Functional specialization and topographic segregation of hippocampal astrocytes

A neuron-glia signalling network in the active brain

Glial cells are active partners of neurons in processing information and synaptic integration. They receive coded signals from synapses and elaborate modulatory responses. The active properties of glia, including long-range signalling and regulated transmitter release, are beginning to be elucidated. Recent insights suggest that the active brain should no longer be regarded as a circuitry of neuronal contacts, but as an integrated network of interactive neurons and glia.

Electrophysiological characteristics of reactive astrocytes in experimental cortical dysplasia

Neocortical freeze lesions have been widely used to study neuronal mechanisms underlying hyperexcitability in dysplastic cortex. Comparatively little attention has been given to biophysical changes in the surrounding astrocytes that show profound morphological and biochemical alterations, often referred to as reactive gliosis. Astrocytes are thought to aid normal neuronal function by buffering extracellular K(+). Compromised astrocytic K(+) buffering has been proposed to contribute to neuronal dysfunction. Astrocytic K(+) buffering is mediated, partially, by the activity of inwardly rectifying K(+) channels (K(IR)) and may involve intracellular redistribution of K(+) through gap-junctions. We characterized K(+) channel expression and gap-junction coupling between astrocytes in freeze-lesion-induced dysplastic neocortex. Whole cell patch-clamp recordings were obtained from astrocytes in slices from postnatal day (P) 16--P24 rats that had received a freeze-lesion on P1. A marked increase in glial fibrillary acidic protein immunoreactivity was observed along the entire length of the freeze lesion. Clusters of proliferative (bromo-deoxyuridine nuclear staining, BrdU+) astrocytes were seen near the depth of the microsulcus. Astrocytes in cortical layer I surrounding the lesion were characterized by a significant reduction in K(IR). BrdU-positive astrocytes near the depth of the microsulcus showed essentially no expression of K(IR) channels but markedly enhanced expression of delayed rectifier K(+) (K(DR)) channels. These proliferative cells showed virtually no dye coupling, whereas astrocytes in the hyperexcitable zone adjacent to the microsulcus displayed prominent dye-coupling as well as large K(IR) and outward K(+) currents. These findings suggest that reactive gliosis is accompanied by a loss of K(IR) currents and reduced gap junction coupling, which in turn suggests a compromised K(+) buffering capacity.

Distribution of chloride channel-2-immunoreactive neuronal and astrocytic processes in the hippocampus

The chloride homeostasis of neurons and non-neuronal cells is maintained in part by a voltage-sensitive inwardly rectifying chloride conductance through the chloride channel-2. This channel is activated by hyperpolarization and extracellular hypotonicity. In the present study, hippocampal sections were immunostained for chloride channel-2, and somata and dendrites of both pyramidal and non-pyramidal cells were found to be immunoreactive. In addition, glial processes in the vicinity of small blood vessels were also immunostained, whereas the neuropil of strata pyramidale and lacunosum-moleculare contained chloride channel-2-positive punctate structures. Electron microscopy and double immunostaining using antibodies against chloride channel-2 and glial fibrillary acidic protein confirmed that the dense network of chloride channel-2-positive processes corresponds to the end feet of astrocytes. The distribution of chloride channel-2-immunoreactive astrocytes was inhomogeneous throughout the hippocampus: strata oriens, pyramidale and lacunosum-moleculare of CA1-CA3 and the outer molecular layer of the dentate gyrus contained the majority of immunoreactive end feet, whereas the other layers showed sparse labeling. Subcellular studies demonstrated that, in addition to astrocytes, chloride channel-2 was localized in the membrane of dendrites, dendritic spines, cell bodies and axon initial segments of neurons, frequently close to, or within active zones of, symmetrical synapses.Thus, chloride channel-2 appears to be involved in transmembrane chloride movements associated with GABAergic synaptic transmission. The specific laminar distribution of chloride channel-2-positive astroglial processes coinciding with that of GABAergic axon terminals suggests that the network of astrocytes may be able to siphon and deliver Cl(-) ions to layers with intense GABAergic transmission, thereby increasing the efficacy of GABA(A) receptor-mediated inhibition.

Identification of an inward rectifier potassium channel gene expressed in mouse cortical astrocytes

These experiments identify an inward rectifier K+ (Kir) channel expressed in mouse cortical and white matter astrocytes at the molecular level. Messenger RNA for one of the known Kir channel genes, Kir4.1, is present at much higher levels in cortical astrocytes in primary culture than the other known Kir family members. In culture, the level of Kir4.1 mRNA is lower in proliferating cells and in cells cultured for 16 h under hypoxic conditions, compared to confluent cells. Partial differentiation of the astrocytes with dibutyryl cAMP or by coculture with neurons has no effect on the Kir4.1 mRNA level. In situ hybridization experiments show that Kir4.1 mRNA is broadly distributed in the adult brain, including the neocortex, the stratum pyrimadale of the hippocampus, and the piriform cortex. Immunostaining confirms that the Kir4.1 protein is expressed by cultured astrocytes and also by cocultured cortical neurons. Astrocytes and neurons display a patchy pattern of immunostaining, raising the possibility that the channels sort themselves in clusters in the plasma membrane. Stellate cells in the neocortex and white matter are immunoreactive for Kir4.1, and double immunofluorescence experiments show colocalization of Kir4.1 and glial acidic fibrillary protein (GFAP) on stellate cells in the white matter. The cloned mouse Kir4.1 cDNA, when expressed heterologously in HEK cells, gives rise to inactivating Kir channels similar to those recorded from cultured astrocytes. These results indicate that the Kir4.1 gene product forms a Kir channel, or is a subunit of the channel, in mouse cortical astrocytes both in culture and in vivo.

Immunocontent and secretion of S100B in astrocyte cultures from different brain regions in relation to morphology

Primary astrocyte cultures prepared from neonatal hippocampus, cerebral cortex and cerebellum were morphologically distinct. Cells from hippocampus and cortex were almost entirely protoplasmic, whereas cerebellar astrocytes had many processes; in the absence of serum these differences were accentuated. We compared the immunocontent and secretion of the mitogenic protein S100B in these cultures. Immunocontent was 2.5 times higher in cerebellar astrocytes than in hippocampal or cortical astrocytes. Cells from all three regions secreted S100B under basal conditions, but the secretion rate was higher in cerebellar astrocytes. Secretion depended on protein synthesis and was increased by incubation with forskolin or lysophosphatidic acid in mechanisms which were additive. The stellate morphology induced by forskolin was reversed by lysophosphatidic acid in hippocampal but not in cerebellar cultures, suggesting that S100B secretion was not associated with a process-bearing phenotype of astrocytes.

Freshly isolated astrocytes from rat hippocampus show two distinct current patterns and different [K(+)](o) uptake capabilities

Whether astrocytes predominantly express ohmic K(+) channels in vivo, and how expression of different K(+) channels affects [K(+)](o) homeostasis in the CNS have been long-standing questions for how astrocytes function. In the present study, we have addressed some of these questions in glial fibrillary acidic protein [GFAP(+)], freshly isolated astrocytes (FIAs) from CA1 and CA3 regions of P7-15 rat hippocampus. As isolated, these astrocytes were uncoupled allowing a higher resolution of electrophysiological study. FIAs showed two distinct ion current profiles, with neither showing a purely linear I-V relationship. One population of astrocytes had a combined expression of outward potassium currents (I(Ka), I(Kd)) and inward sodium currents (I(Na)). We term these outwardly rectifying astrocytes (ORA). Another population of astrocytes is characterized by a relatively symmetric potassium current pattern, comprising outward I(Kdr), I(Ka), and abundant inward potassium currents (I(Kin)), and a larger membrane capacitance (C(m)) and more negative resting membrane potential (RMP) than ORAs. We term these variably rectifying astrocytes (VRA). The I(Kin) in 70% of the VRAs was essentially insensitive to Cs(+), while I(Kin) in the remaining 30% of VRAs was sensitive. The I(Ka) of VRAs was most sensitive to 4-aminopyridine (4-AP), while I(Kdr) of ORAs was more sensitive to tetraethylammonium (TEA). ORAs and VRAs occurred approximately equally in FIAs isolated from the CA1 region (52% ORAs versus 48% VRAs), but ORAs were enriched in FIAs isolated from the CA3 region (71% ORAs versus 29% VRAs), suggesting an anatomical segregation of these two types of astrocytes within the hippocampus. VRAs, but not ORAs, showed robust inward currents in response to an increase in extracellular K(+) from 5 to 10 mM. As VRAs showed a similar current pattern and other passive membrane properties (e.g., RMP, R(in)) to "passive astrocytes"in situ (i.e., these showing linear I-V curves), such passive astrocytes possibly represent VRAs influenced by extensive gap-junction coupling in situ. Thus, our data suggest that, at least in CA1 and CA3 regions from P7-15 rats, there are two classes of GFAP(+) astrocytes which possess different K(+) currents. Only VRAs seem suited to uptake of extracellular K(+) via I(Kin) channels at physiological membrane potentials and increases of [K(+)](o). ORAs show abundant outward potassium currents with more depolarized RMP. Thus VRAs and ORAs may cooperate in vivo for uptake and release of K(+), respectively.

Freshly isolated astrocyte (FIA) preparations: a useful single cell system for studying astrocyte properties

Astrocytes are cell constituents of the mammalian CNS whose intricate relationships with neurons, blood vessels and meninges in situ are well documented. These relationships and their complex morphologies imply numerous functions. Over the past quarter century or so, however, the main experimental basis for determining which roles are likely have been derived from studies on primary astrocyte cultures, usually prepared from neonatal rodent brains. We list a number of examples where these cultures have shown quantitative and qualitative differences from the properties exhibited by astrocytes in situ. The absence of an adequate reliable database makes proposals of likely hypotheses of astrocyte function difficult to formulate. In this article we describe representative studies from our laboratory showing that freshly isolated astrocytes (FIAs), can be used to determine the properties of astrocytes that seem more in concordance with the properties exhibited in situ. Although the cells are most easily isolated from < or =15 day old rat hippocampi they can be isolated from up to 30 day old rats. The examples we describe are that several different types of K(+) currents can be determined by patch clamp electrophysiology, of all the mGluRs only mGluR3 and 5 were detected by single cell RT-PCR, and that single cell Ca(2+) imaging shows that the mGluR5 receptor is functional. It was found that the frequency of cells expressing mGluR5 declines with the age of the animal with the mGluR5b type splice variant replacing the mGluR5a type, as occurs in the intact brain. It is concluded that FIAs can be used to determine the individual characteristics of astrocytes and their properties without the problems of indirect effects inherent in a heterogeneous system such as the slice, and without the problem of cultures unpredictably reflecting the in situ state. The FIAs obviously cannot be used to study interactions of astrocytes with the other CNS components but we propose that they will provide a good database on which hypotheses regarding such interactions can be tested in slices. FIAs can also be isolated from brain slices or intact brain after various pharmacological or electrophysiological perturbations to determine the changes in astrocyte properties that correlate with the perturbations.

Localization of the tandem pore domain K+ channel TASK-1 in the rat central nervous system

Recently, a new family of potassium channels with two pore domains in tandem and four transmembrane segments has been identified. Seven functional mammalian channels have been reported at this time. These channels give rise to baseline potassium currents because they are not gated by voltage and exhibit spontaneous activity at all membrane potentials. Although the physiological role of these ion channels has yet to be determined, three mammalian members of this family (TREK-1, TASK-1, TASK-2) are activated by volatile anesthetics and may therefore contribute to the central nervous system (CNS) depression produced by volatile anesthetics. In this study we used northern blot analysis and immunohistochemical localization to determine the expression of TASK-1 subunits in the CNS. TASK-1 immunoreactivity was prominently found in astrocytes of the hippocampus, in the median eminence, in the choroid plexus, and the granular layer, Purkinje cell layer, and molecular layer of the cerebellum. In the spinal cord, strong TASK-I immunoreactivity was seen in ependymal cells lining the central canal and in white matter. These findings suggest a role for the TASK-1 channel in the production of cerebrospinal fluid and function of hypothalamic neurosecretory cells.

Controversy surrounding the existence of discrete functional classes of astrocytes in adult gray matter

Since 1992, it has been possible to record ionic currents from identified astrocytes in situ, using brain slice technology. Brain slice recordings confirm previous in vitro findings that expression of voltage-gated K(+) and Na(+) channels are a feature of this cell type. In contrast to cultured astrocytes, most investigators found that astrocytes in situ did not contain detectable, or at very best only low, levels of glial fibrillary acidic protein (GFAP). Structural and immunocytochemical investigations determined that these cells are different from oligodendrocyte precursors. In addition to cells with this current pattern, many but not all investigators found a second pool of astrocytes with no voltage-gated ion channels and high GFAP content. These two subpopulations of cells were termed complex and passive astrocytes. The existence of passive astrocytes has been questioned because of possible problems with space clamp conditions and spillage of EGTA-buffered pipette solution around the cells before recordings. Another problem is the fact there is a discrepancy regarding the GFAP content of complex astrocytes. It is of interest that recent immunocytochemical studies suggest the existence of two pools of astrocytes, one with a high GFAP content and one with nondetectable GFAP. Given this, it is tempting to correlate the two (controversial) electrophysiological patterns with immunochemical differences (GFAP) in order to demonstrate two functionally discrete classes of astrocytes in adult gray matter. However, despite evidence that some of the K(+) channels may be involved in proliferation, the role of voltage-gated ion channels in this nonexcitable cell type remains unknown. This is despite the fact that astrocytic Na(+) channels show dramatic changes after pathological events, re-enforcing the notion that the expression of this channel is under tight neuronal control. Several studies suggest that there is a great degree of flexibility and that astrocytes can undergo rapid changes in expression of both membrane ion currents and GFAP. Although it is likely that astrocytes exhibit different structural and membrane properties, this heterogeneity might be a reflection of the flexible plasticity of one astrocyte type under influence of environmental factors rather than of the existence of two distinct and permanent subtypes.

Effects of chloride transport inhibition and chloride substitution on neuron function and on hypoxic spreading-depression-like depolarization in rat hippocampal slices

Chloride fluxes play a crucial role in synaptic inhibition, cell pH regulation, as well as in cell volume control. In many neuropathological processes, cell swelling is a pivotal parameter, since cell volume changes and the dimension of the interstitial space critically modulate synchronized neuronal activity as well as the tissue's susceptibility to seizures or spreading depression. This study therefore focuses on the effects of different Cl(-) transport inhibitors and Cl(-) substitution on neuronal function and hypoxia-induced changes in rat hippocampal tissue slices. Orthodromically evoked focal excitatory postsynaptic potentials were depressed by furosemide (2mM), 4,4'-diisothiocyanatostilbene-2, 2'-disulfonic acid (1mM) and Cl(-) substitution by methylsulfate, but were enhanced by 4,4'-dinitrostilbene-2,2'-disulfonic acid (1mM). All four treatments induced multiple population spike firing in response to single orthodromic volleys, suggesting reduced synaptic inhibition. Antidromic population spikes increased following Cl(-) withdrawal, were unaffected in the presence of furosemide and 4, 4'-dinitrostilbene-2,2'-disulfonic acid, but were abolished by 4, 4'-diisothiocyanatostilbene-2,2'-disulfonic acid. The amplitude of the hypoxic spreading-depression-like extracellular potential shift was reduced by furosemide, 4,4'-diisothiocyanatostilbene-2, 2'-disulfonic acid and Cl(-) withdrawal, i.e. by the same treatments that depressed orthodromically evoked postsynaptic potentials. Furosemide prolonged the time to onset and the duration of the spreading-depression-like extracellular potential shift, while 4, 4'-dinitrostilbene-2,2'-disulfonic acid shortened the time to onset. Spreading-depression-related cell swelling was recorded as the shrinkage of relative interstitial space, which was measured as tetramethylammonium-chloride space. Neither the Cl(-) transport inhibitors nor Cl(-) withdrawal had any detectable effect on spreading-depression-related cell swelling. CA1 pyramidal neurons usually hyperpolarized during drug application and their input resistance decreased. Cl(-) withdrawal increased their input resistance and caused spontaneous burst firing. Hypoxia caused the expected spreading-depression-like rapid, near complete depolarization of single pyramidal neurons and drastically reduced their input resistance. The three Cl(-) transport inhibitors and Cl(-) withdrawal delayed the onset of the hypoxic depolarization. In low Cl(-) solutions, the apparent threshold potential at which spreading depression was triggered shifted to more positive membrane potentials. The final voltage of the hypoxic depolarization was, however, not affected. It appears from these results that the reduction in the hypoxic spreading-depression-like extracellular potential shifts by Cl(-) transport inhibitors is at least partially attributable to desynchronization of depolarization, not to decreased depolarization in individual cells. Other contributing factors could be changes in recording conditions, depression of swelling-induced amino acid release from glial cells and unspecific side-effects of the applied drugs. Desynchronization could also account for the delayed spreading-depression onset. It is concluded that Cl(-) fluxes play a role in the triggering of spreading depression, but the spreading-depression-like depolarization itself or its self-regenerative character is not mediated by Cl(-).

Extracellular pH responses in CA1 and the dentate gyrus during electrical stimulation, seizure discharges, and spreading depression

Since neuronal excitability is sensitive to changes in extracellular pH and there is regional diversity in the changes in extracellular pH during neuronal activity, we examined the activity-dependent extracellular pH changes in the CA1 region and the dentate gyrus. In vivo, in the CA1 region, recurrent epileptiform activity induced by stimulus trains, bicuculline, and kainic acid resulted in biphasic pH shifts, consisting of an initial extracellular alkalinization followed by a slower acidification. In vitro, stimulus trains also evoked biphasic pH shifts in the CA1 region. However, in CA1, seizure activity in vitro induced in the absence of synaptic transmission, by perfusing with 0 Ca(2+)/5 mM K(+) medium, was only associated with extracellular acidification. In the dentate gyrus in vivo, seizure activity induced by stimulation to the angular bundle or by injection of either bicuculline or kainic acid was only associated with extracellular acidification. In vitro, stimulus trains evoked only acidification. In the dentate gyrus in vitro, recurrent epileptiform activity induced in the absence of synaptic transmission by perfusion with 0 Ca(2+)/8 mM K(+) medium was associated with extracellular acidification. To test whether glial cell depolarization plays a role in the regulation of the extracellular pH, slices were perfused with 1 mM barium. Barium increased the amplitude of the initial alkalinization in CA1 and caused the appearance of alkalinization in the dentate gyrus. In both CA1 and the dentate gyrus in vitro, spreading depression was associated with biphasic pH shifts. These results demonstrate that activity-dependent extracellular pH shifts differ between CA1 and dentate gyrus both in vivo and in vitro. The differences in pH fluctuations with neuronal activity might be a marker for the basis of the regional differences in seizure susceptibility between CA1 and the dentate gyrus.

Functional coupling between neurons and glia

Neuronal-glial interactions play an important role in information processing in the CNS. Previous studies have indicated that electrotonic coupling between locus ceruleus (LC) neurons is involved in synchronizing the spontaneous activity. The results of the present study extend the functional electrotonic coupling to interactions between neurons and glia. Spontaneous oscillations in the membrane potential were observed in a subset of glia. These oscillations were synchronous with the firing of neurons, insensitive to transmitter receptor antagonists and disrupted by carbenoxolone, a gap junction blocker. Hyperpolarization of neurons with [Met] (5)enkephalin blocked the oscillations in glia. Selective depolarization of glia with a glutamate transporter substrate (l-alpha-aminoadipic acid) increased the neuronal firing rate, suggesting that changes in the membrane potential of glia can modulate neuronal excitability through heterocellular coupling. Dye-coupling experiments further confirmed that small molecules could be transferred through gap junctions between these distinct cell types. No dye transfer was observed between neurons and oligodendrocytes or between astrocytes and oligodendrocytes, suggesting that the junctional communication was specific for astrocytes and neurons. Finally, immunoelectron microscopy studies established that connexins, the proteins that form gap junctions, were present on portions of the plasmalemma, bridging the cytoplasm of neurons and glia in LC. This heterocellular coupling extends the mechanisms by which glia participate in the network properties of the LC in which the degree of coupling is thought to influence cognitive performance.

Do glia have heart? Expression and functional role for ether-a-go-go currents in hippocampal astrocytes

Potassium homeostasis plays an important role in the control of neuronal excitability, and diminished buffering of extracellular K results in neuronal Hyperexcitability and abnormal synchronization. Astrocytes are the cellular elements primarily involved in this process. Potassium uptake into astrocytes occurs, at least in part, through voltage-dependent channels, but the exact mechanisms involved are not fully understood. Although most glial recordings reveal expression of inward rectifier currents (K(IR)), it is not clear how spatial buffering consisting of accumulation and release of potassium may be mediated by exclusively inward potassium fluxes. We hypothesized that a combination of inward and outward rectifiers cooperate in the process of spatial buffering. Given the pharmacological properties of potassium homeostasis (sensitivity to Cs(+)), members of the ether-a-go-go (ERG) channel family widely expressed in the nervous system could underlie part of the process. We used electrophysiological recordings and pharmacological manipulations to demonstrate the expression of ERG-type currents in cultured and in situ hippocampal astrocytes. Specific ERG blockers (dofetilide and E 4031) inhibited hyperpolarization- and depolarization-activated glial currents, and ERG blockade impaired clearance of extracellular potassium with little direct effect on hippocampal neuron excitability. Immunocytochemical analysis revealed ERG protein mostly confined to astrocytes; ERG immunoreactivity was absent in presynaptic and postsynaptic elements, but pronounced in glia surrounding the synaptic cleft. Oligodendroglia did not reveal ERG immunoreactivity. Intense immunoreactivity was also found in perivascular astrocytic end feet at the blood-brain barrier. cDNA amplification showed that cortical astrocytes selectively express HERG1, but not HERG2-3 genes. This study provides insight into a possible physiological role of hippocampal ERG channels and links activation of ERG to control of potassium homeostasis.

Selective depolarization of interneurons in the early posttraumatic dentate gyrus: involvement of the Na(+)/K(+)-ATPase

Interneurons innervating dentate granule cells are potent regulators of the entorhino-hippocampal interplay. Traumatic brain injury, a leading cause of death and disability among young adults, is frequently associated with rapid neuropathological changes, seizures, and short-term memory deficits both in humans and experimental animals, indicating significant posttraumatic perturbations of hippocampal circuits. To determine the pathophysiological alterations that affect the posttraumatic functions of dentate neuronal networks within the important early (hours to days) posttraumatic period, whole cell patch-clamp recordings were performed from granule cells and interneurons situated in the granule cell layer of the dentate gyrus of head-injured and age-matched, sham-operated control rats. The data show that a single pressure wave-transient delivered to the neocortex of rats (mimicking moderate concussive head trauma) resulted in a characteristic ( approximately 10 mV), transient (<4 days), selective depolarizing shift in the resting membrane potential of dentate interneurons, but not in neighboring granule cells. The depolarization was not associated with significant changes in action potential characteristics or input resistance, and persisted in the presence of antagonists of ionotropic and metabotropic glutamate, and GABA(A) and muscarinic receptors, as well as blockers of voltage-dependent sodium channels and of the h-current. The differential action of the cardiac glycosides oubain and stophanthidin on interneurons from control versus head-injured rats indicated that the depolarization of interneurons was related to the trauma-induced decrease in the activity of the electrogenic Na(+)/K(+)-ATPase. In contrast, the Na(+)/K(+)-ATPase activity in granule cells did not change. Intracellular injection of Na(+), Ca(2+)-chelator and ATP, as well as ATP alone, abolished the difference between the resting membrane potentials of control and injured interneurons. The selective posttraumatic depolarization increased spontaneous firing in interneurons, enhanced the frequency and amplitude of spontaneous inhibitory postsynaptic currents (IPSCs) in granule cells, and augmented the efficacy of depolarizing inputs to discharge interneurons. These results demonstrate that mechanical neurotrauma delivered to a remote site has highly selective effects on different cell types even within the same cell layer, and that the electrogenic Na(+)-pump plays a role in setting the excitability of hippocampal interneuronal networks after injury.

Ion channels in glial cells

Functional and molecular analysis of glial voltage- and ligand-gated ion channels underwent tremendous boost over the last 15 years. The traditional image of the glial cell as a passive, structural element of the nervous system was transformed into the concept of a plastic cell, capable of expressing a large variety of ion channels and neurotransmitter receptors. These molecules might enable glial cells to sense neuronal activity and to integrate it within glial networks, e.g., by means of spreading calcium waves. In this review we shall give a comprehensive summary of the main functional properties of ion channels and ionotropic receptors expressed by macroglial cells, i.e., by astrocytes, oligodendrocytes and Schwann cells. In particular we will discuss in detail glial sodium, potassium and anion channels, as well as glutamate, GABA and ATP activated ionotropic receptors. A majority of available data was obtained from primary cell culture, these results have been compared with corresponding studies that used acute tissue slices or freshly isolated cells. In view of these data, an active glial participation in information processing seems increasingly likely and a physiological role for some of the glial channels and receptors is gradually emerging.

GFAP mRNA positive glia acutely isolated from rat hippocampus predominantly show complex current patterns

Electrophysiologically complex glial cells have been widely identified from different regions of the central nervous system and constitute a dominant glial type in juvenile mice or rats. As these cells express several types of ion channels and neurotransmitter channels that were thought to be only present in neurons, this glial cell type has attracted considerable attention. However, the actual classification of these electrophysiologically complex glial cells remains unclear. They have been speculated to be an immature astrocyte because, although these cells show positive staining for the predominantly astrocytic marker S 100beta, it has not been possible to show staining for the commonly accepted mature astrocytic marker, glial fibrillary acidic protein (GFAP). To address the question of whether these cells might express GFAP at the transcript level, we combined patch-clamp electrophysiological recording with single cell RT-PCR for GFAP mRNA in glial cells acutely isolated from 4 to 12 postnatal day rats. In fresh cell suspensions from the CA1 region, complex glial cells were found to be the dominant cell type (65% total cells). We found that the majority of these electrophysiologically complex cells (74%) were positive for GFAP mRNA. We also showed that the complex cells responded to AMPA and GABA application, and these were also GFAP mRNA positive. We also fixed and stained the preparations for GFAP without electrophysiological recording to better preserve GFAP immunoreactively. In agreement with other studies, only 1.5% of these presumed electrophysiologically complex cells, based on morphology, showed immunoreactivity for GFAP. The expression of GFAP at the transcript level indicates GFAP (-)/GFAP mRNA (+) glial cells have an astrocytic identity. As single cell RT-PCR is able to detect both GFAP (-)/GFAP mRNA (+) and GFAP (+)/GFAP mRNA (+) astrocytic subtypes, the present study also suggests it is a feasible approach for astrocytic lineage studies.

Ion channel expression by astrocytes in situ: comparison of different CNS regions

Patch-clamp recordings were obtained in brain slices from 283 rat astrocytes. The expression of voltage-activated whole-cell currents was compared in four different CNS regions (hippocampus, cerebral cortex, spinal cord, and cerebellum). Our data show that CNS astrocytes do not show significant regional differences in their ion channel complement. With the exception of cerebellar Bergmann glial cells, essentially all astrocytes express a combination of delayed rectifying outward K(+) currents, transient A-type K(+) currents, and small Na(+) currents. Developmentally, an increasing percentage of astrocytes and Bergmann glial cells express inwardly rectifying K(+) currents. We did not observe cells that were passive, i.e., lacking voltage-activated currents. A few cells that appeared "passive" in initial recordings showed voltage-activated K(+) currents after off-line leak subtraction. The heterogeneity observed in the ion channel complement was found to be identical when cell-to-cell variations observed within a given CNS region and between various CNS regions were compared, suggesting a common and fairly stereotypical complement of ion channels in CNS astrocytes. Ion channel expression in Bergmann glial cells differed from that of all other CNS regions studied. These cells typically showed very low input resistances attributable to a significant time- and voltage-independent resting K(+) conductance. However, as with electrophysiologically "passive"-appearing astrocytes, Bergmann glial cells showed expression of delayed rectifying K(+) currents after off-line leak subtraction. Inwardly rectifying K(+) currents were observed in Bergmann glial cells after postnatal day 17. Collectively, our data suggest that all astrocytes contain voltage-gated ion channels that display a common pattern of expression during development.

Endothelial and neuronal nitric oxide synthases are present in the suprachiasmatic nuclei of Syrian hamsters and rats

Nitric oxide (NO) is involved in the transmission of light information to suprachiasmatic nuclei (SCN). By immunocytochemistry, we showed that both neuronal and endothelial NO synthase isoforms (nNOS and eNOS) were present in the SCN of rats and hamsters. nNOS-immunoreactive neurons were located mainly around the SCN with only a few nNOS neurons within the nucleus. By double-label immunocytochemistry, we also found, within the population of SCN glial fibrillary acidic protein (GFAP)-immunoreactive astrocytes, a subpopulation of eNOS-immunoreactive astrocytes. Using Western blot analysis, we detected in SCN protein extracts eNOS and nNOS proteins having the expected 140 and 150 kDa molecular weights, respectively. By in situ hybridization of a 2.4-kb murine eNOS probe, mRNA for eNOS was located in the SCN of rats and hamsters. The transcript was further identified by detection of a RT-PCR product of the predicted size, after amplification of total RNA with primers specific for eNOS. In the SCN and cerebellum, the size of the mRNA for nNOS, detected with a rat probe on Northern blot, was approximately 10.5 kb, corresponding to that previously published. In the same tissues, we found two transcripts, one weakly expressed at approximately 4.0 kb and another more strongly expressed at approximately 2.6 kb, both hybridizing with two non-overlapping murine and rat eNOS probes. These results suggested the existence in the SCN of alternate transcripts for eNOS. We propose that two pathways could link light stimuli and NO release in the SCN: one involving N-methyl-D-aspartate (NMDA) receptors and nNOS in neurons; the other linking alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors and eNOS in astrocytes.

Na(+) and K(+) concentrations, extra- and intracellular voltages, and the effect of TTX in hypoxic rat hippocampal slices

Severe hypoxia causes rapid depolarization of CA1 neurons and glial cells that resembles spreading depression (SD). In brain slices in vitro, the SD-like depolarization and the associated irreversible loss of function can be postponed, but not prevented, by blockade of Na(+) currents by tetrodotoxin (TTX). To investigate the role of Na(+) flux, we made recordings from the CA1 region in hippocampal slices in the presence and absence of TTX. We measured membrane changes in single CA1 pyramidal neurons simultaneously with extracellular DC potential (V(o)) and either extracellular [K(+)] or [Na(+)]; alternatively, we simultaneously recorded [Na(+)](o), [K(+)](o), and V(o). Confirming previous reports, early during hypoxia, before SD onset, [K(+)](o) began to rise, whereas [Na(+)](o) still remained normal and V(o) showed a slight, gradual, negative shift; neurons first hyperpolarized and then began to gradually depolarize. The SD-like abrupt negative DeltaV(o) corresponded to a near complete depolarization of pyramidal neurons and an 89% decrease in input resistance. [K(+)](o) increased by 47 mM and [Na(+)](o) dropped by 91 mM. Changes in intracellular Na(+) and K(+) concentrations, estimated on the basis of the measured extracellular ion levels and the relative volume fractions of the neuronal, glial, and extracellular compartment, were much more moderate. Because [Na(+)](o) dropped more than [K(+)](o) increased, simple exchange of Na(+) for K(+) cannot account for these ionic changes. The apparent imbalance of charge could be made up by Cl(-) influx into neurons paralleling Na(+) flux and release of Mg(2+) from cells. The hypoxia-induced changes in interneurons resembled those observed in pyramidal neurons. Astrocytes responded with an initial slow depolarization as [K(+)](o) rose. It was followed by a rapid but incomplete depolarization as soon as SD occurred, which could be accounted for by the reduced ratio, [K(+)](i)/[K(+)](o). TTX (1 microM) markedly postponed SD, but the SD-related changes in [K(+)](o) and [Na(+)](o) were only reduced by 23 and 12%, respectively. In TTX-treated pyramidal neurons, the delayed SD-like depolarization took off from a more positive level, but the final depolarized intracellular potential and input resistance were not different from control. We conclude that TTX-sensitive channels mediate only a fraction of the Na(+) influx, and that some of the K(+) is released in exchange for Na(+). Even though TTX-sensitive Na(+) currents are not essential for the self-regenerative membrane changes during hypoxic SD, in control solutions their activation may trigger the transition from gradual to rapid depolarization of neurons, thereby synchronizing the SD-like event.

Blood-brain barrier, ion homeostatis and epilepsy: possible implications towards the understanding of ketogenic diet mechanisms

The finding that epileptic seizures alter blood-brain barrier (BBB) properties has stimulated interest into the possibility that phenotypic changes in brain endothelium may constitute a pathological initiator leading to seizures. Recent evidence obtained from epileptic patients undergoing cortical resection, demonstrated abnormal expression of glucose transporter molecules (GLUT1), while [18F]deoxyglucose PET studies demonstrated regions of decreased glucose uptake and hypometabolism in seizure foci. The properties of other 'nonexcitable CNS cells' are also altered in epileptic tissue, and glial cells from epileptic brain displayed diminished capacity for ionic homeostasis; voltage-dependent mechanisms were primarily affected, increasing reliance on energy-dependent mechanisms. Diminished ion homeostasis together with increased metabolic demand of hyperactive neurons may further aggravate the neuropathological consequences of BBB loss of glucose uptake mechanisms. Since ketone bodies can provide an alternative to glucose to support brain energy requirements, it is hypothesized that one of the mechanisms of the ketogenic diet in epilepsy may relate to increased availability of beta-hydroxybutyrate, a ketone body readily transported at the BBB. This hypothesis is supported by the fact that the ketogenic diet is the treatment of choice for the glucose transporter protein syndrome and pyruvate dehydrogenase deficiency, both associated with cerebral energy failure and seizures.

Cesium induces spontaneous epileptiform activity without changing extracellular potassium regulation in rat hippocampus

Cesium has been widely used to study the roles of the hyperpolarization-activated (I(h)) and inwardly rectifying potassium (K(IR)) channels in many neuronal and nonneuronal cell types. Recently, extracellular application of cesium has been shown to produce epileptiform activity in brain slices, but the mechanisms for this are not known. It has been proposed that cesium blocks the K(IR) in glia, resulting in an abnormal accumulation of potassium in the extracellular space and inducing epileptiform activity. This hypothesis has been tested in hippocampal slices and cultured hippocampal neurons using potassium-sensitive microelectrodes. In the present study, application of cesium produced spontaneous epileptiform discharges at physiological extracellular potassium concentration ([K(+)](o)) in the CA1 and CA3 regions of hippocampal slices. This epileptiform activity was not mimicked by increasing the [K(+)](o). The epileptiform discharges induced by cesium were not blocked by the N-methyl-D- aspartate (NMDA) receptor antagonist AP-5, but were blocked by the non-NMDA receptor antagonist CNQX. In the dentate gyrus, cesium induced the appearance of spontaneous nonsynaptic field bursts in 0 added calcium and 3 mM potassium. Moreover, cesium increased the frequency of field bursts already present. In contrast, ZD-7288, a specific I(h) blocker, did not cause spontaneous epileptiform activity in CA1 and CA3, nor did it affect the field bursts in the dentate gyrus, suggesting that cesium induced epileptiform activity is not directly related to blockade of the I(h). When potassium-sensitive microelectrodes were used to measure [K(+)](o), there was no significant increase in [K(+)](o) in CA1 and CA3 after cesium application. In the dentate gyrus, cesium did not change the baseline level of [K(+)](o) or the rate of [K(+)](o) clearance after the field bursts. In cultured hippocampal neurons, which have a large and relatively unrestricted extracellular space, cesium also produced cellular burst activity without significantly changing the resting membrane potential, which might indicate an increase in [K(+)](o). Our results suggest that cesium causes epileptiform activity by a mechanism unrelated to an alteration in [K(+)](o) regulation.

Modulation of burst frequency, duration, and amplitude in the zero-Ca(2+) model of epileptiform activity

Incubation of hippocampal slices in zero-Ca(2+) medium blocks synaptic transmission and results in spontaneous burst discharges. This seizure-like activity is characterized by negative shifts (bursts) in the extracellular field potential and a K(+) wave that propagates across the hippocampus. To isolate factors related to seizure initiation, propagation, and termination, a number of pharmacological agents were tested. K(+) influx and efflux mechanisms where blocked with cesium, barium, tetraethylammonium (TEA), and 4-aminopyridine (4-AP). The effect of the gap junction blockers, heptanol and octanol, on zero-Ca(2+) bursting was evaluated. Neuronal excitability was modulated with tetrodotoxin (TTX), charge screening, and applied electric fields. Glial cell function was examined with a metabolism antagonist (fluroacetate). Neuronal hyperpolarization by cation screening or applied fields decreased burst frequency but did not affect burst amplitude or duration. Heptanol attenuated burst amplitude and duration at low concentration (0.2 mM), and blocked bursting at higher concentration (0.5 mM). CsCl(2) (1 mM) had no effect, whereas high concentrations (1 mM) of BaCl(2) blocked bursting. TEA (25 mM) and low concentration of BaCl(2) (300 microM) resulted in a two- to sixfold increase in burst duration. Fluroacetate also blocked burst activity but only during prolonged application (>3 h). Our results demonstrate that burst frequency, amplitude, and duration can be independently modulated and suggest that neuronal excitability plays a central role in burst initiation, whereas potassium dynamics establish burst amplitude and duration.

Impaired K(+) homeostasis and altered electrophysiological properties of post-traumatic hippocampal glia

Traumatic brain injury (TBI) can be associated with memory impairment, cognitive deficits, or seizures, all of which can reflect altered hippocampal function. Whereas previous studies have focused on the involvement of neuronal loss in post-traumatic hippocampus, there has been relatively little understanding of changes in ionic homeostasis, failure of which can result in neuronal hyperexcitability and abnormal synchronization. Because glia play a crucial role in the homeostasis of the brain microenvironment, we investigated the effects of TBI on rat hippocampal glia. Using a fluid percussion injury (FPI) model and patch-clamp recordings from hippocampal slices, we have found impaired glial physiology 2 d after FPI. Electrophysiologically, we observed reduction in transient outward and inward K(+) currents. To assess the functional consequences of these glial changes, field potentials and extracellular K(+) activity were recorded in area CA3 during antidromic stimulation. An abnormal extracellular K(+) accumulation was observed in the post-traumatic hippocampal slices, accompanied by the appearance of CA3 afterdischarges. After pharmacological blockade of excitatory synapses and of K(+) inward currents, uninjured slices showed the same altered K(+) accumulation in the absence of abnormal neuronal activity. We suggest that TBI causes loss of K(+) conductance in hippocampal glia that results in the failure of glial K(+) homeostasis, which in turn promotes abnormal neuronal function. These findings provide a new potential mechanistic link between traumatic brain injury and subsequent development of disorders such as memory loss, cognitive decline, seizures, and epilepsy.

Apoptotic pathways mobilized in microglia and neurones as a consequence of chromogranin A-induced microglial activation

Senile plaques of Alzheimer's brain are characterized by activated microglia and immunoreactivity for the peptide chromogranin A. We have investigated the mechanisms by which chromogranin A activates microglia, producing modulators of neuronal survival. Primary cultures of rat brain-derived microglia display a reactive phenotype within 24 h of exposure to 10 nM chromogranin A, culminating in microglial death via apoptotic mechanisms mediated by interleukin-1beta converting enzyme. The signalling cascade initiated by chromogranin A triggers nitric oxide production followed by enhanced microglial glutamate release, inhibition of which prevents microglial death. The plasma membrane carrier inhibitor aminoadipate and the type II/III metabotropic glutamate receptor antagonist (RS)-alpha-methyl-4-sulphonophenylglycine are equally protective. A significant amount of the released glutamate occurs from bafilomycin-sensitive stores, suggesting a vesicular mode of release. Inhibition of this component of release affords significant microglial protection. Conditioned medium from activated microglia kills cerebellar granule cells by inducing caspase-3-dependent neuronal apoptosis. Brain-derived neurotrophic factor is partially neuroprotective, as are ionotropic glutamate receptor antagonists, and, when combined with boiling of conditioned medium, full protection is achieved; nitric oxide synthase inhibitors are ineffective.

The mechanism gated by external potassium and sodium controls the resting conductance in hippocampal and cortical neurons

The excitation of densely packed mammalian central neurons is followed by a substantial transitory elevation of external K+ concentration. This phenomenon may have a different functional significance depending on how the resting membrane conductance mechanisms react to the changes in the gradient of these ions. We have found that in the hippocampal and cortex neurons of rat a large fraction of the membrane conductance in the vicinity of the resting potential is provided by the K+ permeability mechanism which is gated by external K+ and Na+. The responses of acutely isolated pyramidal neurons to rapidly altered external [K+] were investigated using the whole-cell patch clamp and concentration clamp techniques. Elevation of [K+]out induced a biphasic inward current at membrane potentials more negative than the reversal potential for K+ ions. This current consisted of an "instantaneously" increased leakage component and a slowly activated current (tau = 48 ms at 21 degrees C) designated below as I(deltaK). The latter demonstrated a first order activation kinetics with a remarkably high Q10 = 7.31. I(deltaK) was absent in the peripheral sensory neurons as well as in the Purkinje neurons. Slow activation of I(deltaK) was critically dependent on [Na+]out: substitution of the extracellular Na+ with choline chloride or Li+ led to the "instantaneous" reaction of the membrane to the changes in [K+]out. By slowing down potassium influx, I(deltaK) may be of importance in preserving densely packed pyramidal neurons from immediate excitation following rapid increases in [K+]out.

Tripartite synapses: glia, the unacknowledged partner

According to the classical view of the nervous system, the numerically superior glial cells have inferior roles in that they provide an ideal environment for neuronal-cell function. However, there is a wave of new information suggesting that glia are intimately involved in the active control of neuronal activity and synaptic neurotransmission. Recent evidence shows that glia respond to neuronal activity with an elevation of their internal Ca2+ concentration, which triggers the release of chemical transmitters from glia themselves and, in turn, causes feedback regulation of neuronal activity and synaptic strength. In view of these new insights, this article suggests that perisynaptic Schwann cells and synaptically associated astrocytes should be viewed as integral modulatory elements of tripartite synapses.

Voltage-gated sodium and potassium channels in radial glial cells of trout optic tectum studied by patch clamp analysis and single cell RT-PCR

Radial glial cells in the visual center of trout were analyzed immunocytochemically and with the whole cell mode of the patch-clamp technique in combination with RT-PCR. By immunostaining with anti-GFAP antibodies radially oriented cell processes spanning the entire width of the tectum were brightly labeled, while with anti-S-100 antiserum the cell bodies residing in a discrete layer close to the ventricular border became most clearly visible. Virtually all radial glial cells examined in brain slices exhibited voltage-gated sodium inward currents that were activated above -40 mV, blocked by micromolar concentrations of TTX and totally eliminated if sodium was substituted for Tris in the bath solution. In contrast with adjacent nerve cells of the same slices radial glial cells did not exhibit spontaneous electrical activity and could not be stimulated to generate action potentials by depolarizing current injections. Two types of voltage-gated potassium outward currents were elicited by depolarizing voltage steps: a sustained current with delayed rectifier properties and a superimposed transient "A"-type current, both being activated at a threshold potential of -40 mV. In cultured radial glial cells subtle differences were noticed regarding current density, inactivation kinetics, and TEA-sensitivity of the potassium currents. Inwardly rectifying potassium currents activating at hyperpolarized voltages were not observed. By single cell RT-PCR the transcripts of two shaker-related potassium channel genes (termed tsha1-a fish homologue to Kv1.2- and tsha3) were amplified, while transcripts for tsha 2 and tsha 4 were not detected.

Optical detection of synaptically induced glutamate transport in hippocampal slices

Although it has long been believed that glial cells play a major role in transmitter uptake at synapses in the CNS, the relative contribution of glial and neuronal cells to reuptake of synaptically released glutamate has been unclear. Recent identification of the diverse glutamate transporter subtypes provides an opportunity to examine this issue. To monitor glutamate transporter activity, we optically detected synaptically induced changes of membrane potential from hippocampal CA1 field in slice preparations using a voltage-sensitive dye, RH155. In the presence of ionotropic glutamate-receptor blockers, synaptic inputs gave rise to a slow depolarizing response (SDR) in the dendritic field. The amplitude of SDR correlated well with presynaptic activities, suggesting that it was related to transmitter release. The SDR was found to be caused by the activities of glutamate transporters because it was not affected by blockers for GABAA, nACh, 5-HT3, P2X, or metabotropic glutamate receptors but was greatly reduced by dihydrokainate (DHK), a specific blocker for GLT-1 transporter, and by D, L-threo-beta-hydroxyaspartate (THA), a blocker for EAAC, GLAST, and GLT-1 transporters. When SDR was detected with RH482 dye, which stains both glial and neuronal cells, 1 mM DHK and 1 mM THA were equally effective in suppressing SDR. The SDR was very small in GLT-1 knockout mice but was maintained in gerbil hippocampi in which postsynaptic neurons were absent because of ischemia. Because GLT-1 transporters are exclusively expressed in astrocytes, our results provide direct evidence that astrocytes play the dominant role in sequestering synaptically released glutamate.