Pharmacologic and therapeutic aspects of the developmentally regulated expression of GABA(A) and GABA(B) receptors: cerebellar granule cells as a model system

The effect of thiocolchicoside on cerebral vasospasm following experimental subarachnoid hemorrhage in the rabbit

BACKGROUND

This study investigated the effects of thiocolchicoside to prevent cerebral vasospasm in a rabbit model of subarachnoid hemorrhage.

METHODS

Twenty-four adult male New Zealand white rabbits were randomly divided into three groups of eight rabbits each: group 1 (control), group 2 (subarachnoid hemorrhage), group 3 (treatment). Thiocolchicoside (4 mg/kg, intraperitoneally) was administered just before intracisternal blood injection and continued for 72 h once a day in the same dose for group 3. Animals were killed 72 h after subarachnoid hemorrhage. Basilar artery cross-sectional areas and arterial wall thicknesses were measured in all groups.

RESULTS

Intraperitoneal administration of thiocolchicoside was found to attenuate cerebral vasospasm after subarachnoid hemorrhage in rabbits. Thiocolchicoside treatment was determined to be effective in increasing the luminal area and reducing the wall thickness of the basilar artery.

CONCLUSIONS

Our findings, for the first time, showed that TCC can prevent vasospasm induced by SAH. Our results also showed that GABAergic activity may play an important role in cerebral vasospasm etiopathogenesis. In conclusion, the thiocolchicoside treatment might be beneficial in preventing vasospasm after subarachnoid hemorrhage, thus showing potential for clinical application.

GABAB receptors in the NTS mediate the inhibitory effect of trigeminal nociceptive inputs on parasympathetic reflex vasodilation in the rat masseter muscle

The present study was designed to examine whether trigeminal nociceptive inputs are involved in the modulation of parasympathetic reflex vasodilation in the jaw muscles. This was accomplished by investigating the effects of noxious stimulation to the orofacial area with capsaicin, and by microinjecting GABAA and GABAB receptor agonists or antagonists into the nucleus of the solitary tract (NTS), on masseter hemodynamics in urethane-anesthetized rats. Electrical stimulation of the central cut end of the cervical vagus nerve (cVN) in sympathectomized animals bilaterally increased blood flow in the masseter muscle (MBF). Increases in MBF evoked by cVN stimulation were markedly reduced following injection of capsaicin into the anterior tongue in the distribution of the lingual nerve or lower lip, but not when injected into the skin of the dorsum of the foot. Intravenous administration of either phentolamine or propranolol had no effect on the inhibitory effects of capsaicin injection on the increases of MBF evoked by cVN stimulation, which were largely abolished by microinjecting the GABAB receptor agonist baclofen into the NTS. Microinjection of the GABAB receptor antagonist CGP-35348 into the NTS markedly attenuated the capsaicin-induced inhibition of MBF increase evoked by cVN stimulation, while microinjection of the GABAA receptor antagonist bicuculline did not. Our results indicate that trigeminal nociceptive inputs inhibit vagal-parasympathetic reflex vasodilation in the masseter muscle and suggest that the activation of GABAB rather than GABAA receptors underlies the observed inhibition in the NTS.

GABA(A) Receptors: Post-Synaptic Co-Localization and Cross-Talk with Other Receptors

γ-Aminobutyric acid type A receptors (GABA(A)Rs) are the major inhibitory neurotransmitter receptors in the central nervous system, and importantly contribute to the functional regulation of the nervous system. Several studies in the last few decades have convincingly shown that GABA can be co-localized with other neurotransmitters in the same synapse, and can be co-released with these neurotransmitters either from the same vesicles or from different vesicle pools. The co-released transmitters may act on post-synaptically co-localized receptors resulting in a simultaneous activation of both receptors. Most of the studies investigating such co-activation observed a reduced efficacy of GABA for activating GABA(A)Rs and thus, a reduced inhibition of the post-synaptic neuron. Similarly, in several cases activation of GABA(A)Rs has been reported to suppress the response of the associated receptors. Such a receptor cross-talk is either mediated via a direct coupling between the two receptors or via the activation of intracellular signaling pathways and is used for fine tuning of inhibition in the nervous system. Recently, it was demonstrated that a direct interaction of different receptors might already occur in intracellular compartments and might also be used to specifically target the receptors to the cell membrane. In this article, we provide an overview on such cross-talks between GABA(A)Rs and several other neurotransmitter receptors and briefly discuss their possible physiological and clinical importance.

Low molecular weight thiols reduce thimerosal neurotoxicity in vitro: modulation by proteins

Thimerosal (TH), an ethylmercury complex of thiosalicylic acid has been used as preservative in vaccines. In vitro neurotoxicity of TH at high nM concentrations has been reported. Although a number of toxicological experiments demonstrated high affinity of mercury to thiol groups of the extracellular amino acids and proteins that may decrease concentration of free TH in the organism, less is known about the role of interactions between proteins and amino acids in protection against TH neurotoxicity. In the present study we examined whether the presence of serum proteins and of l-cysteine (Cys), d,l-homocysteine (Hcy), N-acetyl cysteine (NAC), l-methionine (Met) and glutathione (GSH) in the incubation medium affects the TH-induced changes in the viability, the intracellular levels of calcium and zinc and mitochondrial membrane potential in primary cultures of rat cerebellar granule cells. The cells were exposed to 500 nM TH for 48 h or to 15-25 μM TH for 10 min. Our results demonstrated a decrease in the cells viability evoked by TH, which could be prevented partially by serum proteins, albumin or in a dose-dependent manner by 60, 120 or 600 μM Cys, Hcy, NAC and GSH, but not by Met. This neuroprotection was less pronounced in the presence of proteins. Incubation of neurons with TH also induced the rise in the intracellular calcium and zinc concentration and decrease in mitochondrial membrane potential, and these effects were abolished by all the sulfur containing compounds studied and administered at 600 μM concentration, except Met. The loss of the ethylmercury moiety from TH as a result of interaction with thiols studied was monitored by (1)H NMR spectroscopy. This extracellular process may be responsible for the neuroprotection seen in the cerebellar cell cultures, but also provides a molecular pathway for redistribution of TH-derived toxic ethylmercury in the organism. In conclusion, these results confirmed that proteins and sulfur-containing amino acids applied separately reduce TH neurotoxicity, while their combination modulates in more complex way neuronal survival in the presence of TH.

Synthetic bastadins modify the activity of ryanodine receptors in cultured cerebellar granule cells

Although the interactions of several natural bastadins with the RyR1 isoform of the ryanodine receptor in sarcoplasmic reticulum has been described, their structure-dependent interference with the RyR2 isoform, mainly expressed in cardiac muscle and brain neurons, has not been studied. In this work, we examined calcium transients induced by natural bastadin 10 and several synthetic bastadins in cultured cerebellar granule cells known to contain RyR2. The fluorescent calcium indicator fluo-3 and confocal microscopy were used to evaluate changes in the intracellular Ca(2+) concentration (Ca(i)), and the involvement of ryanodine receptors was assessed using pharmacological tools. Our results demonstrate that apart from the inactive BAST218F6 (a bisdebromo analogue of bastadin 10), synthetic bastadin 5, and synthetic analogues BAST217B, BAST240 and BAST268 (at concentrations >20 microM) increased Ca(i) in a concentration-dependent, ryanodine- and FK-506-sensitive way, with a potency significantly exceeding that of 20 mM caffeine. Moreover, the same active bastadins at a concentration of 5 muM in the presence of ryanodine prevented a thapsigargin-induced increase in Ca(i). These results indicate that bastadins, acting in a structure-dependent manner, modify the activity of RyR2 in primary neuronal culture and provide new information about structure-related pharmacological properties of bastadins.

GABAergic signaling in the developing cerebellum

In the adult central nervous system (CNS), GABA is a predominant inhibitory neurotransmitter that regulates glutamatergic activity. Recent studies have revealed that GABA serves as an excitatory transmitter in the immature CNS and acts as a trophic factor for brain development. Furthermore, synaptic transmission by GABA is also involved in the expression of higher brain functions, such as memory, learning and anxiety. These results indicate that GABA plays various roles in the expression of brain functions and GABAergic roles change developmentally in accordance with alterations in GABAergic transmission and signaling. We have investigated morphologically the developmental changes in the GABAergic transmission system and the key factors important for the formation of GABAergic synapses and networks using the mouse cerebellum, which provides an ideal system for the investigation of brain development. Here, we focus on GABA and GABA(A) receptors in the developing cerebellum and address the processes of how GABA exerts its effect on developing neurons and the mechanisms underlying the formation of functional GABAergic synapses.

Lipid raft localization of GABAA receptor and Na+, K+-ATPase in discrete microdomain clusters in rat cerebellar granule cells

The microdomain localization of the GABA(A) receptor in rat cerebellar granule cells was studied by subcellular fractionation and fluorescence- and immunogold electron microscopy. The receptor resided in lipid rafts, prepared at 37 degrees C by extraction with the nonionic detergent Brij 98, but the raft fraction, defined by the marker ganglioside GM(1) in the floating fractions following density gradient centrifugation, was heterogeneous in density and protein composition. Thus, another major raft-associated membrane protein, the Na(+), K(+)-ATPase, was found in discrete rafts of lower density, reflecting clustering of the two proteins in separate membrane microdomains. Both proteins were observed in patchy "hot spots" at the cell surface as well as in isolated lipid rafts. Their insolubility in Brij 98 was only marginally affected by methyl-beta-cyclodextrin. In contrast, both the GABA(A) receptor and Na(+), K(+)-ATPase were largely soluble in ice cold Triton X-100. This indicates that Brij 98 extraction defines an unusual type of cholesterol-independent lipid rafts that harbour membrane proteins also associated with underlying scaffolding/cytoskeletal proteins such as gephyrin (GABA(A) receptor) and ankyrin G (Na(+), K(+)-ATPase). By providing an ordered membrane microenvironment, lipid rafts may contribute to the clustering of the GABA(A) receptor and the Na(+), K(+)-ATPase at distinct functional locations on the cell surface.

Effect of androgens on sexual differentiation of pituitary gamma-aminobutyric acid receptor subunit GABA(B) expression

Previous work demonstrated a sexually dimorphic ontogenic expression of gamma-aminobutyric acid receptors (GABA(B)R) in rat pituitary. As sex steroids determine sex-specific expression patterns, we now studied the effect of sex hormones on pituitary GABA(B)R expression. GABA(B)R subunits, measured by Western blot and by semi-quantitative RT-PCR and luteinizing hormone (LH), follicle-stimulating hormone (FSH) and testosterone measured by RIA were determined in two experimental designs: First experimental design: 8- and 15-day-old females (8F, 15F); 8F and 15F treated with 100 mug testosterone propionate (TP) on day 1 of life (8F100TP, 15F100TP), 8- and 15-day-old males (8M, 15M) and 8M and 15M castrated on day 1 (8MC, 15MC). Second experimental design: 8-day-old female and male animals: 8F, 8F100TP, 8F treated with 1 mug/day TP on days 1-4 (8F1TP), 8F treated with the androgen antagonist Flutamide (Flut: 2.5 mg/100 g BW of pregnant mother on days E17-E23) (8F-Flut), 8M, 8MC, 8M treated with Flut as above (8M-Flut) and 8MC-Flut. In these animals, in addition, GABA, glutamate, aspartate and taurine were measured by HPLC in hypothalami and cortex. In the first set of experiments, GABA(B1)R mRNA/protein expression was higher in 8F than in 15F, 8M or 15M. In 8F100TP, GABA(B1)R mRNA/protein decreased to male levels. TP treatment did not alter GABA(B1)R expression in 15F. There was no difference in GABA(B1)R expression between 8M and 15M and neonatal castration did not modify its expression. In the second set of experiments, TP (1 mug) or Flut did not modify GABA(B1)R in 8F, while 100 microg TP continued to decrease GABA(B1)R expression. In 8M, Flut, alone or with castration, increased GABA(B1)R mRNA/protein expression to 8F. Hypothalamic GABA content followed the same pattern as pituitary GABA(B)R expression in 8-day-old animals, suggesting a cross-regulation. With regard to hormonal levels, 100 microg, but not 1 microg TP altered gonadotropins at 8 days, although both treatments effectively androgenized females as evidenced by lack of cycling. We conclude that androgens, acting pre- and postnatally, decrease pituitary GABA(B)R subunit expression.

Pathogenesis of peroxisomal deficiency disorders (Zellweger syndrome) may be mediated by misregulation of the GABAergic system via the diazepam binding inhibitor

Background

Zellweger syndrome (ZS) is a fatal inherited disease caused by peroxisome biogenesis deficiency. Patients are characterized by multiple disturbances of lipid metabolism, profound hypotonia and neonatal seizures, and distinct craniofacial malformations. Median live expectancy of ZS patients is less than one year. While the molecular basis of peroxisome biogenesis and metabolism is known in considerable detail, it is unclear how peroxisome deficiency leads to the most severe neurological symptoms. Recent analysis of ZS mouse models has all but invalidated previous hypotheses.

Hypothesis

We suggest that a regulatory rather than a metabolic defect is responsible for the drastic impairment of brain function in ZS patients.

Testing the hypothesis

Using microarray analysis we identify diazepam binding inhibitor/acyl-CoA binding protein (DBI) as a candidate protein that might be involved in the pathogenic mechanism of ZS. DBI has a dual role as a neuropeptide antagonist of GABA(A) receptor signaling in the brain and as a regulator of lipid metabolism. Repression of DBI in ZS patients could result in an overactivation of GABAergic signaling, thus eventually leading to the characteristic hypotonia and seizures. The most important argument for a misregulation of GABA(A) in ZS is, however, provided by the striking similarity between ZS and "benzodiazepine embryofetopathy", a malformation syndrome observed after the abuse of GABA(A) agonists during pregnancy.

Implications of the hypothesis

We present a tentative mechanistic model of the effect of DBI misregulation on neuronal function that could explain some of the aspects of the pathology of Zellweger syndrome.

Hypertension alters GABA receptor-mediated inhibition of neurons in the nucleus of the solitary tract

Previous studies have demonstrated that microinjection of baclofen, a GABA(B) receptor agonist, into the nucleus of the solitary tract (NTS) results in an enhanced pressor response in hypertensive (HT) rats compared with normotensive (NT) rats, suggesting a possible alteration in the responses of neurons in this area to activation of GABA(B) receptors. The following studies were designed to determine whether HT alters the sensitivity of neurons in the NTS to GABA receptor agonists. Sham-operated NT and unilateral nephrectomized, renal-wrap HT Sprague-Dawley rats were anesthetized, and the responses of NTS neurons receiving aortic nerve (AN) afferent inputs to iontophoretic application of GABA, the GABA(A) receptor agonist muscimol, and the GABA(B) agonist baclofen were examined. The AN input was classified as monosynaptic (MSN) if the cell responded to each of two stimuli separated by 5 ms with an action potential. If the cell did not respond, the input was considered polysynaptic (PSN). In MSNs, inhibition of AN-evoked discharge by GABA was not altered in 1 wk of HT but was reduced in 4 wk of HT, whereas in PSNs, sensitivity to GABA was reduced at 1 and 4 wk of HT. In HT rats, inhibition of AN-evoked discharge by baclofen was enhanced in MSNs, but not in PSNs, after 1 and 4 wk of HT, whereas inhibition by muscimol was reduced in MSNs and PSNs at 1 and 4 wk of HT. Changes in sensitivity to muscimol and baclofen within MSNs were the same whether the MSN received a slowly or a rapidly conducted AN afferent input. The results demonstrate that early in HT the sensitivity of NTS neurons to inhibitory amino acids is altered and that these changes are maintained for > or =4 wk. The alterations are dependent on the subtype of GABA receptor being activated and whether the neuron receives a mono- or polysynaptic baroreceptor afferent input.

Normal formation of the postsynaptic elements of GABAergic synapses in the reeler cerebellum

Synaptic transmission mediated by gamma-amino butyric acid (GABA) plays an important role in inhibition of glutamatergic excitatory transmission and expression of higher brain functions, such as memory, learning and anxiety. To elucidate mechanisms underlying formation of the postsynaptic elements for GABAergic transmission, we employed the reeler mutant mice in this study. In the reeler cerebellum, abnormal cytoarchitecture and an aberrant environment affect the formation of neural networks and maturation of neurons. We examined the expression and localization of GABA(A) receptor alpha subunits in the reeler cerebellum and determined whether various abnormalities in the reeler mice affected formation of the postsynaptic elements. In situ hybridization analysis revealed that the specific expression of alpha subunit mRNAs in each neuronal type was preserved. Abnormal expression of alpha subunits was not detected, although GABAergic networks were altered and neuronal maturation was severely disturbed. Immunohistochemistry for the alpha1 and alpha6 subunits, which were expressed abundantly in the reeler cerebellum, revealed that both subunit proteins accumulated at positions adjacent to GABAergic terminals. These results, taken together, suggested that expression of the GABA(A) receptor subunits in postsynaptic neurons might be genetically determined, but trafficking and accumulation of the subunit proteins at the GABAergic synapse may be induced by GABAergic innervation.

Loss of GABAergic neuronal phenotype in primary cerebellar cultures following blockade of glutamate reuptake

Prolonged inhibition of glutamate reuptake by L-trans-pyrrolidine-2,4-dicarboxylate (PDC), a specific glutamate transporter blocker, reduced the number of GABA positive neurons in a primary cerebellar culture by 54%. The disappearance of immunostaining for GABA was gradual and was partially prevented by the N-methyl-D-aspartate (NMDA) receptor blocker, MK-801, and the alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA) receptor antagonist, NBQX. Combined blockade of NMDA and AMPA receptors restored the original proportion of GABAergic neurons observed in control cultures. Following the PDC exposure, expression of other GABAergic markers, such as glutamic acid decarboxylase (GAD) and vesicular GABA transporter (VGAT) was also dramatically decreased in an AMPA receptor-dependent manner. Loss of GABA or GAD immunostaining is commonly regarded as a sign of degeneration of GABAergic neurons. However, none of the GABAergic neurons were positive for propidium iodide uptake or showed abnormal nuclear morphology. Based on the above data we conclude that prolonged activation of ionotropic glutamate receptors by endogenously released glutamate was not toxic to cerebellar GABAergic neurons, but lead to the loss of their characteristic neurotransmitter phenotype.

Amobarbital inhibits K(+)-stimulated glucose oxidation in cerebellar granule neurons by two mechanisms

The study aimed at determining mechanism(s) by which amobarbital (amytal) suppresses glucose oxidation in cerebellar granule neurons in primary cultures, a glutamatergic preparation. When challenged with a depolarizing K(+) concentration (55 mM), the cells doubled their rate of glucose oxidation (production of 14CO(2) from U-[14C]glucose) and glycolysis (lactate accumulation). At normal K(+) concentration, amobarbital reduced 14CO(2) production with half-maximum effect at 0.5-1 mM; at 55 mM K(+), the inhibition was more potent, with more than half of the K(+)-induced stimulation abolished at 50 microM. Dixon plot analysis showed a single inhibitory mechanism at 5.4 mM K(+), but at 55 mM K(+), two kinetically different mechanisms could be distinguished. A more pronounced compensatory amobarbital-induced increase in glycolysis at 5.4 than at 55 mM K(+) suggested that amobarbital in addition to its inhibition of mitochondrial respiration inhibited K(+)-induced increase in energy demand, probably by its known suppression of stimulated glutamate release.

Neurotoxicity of lindane and picrotoxin: neurochemical and electrophysiological correlates in the rat hippocampus in vivo

In the present study, we compared in vivo changes of extracellular amino acid levels and nucleotide derivatives to a single ip dose of lindane (10-60 mg/kg) and picrotoxin (5 mg/kg) in the hippocampus of halothane anaesthetized rat by microdialysis-coupled HPLC analysis. Brain activity was monitored by EEG. The effects of lindane and picrotoxin on EEG pattern of rats as well as on hippocampal amino acid and nucleotide status were studied in 0-50 min, 50-100 min and 100-150 min periods post-dosing. Significant decreases in Glu and Asp were found after picrotoxin treatment. After 50-100 min post-dosing, hippocampal hypoxanthine and inosine levels increased to both lindane (10 mg/kg) and picrotoxin whereas xanthine and uridine levels increased to picrotoxin, only. Lindane elicited a dose-dependent occurrence of negative spikes accompanied with rhythmic activity at 4-5 Hz. The picrotoxin-induced 4-5 Hz activity did not display negative sharp waves and was accompanied by 10 Hz oscillations.

The role of amino acid neurotransmitters in the regulation of pituitary gonadotropin release in fish

Both glutamate and γ-aminobutyric acid (GABA) are involved in pituitary hormone release in fish. Glutamate serves 2 purposes, both as a neurotransmitter and as a precursor for GABA synthesis. Glutamate can be catabolized to GABA by the actions of 2 distinct but related enzymes, glutamate decarboxylase 65 (GAD65) and GAD67. They derive from 2 different genes that likely arose from an early gene duplication prior to the emergence of teleosts more than 400 million years ago. There is good evidence for the involvement of GABA in luteinizing hormone (LH) release in fish. The mechanism of GABA action to stimulate LH release appears to be a combination of effects on GnRH release, potentiation of gonadotropin hormone-releasing hormone (GnRH) action, and in some cases directly at the LH cell. These actions appear to be dependent on such factors as sex or sex steroid levels, and there may also be species differences. Nevertheless, the stimulatory effects of GABA on LH are present in at least 4 fish species. In contrast, convincing data for the inhibitory effects of GABA on LH release have only been observed in 1 fish species. The sites and mechanisms of action of amino acid neurotransmitters on LH release have yet to be fully characterized. Both N-methyl-D-aspartic acid (NMDA) and S-α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) type glutamate receptors are likely to have important roles. We suggest that it is a receptor similar to the GABAA type which mediates the effects of GABA on LH release in fish, at least partially acting on the GnRH neuron, but likely directly acting at the gonadotroph as well. GABA may also be involved in regulating the release of other pituitary hormones in fish, namely follicle stimulating hormone (FSH = GTH-I), prolactin, and growth hormone. Based on the findings described in this review, a working model for the involvement of glutamate and GABA in the regulation of LH release in teleost fish is proposed.

Key words: glutamate, GABA, luteinizing hormone, muscimol, patch clamp electrophysiology, reproduction, fish.

Recent advances in GABA research

In this article I throw attention on to this GABA issue by outlining several aspects of current interest in the field of GABA research. The theme was selected in association with the Pharmacology and Therapeutical Potential of the GABA System symposium of the Second European Congress of Pharmacology held in July 1999 in Budapest, Hungary. A wide range of topics relating to the GABA system were outlined, including new members of the GABAA receptor gene family, subunit composition of native GABA(A) receptors, surface expression and clustering of GABA(A) receptor subunits, allosteric modulation of GABA(A) receptors, localization of agonist binding sites, GABA release, GABA(A)-GABA(B) receptor crosstalk, GABA(A) and GABA(B) receptor functions in different brain areas, altered transport and GABA(A) receptor pattern in different models of epilepsy.

Separation of carrier mediated and vesicular release of GABA from rat brain slices

In this study the temperature dependence of [3H]GABA release from brain slices evoked by electrical field stimulation and the Na+/K+ ATPase inhibitor ouabain was investigated. [3H]GABA has been taken up and released from hippocampal slices at rest and in response to electrical field stimulation (20 V, 10 Hz, 3 msec, 180 pulses) at 37 degrees C. When the bath temperature was cooled to 7 degrees C, during the sample collection period, the tissue uptake and the resting outflow of [3H]GABA were not significantly changed. In contrast, the stimulation-induced tritium outflow increased both in absolute amount (Bq/g) and in fractional release and the S2/S1 ratio was also higher at 7 degrees C. Perfusion of the slices with tetrodotoxin (TTX, 1 microM) inhibited stimulation-induced [3H]GABA efflux indicating that exocytotic release of vesicular origin is maintained under these conditions. 15 min perfusion with ouabain (10-20 microM) induced massive tritium release both in hippocampal and in striatal slices. However, the fraction of [3H]GABA outflow evoked by ouabain was much higher in the hippocampus than in the striatum. Sequential lowering the bath temperature from 37 degrees C to 17 degrees C completely abolished ouabain-induced [3H]GABA release in both brain regions, indicating that it is a temperature-dependent, carrier-mediated process. When the same experiments were repeated under Ca2+ free conditions, cooling the bath temperature to 17 degrees C, although substantially decreased the release but failed to completely abolish the tritium outflow evoked by ouabain, a significant part was maintained. Our results show that vesicular (field stimulation-evoked) and carrier-mediated (ouabain-induced) release of GABA is differentially affected by low temperature: while vesicular release is unaffected, carrier-mediated release is abolished at low bath temperature. Therefore, lowering the temperature offers a reliable tool to separate these two kinds of release and makes possible to study exclusively the pure neuronal release of GABA of vesicular origin.