Influence of Sodium-Independent γ-Aminobutyric Acid Binding on the Assay of Sodium-Dependent γ-Aminobutyric Acid Binding in a Membrane Preparation of Rat Brain

A large amount of [³H]GABA was bound to crude synaptic membrane fractions of rat. by sodium-independent process in a medium that contained 100 μM [³H]GABA used for assaying GABA uptake site. This [³H]-GABA binding was different from receptor binding of GABA. It was confirmed that this sodium-independent [²H]GABA binding scarcely occurred in the presence of a physiological concentration of sodium chloride, and that sodium-independent GABA binding had a negligible influence on sodium-dependent GABA binding.     

STRUCTURE-ACTIVITY STUDIES ON THE INHIBITION OF GABA BINDING TO RAT BRAIN MEMBRANES BY MUSCIMOL AND RELATED COMPOUNDS

Abstract— A series of compounds structurally related to muscimol (5-aminomethyl-3-isoxazolol) was tested as inhibitors of the sodium-independent binding of GABA to membranes from rat brain. Muscimol, 5-(l-aminoethyl)-3-isoxazolol, 5-(2-aminoethyl)-3-isoxazolol (homomuscimol), and the bicyclic derivative 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol (THIP) were relatively potent inhibitors of GABA binding. THIP is an analogue of muscimol locked in a folded conformation. The structurally related compound 1,2,3,6-tetrahydropyridine-4-carboxylic acid (isoguvacine), a semirigid analogue of trans-4-aminocrotonic acid, was also a potent inhibitor of GABA binding. Apart from muscimol, these inhibitors of GABA binding did not influence the sodium-dependent,'high-affinity' uptake of GABA in rat brain slices, whereas the potent GABA uptake inhibitors guvacine and nipecotic acid did not influence GABA binding. The present results support previous findings that different conformational modes of GABA interact with GABA postsynaptic receptors and the neuronal GABA transport system in rat brain, and indicate that the 'active conformation' of GABA with respect to the receptors is partially folded and almost planar. Based on a comparison of the present results with previous in vivo studies the structural requirements for GABA-like activity in rat cerebral cortex and cat spinal cord seem to be somewhat different.     

Developmental changes in high-affinity uptake of GABA by cultured neurons

High-affinity uptake of [³H]-aminobutyric acid (GABA) was studied in cultures of neonatal rat cortical neurons grown on pre-formed monolayers of non-neuronal (glial) cells. Both the maximum rate (V _max) and, to a smaller extent, theK _m of [³H]GABA uptake increased with time. In addition, in parallel with these changes, 2,4-diaminobutyric acid and cis-3-aminocyclohexane-1-carboxylic acid (ACHC), compounds which are considered typical substrate/inhibitors of GABA uptake in neurons, became progressively stronger inhibitors of [³H]GABA uptake. Consequently, the present results may mean that the studies using uptake, of [³H]GABA, [³H]ACHC, or [³H]DABA as a specific marker for GABAergic neurons differentiating during the ontogenetic development of the central nervous system may have to be interpreted with caution.     

THE UPTAKE OF [3H]GABA BY SLICES OF RAT CEREBRAL CORTEX

—A rapid accumulation of [³H]GABA occurs in slices of rat cerebral cortex incubated at 25° or 37° in a medium containing [³H]GABA. Tissue medium ratios of almost 100:1 are attained after a 60 min incubation at 25°. At the same temperature no labelled metabolites of GABA were found in the tissue or the medium. The process responsible for [³H]GABA uptake has many of the properties of an active transport mechanism: it is temperature sensitive, requires the presence of sodium ions in the external medium, is inhibited by dinitrophenol and ouabain, and shows saturation kinetics. The estimated K_m value for GABA is 2·2 × 10^?5m , and V_max is 0·115 μmoles/min/g cortex. There is only negligible efflux of the accumulated [³H]GABA when cortical slices are exposed to a GABA-free medium. [³H]GABA uptake was not affected by the presence of large molar excesses of glycine, l -glutamic acid, l -aspartic acid, or β-aminobutyrate, but was inhibited in the presence of l -alanine, l -histidine, β-hydroxy-GABA and β-guanidinopropionate. It is suggested that the GABA uptake system may represent a possible mechanism for the inactivation of GABA or some related substance at inhibitory synapses in the cortex.     

EFFECTS OF GABA ANALOGUES OF RESTRICTED CONFORMATION ON GABA TRANSPORT IN ASTROCYTES AND BRAIN CORTEX SLICES AND ON GABA RECEPTOR BINDING

—The effects of a variety of acyclic or heterocyclic GABA analogues on GABA receptor binding and on high affinity transport of GABA in cultured astrocytes and mini-slices of brain cortex were studied. The receptor and transport sites were found to be stereospecific and they exhibited opposite stereoselectivity for (R)- and (S)-trans-4-amino-4-methylcrotonic acid and (R)- and (S)-β-proline. The most potent inhibitors of GABA binding were (RS)-4, 5-dihydromuscimol, muscimol, GABA, isoguvacine and isonipecotic acid with IC₅₀values of, respectively, 0.009, 0.006, 0.033, 0.037 and 0.33 μM. Under the present experimental conditions the following compounds inhibited preferentially the glial transport system: (3RS, 4SR)-4-hydroxynipecotic acid, guvacine, (RS)-N-methylnipecotic acid, (RS)-β-proline and β-alanine (IC₅₀ values 10, 25, 70, 320 and 1000 μM, respectively vs. 200, 100, 300, 1200 and >5000 for neuronal transport). On the other hand, (R)-trans-4-amino-4-methylcrotonic acid, (3RS, 4SR, 5SR)-4-hydroxy-5-methymipecotic acid and (RS)-3-hydroxy-5-aminovaleric acid preferentially inhibited neuronal transport as studied in mini-slices of brain cortex (IC₅₀ values 160, 300 and 430 μM, respectively vs. 500, > 5000 and 1400 μM for glial transport).     

Function of taurine transporter (Slc6a6/TauT) as a GABA transporting protein and its relevance to GABA transport in rat retinal capillary endothelial cells

The purpose of this study was to identify the uptake mechanism of γ-aminobutyric acid (GABA) via taurine transporter (Slc6a6/TauT) and its relationship with GABA transport at the inner BRB. Rat Slc6a6/TauT-transfected HeLa cells exhibited Na⁺-, Cl⁻-, and concentration-dependent [³H]GABA uptake with a K_m of 1.5 mM. Taurine, β-alanine, and GABA markedly inhibited Slc6a6/TauT-mediated uptake of [³H]GABA. The uptake of [³H]GABA by a conditionally immortalized rat retinal capillary endothelial cell line (TR-iBRB2) was Na⁺-, Cl⁻-, and concentration-dependent with a K_m of 2.0 mM. This process was more potently inhibited by substrates of Slc6a6/TauT, taurine and β-alanine, than those of GABA transporters, GABA and betaine. In the presence of taurine, there was competitive inhibition with a K_i of 74 μM. [³H]Taurine also exhibited competitive inhibition with a K_i of 1.8 mM in the presence of GABA. In conclusion, rat Slc6a6/TauT has the ability to use GABA as a substrate and Slc6a6/TauT-mediated GABA transport appears to be present at the inner BRB.     

Properties of [3H] ß-carboline-3-carboxylate ethyl ester binding to the benzodiazepine receptor

β-Carbolines are inhibitors of [³H] diazepam binding with the most potent inhibitor being β-carboline-3-carboxylate ethyl ester (β-CCE). In this report the binding of [³H] β-CCE to extensively washed rat forebrain membranes is characterized. [³H] ß-CCE binds with high affinity (K_D = 1.4 nM) to an apparently homogenous population of benzodiazepine receptor. The rank order of potency for inhibition of [³H] ß-CCE binding by different benzodiazepines is clonazepam > diazepam > chlordiazepoxide, which is similar to that observed for inhibition of [³H] diazepam binding. In marked contrast to [³H] diazepam, the binding of [³H] ß-CCE is not modulated by GABA since concentrations of GABA as high as 10^?3 M had no effect. [³H] ß-CCE is also less potent than [³H] diazepam in its interaction with the peripheral type kidney benzodiazepine receptor indicating that this ligand has a higher degree of specificity for the central brain type benzodiazepine receptor.     

Binding sites for [3H]GABA, [3H]flunitrazepam and [35S]TBPS in insect CNS

The CNS of the cockroach Periplaneta americana contains saturable, specific binding sites for [³H]GABA, [³H]flunitrazepam and [³⁵S]TBPS. The [³H]GABA binding site exhibits a pharmacological profile distinct from that reported for mammalian GABA_A and GABA_B receptors. The most potent inhibitors of [³H]GABA binding were GABA and muscimol, whereas isoguvacine, thiomuscimol and 3-aminopropane sulphonic acid were less effective. Bicuculline methiodide and baclofen were ineffective. Binding of [³⁵S]TBPS was partially inhibited by 1.0 × 10⁻⁶ M GABA, whilst binding of [³H]flunitrazepam was enhanced by 1.0 × 10⁻⁷ M GABA. The pharmacological profile of the [³H]flunitrazepam binding site showed some similarities with the peripheral benzodiazepine binding sites of vertebrates, with Ro-5-4864 being a far more effective inhibitor of binding than clonazepam. Thus a class of GABA receptors with pharmacological properties distinct from mammalian GABA receptor subtypes is present in insect CNS.     

RELEASE OF [3H]GABA FROM RAT CORTICAL SLICES: NEURONAL VS GLIAL ORIGIN 1

Abstract— The effect of L-2,4 diaminobutyric acid (DABA) and β-alanine on the K⁺ stimulated release of [³H]GABA was examined using a continuous superfusion system in which a carrier mediated exchange diffusion could be demonstrated between [³H]GABA in preloaded rat cortical slices and unlabeled DABA and β-alanine in the superfusion medium. These structurally related amino acids were chosen to investigate the source of releasable [³H]GABA because of evidence suggesting they may have differing affinities for the GABA carrier transport system that are specific for neurons and glia, DABA having a greater affinity for the neuronal GABA system and β-alanine for the glial. Five millimolars-DABA in the superfusion medium nearly abolished the K⁺ stimulated release of [³H]GABA whereas β-alanine had little effect. The results and conclusions are discussed in terms of a postulated carrier mediated exchange of unlabeled DABA with a specific neuronal pool of [³H]GABA interfering with the K⁺ stimulated release of the radiolabeled GABA. The results provide indirect evidence in favor of a neuronal pool as the source of releasable [³H]GABA in this system.     

The metabolism of gamma-aminobutyric acid (GABA) in the lobster nervous system--uptake of GABA in the nerve-muscle preparations

The lack of information on the mechanism of inactivation of the crustacean neuromuscular inhibitory transmitter compound prompted a study of the disposition of radioactive γ-aminobutyric acid (GABA) in lobster nerve-muscle preparations. A specific GABA transport system was found. Radioactive GABA was concentrated by the tissues to levels several times those in the medium, and net uptake could be demonstrated. The process was dependent on sodium ions in the medium; neither lithium nor choline could substitute for sodium. Incubations with increasing GABA concentrations indicated that uptake was a saturable mechanism with an apparent K_m of 5.8 × 10⁻⁵m . Of many compounds tested, only desmethylimipramine, chlopromazine (and several related compounds), and certain close structural analogues (guanidinoacetic acid, β-guani-dinopropionic acid and,β-hydroxy-GAB A) were effective inhibitors of uptake. The inhibition with all these compounds, however, was at high concentrations (5 × 10⁻⁴ to 10⁻³m ) which limited their usefulness for physiological studies. A separate uptake mechanism for glutamate was found in the lobster nerve-muscle preparations. This process was not described in detail, but certain properties are similar to those of the GABA transport system. The cellular location of the GABA uptake system remains unknown. By analogy with noradrenaline inactivation, however, it is postulated that uptake could serve to terminate the physiological actions of GABA by rapidly removing it from its sites of action in synaptic clefts.     

Ubiquitin-Directed Antibodies Inhibit Neuronal Transporters in Rat Brain Synaptosomes

Affinity-purified antibodies specific for ubiquitin were found to inhibit the sodium-dependent uptake of [3H]choline, gamma-[3H]aminobutyric acid [( 3H]GABA), [3H]glutamate, [3H]norepinephrine, [3H]aspartate, and [3H]serotonin in rat cerebral cortical synaptosomes at a low concentration (10 micrograms/ml). These antibodies (termed anti-Ub) had no effect on the sodium-independent uptake of these substances or their calcium-dependent efflux. Synaptosomal [3H]deoxyglucose uptake was not affected in normal Krebs Ringer buffer containing 10 mM glucose, but was inhibited in glucose-free medium. Other nonneuronal sodium-dependent transport processes were found to be unaffected by 10 micrograms/ml anti-Ub, suggesting that anti-Ub does not bind indiscriminantly to sodium-binding sites on sodium-dependent organic solute transporters. Finally, anti-Ub inhibited sodium-dependent [3H]GABA and [3H]glutamate uptake in plasma membrane ghosts, devoid of membrane potential, which were derived from rat cerebral cortical synaptosomes. These results suggest that neuronal transporters or sites proximal to them may be ubiquitinylated on the plasma membrane surface.     

A role for calcium/calmodulin kinase(s) in the regulation of GABA exocytosis

A possible role for protein kinases in the regulation of GABA exocytosis in nerve endings was investigated. The effect on the release of the radioactive neurotransmitter ([³H]GABA) from mouse brain synaptosomes of several protein kinase inhibitors was estimated after treatment with 37 mM K⁺ in the absence of external Na⁺, a condition under which [³H]GABA release is completely Ca²⁺ dependent. Among the inhibitors one group inhibit the kinases by binding to the catalytic site (i.e. staurosporine and H7) and others (TFP, sphingosine and W7) act on the regulatory site of protein kinases. The compounds of the second group, which are reported to inhibit calmodulin dependent events and the increase in cytosolic Ca²⁺ (Ca _i ) induced by high K⁺ depolarization, were the most efficient inhibitors of [³H]GABA release. The selective inhibitor of CaMPK II, KN-62, also markedly diminished [³H]GABA release as well as the increase in Ca _i induced by high K⁺. The kinase inhibitors from the first group that are unable to diminish the increase in Ca _i induced by high K⁺ were also less efficient inhibitors of [³H]GABA release even at high concentrations. The present results indicate that at the doses tested all the drugs inhibit to some extent the release of the Ca²⁺ dependent fraction of [³H]GABA perhaps by inhibiting a CaMPK II mediated phosphorylation step triggered by depolarization and facilitated by the elevation of Ca _i . In addition, the second group of antagonists and KN-62 inhibit the elevation of Ca _i to high K⁺ thus exhibiting a higher efficiency on [³H]GABA release than the first group of antagonists.     

[3H]muscimol binding in the rat retina

Crude membrane fractions were prepared from rat retinae and used to study the specific binding of [³H]muscimol, a potent GABA agonist. Specific [³H]muscimol binding was enhanced 2–3 fold by pretreatment of the membranes with 0.025% Triton X-100. Two muscimol binding sites were demonstrated with K_D values of 4.4 and 12.3 nM. GABA, muscimol, and 3-aminopropanesulfonic acid were the most potent inhibitors of specific [³H]muscimol binding with K_I values of 15, 10, and 50 nM, respectively. These data are consistent with binding to the synaptic GABA receptor.     

Uptake,exchange, and release of GABA by cerebellar glomeruli

Glomerular particles were isolated from the bovine cerebellar vermis and studied in vitro to further assess the possibility that -aminobutyric acid (GABA) is utilized as a neurotransmitter in this synaptic complex. Cerebellar glomeruli accumulated [³H]GABA at two different high affinity sites, with affinities (K _T) of 2.2×10^–6M and 3.0×10^–5M. These uptake sites could not be distinguished on the basis of their temperature sensitivities, sodium dependence, substrate specificities or responses to metabolic inhibitors. Although an exchange process contributed to the uptake measured in these experiments, a considerable amount of the [³H]GABA accumulated by glomerular particles was stored in an osmotically-sensitive, nonexchangeable pool. Glomerular particles preloaded with [³H]GABA exhibited a Ca²⁺-independent release of this amino acid in response to membrane depolarization. However, when preloaded glomerular particles were exposed to unlabeled GABA, which presumably displaced [³H]GABA from the exchangeable pool, a K^–-evoked and Ca²⁺-dependent release of the remaining [³H]GABA occurred. The observed net uptake, together with the depolarization-induced and Ca²⁺-dependent release, of [³H]GABA from glomerular particles supports the suggestion that functionally active GABAergic synapses are present in these structures.     

Pharmacology of Sodium-Dependent High-Affinity l-[3H]Glutamate Transport in Glial Cultures

Abstract: Pharmacological and molecular biological studies provide evidence for subtypes of sodium-dependent high-affinity glutamate (Glu) transport in the mammalian CNS. At least some of these transporters appear to be selectively expressed in different brain regions or by different cell types. In the present study, the properties of l -[³H]Glu transport were characterized using astrocyte-enriched cultures prepared from cerebellum and cortex. In both brain regions, the kinetic data for sodium-dependent transport were consistent with a single site with K_m values of 91 ± 17 µM in cortical glial cells and 66 ± 23 µM in cerebellar glial cells. The capacities were 6.1 ± 1.6 nmol/mg of protein/min in cortical glial cells and 8.4 ± 0.9 nmol/mg of protein/min in cerebellar glial cells. The potencies of ～40 excitatory amino acid analogues for inhibition of sodium-dependent transport into glial cells prepared from cortex and cerebellum were examined, including compounds that are selective inhibitors of transport in synaptosomes prepared from either cerebellum or cortex. Of the analogues tested, 14 inhibited transport activity by >50% at 1 mM concentrations. Unlike l -[³H]Glu transport in synaptosomes prepared from cerebellum or cortex, there were no large differences between the potencies of compounds for inhibition of transport measured in glial cells prepared from these two brain regions. With the exception of (2S,1′R,2′R)-2-(carboxycyclopropyl)glycine and l -α-aminoadipate, all of the compounds examined were ～10–200-fold less potent as inhibitors of l -[³H]Glu transport measured in glial cells than as inhibitors of transport measured in synaptosomes prepared from their respective brain regions. The pharmacology of transport measured in these glial cells differs from the reported pharmacology of the cloned Glu transporters, suggesting the existence of additional uncloned Glu transporters or Glu transporter subunits.     

The Binding of the GAB A Agonist [3H]THIP to Rat Brain Synaptic Membranes

Abstract: THIP (4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol) is a specific GABA agonist with potent analgesic properties. The binding of radioactive THIP to thoroughly washed, frozen, and thawed membranes isolated from rat brains has been studied at 2°C under sodium ion-free conditions and compared with the binding of [³H]GABA and [³H]piperidine-4-sulphonic acid ([³H]P4S). The best computer fits to the experimental data were in all cases attained with a receptor model based on three independent binding sites, of which only the high- and medium-affinity sites could be characterised satisfactorily. While the K_D values were found to be comparable for all three ligands employed, the density of the high-affinity binding site (B_M1) was, with the exception of the membranes from the cerebellum, considerably lower for [³H]THIP than for [³H]GABA and [³ H]P4S. The regional distribution of the GABA receptors, which bind [³H]THIP, was different from those recognizing [³H]GABA and [³H]P4S. A number of analogues, including asymmetric compounds with known configuration, were tested as inhibitors of the binding of [³H]GABA, [³H]muscimol, [³H]THIP, [³H]isoguvacine, and [³H]P4S. The concentrations of the asymmetric compounds required for the inhibition of [³H]P4S binding were much higher than those required for the displacement of [³H]GABA, [³H]muscimol, [³H]THIP, and [³H]isoguvacine. The comparable relative potencies of inhibitors do, however, indicate that all of the ligands bind to the GABA receptors.     

Substrate Specificity of [3H] Muscimol Binding to a Particulate Fraction of a Neuron-enriched Culture of Embryonic Rat Brain

Abstract: The effects of some GABA analogues and some drugs on the binding of [³H]muscimol (3.08 nM) to thoroughly washed subcellular particles prepared from a neuron-enriched culture of embryonic rat brain were examined using Na+-free Tris-citrate medium and a centrifugation method. Competition for [³H]muscimol binding sites by excess(10^?5 M) unlabelled GABA provided estimates of “specific” binding. In accord with in vivo neuropharmacological studies on GABA receptors and with in vitro studies on cerebral membrane preparations, [³H]muscimol binding was potently inhibited by muscimol itself (IC₅₀, 2.5 nM), GABA (1C₅₀, 43 nM), isoguvacine (IC₅₀, 61 nM), and 3-aminopropanesulphonic acid (IC₅₀, 160 nM), and less potently inhibited by the GABA antagonist bicuculline methobromide (IC₅₀, 800 nM). δ- Aminovaleric acid (IC₅₀, 2.6 μM), the glycinelp-alanine antagonist strychnine (IC₅₀, 6.6 μM), and the predominantly glial GABA uptake inhibitors β-alanine (IC₅₀, 23 μM) and p-proline (IC₅₀, 66 μM) also inhibited [³H]muscimol binding. Other inhibitors of Na+-dependent GABA uptake, (±)-nipecotic acid, L- 2,4-diaminobutyric acid, and guvacine, as well as picrotoxinin, were relatively inactive as inhibitors of [³H]muscimol binding (IC₅₀≥ 1 mM). In addition to revealing that GABA receptors are present on neuronal membranes before the formation of most synapses, this binding of [³H]muscimol that occurs to neuronal, but not to glial, membranes might be useful as a “neuronal marker” and for the further characterization and isolation of GABA receptors.     

Identification and characterization of GABA, proline and quaternary ammonium compound transporters from Arabidopsis thaliana

Arabidopsis thaliana grows efficiently on GABA as the sole nitrogen source, thereby providing evidence for the existence of GABA transporters in plants. Heterologous complementation of a GABA uptake-deficient yeast mutant identified two previously known plant amino acid transporters, AAP3 and ProT2, as GABA transporters with Michaelis constants of 12.9±1.7 and 1.7±0.3 mM at pH 4, respectively. The simultaneous transport of [1-¹⁴C]GABA and [2,3-³H]proline by ProT2 as a function of pH, provided evidence that the zwitterionic state of GABA is an important parameter in substrate recognition. ProT2-mediated [1-¹⁴C]GABA transport was inhibited by proline and quaternary ammonium compounds.     

Na(+)- and Cl(-)-dependent [3H]GABA binding to catfish brain particles

The uptake of [3H]GABA by homogenates of catfish brain was previously shown to be temperature-sensitive and sodium-dependent, and to display saturation kinetics. The present study is a continuation of this work and was undertaken to characterize the initial binding of [3H]GABA to its transport system. [3H]GABA binding to catfish brain particles at 4 degrees C displayed saturability and was totally dependent on both Na+ and Cl-, the optimum concentrations of which were 150 mM and 75 mM, respectively. The effects of a number of drugs on binding were established. Unlabelled GABA was the most potent inhibitor (IC50 = 3.2 microM). The structural analogues nipecotic acid and guvacine were also strongly inhibitory. Interestingly, verapamil, a Ca2+ channel blocker, also inhibited [3H]GABA binding (IC50 = 38 microM). Harmaline, known to compete for Na+ binding in other transport systems, did not appear to influence Na+ binding but was effective at displacing [3H]GABA. These results suggest that the interaction of GABA with its carrier is similar to that found in the mammalian nervous system and is further evidence that GABA is involved in neurotransmission in catfish brain.     

Study of the Mechanism of Release of [3H]GABA from a Teleost Retina In Vitro

Abstract: γ-Aminobutyric acid (GABA) is thought to be a neurotransmitter in the vetebrate retina. We studied the voltage and Ca²⁺ dependency of the process of release of [³H]GABA from the retina of the teleost Eugenes plumieri, using a microsuperfusion technique. Two depolarizing agents, veratridine and high potassium, produced a concentration-dependent release of [³H]GABA. The veratridine effect was inhibited in Na⁺-free solution, but was not affected by 1 μM tetrodotoxin. A substantial inhibition (about 75%) of the veratridine-and potassium-stimulated release of [³H] GABA occurred in Ca²⁺-free medium. Inhibitors of the Ca²⁺ channel, such as Mg²⁺(20 mM), La³⁺ (0.1 mM), and methoxy-verapamil (4 μM-0.4 mM), inhibited the veratridine-and K⁺-stimulated release. However, Co²⁺ and Cd²⁺ caused a potentiation and no change of the K⁺-and veratridine-stimulated release, respectively. This release process is apparently specific, since both depolarizing agents were unable to release [³H]methionine, a nontransmitter amino acid, under the same experimental conditions. Autoradio-graphic studies with [³H]GABA, using the same incubation conditions as for the release experiments, showed a high density of silver grains over the horizontal cells with almost no accumulation by amacrine cells and Muller cells. β-Alanine and nipecotic acid were used as two relative specific inhibitors of the glial and neuronal GABA uptake mechanisms, respectively. Only a small heteroexchange with [³H]GABA was found with β-alanine, and no inhibition of the subsequent veratridine-stimulated release. On the other hand, nipecotic acid produced a strong heteroexchange with [³H]GABA and lacked the capacity to induce the veratridine-stimulated release of [³H]GABA. These results suggest a voltage-and Ca²⁺-dependent neuronal release of [³H]GABA from retina.