Studies on the release of exogenous and endogenous GABA and glutamate from rat brain synaptosomes

Inhibition by anandamide and synthetic cannabimimetics of the release of [3H]D-aspartate and [3H]GABA from synaptosomes isolated from the rat hippocampus

Cannabinoids (CB) can act as retrograde synaptic mediators of depolarization-induced suppression of inhibition or excitation in hippocampus. This mechanism may underlie the impairment of some cognitive processes produced by these compounds, including short-term memory formation in the hippocampus. In this study, we investigated several compounds known to interact with CB receptors, evaluating their effects on K(+)-evoked release of [3H]D-aspartate ([3H]D-ASP) and [3H]GABA from superfused synaptosomes isolated from the rat hippocampus. [3H]D-ASP and [3H]GABA release were inhibited to different degrees by the synthetic cannabinoids WIN 55,212-2; CP 55,940, and arachidonyl-2'-chloroethylamide/N-(2-chloroethyl)-5Z,8Z,11Z,14Z-eicosatetraenamide (ACEA), as well as by the endocannabinoids, anandamide (AEA), and 2-arachidonoylglycerol (2-AG). Both types of release were also inhibited by capsaicin. The inhibition produced by each of the cannabinoid compounds and capsaicin was unaffected by capsazepine or by the CB1-receptor antagonists AM-251 and SR141716A. The mechanism underlying AEA- and synthetic CB-induced inhibition of the release of [3H]GABA and [3H]D-ASP from rat hippocampal synaptosomes might not involve activation of presynaptic CB1 receptors.

Effect of chronic treatment with the GABA transaminase inhibitors gamma-vinyl GABA and ethanolamine O-sulphate on the in vitro GABA release from rat hippocampus

1. The effects of 2, 8 and 21 day oral treatment with the specific gamma-aminobutyric acid transaminase (GABA-T) inhibitors gamma-vinyl GABA (GVG) and ethanolamine O-sulphate (EOS) on brain GABA levels, GABA-T activity, and basal and stimulated GABA release from rat cross-chopped brain hippocampal slices was investigated. 2. Treatment with GABA-T inhibitors lead to a reduction in brain GABA-T activity by 65-80% compared with control values, with a concomitant increase in brain GABA content of 40-100%. 3. Basal hippocampal GABA release was increased to 250-450% of control levels following inhibition of GABA-T activity. No Ca2+ dependence was observed in either control or treated tissues. 4. GVG and EOS administration led to a significant elevation in the potassium stimulated release of GABA from cross-chopped hippocampal slices compared with that of controls. Although stimulated GABA release from control tissues was decreased in the presence of a low Ca2+ medium, GVG and EOS treatment abolished this Ca2+ dependency. 5. GABA compartmentalization, Na+ and Cl- coupled GABA uptake carriers and glial release may provide explanations for the loss of the Ca2+ dependency of stimulated GABA release observed following GVG and EOS treatment. 6. Administration of GABA-T inhibitors led to increases in both basal and stimulated hippocampal GABA release. However, it is not clear which is the most important factor in the anticonvulsant activity of these drugs, the increased GABA content 'leaking' out of neurones and glia leading to widespread inhibition, or the increase in stimulated GABA release which may occur following depolarization caused by an epileptic discharge.

Membrane lipid peroxidation induces changes in gamma-[3H]aminobutyric acid transport and calcium uptake by synaptosomes

In the present study, we analyze the effect of Fe2+/ascorbate-induced lipid peroxidation on Ca(2+)-dependent and Ca(2+)-independent release and on the uptake of gamma-[3H]aminobutyric acid (GABA) by sheep brain synaptosomes. In addition, we study the effect of lipid peroxidation on the levels of cytosolic calcium and on the uptake of calcium (45Ca2+). After membrane lipid peroxidation, a decrease in the uptake of GABA is observed. After ascorbate/Fe(2+)-induced membrane lipid peroxidation, a significant decrease in [3H]GABA release in response to K(+)-depolarization occurs, in the absence and in the presence of Ca2+. The influx of 45Ca2+ induced by K(+)-depolarization is significantly depressed under peroxidative conditions, while basal calcium uptake is inhibited to a much lesser degree. The levels of free ionic calcium [Ca2+]i, as determined by the fluorescent dye Indo-1, are increased after synaptosomes were submitted to the ascorbate/Fe2+ oxidative stress. It is concluded that membrane lipid peroxidation induces a decrease in Ca(2+)-dependent and Ca(2+)-independent efflux of accumulated [3H]GABA in response to elevated K+ pulses (60 mM) and in the depolarization-induced calcium influx, while free ionic calcium levels increase. The Ca(2+)-dependent efflux is interpreted to reflect stimulus-secretion coupling process and the Ca(2+)-independent efflux may reflect membrane transport processes. Thus, the results suggest a possible relationship between a reduced calcium movement across the membrane, the decrease in neurotransmitters uptake and release and oxidative stress.

Calcium-dependent and -independent release of glutamate from synaptosomes monitored by continuous fluorometry

An enzyme-linked fluorometric assay is described for the continuous monitoring of the unidirectional efflux of glutamate from guinea-pig synaptosomes. Glutamate efflux from freshly suspended, polarized synaptosomes occurs at 0.35-0.39 nmol min-1 mg of protein-1 and is not significantly affected by external Ca2+. KCl depolarization (30 mMKCl) in the absence of Ca2+ doubles this rate, whereas in the presence of Ca2+, the initial kinetics of the assay are consistent with the release in the first 5 s of 0.6 nmol mg of protein-1. The final extent of Ca2+-dependent release amounts to 1.9 nmol mg of protein-1, or 8.5% of the total intrasynaptosomal glutamate content. Preincubation of synaptosomes at 30 degrees C for 2 h before depolarization leads to a decrease in Ca2+-independent release and an increase in Ca2+-dependent release, consistent with an intrasynaptosomal relocation of the amino acid.

4-Aminobutyrate can be released exocytotically from guinea-pig cerebral cortical synaptosomes

Guinea-pig synaptosomes possess two functional pools of 4-aminobutyrate (GABA). One is rapidly labelled by added [14C]GABA, is steadily released in a Ca2+-independent manner when the Na+ electrochemical potential across the plasma membrane is collapsed, and is depleted by the GABA analogue 2,4-diaminobutyrate (DABA), all of which is consistent with a cytosolic location. A second, noncytosolic compartment only slowly equilibrates with exogenous [14C]GABA, is not depleted by DABA, but can release 350 pmol of endogenous GABA/mg of protein (8% of the total intrasynaptosomal GABA) within 15 s of depolarization in the presence of Ca2+. Ca2+-independent release occurs by thermodynamic reversal of the plasma membrane uptake pathway following artifactually prolonged depolarization, whereas Ca2+-dependent release is consistent with physiological exocytosis from vesicular stores.

Is the concentration of gamma-aminobutyric acid in the nerve terminal regulated via product inhibition of glutamic acid decarboxylase?

Fractions of synaptosomes were used to study the regulation of gamma-aminobutyric acid (GABA) synthesis. The isolated synaptosomes were superfused in media of various compositions. [3H]GABA and GABA released into the medium or remaining in the synaptosomes were analyzed by liquid scintillation and HPLC techniques. Different conditions, designed to increase the GABA efflux rate were used: the rate of superfusion was varied and the concentrations of K+ and Ca2+ were altered. Stimulation of GABA efflux was paralleled with an increased synthesis of GABA, since, in spite of the increased GABA efflux, a relatively constant intraterminal level was found. The findings suggest that the intraterminal concentration of GABA and thus also its synthesis is regulated via product inhibition. In addition, [3H]GABA, exogenous, and GABA, endogenous, responded to external stimulae (Ca2+, veretradine, various GABA concentrations and the glutaminase inhibitor diazo-nor-leucine) in a way which was compatible with them being localized in and/or released from different compartments.

Compartmentation and release of exogenous GABA in sheep brain synaptosomes

Exogenous tritiated gamma-aminobutiric acid ([3H]GABA) is retained in two compartments in sheep cortex synaptosomes, corresponding to cytoplasmic and vesicular spaces, assuming that freeze-thawing the synaptosomes loaded with [3H]GABA releases the cytoplasmic [3H]GABA (81 +/- 3.9%), and that subsequent solubilization of the synaptosomes with 1% sodium cholate releases the vesicular [3H]GABA (19 +/- 3.9%). Depolarization of synaptosomes with 40 mM K+ in a Na+-medium, in the absence of Ca2+, releases 20.3 +/- 2.7% of the [3H]GABA retained in the synaptosomes. The [3H]GABA released under these conditions comes predominantly from the cytoplasm. The presence of 1 mM Ca2+ during depolarization releases an additional 13% (a total of about 33.5 +/- 9.9%) of the releasable [3H]GABA, and the [3H]GABA release which is Ca2+-dependent also comes mostly from the cytoplasmic compartment. When choline replaces external Na+, the [3H]GABA release is absolutely Ca2+-dependent, and the [3H]GABA released also comes mostly from the cytoplasmic pool. Therefore, it appears that [3H]GABA taken up by synaptosomes is accumulated mostly in the cytoplasmic compartment from which it is released upon depolarization. The technique described permits distinguishing the effect of different factors on the two pools of accumulated [3H]GABA.

Synaptosomes possess an exocytotic pool of glutamate

There is increasing evidence that L-glutamate is a major excitatory neurotransmitter in the central nervous system. Immunocytochemical studies indicate that glutamate within nerve terminals may be concentrated in vesicles and glutamate-accumulating vesicles have recently been isolated. Exocytotic release of glutamate from synaptosomes (isolated nerve terminals) has not been convincingly demonstrated, however, and remains highly controversial. In order to study the kinetics of release of endogenous L-glutamate from guinea pig cerebral cortical synaptosomes we have devised a continuous enzymatic assay. This has enabled us to identify a pool, equivalent to 15-20% of the total synaptosomal glutamate, which is capable of rapid Ca2+-dependent exocytotic release.

A superfusion apparatus for the examination of neurotransmitter release from synaptosomes

A superfusion apparatus for examining the release of neurotransmitter from synaptosomes is described. The apparatus provides for accurate temperature control, rapid switching and continuous gassing of superfusing buffers, short (i.e. 1-s) pulses of high-potassium-containing buffer for examining depolarized release, rapid collection of superfusing eluate, and low tissue chamber dead volume. Depolarization (high-potassium) evoked release and to some degree spontaneous release of [3H]dopamine from rat striatal synaptosomes are dependent on buffer calcium concentration. The putative calcium channel blocker nickel reduces both spontaneous and high-potassium-evoked [3H]dopamine release.

Effects of inhibitors of Na+,K+-ATPase on the membrane potentials and neurotransmitter efflux in rat brain slices

The potassium potential EK, of rat brain slices was estimated by determining the uptake of 86Rb+. The ERb was the same for slices prepared from five rostral brain regions, the average value being 66.4 mV. The ERb values in the presence of 20 microM ouabain were only slightly lower than the resting values; increasing concentrations of ouabain above 20 microM resulted in a graded depolarization in all five brain regions. High concentrations (1 mM) of two other inhibitors of Na+,K+-ATPase, dihydro-ouabain and strophanthidin, produced no more depolarization than did 20 microM ouabain. Competitive binding studies indicated that the differential effects were due to the relative binding to brain slices. Erythrosin B, an inhibitor of Na+,K+-ATPase, had no measurable effect on ERb. Intermediate concentrations of the Na+/H+ ionophore monensin slightly hyperpolarized striatal slices, whereas the same monensin concentrations plus 20 microM ouabain, 1 mM strophanthidin or 70 microM erythrosin B resulted in marked depolarization. Measurement of the membrane potential via uptake of methyltriphenylphosphonium cation indicated that ERb was indeed a valid estimation of the membrane potential. EK was measured directly by monitoring 42K+ uptake in striatal slices and was found to be essentially identical to ERb. Uptake of 22Na+ was consistent with the values for ERb or EK. Several conditions that resulted in little or no measurable depolarization of striatal slices did induce efflux of exogenously loaded GABA and dopamine; these conditions included 20 microM ouabain, 1 mM dihydro-ouabain or strophanthidin, and 70 microM erythrosin B. Neurotransmitter efflux in the absence of general cell depolarization was not accompanied by altered rates of respiration or decreased ATP levels.

A kinetic analysis of the release of acidic amino acids from rat cortical synaptosomes following pre-loading with [14C]glutamic acid

The kinetics of the release of acidic amino acids have been studied in rat cortical synaptosomes. After pre-loading for 30 minutes in labeled glutamate, labeled glutamate and aspartate appear to be totally releasable. However, extra-synaptosomal Ca does not facilitate release. When elevated [K]0 was used to depolarize, release was unaffected by removal of Ca from the incubation medium. When veratridine was used as a depolarizing agent, presence of Ca in the incubation medium inhibited release. In all solutions, semi-log plots of synaptosomal label content as a function of time were non-linear, which is incompatible with release from a single compartment. Previous studies of the effect of membrane potential on transport led to the development of a carrier model which should participate in depolarization induced release (19). Under the conditions used in the present studies, this carrier should be saturated. When the data were fitted to a two compartment model, with release from compartment A linear with compartment size and release from compartment B via a saturated carrier, an excellent fit was obtained. Under control conditions, about 90% of the labeled amino acid is in compartment B and about 70% of the total release is from this compartment. Rate of release is greatly accelerated in depolarizing solutions. Under depolarizing conditions, there is a large shift of labeled amino acid from compartment B to compartment A and release from compartment A predominates. Analysis of the results under the several depolarizing conditions used shows that the present results are consistent with the predictions of the carrier model which has been developed from previous studies of the Na and membrane potential dependence of glutamate transport.

On the existence of two GABA pools associated with newly synthesized GABA and with newly taken up GABA in nerve terminals

[14C]Glutamic acid and [3H]GABA were injected into the lateral ventricle of mouse and then [14C]GABA and [3H]GABA in synaptosomes isolated from the animals were analysed. The [14C]GABA was interpreted to be newly synthesized GABA from [14C]glutamic acid while the [3H]GABA to be newly taken up GABA. We have obtained the following results: (1) when the animals were pretreated with aminooxyacetic acid and thus the GABA content in synaptosomes increased to about 2 times of the control level, only the [3H]GABA was enhanced to 3 times of the control level without any change of [14C]GABA, (2) the release of [14C]GABA from synaptosomes by high K+ depolarization was 1.5 times greater than that of [3H]GABA, (3) the releases of both [14C]GABA and [3H]GABA were increased in the presence of cold GABA, L-2,4-diaminobutyric acid or gamma-amino-beta-hydroxybutyric acid, but only slightly increased in the presence of beta-alanine. These results would suggest that newly synthesized GABA and newly taken up GABA localized individually in different pools, which might localize either in different nerve terminals or separately in the same nerve terminal.

Putative acidic amino acid transmitters in the cerebellum. I. Depolarization-induced release

In the present investigation we studied the autoradiographic localization and the characteristics of the depolarization-induced release of acidic amino acids in in vitro rat cerebellar preparations. Light microscopy autoradiography of cerebellar slices preincubated in the presence of the non-metabolized glutamate analogue D-[3H]aspartate showed a large accumulation of radioactivity over glial cells, and very little labelling of the granule cells, whose putative neurotransmitter may be glutamate. In spite of its predominant localization in glia, D-[3H]aspartate (and [14C]glutamate) was released from cerebellar slices depolarized with high [K+] in a Ca2+-dependent way, and the release elicited by veratrine was prevented by TTX. These findings, together with the observation that freshly isolated or cultured glial cells did not show any Ca2+-dependent, depolarization-induced release of D-[3H]aspartate, suggest that the radioactive amino acid released from slices has a neuronal origin. The high [K+]-induced release of exogenous radioactive acidic amino acids from superfused cerebellar synaptosomal preparations exhibited, as best, a modest Ca2+-dependence, a result probably due to the existence of a substantial non-Ca2+-dependent release of the amino acid from glial fragments contaminating the preparation. However, both the K+-evoked release of endogenous glutamate, and that of [14C]glutamate previously synthesized from [14C]glutamine were largely Ca2+-dependent, suggesting that nerve endings are the main sites involved in the stimulus-coupled secretion. In the experiments in which synaptosomes had been prelabelled with [14C]glutamine, a study of the specific radioactivity of the glutamate released and of that present in synaptosomes at the beginning and at the end of superfusion period provided evidence in favour of a preferential release of the newly synthesized [14C]glutamate. In contrast to glutamate, endogenous aspartate was not released in a Ca2+-dependent manner, and the efflux of newly formed [14C]aspartate was only slightly potentiated by Ca2+, which suggests that glutamate and aspartate are not released from the same sites. Studies on preparations (slices and synaptosomes) from immature, 8-day-old cerebella showed that neither the K+-evoked release of D-[3H]aspartate, nor that of endogenous glutamate was Ca2+-dependent. In conclusion, the data presented are consistent with the proposition that glutamate has a neurotransmitter role in the cerebellum.U