The sodium-activated potassium channel Slack is modulated by hypercapnia and acidosis.

Slack (Slo 2.2), a member of the Slo potassium channel family, is activated by both voltage and cytosolic factors, such as Na(+) ([Na(+)](i)) and Cl(-) ([Cl(-)](i)). Since the Slo family is known to play a role in hypoxia, and since hypoxia/ischemia is associated with an increase in H(+) and CO(2) intracellularly, we hypothesized that the Slack channel may be affected by changes in intracellular concentrations of CO(2) and H(+). To examine this, we expressed the Slack channel in Xenopus oocytes and the Slo 2.2 protein was allowed to be inserted into the plasma membrane. Inside-out patch recordings were performed to examine the response of Slack to different CO(2) concentrations (0.038%, 5%, 12%) and to different pH levels (6.3, 6.8, 7.3, 7.8, 8.3). In the presence of low [Na(+)](i) (5 mM), the Slack channel open probability decreased when exposed to decreased pH or increased CO(2) in a dose-dependent fashion (from 0.28+/-0.03, n=3, at pH 7.3 to 0.006+/-0.005, n=3, P=0.0004, at pH 6.8; and from 0.65+/-0.17, n=3, at 0.038% CO(2) to 0.22+/-0.07, n=3, P=0.04 at 12% CO(2)). In the presence of high [Na(+)](i) (45 mM), Slack open probability increased (from 0.03+/-0.01 at 5 mM [Na(+)](i), n=3, to 0.11+/-0.01, n=3, P=0.01) even in the presence of decreased pH (6.3). Since Slack activity increases significantly when exposed to increased [Na(+)](i), even in presence of increased H(+), we propose that Slack may play an important role in pathological conditions during which there is an increase in the intracellular concentrations of both acid and Na(+), such as in ischemia/hypoxia.

O(2) deprivation inhibits Ca(2+)-activated K(+) channels via cytosolic factors in mice neocortical neurons.

O(2) deprivation induces membrane depolarization in mammalian central neurons. It is possible that this anoxia-induced depolarization is partly mediated by an inhibition of K(+) channels. We therefore performed experiments using patch-clamp techniques and dissociated neurons from mice neocortex. Three types of K(+) channels were observed in both cell-attached and inside-out configurations, but only one of them was sensitive to lack of O(2). This O(2)-sensitive K(+) channel was identified as a large-conductance Ca(2+)-activated K(+) channel (BK(Ca)), as it exhibited a large conductance of 210 pS under symmetrical K(+) (140 mM) conditions, a strong voltage-dependence of activation, and a marked sensitivity to Ca(2+). A low-O(2) medium (PO(2) = 10-20 mmHg) markedly inhibited this BK(Ca) channel open probability in a voltage-dependent manner in cell-attached patches, but not in inside-out patches, indicating that the effect of O(2) deprivation on BK(Ca) channels of mice neocortical neurons was mediated via cytosol-dependent processes. Lowering intracellular pH (pH(i)), or cytosolic addition of the catalytic subunit of a cAMP-dependent protein kinase A in the presence of Mg-ATP, caused a decrease in BK(Ca) channel activity by reducing the sensitivity of this channel to Ca(2+). In contrast, the reducing agents glutathione and DTT increased single BK(Ca) channel open probability without affecting unitary conductance. We suggest that in neocortical neurons, (a) BK(Ca) is modulated by O(2) deprivation via cytosolic factors and cytosol-dependent processes, and (b) the reduction in channel activity during hypoxia is likely due to reduced Ca(2+) sensitivity resulting from cytosolic alternations such as in pH(i) and phosphorylation. Because of their large conductance and prevalence in the neocortex, BK(Ca) channels may be considered as a target for pharmacological intervention in conditions of acute anoxia or ischemia.

Human neocortical excitability is decreased during anoxia via sodium channel modulation.

When the central nervous system in humans is deprived of oxygen, the effects are potentially disastrous. Electroencephalographic activity is lost and higher brain function ceases rapidly. Despite the importance of these effects, the mechanisms underlying the loss of cortical activity are poorly understood. Using intracellular recordings of human neocortical neurons in tissue slices, we show that, whereas anoxia produces a relatively small depolarization and modest alterations in passive properties, it causes a major decrease in excitability. Whole-cell voltage-clamp studies of acutely isolated human neocortical pyramidal neurons demonstrate that anoxia and metabolic inhibition produce a large negative shift in the steady-state inactivation [h infinity (V)] curve for the voltage-dependent sodium current (INa). Inclusion of ATP in the patch pipette decreased the shift of the h infinity (V) curve by two-thirds. Because increased inactivation of INa decreases cellular metabolic demand, we postulate that this promotes neuronal survival during periods of oxygen deprivation. These data show a novel mechanism by which anoxia links metabolism to membrane ionic conductances in human cortical neurons.

Intracellular pH regulation of CA1 neurons in Na(+)/H(+) isoform 1 mutant mice.

To understand the role of Na(+)/H(+) exchanger 1 (NHE1) in intracellular pH (pH(i)) regulation and neuronal function, we took advantage of natural knockout mice lacking NHE1, the most ubiquitously and densely expressed NHE isoform in the central nervous system (CNS). CA1 neurons from both wild-type (WT) and NHE1 mutant mice were studied by continuous monitoring of pH(i), using the fluorescent indicator carboxy-seminaphthorhodafluor-1 (SNARF-1) and confocal microscopy. In the nominal absence of CO(2)/HCO(3)(-), steady-state pH(i) was higher in WT neurons than in mutant neurons. Using the NH(4)Cl prepulse technique, we also show that H(+) flux in WT neurons was much greater than in mutant neurons. The recovery from acid load was blocked in WT neurons, but not in mutant neurons, by removal of Na(+) from the extracellular solution or by using 100 microM 3-(methylsulfonyl-4-piperidino-benzoyl)-guanidine methanesulfonate (HOE 694) in HEPES buffer. Surprisingly, in the presence of CO(2)/HCO(3)(-), the difference in H(+) flux between WT and mutant mice was even more exaggerated, with a difference of more than 250 microM/s between them at pH 6.6. H(+) flux in CO(2)/HCO(3)(-) was responsive to diisothiocyanato-stilbene-2, 2'-disulfonate (DIDS) in the WT but not in the mutant. We conclude that (a) the absence of NHE1 in the mutant neurons tended to cause lower steady-state pH(i) and, perhaps more importantly, markedly reduced the rate of recovery from an acid load; and (b) this difference in the rate of recovery between mutant and WT neurons was surprisingly larger in the presence, rather than in the absence, of HCO(3)(-), indicating that the presence of NHE1 is essential for the regulation and/or functional expression of both HCO(3)(-)-dependent and -independent transporters in neurons.

Mutation in pre-mRNA adenosine deaminase markedly attenuates neuronal tolerance to O2 deprivation in Drosophila melanogaster.

O2 deprivation can produce many devastating clinical conditions such as myocardial infarct and stroke. The molecular mechanisms underlying the inherent tissue susceptibility or tolerance to O2 lack are, however, not well defined. Since the fruit fly, Drosophila melanogaster, is extraordinarily tolerant to O2 deprivation, we have performed a genetic screen in the Drosophila to search for loss-of-function mutants that are sensitive to low O2. Here we report on the genetic and molecular characterization of one of the genes identified from this screen, named hypnos-2. This gene encodes a Drosophila pre-mRNA adenosine deaminase (dADAR) and is expressed almost exclusively in the adult central nervous system. Disruption of the dADAR gene results in totally unedited sodium (Para), calcium (Dmca1A), and chloride (DrosGluCl-alpha) channels, a very prolonged recovery from anoxic stupor, a vulnerability to heat shock and increased O2 demands, and neuronal degeneration in aged flies. These data clearly demonstrate that, through the editing of ion channels as targets, dADAR, for which there are mammalian homologues, is essential for adaptation to altered environmental stresses such as O2 deprivation and for the prevention of premature neuronal degeneration.

Prostaglandin transporter expression in mouse brain during development and in response to hypoxia.

Prostaglandins (PGs) are bioactive lipid mediators released following brain hypoxic-ischemic injury. Clearance and re-uptake of these prostaglandins occur via a transmembrane prostaglandin transporter (PGT), which exchanges PG for lactate. We used Western blot analyses to examine the PGT developmental profile and its regional distribution as well as changes in transporter expression during chronic hypoxia in the neonatal mouse brain. Microsomal preparations from four brain regions (cortex, hippocampus, cerebellum and brainstem/diencephalon) showed gradual increases in prostaglandin transporter expression in all brain regions examined from postnatal day 1 till day 30. There was a significant regional heterogeneity in the prostaglandin transporter expression with highest expression in the cortex, followed by cerebellum and hippocampus, and least expressed in the brainstem/diencephalon. To further delineate the pattern of prostaglandin transporter expression, separate astrocytic and neuronal microsomal preparations were also examined. In contrast to neurons, which had a robust expression of prostaglandin transporters, astrocytes had very little PGT expression under basal conditions. In response to chronic hypoxia, there was a significant decline in PGT expression in vivo and in neurons in vitro, whereas cultured astrocytes increased their PGT expression. This is the first report on PGT expression in the CNS and our studies suggest that PGTs have 1) a widespread distribution in the CNS; 2) a gradual increase and a differential expression in various regions during brain development; and 3) striking contrast in expression between glia and neurons, especially in response to hypoxia. Since PGTs play a role as prostaglandin-lactate exchangers, we hypothesize that PGTs are important in the CNS during stress such as hypoxia.

The calcium-sensitive large-conductance potassium channel (BK/MAXI K) is present in the inner mitochondrial membrane of rat brain.

Large-conductance voltage- and calcium-sensitive channels are known to be expressed in the plasmalemma of central neurons; however, recent data suggest that large-conductance voltage- and calcium-sensitive channels may also be present in mitochondrial membranes. To determine the subcellular localization and distribution of large-conductance voltage- and calcium-sensitive channels, rat brain fractions obtained by Ficoll-sucrose density gradient centrifugation were examined by Western blotting, immunocytochemistry and immuno-gold electron microscopy. Immunoblotting studies demonstrated the presence of a consistent signal for the alpha subunit of the large-conductance voltage- and calcium-sensitive channel in the mitochondrial fraction. Double-labeling immunofluorescence also demonstrated that large-conductance voltage- and calcium-sensitive channels are present in mitochondria and co-localize with mitochondrial-specific proteins such as the translocase of the inner membrane 23, adenine nucleotide translocator, cytochrome c oxidase or complex IV-subunit 1 and the inner mitochondrial membrane protein but do not co-localize with calnexin, an endoplasmic reticulum marker. Western blotting of discrete subcellular fractions demonstrated that cytochrome c oxidase or complex IV-subunit 1 was only expressed in the mitochondrial fraction whereas actin, acetylcholinesterase, cadherins, calnexin, 58 kDa Golgi protein, lactate dehydrogenase and microtubule-associated protein 1 were not, demonstrating the purity of the mitochondrial fraction. Electron microscopic examination of the mitochondrial pellet demonstrated gold particle labeling within mitochondria, indicative of the presence of large-conductance voltage- and calcium-sensitive channels in the inner mitochondrial membrane. These studies provide concrete morphological evidence for the existence of large-conductance voltage- and calcium-sensitive channels in mitochondria: our findings corroborate the recent electrophysiological evidence of mitochondrial large-conductance voltage- and calcium-sensitive channels in glioma and cardiac cells.

Mechanisms underlying hypoxia-induced neuronal apoptosis.

In vivo models of cerebral hypoxia-ischemia have shown that neuronal death may occur via necrosis or apoptosis. Necrosis is, in general, a rapidly occurring form of cell death that has been attributed, in part, to alterations in ionic homeostasis. In contrast, apoptosis is a delayed form of cell death that occurs as the result of activation of a genetic program. In the past decade, we have learned considerably about the mechanisms underlying apoptotic neuronal death following cerebral hypoxia-ischemia. With this growth in knowledge, we are coming to the realization that apoptosis and necrosis, although morphologically distinct, are likely part of a continuum of cell death with similar operative mechanisms. For example, following hypoxia-ischemia, excitatory amino acid release and alterations in ionic homeostasis contribute to both necrotic and apoptotic neuronal death. However, apoptosis is distinguished from necrosis in that gene activation is the predominant mechanism regulating cell survival. Following hypoxic-ischemic episodes in the brain, genes that promote as well as inhibit apoptosis are activated. It is the balance in the expression of pro- and anti-apoptotic genes that likely determines the fate of neurons exposed to hypoxia. The balance in expression of pro- and anti-apoptotic genes may also account for the regional differences in vulnerability to hypoxic insults. In this review, we will examine the known mechanisms underlying apoptosis in neurons exposed to hypoxia and hypoxia-ischemia.

Altered iPSC-derived neurons' sodium channel properties in subjects with Monge's disease.

Monge's disease, also known as chronic mountain sickness (CMS), is a disease that potentially threatens more than 140 million highlanders during extended time living at high altitudes (over 2500m). The prevalence of CMS in Andeans is about 15-20%, suggesting that the majority of highlanders (non-CMS) are rather healthy at high altitudes; however, CMS subjects experience severe hypoxemia, erythrocytosis and many neurologic manifestations including migraine, headache, mental fatigue, confusion, and memory loss. The underlying mechanisms of CMS neuropathology are not well understood and no ideal treatment is available to prevent or cure CMS, except for phlebotomy. In the current study, we reprogrammed fibroblast cells from both CMS and non-CMS subjects' skin biopsies into the induced pluripotent stem cells (iPSCs), then differentiated into neurons and compared their neuronal properties. We discovered that CMS neurons were much less excitable (higher rheobase) than non-CMS neurons. This decreased excitability was not caused by differences in passive neuronal properties, but instead by a significantly lowered Na(+) channel current density and by a shift of the voltage-conductance curve in the depolarization direction. Our findings provide, for the first time, evidence of a neuronal abnormality in CMS subjects as compared to non-CMS subjects, hoping that such studies can pave the way to a better understanding of the neuropathology in CMS.

Neuronal tolerance to O2 deprivation in drosophila: novel approaches using genetic models.

In spite of many advances in monitoring oxygenation and preventing cerebro-vascular accidents, there is still considerable morbidity and mortality from conditions with cerebral blood flow impairment and O2 deprivation leading to hypoxic/ischemic brain injury. Part of this failure is related to the complexity of the cascade of events that ensue after hypoxia or ischemia, but also part of it may be related to the fact that most research in the previous few decades has focused, justifiably, on cerebral vessel disease. However, an important aspect of the cascade is dependent on many factors that are inherent to the nature and response of the tissue itself. Hence, there is more need now for a two-pronged approach to hypoxic/ischemic brain injury, one focusing on vessel disease, its prevention, and treatment, and the other centering on the brain tissue itself and the factors that render neurons and glia more susceptible or more tolerant to a lack of oxygenation. In the past several years, a number of methods, techniques, and animal models have been used to address the response of neurons and glia to lack of oxygen. In this review, we highlight some novel ideas and some results that we and others have obtained, mostly pertaining to the genetic endowment and responses of the central nervous system to O2 deprivation. The role and importance of genetic models, such as the Drosophila melanogaster, are discussed, and an example illustrating how to harness the power of Drosophila genetics is detailed.

Congenital failure of automatic control of ventilation, gastrointestinal motility and heart rate.

A new congenital syndrome characterized by the simultaneous failure of control of ventilation (Ondine's curse) and intestinal motility (Hirschsprung's disease) is reported in three infants, all of whom died in the first few months of life; two were siblings. Detailed studies in one also revealed markedly decreased esophageal motility and abnormal control of heart rate. In one infant, minute ventilation was lower in quiet than in REM sleep and lower in both states of sleep than in wakefulness. Although the mean inspiratory flow was decreased in quiet sleep, the hypoventilation resulted primarily from a decrease in respiratory frequency. Intravenous doxapram increased ventilation but did not reverse respiratory failure. Aminophyllin, progesterone, physostigmine and chlorpromazine did not change ventilation significantly; imipramine resulted in a significant decrease. Both long and short-term variability of the heart rate were markedly decreased when compared with the normal infant. Although neuropathologic studies postmorten did not reveal an anatomic defect, we postulate that a developmental abnormality in serotonergic neurons is responsible for this new syndrome.

Neuroanatomical distribution and binding properties of saxitoxin sites in the rat and turtle CNS.

Since saxitoxin (STX) binds to voltage-sensitive sodium channels and blocks their function, it has been widely used in the study of these channels. There is, however, limited information on STX binding properties and the neuroanatomical distribution of the Na+ channel as a function of brain region in the rat and in lower vertebrates such as the turtle. In the present study, we used a broad range of 3H-STX concentration (up to 64 nM) to examine saturation profiles and density distribution in both adult rat and turtle brains. We found that (1) STX sites do not vary greatly in affinity (most Kds = 2 to 5 nM) in various regions of the adult rat brain; (2) STX binding distribution was very heterogeneous in the rat with much higher density in the cortex, hippocampus, amygdala, and cerebellum than in the brainstem and spinal cord; (3) STX sites are mostly localized in layers made mostly of neurons with low density in white matter; and (4) turtle brain STX sites had similar binding properties, but its brain had much fewer STX sites than the rat, especially in the cerebellum and rostral areas such as the cortex. We conclude that (a) adult brain sodium channels have similar STX binding affinity in spite of the existence of multiple sodium channel subtypes; (b) the brainstem is very different from rostral brain areas in channel density; and (c) the turtle brain has a much lower sodium channel density than the rat brain.

Major differences in CNS sulfonylurea receptor distribution between the rat (newborn, adult) and turtle.

Our previous results have shown that KATP channels play an important role in K+ efflux and extracellular K+ accumulation in the rat brain, and this role was quantitatively more important in the adult than in the newborn brain. The purpose of this study was to localize by autoradiographic techniques the binding sites of glibenclamide, a potent sulfonylurea ligand that targets KATP channels, in the adult and newborn rat central nervous system (CNS). Since the adult turtle is resistant to anoxia, we also compared the rat to the turtle brain sulfonylurea receptor distribution. In all three animal groups (newborn rat, adult rat, adult turtle), specific glibenclamide binding was saturable. Scatchard plots were curvilinear in the rat, thus suggesting that glibenclamide binds to two types of sites, i.e., high and low affinity sites. Scatchard analysis on turtle brain tissue showed evidence of one binding site only. We also found that the distribution of glibenclamide binding sites was heterogeneous in the adult rat CNS with a higher density in rostral than in caudal regions. The highest binding densities were seen in the cortex, hippocampus, cerebellum, substantia nigra, and a few thalamic nuclei; intermediate densities were observed in the basal ganglia, septum, thalamus, and the hypoglossal nucleus. There was a low density in most areas of the hypothalamus, midbrain, brainstem, and spinal cord. Compared with the adult rat, the newborn had a very homogeneous distribution of binding sites and densities were very low throughout the CNS; the level of binding density was even lower in some regions undetectable in the adult turtle. Our results indicate that (1) there are high and low affinity sulfonylurea receptors in the rat CNS, (2) there is a striking heterogeneity in the distribution and density of sulfonylurea receptors in the adult rat CNS and this is in sharp contrast to the homogeneous distribution and low density in both newborn rat and adult turtle; (3) sulfonylurea receptors increase in number postnatally in the rat since binding density increases and the Kd in the newborn rat is similar to that in the adult rat. We speculate that KATP channels and sulfonylurea receptors are poorly developed in the turtle and develop mostly after birth in the rat, reaching highest density in adulthood.

Postnatal development of voltage-sensitive Na+ channels in rat brain.

Previous studies have shown that mammalian neuronal excitability increases with age, and this excitability may be related to development of Na+ channels. In addition, evidence suggests that Na+ channels are involved in the neuronal response to O2 deprivation. Because of this, we wished to examine the pharmacologic properties and neuroanatomical distribution of the Na+ channels in newborn brain and as a function of age. In this study, we used ligand-binding techniques and autoradiography with 3H-saxitoxin (STX) to investigate Na(+)-channel distribution in brains of rats at postnatal days 0, 3, 10, 21, 35, and 120. We found that (1) in each area examined, the Scatchard plots for STX binding were linear in both immature and mature brains in a ligand concentration range of 0.4-64 nM; the slopes, however, were different between areas or ages, with Kd values ranging between 1 and 5 nM; (2) STX-binding density was more than tenfold lower in the rostral brain and cerebellum at birth than in the adult and increased with age; (3) binding density in the newborn brainstem was higher than in other areas such as the cortex and cerebellum, which is opposite to the distribution in the adult; and (d) the brainstem had a different developing pattern with an early-peak density level (P10-21) and a lower adult level.(ABSTRACT TRUNCATED AT 250 WORDS)

Major difference in the expression of delta- and mu-opioid receptors between turtle and rat brain.

The reptilian turtle brain has a remarkably higher endurance for anoxia than mammalian brains. Since the response to O(2) deprivation is dependent in a major way on the expression and regulation of membrane proteins, differences in such proteins may play a role in the species-related differences in hypoxic responses. Because opioid system is involved in the regulation of hypoxic responses, we asked whether there are differences between rat and turtle brains in terms of opioid receptor expression. In this work, we compared the expression and distribution of delta-and mu-opioid receptors in the turtle and rat brains. Our results show that (1) the dissociation constant (K(d)) for delta-receptor binding was approximately four times lower and B(max) was more than double in the turtle brain homogenates than in rat ones; (2) the delta-receptor binding density was heterogeneously distributed in the turtle brain, with a higher density in the rostral regions than in the brainstem and spinal cord, and was generally much higher than in rat brains from the cortex to spinal cord; (3) the delta-opioid receptors in the rat brains were mostly located in the cortex, caudate putamen, and amygdala with an extremely low density in most subcortical (e.g., hippocampus and thalamus) and almost all brainstem regions; and (4) in sharp contrast to delta-opioid receptors, mu-opioid receptor density was much lower in all turtle brain regions compared with the rat ones. Our results demonstrate that the turtle brain is actually an organ of delta-opioid receptors, whereas the rat brain has predominantly mu-opioid receptors. Because we have recently found that delta-opioid receptors protect neurons against glutamate and hypoxic stress, we speculate that the unique pattern of delta-receptor receptor expression and distribution plays a critical role in the tolerance of turtle brain to stressful situations characterized by glutamate excitotoxicity.

Ontogeny and distribution of GABAA receptors in rat brainstem and rostral brain regions.

Previous studies from our laboratory and others have shown that there are major age-related differences in brainstem neuronal function. Since GABAA receptors are major targets for GABA-mediated inhibitory modulation and play a key role in regulating cardiorespiratory function, especially during O2 deprivation, we examined differences in GABAA receptor density and distribution during postnatal development. Using quantitative receptor autoradiography, the present study was performed to examine the postnatal expression of GABAA receptors in the rat brainstem and rostral brain areas at five ages, i.e. postnatal day 1 (P1), P5, P10, P21 and P120. Ten-micrometer brain sections at different brain levels were labelled with [3H]muscimol in Tris-citrate buffer. We found that (i) GABAA receptors appeared very early in almost all the brainstem as well as rostral areas; (ii) at P1, the brainstem had a higher GABAA receptor binding density than rostral areas and its density peaked at P5 or P10; and (iii) receptor densities of the cerebellum and rostral brain areas such as cortex, thalamus and dentate gyrus increased with age, especially between P10 and P21, but most other subcortical areas like caudate-putamen and hippocampal CA1 area did not increase remarkably after birth. We conclude that: (i) GABAA receptors exist in most brain areas at birth; (ii) there are several patterns of postnatal development of GABAA receptors in the CNS with dramatic differences between the brainstem and cortex; (iii) brainstem functions rely more on GABAA receptors in early postnatal life than at more mature stages. We speculate that GABAA receptors develop earlier in phylogenetically older structures (such as brainstem) than in newer brain regions (such as cortex).

Expression and localization of Na+/H+ exchangers in rat central nervous system.

Neurons in the central nervous system regulate their intracellular pH using particular membrane proteins of which two, namely the Na+-dependent Cl-/HCO3- exchanger and the Na+/H+ exchanger, are essential. In this study we examined messenger RNA expression and distribution of Na+/H+ exchanger in the newborn rat central nervous system and with maturation using Northern blot analysis and in situ hybridization. Our study clearly shows that each Na+/H+ exchanger has a different expression pattern in the rat central nervous system. As in non-excitable tissues, Na+/H+ exchanger 1 is by far the most abundant of all Na+/H+ exchangers in the rat central nervous system. Its expression is ubiquitous although its messenger RNA appears at higher levels in the hippocampus, in the 2nd/3rd layers of periamygdaloid cortex and in the cerebellum. The low level of messenger RNAs encoding Na+/H+ exchanger 2 and 4 is mainly expressed in the cerebral cortex and in the brainstem-diencephalon, while Na+/H+ exchanger 3 transcripts are found only in the cerebellar Purkinje cells. From a developmental point of view, Na+/H+ exchanger 1, 2 and 4 showed an increased level in their transcripts in the cerebral cortex while an opposite trend existed in the cerebellum from postnatal day 0 to postnatal day 30. The messenger RNA for Na+/H+ exchanger 3, however, increased its level with age in cerebellum. From our data we conclude that: i) the expression of the Na+/H+ exchanger is age-, region-, and subtype-specific, with Na+/H+ exchanger 1 being the most prevalent in the rat central nervous system; ii) specialization of groups of neurons with respect to the type of Na+/H+ exchanger is clearly illustrated by Na+/H+ exchanger 3 which is almost totally localized in cerebellar Purkinje cells; and iii) the developmental increase in the messenger RNA for Na+/H+ exchanger 1 in the cerebral cortex and hippocampus is consistent with our previous studies on intracellular pH physiology in neonatal and mature neurons. Together this study indicates that expression of each Na+/H+ exchanger messenger RNA is differentially regulated both during development and in the different regions of rat central nervous system.

Effect of prolonged O2 deprivation on Na+ channels: differential regulation in adult versus fetal rat brain.

Neuronal Na+ channels are functionally inhibited in the adult in response to acute O2 deprivation. Since prolonged hypoxia may not only affect channel function, but also its expression, we hypothesized that long-term hypoxia alters Na+ channel density. This alteration may depend on age, because we have found major differences in neuronal responses to hypoxia between the immature and adult. In the present work, we used northern blots, slot blots, saxitoxin binding and autoradiography to ask whether: (i) prolonged hypoxia alters Na+ channel messenger RNA and protein levels in the brain; (ii) there is a difference between the adult and prenatal brains regarding Na+ channel expression with hypoxic exposure; and (iii) regional differences in Na+ channel expression occur in hypoxia-exposed brains. Our results show the following. (1) Na+ channel messenger RNA and saxitoxin binding density decreased after prolonged hypoxia in adult brain homogenates; this is in sharp contrast to the changes observed in fetal brains, which tended to increase Na+ channel messenger RNA and protein after hypoxia. (2) Changes in saxitoxin binding density are related to alterations in the number of saxitoxin binding sites and not to binding affinity, since there was no major change in Kd values between the hypoxia and naive groups. (3) The hypoxia-induced Na+ channel expression was heterogeneous, with major differences between rostral regions (e.g., the cortex) and caudal regions (e.g., the medulla and pons). We speculate that down-regulation of Na+ channels during long-term hypoxia in mature brains is an adaptive cellular response, aimed at minimizing the mismatch between energy supply and demand, since maintenance of Na+ gradients is a major energy-requiring process. However, the prenatal brain does not depend on this adaptive mechanism in response to hypoxic stress.

The role of HCO3(-)-dependent mechanisms in pHi regulation during O2 deprivation.

We have reported in our previous work that, in the absence of HCO(3)(-), Na(+)/H(+) exchanger is responsible for an anoxia-induced alkalinization in hippocampal CA1 neurons. HCO(3)(-)-dependent mechanisms have been reported to play a key role in pH(i) regulation in nerve cells, but how their function is affected by O(2) deprivation has not been well studied. In this work, pH(i) measurements (obtained from dissociated neurons loaded with carboxy-seminaphthorhodafluor-1 and using confocal microscopy) and whole-cell patch clamp recording techniques were used to investigate the role of HCO(3)(-)-dependent membrane exchangers on CA1 neurons during O(2) deprivation. Anoxia (5 min) induced a small acidification in neurons in the presence of HCO(3)(-) and this acidification was changed to a significant alkalinization when neurons were bathed with Hepes buffer or when 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid was applied in a HCO(3)(-) solution, indicating that HCO(3)(-)-dependent mechanisms were involved. A marked anoxia-induced acidification (0.33+/-0.11 pH unit) was seen when the Na(+)/H(+) exchange was blocked with 3-(methylsulfonyl-4-piperidino-benzoyl)-guanidine methanesulfonate in the presence of HCO(3)(-), but the same anoxia did not cause a significant pH(i) change in a Na(+) free, HCO(3)(-) solution, suggesting that the anoxia-induced acidification in the presence of 3-(methylsulfonyl-4-piperidino-benzoyl)-guanidine methanesulfonate is dependent on both Na(+) and HCO(3)(-). Furthermore, anoxia did not cause a significant pH(i) change when both 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid and 3-(methylsulfonyl-4-piperidino-benzoyl)-guanidine methanesulfonate were present. Current clamp recordings showed a significant membrane depolarization following anoxia in HCO(3)(-) solution but not in Hepes buffer. Our data suggest that, in hippocampal neurons: a) pH(i) regulation during O(2) deprivation is affected not only by metabolism but also by membrane exchangers, and b) besides the activation of Na(+)/H(+) exchange, anoxia activates a 4,4'-diisothiocyanostilbene-2,2'-disulfonic acid-sensitive, Na(+)-dependent acid loader (possibly electrogenic).

Activation of voltage-sensitive sodium channels during oxygen deprivation leads to apoptotic neuronal death.

Sodium (Na(+)) entry into neurons during hypoxia is known to be associated with cell death. However, it is not clear whether Na(+) entry causes cell death and by what mechanisms this increased Na(+) entry induces death. In this study we used cultures of rat neocortical neurons to show that an increase in intracellular sodium (Na(i)(+)) through voltage-sensitive sodium channels (VSSCs), during hypoxia contributes to apoptosis. Hypoxia increased Na(i)(+) and induced neuronal apoptosis, as assessed by electron microscopy, annexin V staining, and terminal UDP nick end labeling staining. Reducing Na(+) entry with the VSSC blocker, tetrodotoxin (TTX), attenuated apoptotic neuronal death via a reduction in caspase-3 activation. Since the attenuation of apoptosis by TTX during hypoxia suggested that the activation of VSSCs and Na(+) entry are crucial events in hypoxia-induced cell death, we also determined whether the activation of VSSCs per se could lead to apoptosis under resting conditions. Increasing Na(+) entry with the VSSC activator veratridine also induced neuronal apoptosis and caspase-3 activation. These data indicate that a) Na(+) entry via VSSCs during hypoxia leads to apoptotic cell death which is mediated, in part, by caspase-3 and b) activation of VSSCs during oxygen deprivation is a major event by which hypoxia induces cell death.