pathophysiology of ischemia. Exocytosis itself does not have an essential role in volume regulation.

Comparison of pancreatic and hypophysiotropic TRH systems.

The thyrotropin-releasing hormone (TRH) is a molecule with widespread distribution through many organ systems. The function of TRH is probably not identical in each system so that TRH synthesis and secretion may be unique for each system under specific experimental conditions. The present study was designed to explore the common and diverse features of the regulation of TRH encoded with the same gene in two different organs: hypophysiotropic hypothalamus and pancreatic islets. During in vitro incubation, the TRH content in hypothalamic structures remained stable while that in isolated pancreatic islets increased sharply. In contrast to the pancreatic islets, exposure to different concentrations of D-glucose did not affect TRH release from the hypothalamic paraventricular nucleus or median eminence. This divergence in the regulation of the hypophysiotropic and pancreatic TRH systems may be related to differences in the role of TRH produced in these tissues.

Four-week ethanol drinking increases both thyrotropin-releasing hormone (TRH) release and content in rat pancreatic islets.

Ethanol exerts profound effects on the endocrine and exocrine pancreas. Some effects of chronic alcohol consumption on insulin secretion in response to glucose load are similar to those of TRH gene disruption. TRH is present in insulin-producing B-cells of the islets of Langerhans; its role in this location is still not fully explored. To examine the possible effect of long-term in vivo ethanol treatment on pancreatic TRH we compared three groups of rats: a 10% (wt:vol) ethanol-drinking group (E), absolute controls (AC) and pair-fed (PF) group with solid food intake corresponding to that of E. The fluidity of pancreatic membranes was not affected by chronic in vivo exposure of rats to ethanol, but was significantly decreased in PF group. Four-week treatment resulted in significantly higher TRH content in isolated islets of the E group and increased basal and 80 mM isotonic ethanol-induced secretion compared to AC and PF. Plasma levels of insulin, C-peptide, IGF-I, and glycemia were, however, not affected by ethanol treatment. Cell swelling, which can be induced by the presence of permeants (e.g. ethanol) in an isotonic extracellular medium, is a strong stimulus for secretion in various types of cells. In the present study, isosmotic ethanol (40, 80, and 160 mM) induced dose-dependent release of TRH and insulin from adult rat pancreatic islets in vitro. The same concentrations were not effective when applied in a hyperosmotic medium (addition of ethanol directly to the medium), thus indicating the participation of cell swelling in the ethanol-induced secretion. In conclusion, chronic ethanol treatment significantly affected pancreatic TRH and this effect might be mediated by cell swelling. The role of these changes in the profound effect of ethanol on the endocrine and exocrine pancreas remains to be established.

Glucose stimulates and insulin inhibits release of pancreatic TRH in vitro.

OBJECTIVE: Pancreatic TRH is present in insulin-producing B-cells of the islets of Langerhans. There is fragmentary evidence that it may be involved in glucoregulation. The aim of our present study was to analyze how glucose and insulin affect TRH secretion by the pancreatic islets. DESIGN: Isolated pancreatic islets were incubated with different concentrations of glucose, insulin and glucagon, and TRH release was measured. RESULTS: In the present study, 6 and 12mmol/l d-glucose caused significant TRH release from isolated adult rat pancreatic islets when compared with that in the presence of the same concentrations of biologically ineffective l-glucose. Thirtymmol/l d-glucose was also ineffective, but this was not due to depression of secretion by hyperosmolarity since isosmotic compensation for the high glucose addition did not restore its stimulatory effect. Five micromol/l dibutyryl cyclic 3',5'-adenosine monophosphate (db-cAMP) increased both basal and glucose-stimulated TRH release, but this effect was not seen with 50micromol/l db-cAMP. Stimulation of phosphodiesterase by imidazole resulted in decreased basal but not glucose-stimulated release of TRH. Glucagon (10(-7)mol/l) did not affect either basal or glucose-stimulated release of TRH, while insulin (10(-7) and 10(-6)mol/l) inhibited both. CONCLUSION: Our present data showing that glucose stimulates and insulin inhibits pancreatic TRH release are compatible with the possibility that this substance may play a role in glucoregulation.

Hypo-osmolarity stimulates and high sodium concentration inhibits thyrotropin-releasing hormone secretion from rat hypothalamus.

The hypothalamic paraventricular nucleus, representing cell bodies in which thyrotropin-releasing hormone is synthesized, and the median eminence, representing nerve terminals, were incubated in vitro. Various hypo- and hyperosmotic solutions were tested to determine osmotic sensitivity of thyrotropin-releasing hormone secretion. High KCl (56 mM) causing membrane depolarization was used as a non-specific control stimulus to induce thyrotropin-releasing hormone secretion. A 30% decrease of medium osmolarity (from 288 to 202 mOsmol/l) increased thyrotropin-releasing hormone secretion from both the paraventricular nucleus and median eminence. A 30% decrease of medium NaCl content by its replacement with choline chloride did not affect basal thyrotropin-releasing hormone secretion. Increasing medium osmolarity with biologically inactive L-glucose did not affect basal or KCl-induced thyrotropin-releasing hormone secretion from either structure. Medium made hyperosmotic (350-450 mOsmol/l) by increasing the NaCl concentration resulted in a dose-dependent decrease of basal thyrotropin-releasing hormone secretion and abolished KCl-induced thyrotropin-releasing hormone secretion. If an osmotically equivalent amount of choline chloride was substituted for NaCl, there was no effect on thyrotropin-releasing hormone secretion, indicating a specific action of Na+. This study indicates a specific sensitivity to high concentrations of Na+ ions of both thyrotropin-releasing hormone-producing parvocellular paraventricular neurons and thyrotropin-releasing hormone-containing nerve terminals in the median eminence.