Glutamate as a CNS transmitter. I. Evaluation of glucose and glutamine as precursors for the synthesis of preferentially released glutamate

Glutaminase-like immunoreactivity in the lower brainstem and cerebellum of the adult rat

Distribution of putative glutamatergic neurons in the lower brainstem and cerebellum of the rat was examined immunocytochemically by using a monoclonal antibody against phosphate-activated glutaminase, which has been proposed to be a major synthetic enzyme of transmitter glutamate and so may serve as a marker for glutamatergic neurons in the central nervous system. Intensely-immunolabeled neuronal cell bodies were densely distributed in the main precerebellar nuclei sending mossy fibers to the cerebellum; in the pontine nuclei, pontine tegmental reticular nucleus of Bechterew, external cuneate nucleus, and lateral reticular nucleus of the medulla oblongata. Phosphate-activated glutaminase-immunoreactive granular deposits were densely seen in the brachium pontis and restiform body, suggesting the immunolabeling of mossy fibers of passage. In the cerebellum, neuropil within the granule cell layer of the cerebellar cortex displayed intense phosphate-activated glutaminase-immunoreactivity, and that within the deep cerebellar nuclei showed moderate immunoreactivity. These results indicate that many mossy fiber terminals originate from phosphate-activated glutaminase-containing neurons and utilize phosphate-activated glutaminase for the synthesis of transmitter glutamate. Intensely-immunostained neuronal cell bodies were further observed in other regions which have been reported to contain neurons sending mossy fibers to the cerebellum; in the dorsal part of the principal sensory trigeminal nucleus, dorsomedial part of the oral subnucleus of the spinal trigeminal nucleus, interpolar subnucleus of the spinal trigeminal nucleus, paratrigeminal nucleus, supragenual nucleus, regions dorsal to the abducens nucleus and genu of the facial nerve, superior and medial vestibular nuclei, cell groups f, x and y, hypoglossal prepositus nucleus, intercalated nucleus, nucleus of Roller, reticular regions intercalated between the motor trigeminal and principal sensory trigeminal nuclei, linear nucleus, and gigantocellular and paramedian reticular formation. Neuronal cell bodies with intense phosphate-activated glutaminase-immunoreactivity were also found in other brainstem regions, such as the paracochlear glial substance, posterior ventral cochlear nucleus, and cell group e. Although it is still controversial whether all glutamatergic neurons use phosphate-activated glutaminase in a transmitter-related process and whether phosphate-activated glutaminase is involved in other metabolism-related processes, the neurons showing intense phosphate-activated glutaminase-immunoreactivity in the present study were suggested to be putative glutamatergic neurons.

Changes in the activity of glutamate related enzymes in cerebral cortex, during insulin-induced seizures

The activity of glutamate related enzymes and the concentration of glutamine, glutamate and gamma-amino n-butyric acid (GABA) were investigated in the cerebral cortex of rats, in different stages of insulin-induced hypoglycemia. Hypoglycemia was produced by intraperitoneal injection of insulin 0.05-100 units per kg body weight. The minimum required dose to produce irreversible severe hypoglycemia was 0.5 units/kg. In 85% of the cases an insulin induced hypoglycemic convulsion, was achieved 130-150 minutes after injection. Blood glucose levels during insulin induced seizures ranged between 8-15 mg%. In the range of 0.5-100 u insulin/kg the degree of hypoglycemia and the onset of convulsions were identical. The concentration of glutamine was significantly reduced during convulsive and postconvulsive stages. Glutamate and GABA concentrations were reduced significantly in all stages of insulin-induced hypoglycemia. The decrease in glutamine concentration was concurrent with an increase in the activity of its degradative enzyme, glutaminase. This was apparent at the preconvulsive, convulsive and postconvulsive stages. The activity of other enzymes related to energy production such as glutamate dehydrogenase (GDH), glutamate transaminase (GPT) and aspartate aminotransferase (AAT) were also increased. The activity of glutamine synthase (GS) was unaffected by hypoglycemia. Insulin induced changes in glutamine, glutamate and their related enzymes could not be attributed to convulsion since a similar pattern of changes was observed in the preconvulsive and postconvulsive stages, and no changes were detected following picrotoxin-induced seizures.

Isolation of metabolically distinct synaptosomes on Percoll gradients

Synaptosomes were prepared from whole rat brain by six different methods based on gradients of sucrose, Ficoll or Percoll. In these, the synthesis and calcium-specific release of amino acids were assessed by two different procedures. Preparations based on sucrose showed the least calcium-specific release, followed by Ficoll-derived synaptosomes. As previously described, Percoll gave two separate populations of synaptosomes, both very active in terms of release of aspartate, glutamate, and GABA. The data involving release and synthesis were not identical, but did agree in the following: in low-density synaptosomes, haloperidol blocked both the release and synthesis of glutamate, but was without effect in the heavier population. 2-chloroadenosine and 2-oxoglutarate affected both release and synthesis only in the high-density population. Dopamine blocked aspartate release and synthesis only in the high-density population. These results suggest that haloperidol interferes with glutamate release and synthesis via a mechanism which may not involve adenosine, serotonin, or dopamine.

Hypoxic neuronal injury in vitro depends on extracellular glutamine

Hypoxic neuronal injury (HNI) in cortical cell cultures was enhanced in a concentration-dependent fashion by the presence of 500 microM to 2 mM (EC50 about 500 microM) glutamine in the medium, concentrations approximating those normally present in cerebrospinal fluid (CSF). Regardless of the glutamine concentration, glutamate receptor antagonists 2-amino-5-phosphonovalerate or dextrorphan could substantially reduce HNI. Thus, the availability of extracellular glutamine could be a determinant of hypoxic neuronal injury in vivo, most likely reflecting its importance in the synthesis of the neurotransmitter excitotoxins glutamate and aspartate.

Importance of glutamine for gamma-aminobutyric acid synthesis in rat neostriatum in vivo

This work was carried out to evaluate the importance of glial cells in providing precursors for the in vivo synthesis of gamma-aminobutyric acid (GABA). Fluorocitrate, which selectively inhibits the tricarboxylic acid cycle in glial cells, was administered locally in rat neostriatum. Inhibition of the glial cell tricarboxylic acid cycle led to a decrease both in glutamine level and in gamma-vinyl GABA (GVG)-induced GABA accumulation, an observation indicating reduced GABA synthesis. The role of glutamine, which is synthesized in glial cells as a precursor for GABA, was further investigated by inhibition of glutamine synthetase with intrastriatally administered methionine sulfoximine. In this case, the glutamine level was reduced to near zero values, and the GVG-induced GABA accumulation was only half that of normal. The results show that glutamine is an important precursor for GABA synthesis, but it cannot be the sole precursor because it was not possible to depress the GVG-induced GABA accumulation completely.

Astrocyte metabolism of [15N]glutamine: implications for the glutamine-glutamate cycle

The metabolism of glutamine was studied in cultured astrocytes by incubating these cells with [2-15N]-glutamine and using gas chromatography-mass spectrometry to quantitate the transfer of 15N to other amino acids. We found that astrocytes simultaneously synthesize and consume [2-15N]glutamine, with the respective synthetic and utilization rates being approximately equal (ca. 13.0 nmol min-1 mg protein-1). Considerable 15N was transferred to alanine and a significant amount to the essential amino acids leucine, tyrosine, and phenylalanine, the latter process denoting active reamination of cognate ketoacids. A net export of alanine into the medium was noted. Astrocyte glutamine utilization appeared to be mediated via both the phosphate-activated glutaminase (PAG) pathway and the glutamine aminotransferase pathway, the activity of which was about half that of PAG. The glutamine concentration in the incubation medium determined whether net synthesis or utilization of this amino acid occurred. When glutamine was omitted from the medium, net synthesis occurred. When it was present at a high (5 mM) level, net consumption was observed. At a physiologic (0.5 mM) concentration, neither net synthesis nor consumption was noted, although the 15N data indicated that glutamine was actively metabolized. An implication of this work is that astrocytes clearly are capable of both synthesizing and utilizing glutamine, and current concepts of a glutamate-glutamine cycle functioning stoichiometrically between astrocytes and neurons may be an oversimplification.

Metabolism of glutamate and ammonia in astrocyte: an immunocytochemical study

Alpha-ketoglutarate (alpha-KG) reductive amination activity in rat brain was found to be mostly absorbed with an antibody against liver glutamate dehydrogenase. With this and anti-glutamine synthetase antibodies, alpha-KG reductive amination activity was immunocytochemically shown to coexist with glutamine synthetase activity in astrocytes. The results suggest that astrocytes de novo synthesize glutamate from alpha-KG and ammonia, and metabolize it to glutamine.

Evidence that aspartate aminotransferase activity and ketodicarboxylate carrier function are essential for biosynthesis of transmitter glutamate

Based on the selective inhibition of glutamate release in cerebellar granule cells in primary cultures by the aspartate aminotransferase inhibitor, aminooxyacetic acid, and by the ketodicarboxylate carrier inhibitor, phenylsuccinate, a novel model for synthesis of transmitter glutamate is suggested: Glutamate is formed from glutamine in the mitochondrial intramembrane space by phosphate-activated glutaminase, transported across the inner membrane in exchange with aspartate, transaminated in the matrix to alpha-ketoglutarate, which via the ketodicarboxylate carrier is transferred to the cytoplasm, and transaminated to form transmitter glutamate. Such a mechanism would explain the functional role of aspartate aminotransferase in glutamatergic neurons.

The effect of fluorocitrate on transmitter amino acid release from rat striatal slices

In order to study the role of glutamine from glial cells for the synthesis of transmitter amino acids, the effect of the gliotoxic-substance fluorocitrate on amino acid release from slices was investigated. In vivo treatment with 1 nmol fluorocitrate reduced the Ca2+ dependent K+ evoked release of endogenous glutamate and GABA from the slices, whereas the glutamine efflux decreased and alanine efflux increased. The K+ evoked release of [3H]D-aspartate increased during fluorocitrate treatment. The latter is consistent with an inhibited uptake of D-aspartate into glial cells. Incubation of striatal slices with fluorocitrate (0.1 mM) decreased the glutamine efflux and increased the alanine efflux. Similar to the in vivo condition, fluorocitrate increased the K+ evoked [3H]D-aspartate release, but the K+ evoked release of endogenous glutamate and GABA increased rather than decreased. The ratio between the K+ evoked release of exogenous D-aspartate to endogenous glutamate increased in both cases. The results suggest an important role of glial cells in the synthesis and inactivation of transmitter amino acids.

Effects of neuroleptics on glutaminase from rat synaptosomes

Phosphate-activated glutaminase was isolated from synaptosomes from three areas of rat brain. Glutamine utilization phosphate activation and inhibition by glutamate or ammonia were assessed in the absence or presence of haloperidol, chlorpromazine, or clozapine. All three drugs (at 1 micromolar concentration) elevated the Km for glutamine using preparations from the amygdala, hippocampus, or striatum. They interfered with phosphate activation only in the amygdala preparation. No drug affected end-product inhibition. The data suggest that neuroleptics may depress the release of glutamic acid from synaptosomes by interfering with the activation of glutaminase by phosphate.

Histochemically demonstrable activity of phosphate-activated glutaminase in the postnatally developing rat hippocampus

Phosphate-activated glutaminase (PAG) mediating the conversion of glutamine to glutamate and ammonia, appears to be the major glutamate metabolizing enzyme in brain. The functional relevance of PAG in postnatally maturing glutamatergic/aspartatergic structures of the rat hippocampus was studied by means of quantitative enzyme histochemistry as an alternative to immunocytochemical techniques. The calibration of the histochemical PAG reaction as well as several control experiments for specificity were carried out to ensure reliability of findings. PAG activity increased markedly during the first weeks of life with a drastic rise between postnatal days 12 and 15. On the other hand, activity of NADH diaphorase involved in the histochemical PAG assay as an auxiliary enzyme, showed a different distribution pattern as well as a different developmental sequence with high levels early in ontogenesis. The topographical and temporal parallelisms of PAG activity to several other parameters which are putatively associated with postnatally maturing glutamatergic/aspartatergic transmission processes, mutually indicate their significance in such a functional context.

Isolation of a cDNA for rat brain glutaminase

A single phage was isolated from a lambda gt11 rat brain cDNA library by screening with antibodies prepared against rat renal glutaminase. Partial proteolysis of the fusion protein produced by a lysogen of the isolated phage generated a series of immunoreactive peptides that co-migrated with those derived from the purified brain glutaminase. The cDNA has a single open reading frame which encodes 326 amino acids that are in frame with beta-galactosidase. A 72-kDa protein, corresponding in size to the precursor of mitochondrial glutaminase, was immunoprecipitated from the translation products of rat renal mRNA that selectively hybridized to the cDNA. A probe made from the glutaminase cDNA detected an mRNA about 6 kb in length. This mRNA was present in rat brain and normal kidney RNA, increased 6-fold in acidotic kidney RNA, but was not detectable in liver RNA.

Relationships between glutamate and monoamine metabolites in cerebrospinal fluid and serum in healthy volunteers

The concentrations of homovanillic acid (HVA), 5-hydroxyindoleacetic acid (5-HIAA), 4-hydroxy-3-methoxyphenylglycol (HMPG), and glutamate were determined in cerebrospinal fluid (CSF) and serum in 10 healthy volunteers. The monoamine metabolites were measured by mass fragmentography and the glutamate by high-performance liquid chromatography. The level of glutamate in CSF was low (0.34 +/- 0.14 nmol/ml) in comparison with previously published values. The concentrations of monoamine metabolites in CSF were in close agreement with earlier findings. There were negative correlations between the concentrations of HVA (r = -0.82, p less than 0.01) and 5-HIAA (r = -0.77, p less than 0.01) and glutamate in CSF, but not in serum. The serum levels of HMPG and glutamate were negatively correlated (r = -0.95, p less than 0.001), but not the CSF levels. The HMPG levels in serum and CSF were positively correlated (r = 0.94, p less than 0.001), but not the HVA and the 5-HIAA levels. The serum and CSF levels of glutamate were positively correlated (r = 0.67, p less than 0.05). The results indicate relationships among the metabolism of dopamine, serotonin, and glutamate in the brain and between the metabolism of noradrenaline and glutamate in peripheral tissue.

In vivo analysis of cortical amino acid neurotransmitters collected in the rat by a new double lumen push-pull catheter system

The release of both endogenous and newly synthesized amino acid neurotransmitters was examined simultaneously in different areas of the cerebral cortex in the freely moving rat. An array of push-pull guide tubes was implanted permanently to rest above the frontal, parietal, temporal and occipital areas of the cortex of each rat. Then a new double-lumen catheter system, specially adapted for localized push-pull perfusion of the conscious animal, was used to perfuse an artificial cerebrospinal fluid at each cortical site. For the new synthesis experiments, 0.5 microCi of [14C]glucose in a volume of 2.0 microliter was first microinjected into the perfusion site as a precursor to label amino acids. After the site was perfused at a rate of 12.0 microliter/min, each of the samples was assayed by two-dimensional thin-layer chromatography. In a second analysis, the content of six endogenous amino acids present in unlabeled samples of push-pull perfusate was quantified by high-performance liquid chromatography analysis with electrochemical detection. The results showed a notable homogeneity among each of the four cortical areas in the content of four of the six amino acids examined. Endogenous glutamine exhibited the highest proportional content in the cortical perfusates, whereas glutamic acid was proportionally higher in terms of new synthesis. An anatomical analysis revealed that the level of endogenous glutamic acid in the frontal area was significantly lower than that found in the occipital or temporal regions of the rat's cortex. An opposite result was obtained when the proportional synthesis of glutamic acid from [14C] glucose was compared in different cortical regions in that a statistically higher release occurred in the frontal than in the occipital cortex.

Immunohistochemical study of glutaminase-containing neurons in the cerebral cortex and thalamus of the rat

In an attempt to identify glutamatergic neurons, the cerebral cortex and thalamus of the rat were examined immunohistochemically by using a monoclonal antibody against phosphate-activated glutaminase (PAG), a major synthetic enzyme of transmitter glutamate in the central nervous system. In both the neocortex and mesocortex, pyramidal cells in layers V and VI showed intense PAG-like immunoreactivity (PAG-LI), whereas neuronal cell bodies in layers I-IV showed weak PAG-LI. At the deep border of layer VI, neurons with horizontally elongated cell bodies showed PAG-LI. In the pyriform and entorhinal cortices, neurons with intense to moderate PAG-LI were seen in layer II as well as in the deeper layers. In the hippocampal formation, pyramidal cells in CA1, CA2, and CA3 and polymorphic cells in CA4 showed PAG-LI; PAG-LI was most intense in pyramidal cells of CA3. Fine granules with weak PAG-LI were also seen on and/or within the cell bodies of granule cells in the dentate gyrus. In the thalamus, neurons with PAG-LI were distributed in all nuclei, although regional differences were observed in the distribution pattern of neurons with PAG-LI and in the intensity of PAG-LI in individual neurons. The largest neurons in each thalamic nucleus showed intense PAG-LI; these were considered to be projection neurons. In addition to perikaryal labeling, many fine, PAG-like immunoreactive granules were distributed in the neuropil of both the cerebral cortex and thalamic nuclei. Some of these fine granules with PAG-LI in the neuropil were assumed to represent fiber terminals with PAG-LI, because the distribution pattern of the deposits in the primary somatosensory and primary visual cortices resembled that of thalamocortical fiber terminals. Glutamate is rather ubiquitous in the mammalian central nervous system, and it is still debatable whether the monoclonal antibody to PAG from brain mitochondria can distinguish transmitter-related glutaminase from the other metabolism-related ones. In the present study, however, large neurons in the thalamic nuclei, as well as pyramidal neurons in the cerebral cortex, showed PAG-LI most intensely, supporting the assumption that projection neurons of the cerebral cortex and thalamus are primarily glutamatergic.

Tissue culture of retinal glial cells

In order to be able to investigate the properties and characteristics of glia in the retina, a monotypic culture of retinal glial cells is likely to be an important research vehicle. Several techniques are now available to produce cultures of glial cells from the retina. These methods generally result in cultures of Mueller cells rather than retinal astrocytes. Publications on glial cultures involved complex procedures for the isolation of the target cell. Recent developments have resulted in simpler procedures with the advantage that large numbers of cultures can be established quickly and easily. Glial cultures have already been used in a variety of studies, simpler methods of culture, particularly if these can be adapted for culture of human glial cells, will probably result in more extensive use of cultures to unravel the properties of retinal glia.

A novel chloride-dependent L-[3H]glutamate binding site in astrocyte membranes

Membrane fractions prepared from astrocytes grown in culture exhibit a specific binding site for L-[3H]glutamate that is Cl--dependent and Na+-independent. The binding site is a single saturable site with a KD of about 0.5 microM, is inhibited by L-aspartate, L-cysteate, and quisqualate, and is insensitive to kainate, N-methyl-D-aspartate, alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate, and 2-amino-4-phosphonobutyrate. The pharmacological characteristics of the binding site indicate that it is distinct from any site previously described in synaptic membrane preparations. Comparisons of ionic requirements, ligand specificity, and inhibitor sensitivities, however, suggest the described binding is the first step in a Cl--dependent high-affinity glutamate uptake system. Such binding studies provide a useful model system in which to investigate the close association between excitatory amino acids, astrocytes, the termination of glutamate's excitatory action by high-affinity uptake, and the excitotoxic action of acidic amino acids in membranes of a single cell type.

Increase in the stimulation-induced overflow of glutamate by fluoroacetate, a selective inhibitor of the glial tricarboxylic cycle

Fluoroacetate is known to be taken up selectively by glia, where after forming fluorocitrate, it inhibits the tricarboxylic acid cycle. Since uptake into glia has a major role in the inactivation of synaptically released glutamate, the effect of fluoroacetate on the overflow of glutamate evoked by electrical field stimulation in slices of rat hippocampus was investigated. In agreement with previous reports, 1 mM fluoroacetate reduced the release and content of glutamine, but increased only slightly the overflow of glutamate induced by stimulation. If, however, 0.5 mM glutamine was added to the superfusion fluid, fluoroacetate nearly tripled the overflow of glutamate evoked by electrical field stimulation. The large glutamate overflow due to field stimulation in the presence of fluoroacetate was fully Ca2+ -dependent. Results confirm the major role of glia in the inactivation of glutamate. The absence of such an uptake may contribute to the in vivo convulsive effect of fluoroacetate.

In vivo studies on the extracellular, and veratrine-releasable, pools of endogenous amino acids in the rat striatum: effects of corticostriatal deafferentation and kainic acid lesion

The effects of corticostriatal deafferentation (decortication) and destruction of intrinsic neurons (intrastriatal kainate injection) on the extracellular concentration, and veratrine-releasable pools, of endogenous amino acids in the rat striatum were examined using the in vivo brain dialysis technique. Intracellular amino acid content was also determined. Decortication reduced selectively intra- and extracellular levels of glutamate (Glu) and aspartate (Asp). Extracellular changes were more pronounced than those in tissue content. gamma-Aminobutyric acid (GABA), taurine (Tau), and phosphoethanolamine (PEA) levels were not affected, whereas nonneuroactive amino acids were increased at 1 week but not at 1 month post-lesion. The intracellular pool of Glu and Asp was also reduced in kainate-lesioned striata. However, extracellular levels of these compounds were not affected significantly by this treatment. The tissue content of all other amino acids was decreased, the most prominent change being in the concentration of GABA. Extracellular GABA concentration was also reduced dramatically, whereas the concentrations of noneuroactive amino acids were increased to varying degrees. These data suggest that transmitter pools of neuroactive amino acids are an important supply for their extracellular pools. Lesion-induced alterations in nonneuroactive amino acids are discussed with regard to the loss of metabolic pools, glial reactivity, and changes in blood-brain barrier transport. Veratrine induced a massive release of neuroactive amino acids such as Glu, Asp, GABA, and Tau into the extracellular fluid, and a delayed increase in PEA. Extracellular levels of neuroactive amino acids were raised slightly. Decortication reduced, selectively, the amounts of Glu and Asp released by veratrine.(ABSTRACT TRUNCATED AT 250 WORDS)

Effect of cortical kindling on [3H]D-aspartate uptake and glutamate metabolism in rats

The effects of cortical kindling in rats on [3H]D-aspartate uptake and on glutaminase and glutamine synthetase activities has been studied. The high affinity uptake of [3H]D-aspartate in control cortical tissue (Km approximately 2 microM) was undetectable in the kindled tissue, whilst the enzyme activities were unchanged. A loss of the high-affinity uptake sites for excitatory amino acids may be a contributing factor to the kindling phenomenon.

Serum effects on substrate oxidation by dissociated brain cells: possible sites of action

This report is an extension of recent studies indicating the presence of a factor in serum that preferentially inhibits 14CO2 production from labeled glucose. Experiments with dissociated cells revealed that the inhibitory effects of serum were only slightly changed over more than a 50-fold range in initial glucose concentration. Serum had no effect on the rate of glucose transport (uptake of 1,3[3H]2-deoxyglucose). The inhibitory effect of serum was greater on 14CO2 production from [6-14C]glucose than [1-14C]glucose. Other studies revealed that 14CO2 production from [1-14C]pyruvate was more than 5 times the rate obtained using [3-14C]pyruvate; however, the inhibitory effect of serum was much greater on the latter (20% vs 60% inhibition respectively) at 2 mM pyruvate and in the presence of 1% fetal bovine serum. Attempts to characterize the factor using Amicon filtration showed the highest inhibitory activity in a 10,000 mol. wt. fraction, although some inhibitory activity was found in commercial preparations of bovine serum albumin. Delipidation of serum had no effect. Based on these results, we postulate that the observed decrease in labeled CO2 production reflects the regulation of substrate utilization at the pyruvate carboxylase step by one or more factors in serum (with a mol. wt. of approximately 10,000).

Immunocytochemical characterization of glutamate dehydrogenase in the cerebellum of the rat

The immunocytochemical distribution of glutamate dehydrogenase was studied in the cerebellum of the rat using antibodies made in rabbit and guinea pig against antigen purified from bovine liver. Antiserum was found to block partially enzymatic activity both of the purified enzyme and of extracts of the rat cerebellum. Using immunoblots of proteins of rat cerebellum, a major immunoreactive protein and several minor immunoreactive proteins were detected with antiserum. Only a single immunoreactive protein was detected using affinity-purified antibody preparations. This protein migrates with a molecular weight identical to that of the subunit of glutamate dehydrogenase. Further evidence that the antibodies were selective for glutamate dehydrogenase in rat cerebellum was obtained through peptide mapping. Purified glutamate dehydrogenase and the immunoreactive protein from rat cerebellum generated similar patterns of immunoreactive peptides. No significant cross-reaction was observed with glutamine synthetase. Immunocytochemistry was done on cryostat- and Vibratome-cut sections of the cerebellum of rats that had been perfused with cold 4% paraformaldehyde. Glial cells were found to be the most immunoreactive structures throughout the cerebellum. Most apparent was the intense labeling of Bergmann glial cell bodies and fibers. In the granule cell layer, heavy labeling of astrocytes was seen. Purkinje and granule cell bodies were only lightly immunoreactive, whereas stellate, basket, and Golgi cells were unlabeled. Labeling of presynaptic terminals was not apparent. These findings suggest that glutamate dehydrogenase, like glutamine synthetase, is enriched in glia relative to neurons.

A small subset of cortical astrocytes in culture accumulates glycogen

We are interested in identifying the target cells for norepinephrine in cerebral cortex and in characterizing the effects of norepinephrine on these target cells. Norepinephrine inhibits the incorporation of tritiated glucose into glycogen in rat cerebral cortex in dissociated cell culture. To identify which cells store glycogen in these cultures we combined glycogen cytochemistry with glial fibrillary acidic protein immunocytochemistry. Using this technique we show that cytochemically detectable glycogen is restricted to a small subset of astrocytes as well as an unidentified cell type which does not contain glial fibrillary acidic protein. These results demonstrate that only a minority of astrocytes in cortical cultures accumulate glycogen. Therefore cortical astrocytes are differentiated with respect to glycogen accumulation, an important metabolic function. We do not know if glycogen accumulation in astrocytes is a constitutive or facultative property. In either case the subset of astrocytes which accumulates glycogen might be one of the major cellular targets for norepinephrine in cerebral cortex.

Reversible shifts in the Ca2+-dependent release of aspartate and glutamate from hippocampal slices with changing glucose concentrations

It is known that low glucose concentrations increase the aspartate and decrease the glutamate content of brain tissue both in vivo and in vitro. To see whether these changes occur in the transmitter compartment or not, the release of aspartate and glutamate evoked by electrical-field stimulation or by high K+ was followed in slices of rat hippocampus superfused with 5 or 0.2 mM glucose. Superfusion with 0.2 mM glucose increased the evoked release of aspartate about ten times and that of glutamate about threefold. This shift in the ratio of aspartate to glutamate released was accompanied by a similar increase in the relative amount of aspartate contained in the slices. The high evoked release of aspartate and glutamate was well maintained, provided 0.5 mM glutamine was added to the medium. Changing the concentration of glucose after the first period of stimulation rapidly altered the relative amounts of aspartate and glutamate released but not the enhanced release of glutamate. The large evoked release of both aspartate and glutamate in 0.2 mM glucose was almost entirely Ca2+-dependent. The relative amounts of aspartate and glutamate released by 50 mM K+ also changed when the glucose concentration was reduced. Results suggest two effects of low glucose concentrations: an increase in the overflow of synaptically released glutamate due to a decreased uptake and an increase in the proportion of aspartate to glutamate formed and released from the transmitter pool. These observations are consistent with the interpretation that these two transmitters can be released in different proportions from the same terminals.

Effect of thyroid deficiency on the regional development of glutaminase, a glutamatergic neuron marker, in the rat brain

The effect of thyroid deficiency on the activity of phosphate-activated glutaminase (the marker for glutamatergic neurons) was studied in different parts of the rat brain at ages 5, 10, 15 and 25 days, and at day 130 following 102 days of rehabilitation. The brain regions investigated were the cerebral cortex, basal forebrain, hippocampus and cerebellum. During normal development, the activity of glutaminase increased relatively earlier in the cerebral cortex and hippocampus than in the cerebellum, while the absolute value reached a much higher level in the hippocampus than in other brain regions. In the basal forebrain, the developmental pattern of glutaminase was bimodal, and the rise in enzyme activity after 15 days coincided with the decrease in the cerebral cortex. These regional developmental changes in glutaminase activity correlated well with known information on the formation of glutamatergic cells and pathways in the brain. Neonatal thyroid deficiency had little effect on the developmental patterns of enzyme activity, the exception being a transient decrease in 10-day-old hypothyroid hippocampus. The present results, together with previous findings, indicate that the effect of thyroid hormone on neural maturation is cell-type specific and the glutamatergic neurons are not the main targets of thyroid hormone action.

2-Oxo-[14C]glutarate is taken up by glutamatergic nerve terminals in the rat striatum

High affinity uptake of [14C]glutamate into rat striatal synaptosomes was reduced by 33% after bilateral cortical ablation. The lesion had no effect on striatal [14C]GABA uptake, but reduced 2-oxo-[14C]glutarate uptake by 67%. The results demonstrate the existence of a high-affinity uptake site for 2-oxoglutarate on glutamatergic nerve terminals and support the contention that this Krebs cycle intermediate may be used to replenish the neuronal pool of neurotransmitter glutamate. 2-Oxo-[14C]glutarate uptake may serve as a selective marker for glutamatergic neurones.

Effects of glutaminase inhibition on release of endogenous glutamic acid

The effects of four inhibitors of glutamine hydrolysis on synaptosomes derived from several regions of the brain were studied. The calcium-specific release of endogenous glutamic acid was determined in the presence of varying concentrations of 6-diazo-5-oxo-norleucine (DON), N-ethyl-maleimide (NEM), 2-chloroadenosine (2-CA) or haloperidol. Both DON and NEM reduced the calcium-specific release in a concentration-dependent manner, equally in all regions tested. 2-Chloroadenosine also decreased release and the effect was most evident in the amygdala. As reported earlier, haloperidol blocked release of glutamic acid only in the amygdala. In synaptosomes from the amygdala, both DON and NEM failed to affect the calcium-specific release of aminobutyric acid (GABA), glycine or serotonin at concentrations which reduced release of glutamate by 50%; NEM, but not DON, elevated the release of dopamine. Dopamine itself affected neither the release of glutamate nor its blockade by haloperidol even in extremely large concentrations.

Neurotransmitter glutamate: its clinical importance

Excitatory amino acid glutamate has several important functions in the mammalian central nervous system (CNS). This review focuses on the transmitter role of glutamate and discusses anatomical and pharmacological data of clinical neurological relevance. Experimental and clinical conditions which have been associated with altered content, uptake, membrane binding or release of glutamate in the CNS are discussed. Such conditions include, epilepsy, disorders of the basal ganglia, cerebral ischemia, hypoxia, hypoglycemia, metabolic encephalopathies, olivopontocerebellar atrophy and cerebellar ataxias, amino acidopathies, mental and other neurological disorders. With the exception of a few fibre systems, it is very difficult to differentiate between glutamate and aspartate as CNS transmitters. The term glutamate is, thus, used in the sense glutamate and/or aspartate unless specifically stated.

Effect of insulin-induced hypoglycemia on the concentrations of glutamate and related amino acids and energy metabolites in the intact and decorticated rat neostriatum

The glutamate (Glu) terminals in rat neostriatum were removed by a unilateral frontal decortication. One to two weeks later the effects of insulin-induced hypoglycemia on the steady-state levels of amino acids [Glu, glutamine (Gln), aspartate (Asp), gamma-aminobutyric acid (GABA), taurine] and energy metabolites (glucose, glycogen, alpha-ketoglutarate, pyruvate, lactate, ATP, ADP, AMP, phosphocreatine) were examined in the intact and decorticated neostriatum from brains frozen in situ. The changes in the metabolite levels were examined during normoglycemia, hypoglycemia with burst-suppression (BS) EEG, after 5 and 30 min of hypoglycemic coma with isoelectric EEG, and 1 h of recovery following 30 min of isoelectric EEG. In normoglycemia Glu decreased and Gln and glycogen increased significantly on the decorticated side. During the BS period no significant differences in the measured compounds were noted between the two sides. After 5 min of isoelectric EEG Glu, Gln, GABA, and ATP levels were significantly lower and Asp higher on the intact than on the decorticated side. No differences between the two sides were found after 30 min of isoelectric EEG. After 1 h of recovery from 30 min of isoelectric EEG Glu, Gln, and glycogen had not reached their control levels. Glu was significantly lower, and Gln and glycogen higher on the decorticated side. The Asp and GABA levels were not significantly different from control levels. The results indicate that the turnover of Glu is higher in the intact than in decorticated neostriatum during profound hypoglycemia.

Excitatory amino acid neurotransmitters in the corticostriate pathway: studies using intracerebral microdialysis in vivo

The concentration of extracellular excitatory amino acids in the striatum of conscious, unrestrained rats was measured using intracerebral microdialysis, during chemical stimulation of the striatum in intact and hemidecorticate animals. Chemical stimulation of the striatum with tityustoxin (0.1 microM) evoked a rise in dialysate concentration of glutamate (to 383% of basal) and aspartate (to 156% of basal), accompanied by a drop in glutamine (to 55% of basal). These changes showed significant attenuation after treatment with L-proline (1 mM) or 2-chloroadenosine (15 microM). Unilateral degeneration of the corticostriate pathway, produced by frontal hemidecortication, caused a reduction in both basal and stimulated levels of glutamate in the lesioned side, whereas no effect was observed in the intact side. Similarly, basal and stimulated levels of glutamine were unchanged in the intact side, but were increased in the lesioned side. These results provide in vivo evidence for glutamate and possibly aspartate being neurotransmitters in the corticostriate pathway. In addition they lend support to previous studies in vitro, which implicated glutamine as the principal precursor for neurotransmitter glutamate.

Regional distribution and extracellular levels of amino acids in rat central nervous system

The extracellular levels of aspartate, glutamate, serine, glutamine, glycine, alanine and GABA were studied in vivo with the microdialysis technique in 15 different regions of the rat brain. The effect of high K+ on the overflow of these amino acids was also studied. These results were compared with those from a regional dissection of 17 brain regions in which the tissue content of the same amino acids was determined. The in vivo data showed an unevenly distributed KCl response of aspartate, glutamate, taurine and GABA, all of which are putative neurotransmitters. It was not possible to predict the response to high K+ from the magnitude of the unstimulated overflow. Glutamine overflow was inversely related to that of glutamate during the high K+ stimulus, which is consistent with glutamine being the main precursor of glutamate. Only for GABA and alanine was overflow proportional to the tissue level in the different regions studied.

Non-linear kinetics of glutamyl-tRNA synthesis catalyzed by high molecular weight complexes from rat brain neuronal cells but not from glial cells

High molecular weight complexes of aminoacyl-tRNA synthetases isolated from rat brain catalyze the formation of glutamyl-tRNA with an initial lag time of the order of 1 min, as previously reported for the formation of glutamyl-tRNA and glutaminyl-tRNA catalyzed by similar complexes from bovine brain (Vadeboncoeur and Lapointe, Eur. J. Biochem., 109 (1980) 581-587). To determine the type(s) of brain cell(s) where this phenomenon occurs, we have studied the kinetics of glutamyl-tRNA formation catalyzed by high molecular weight complexes of aminoacyl-tRNA synthetases isolated from neuronal and from glial cells, either transformed (Neuro-2A and C6), or from primary cultures, or isolated from rat brain. The delay in the formation of glutamyl-tRNA was observed only in the case of neuronal cells isolated from rat brain, whereas a delay in the formation of glutaminyl-tRNA was also seen in these cells, as well as in neuronal cells in primary culture and in synaptosomes. The kinetics of formation of aspartyl-tRNA and valyl-tRNA catalyzed by high molecular weight complexes from all these cells was linear.

Exogenous glutamate is metabolized to glutamine and exported by rat primary astrocyte cultures

Rat cortical astrocytes in primary culture were examined for their capacity to transport and metabolize exogenous L-[U-14C]glutamate. After incubation for time periods up to 120 min, cells and incubation media were analyzed for labelled and endogenous glutamate and its metabolic products by HPLC coupled with fluorescence detection and liquid scintillation counting. Glutamine was the major labelled metabolite after 120 min, accounted for 38% of the original glutamate label, and was found primarily in the incubation medium. A further 13.5% of the label was recovered in deaminated metabolites of glutamate, 1.2% was associated with aspartate, 23% remained in glutamate, and 10.2% was found in an acid-precipitated cell fraction. More than 84% of the label was recovered in these fraction. suggesting that the maximum possible formation and loss of 14CO2 was 16%. The rate of total glutamine synthesis was 1.1 nmol X mg protein-1 X min-1 when 9 microM exogenous glutamate was present. The total amount of glutamine synthesized greatly exceeded the consumption of glutamate, indicating that a substantial proportion of glutamine was synthesized from other carbon sources. Almost all of the newly formed glutamine was exported into the medium. These results indicate that astrocytes in primary culture, by accumulating glutamate, producing glutamine, and exporting it, are capable of carrying out the glial component of the glutamine cycle.

Glutamate and the pathophysiology of hypoxic--ischemic brain damage

Information obtained over the past 25 years indicates that the amino acid glutamate functions as a fast excitatory transmitter in the mammalian brain. Studies completed during the last 15 years have also demonstrated that glutamate is a powerful neurotoxin, capable of killing neurons in the central nervous system when its extracellular concentration is sufficiently high. Recent experiments in a variety of preparations have shown that either blockade of synaptic transmission or the specific antagonism of postsynaptic glutamate receptors greatly diminishes the sensitivity of central neurons to hypoxia and ischemia. These experiments suggest that glutamate plays a key role in ischemic brain damage, and that drugs which decrease the accumulation of glutamate or block its postsynaptic effects may be a rational therapy for stroke.

Heterogeneity of sodium-dependent excitatory amino acid uptake mechanisms in rat brain

The pharmacologic and kinetic characteristics of sodium-dependent uptake of [3H]L-glutamate, [3H]D-aspartate, and [3H]L-aspartate into crude synaptosomal preparations of rat corpus striatum and cerebellum have been examined in vitro. In cerebellum the apparent Kts and Vmax for the three excitatory amino acids were identical whereas in striatal synaptosomes, the Vmax for [3H]L-glutamate was 30% greater (P less than or equal to .001) than for [3H]D-aspartate and 50% greater (P less than or equal to .001) than for [3H]L-aspartate. L-Amino adipic acid inhibited the uptake of the three amino acids in both regions of brain was 15- to 20-fold more potent in cerebellum than in striatum. In contrast, dihydrokainic acid inhibited transport processes in the corpus striatum but was without activity in cerebellar preparations. The neurotoxin kainic acid blocked only a portion (60%) of [3H]L-glutamate and [3H]D-aspartate uptake in cerebellum while completely inhibiting amino acid transport in corpus striatum. Three days post kainic acid lesion, [3H]D-aspartate uptake was attenuated more than [3H]L-glutamate uptake in the corpus striatum; destruction of corticostriatal afferents reduced [3H]L-glutamate to a greater extent than [3H]D-aspartate. Various lesions of the cerebellum affected excitatory amino acid transport processes to a similar extent. These results suggest that excitatory amino acid transport systems are pharmacologically distinct in different brain regions and may be heterogeneous within a single region.

Potassium-stimulated efflux of radiolabeled products formed from L-[14C(U)]-glutamine in vitro by the saccule of the rainbow trout (Salmo gairdnerii R.)

Samples of saccular macula from the rainbow trout were incubated in vitro with uniformly-labeled L-[14C]-glutamine, and radiolabeled products, released by potassium-induced depolarization in the presence of calcium, were examined. Most of the effluxed radioactivity was distributed in six (of 17) thin-layer chromatographic fractions. Fractions corresponding to aspartate and glutamate showed highly significant increases in radioactivity (as percent of total recovered radioactivity) during high-potassium treatment. Radioactivity in a fraction with an RF close to that of ornithine also significantly increased during potassium, and dropped sharply after potassium. The origins of the thin-layer fractions, with respect to sensory and neural elements in the saccular macula samples, are discussed.

Properties of the uptake and release of neurotransmitter glutamate in cerebral cortical tissue of guinea pigs

In order to investigate the dynamics of glutamate as a neurotransmitter and to avoid a complication by its metabolism, we studied the uptake and release of labeled non-metabolizable D-isomers of aspartate and glutamate in cerebral cortical slices and synaptosome preparation from guinea-pigs. The rate of uptake of D-aspartate and glutamate was mutually inhibited in a non-competitive fashion, indicating that their uptake mechanisms are not exactly the same. By ouabain (0.05 mM), the uptake of D-aspartate and glutamate into synaptosome preparation was less inhibited than that into cerebral slices. In synaptosome preparation most of the preloaded D-aspartate and glutamate was released by high-potassium (50 mM) stimulation, whereas in cerebral slices only a slight release was observed. However, when the slices were superfused with a medium free of sodium ions, which are absolutely necessary for the uptake, after preloaded with the labeled amino acids in the standard medium, a distinct release of radioactivity was induced by high-potassium stimulation. This potassium-induced release corresponded to only about 20% of the radioactivity accumulated in the slices. The accumulation of D-aspartate and glutamate into cerebral slices was much larger on the basis of their protein content than that into synaptosome preparation, when a high concentration (1 mM) of the amino acids was added to the medium. These observations suggest that the uptake system of D-aspartate and glutamate in cerebral slices is quite different from that in synaptosome preparation, and that the accumulation into cerebral slices is mainly localized in glial cells. In vivo the glial cell uptake is probably more important in removing the released neurotransmitter glutamate.

The effect of methioninesulfoximine, an inhibitor of glutamine synthetase, on the levels of amino acids in the intact and decorticated rat neostriatum

The glutamate terminals in rat rostral neostriatum were removed on one side by ipsilateral frontal cortical ablation. Intraperitoneal administration of L-methioninesulfoximine (MSO), an inhibitor of glutamine synthetase, induced different changes in amino acid concentration in the rostral neostriatum on the intact and decorticated sides. Glutamine was more reduced on the decorticated side than on the intact side. Glutamate changed less, but decreased more on the intact than on the lesioned side. There seems to be a correlation between the decrease in glutamine and the increase in alanine on both sides. GABA and taurine did not change significantly. The results indicate that the formation of glutamine probably formed from released glutamate is lower on the decorticated side than on the intact side.

A comparative analysis of glial and neuronal markers in the retina of fish: variable character of horizontal cells

The immunohistochemical localizations of the enzymes glutamine synthetase, carbonic anhydrase-C, and the intermediate filament protein GFA were examined for potential neuroglial specificity in the retinas of several types of fish. Both glutamine synthetase and carbonic anhydrase-C appear to be characteristic markers for retinal Müller cells. However, the horizontal neurons of most fish examined also possess high levels of carbonic anhydrase. Furthermore, GFA, the characteristic marker for fibrous astroglia in higher vertebrates, was found specifically localized in the horizontal neurons of several teleost fish. The identity of the GFA antigens was qualified by immunochemical as well as cytological examinations. Furthermore, specific antisera to other intermediate filament proteins, including neurofilaments, validated and contrasted with the observations made with antisera to GFA. The presence of GFA in horizontal neurons of fish is widespread but not typical of all fish. These observations indicate an evolutionary constancy of retinal Müller glial cells. However, these results again focus attention on whether horizontal cells are truly neurons or rather represent an intermediate cell type that may prove useful in studying the evolution, ontogeny, and functional significance of the neuronal-glial phenotypic dichotomy.

Effect of glutamine on glutamate release from hippocampal slices induced by high K+ or by electrical stimulation: interaction with different Ca2+ concentrations

To characterize the effect of glutamine on the release of glutamate, aspartate, and gamma-aminobutyric acid (GABA), rat hippocampal slices were superfused with different concentrations of glutamine or Ca2+. Amino acids released and retained were analyzed by HPLC. Glutamine (0.5 mmol/L) increased more than threefold the release of glutamate evoked by 50 mmol/L K+ in the presence of 2.6 mmol/L Ca2+ without a corresponding increase in glutamate content, while the release of aspartate was increased less and that of GABA not at all by glutamine. The evoked release of all three amino acids, including the enhanced release of glutamate in the presence of glutamine, was strongly dependent on Ca2+ concentrations between 0.1 and 2.6 mmol/L. The potentiation of glutamate release by glutamine reached a plateau at 0.25 mmol/L glutamine. Intermittent electrical field stimulation increased the release of only glutamate and this release was nearly doubled by glutamine. The increased release was Ca2+ dependent and tetrodotoxin (TTX) sensitive. Results suggest that extracellular glutamine promotes primarily the formation of releasable glutamate and this enhancement is dependent on extracellular Ca2+.

Potassium-induced release of endogenous amino acids in the guinea pig cochlea

Guinea pig cochleae were perfused with high-potassium solutions to depolarize hair cells artificially and induce the release of afferent neurotransmitter. Sequential injections of artificial perilymph containing 5 mM KCl, then 50 mM KCl, and finally 5 mM KCl were made into the scala tympani. This injection sequence was conducted under either normal divalent-cation conditions (2.0 mM CaCl2, 1.0 mM MgCl2) or calcium-deficient conditions intended to antagonize evoked transmitter release (0.1 mM CaCl2, 20.0 mM MgCl2). The levels of 21 endogenous primary amines in effluent collected from the scala vestibuli were determined by gradient-elution, reverse-phase HPLC using o-phthaldialdehyde-thiol adducts with fluorescence detection. Analyses indicated effluent concentrations of glutamate, taurine, and a coeluting taurine-gamma-aminobutyrate (GABA) fraction (but not GABA alone) increased significantly after exposure to 50 mM KC1 and returned to baseline levels after reintroduction of 5 mM KC1 under normal divalent-cation conditions. Correspondent changes in the release of these constituents were significantly attenuated under calcium-deficient conditions. This was not the case for potassium-induced changes in the release of arginine, aspartate, and isoleucine. These data are consistent with the hypothesis that the receptoneuronal transmitter is glutamate and further suggest a calcium-dependent mechanism involving taurine.

Immunocytochemical localization of glutaminase-like and aspartate aminotransferase-like immunoreactivities in the rat and guinea pig hippocampus

There is considerable evidence that pathways of the hippocampus use an excitatory amino acid as transmitter. We have attempted to immunocytochemically identify excitatory amino acid neurons in the hippocampus of the rat and guinea pig using antiserum to glutaminase and antiserum to aspartate aminotransferase, which have been proposed as markers for aspartergic/glutamergic neurons. Glutaminase-like immunoreactivity was seen in granule cells in the dentate gyrus and fibers and puncta associated with the mossy fiber pathway in the hilus and stratum lucidum of the hippocampus. At the ultrastructural level, glutaminase-like immunoreactivity was observed in mossy fiber terminals in the stratum lucidum. Glutaminase-like immunoreactivity was also seen in pyramidal cells in regio inferior and regio superior and in cells in layer two of the entorhinal cortex. Schaffer collateral terminals, commissural fiber terminals and perforant pathway terminals were not seen at the light microscopic level. Glutaminase-like immunoreactivity is thus found in the cell bodies of proposed excitatory amino acid neurons of hippocampal pathways, but does not appear to label all terminals. Aspartate aminotransferase-like immunoreactivity was not seen in any cells, fibers or terminals in the rat or guinea pig hippocampus.

Ornithine as a precursor of neurotransmitter glutamate: effect of canaline on ornithine aminotransferase activity and glutamate content in the septum of rat brain

Local injections of L-canaline into the septum produce a rapid and almost complete inhibition of ornithine aminotransferase activity followed by a decrease in glutamate content in this region. The time-course of canaline action shows the existence of two glutamate pools with different sizes and half-life values. Surgical lesions of the hippocampal-septal glutamatergic pathway affected the site and kinetics of the small pool of glutamate in the septum, suggesting the participation of ornithine aminotransferase in the synthesis of this pool. This indicates a possible role of ornithine as a precursor of the transmitter glutamate. The localization of ornithine aminotransferase does not seem, however, to be specific for the nerve-terminal compartment. The data obtained allow estimation of the turnover rate of the specific pool of neurotransmitter glutamate.