Showing papers on "Kainate receptor published in 1985"

Glutamate stimulates inositol phosphate formation in striatal neurones

Abstract: The major excitatory amino acids, glutamate (Glu) and aspartate (Asp), are thought to act at three receptor subtypes in the mammalian central nervous system (CNS). These are termed quisqualate (QA), N-methyl-D-aspartate (NMDA) and kainate (KA) receptors according to the specific agonist properties of these compounds revealed by electrophysiological studies. Although Glu has been shown to stimulate cyclic GMP formation in brain slices, direct regulation of second messenger systems (cyclic AMP, Ca2+ or inositol phosphates) subsequent to activation of excitatory amino-acid receptors, has not been extensively studied. Here we demonstrate that in striatal neurones, excitatory amino acids, but not inhibitory or non-neuroactive amino acids, induce a three- to fourfold increase in inositol mono-, di- and triphosphate (IP, IP, IP) formation with the relative potency QA greater than Glu greater than NMDA, KA. The Glu-evoked formation of inositol phosphates appears to result principally from actions at QA as well as NMDA receptors on striatal neurones. Our results suggest that excitatory amino acids stimulate inositol phosphate formation directly, rather than indirectly by the evoked release and subsequent actions of adenosine or acetylcholine.

The neurotoxicity of excitatory amino acids is produced by passive chloride influx

Abstract: In the 15 years since the neurotoxic properties of glutamate and related amino acids were first described, there has been no thoroughly convincing explanation of the pathophysiology of excitatory amino acid-induced neuronal death. These substances depolarize central neurons, increase the frequency of neuronal discharge, and augment synaptic activity, leading to the suggestion that one or more of these properties may in some way be responsible for toxicity. More recently, an excessive calcium influx triggered by amino acids has been implicated in this process. As isolation of the different factors potentially involved in amino acid neurotoxicity is virtually impossible in vivo, dispersed hippocampal cultures were used to define the pathophysiology of this process in vitro. The toxicity of glutamate, N-methyl-D-aspartate, and kainate was unaffected when calcium was deleted and tetrodotoxin added to the balanced salt solution bathing the cultures. In parallel experiments, the calcium ionophore A23187 was not toxic in the presence of calcium. These experiments failed to confirm a role for neuronal activity or calcium influx in this process. However, when depolarization was blocked by deleting sodium from the control salt solution, neither glutamate, N-methyl-D-aspartate, nor kainate produced obvious changes. Alternately, when passive chloride influx was prevented by largely deleting chloride from the bath, the cells were also unchanged by the amino acids. Further experiments showed that depolarization produced by high external potassium concentrations or veratridine was also toxic, but only in the presence of external chloride. These experiments suggest that the pathophysiology of amino acid neurotoxicity may be rather straightforward.(ABSTRACT TRUNCATED AT 250 WORDS)

Quantitative studies on some antagonists of N-methyl D-aspartate in slices of rat cerebral cortex.

Abstract: Coronal sections of rat brain (500 micron thick) were trimmed to form 'wedges' of tissue consisting of cerebral cortex and corpus callosum. When these slices were placed in a two-compartment bath, the cortical tissue could be depolarized, relative to the corpus callosum, by superfusions of high K+, or by amino acids such as L-glutamate, L-aspartate, quisqualate, kainate and N-methyl D-aspartate (NMDA). Responses to NMDA were reduced by magnesium ions, by the organic antagonists (-)-2-amino 5-phosphonovalerate (APV) and 2-amino 7-phosphonoheptanoate (APH), and by the dissociative anaesthetic ketamine. In this preparation, all these antagonists shifted the NMDA dose-response curve to the right in a parallel manner. A Schild plot for Mg2+ had a slope significantly less than unity, indicative of a non-competitive action, whilst Schild plots for (-)-APV, APH and ketamine appeared linear and had slopes of approximately 1. Analysis of the results of combination experiments suggested that the presumed competitive antagonists, (-)-APV and APH, share a common site of action as NMDA antagonists, and that this site is distinct from that at which ketamine exerts its action. The action of Mg2+ is clearly different from that of either (-)-APV or ketamine. It is concluded that ketamine is a non-competitive antagonist of NMDA and may act at an allosteric site on the NMDA receptor complex to influence its function.

Autoradiographic characterization of N-methyl-D-aspartate-, quisqualate- and kainate-sensitive glutamate binding sites.

Abstract: Quantitative autoradiography was used to characterize the pharmacological specificity and anatomical distributions of subtypes of L-[3H]glutamate binding sites in rat brain. One population of sites was sensitive to N-methyl-D-aspartate (NMDA) and other compounds thought to be specific for the NMDA receptor. This site was enriched in stratum radiatum of hippocampus (CA1) where it constituted about 80% of glutamate binding sites and it represented a variable portion of glutamate binding sites throughout the brain. A second population of sites had a high affinity for quisqualate. Approximately 80% of glutamate binding sites in cerebellar molecular layer were of the high affinity quisqualate type. The number of these sites was greatly increased in the presence of Cl- and Ca++ ions. A subset of the high affinity quisqualate sites was sensitive to competition by kainate, particularly in stratum lucidum of hippocampus; the density of these high affinity kainate-sensitive sites was decreased in the presence of Ca++ but not Cl- ions. At high concentrations quisqualate competes for all glutamate binding sites, as reported previously. There was a good correspondence between the density and distribution of low affinity quisqualate sites and NMDA-sensitive sites. Pharmacological analysis suggested that the low affinity quisqualate site and the NMDA site are equivalent. Anatomical and pharmacological evidence suggests that the NMDA-, (high affinity) quisqualate- and kainate-sensitive glutamate binding sites may correspond to the physiologically defined NMDA, quisqualate and kainate receptors.

Dual-component amino-acid-mediated synaptic potentials: excitatory drive for swimming in Xenopus embryos.

Abstract: The neuronal basis of the excitation received by motoneurones during swimming in curarized Xenopus embryos has been investigated further. Extracellular stimulation of axons in the fibre tracts of the spinal cord has been used to evoke unitary excitatory post-synaptic potentials (p.s.p.s) in motoneurones. The p.s.p.s. had a rise time of 3-5 ms and a long falling phase lasting up to 200 ms. These potentials consist of two components: a 'fast' p.s.p. which is insensitive to 50 microM-(+/-)-2-amino-5-phosphonovaleric acid (APV) but is blocked by 2 mM-cis-2,3-piperidine dicarboxylic acid (PDA) and is therefore probably mediated by kainate/quisqualate receptors, and a 'slow' p.s.p. which is blocked by both APV and PDA and is therefore probably mediated by N-methyl-D-aspartate (NMDA) receptors. Paired intracellular recordings from motoneurones and interneurones have revealed a class of spinal cord interneurone which makes descending excitatory amino-acid-dependent synapses onto motoneurones and commissural interneurones. The p.s.p.s evoked by intracellular stimulation of these excitatory interneurones consist of 'fast' and 'slow' components identical in shape and pharmacological properties to those of the extracellularly evoked potentials. One neurone may, therefore, be able to release a transmitter which activates both NMDA and non-NMDA receptors on the same post-synaptic neurone generating fast and slow post-synaptic potentials. The excitatory interneurones play an important role in the generation of the swimming pattern in the curarized Xenopus embryo. Like motoneurones, they fire once per swimming cycle in phase with the ipsilateral motoneurones and receive a background excitation during swimming that is excitatory amino acid mediated. They are therefore part of the swimming rhythm generator. The temporal summation of the extracellularly evoked p.s.p.s shows that these excitatory interneurones are sufficient to generate the excitatory drive received by motoneurones during swimming.

Pharmacological properties of gamma-aminobutyric acid-, glutamate-, and aspartate-induced depolarizations in cultured astrocytes.

Abstract: Differentiated glial fibrillary acidic protein-positive astrocytes in homogeneous cultures of early postnatal rat cerebral hemispheres respond by membrane depolarization to gamma-aminobutyric acid (GABA), glutamate, and aspartate with a threshold concentration of approximately 10(-5) M. The GABA-induced depolarization is antagonized by two blockers of the neuronal GABAA receptor, picrotoxin and bicuculline, but is not affected by the uptake blockers beta-alanine or nipecotic acid. An agonist of the GABAA receptor, muscimol, produces a dose-response curve similar to that of GABA, whereas the agonist of the GABAB receptor, baclofen, did not alter the membrane potential. When repetitive pulses of GABA are given to one cell, its responsiveness depends on the time interval between pulses. Within 30 sec after termination of the first pulse the cell remains unresponsive to the second pulse. With increased time intervals between the pulses, reactivity toward GABA recovers. Five minutes after the first pulse the cell regains 75% of its initial depolarization peak. Aspartate results in a depolarization similar in size and time course to that induced by glutamate. The glutamate agonists, quisqualate and ibotenate, and kainate are less potent than glutamate. N-Methyl-D-aspartate has no effect on the membrane potential of astrocytes. The pharmacological features of the glutamate response are therefore similar to those of the receptor mediating neuronal glutamate transport.

Cellular uptake disguises action of L-glutamate on N-methyl-D-aspartate receptors. With an appendix: diffusion of transported amino acids into brain slices.

Abstract: Pharmacological properties of the guanosine 3'5'-cyclic monophosphate (cyclic GMP) responses to excitatory amino acids and their analogues were compared in slices and dissociated cells from the developing rat cerebellum maintained in vitro. The intention was to determine the extent to which cellular uptake might influence the apparent properties of receptor-mediated actions of these compounds. In slices, the potencies of the weakly (or non-) transported analogues, N-methyl-D-aspartate (NMDA) and kainate (KA) (EC50 = 40 microM each) were higher than those of the transported amino acids, D- and L-aspartate (EC50 = 250 microM and 300 microM) and D- and L-glutamate (EC50 = 540 microM and 480 microM). Quisqualate (up to 300 microM) failed to increase cyclic GMP levels significantly. The sensitivity of agonist responses to the NMDA receptor antagonist, DL-2-amino-5-phosphonovalerate (APV), was in the order NMDA greater than L-aspartate greater than L-glutamate, KA. In dissociated cells, L-glutamate was 280 fold more potent (calculated EC50 = 1.7 microM). L- and D-aspartate (calculated EC50 = 13 microM) and D-glutamate (EC50 = 130 microM) were also more effective than in slices. The potencies of NMDA and KA were essentially unchanged. Responses to NMDA, L-glutamate and L-aspartate under these conditions were equally sensitive to inhibition by APV but the response to KA remained relatively resistant to this antagonist. The implications of these results are that, in slices, cellular uptake is responsible for (i) the dose-response curves to L-glutamate, L- and D-aspartate bearing little or no relationship to the true (or relative) potencies of these amino acids; (ii) the potency of APV towards the actions of transported agonists acting at NMDA receptors being reduced and (iii) a differential sensitivity to APV of responses to L-glutamate and L-aspartate being created, the consequence being that a potent action of L-glutamate on NMDA receptors is disguised. These conclusions are supported by theoretical considerations relating to the diffusion of transported amino acids into brain slices, as elaborated in the Appendix.

Synaptic transmission between dorsal root ganglion and dorsal horn neurons in culture: antagonism of monosynaptic excitatory postsynaptic potentials and glutamate excitation by kynurenate

Abstract: Intracellular recording techniques have been used to provide information on the identity of excitatory sensory transmitters released at synapses formed between dorsal root ganglion (DRG) and dorsal horn neurons maintained in cell culture. Explants of embryonic rat DRG were added to dissociated cultures of embryonic dorsal horn neurons and synaptic potentials were recorded intracellularly from dorsal horn neurons after DRG explant stimulation. More than 80% of dorsal horn neurons within 1 mm of DRG explants received at least one fast, DRG-evoked, monosynaptic input. In the presence of high divalent cation concentrations, the acidic amino acid receptor agonists, L-glutamate, kainate, and quisqualate excited all dorsal horn neurons which received a monosynaptic DRG neuron input, whereas aspartate and N-methyl-D-aspartate (NMDA) had little or no action. Several compounds reported to antagonize the actions of acidic amino acids were tested for their ability to block DRG-evoked synaptic potentials and glutamate-evoked responses in dorsal horn neurons. 2-Amino-5-phosphonovalerate, a selective NMDA receptor antagonist, was relatively ineffective at antagonizing DRG-evoked synaptic potentials and glutamate-evoked responses. In contrast, kynurenate was found to be a potent antagonist of amino acid-evoked responses and of synaptic transmission at all DRG-dorsal horn synapses examined. The blockade of synaptic transmission by kynurenate appeared to result from a postsynaptic action on dorsal horn neurons. These findings indicate that glutamate, or a glutamate-like compound, but not aspartate, is the excitatory transmitter that mediates fast excitatory postsynaptic potentials at the DRG-dorsal horn synapses examined in this study.

Roles of aspartate and glutamate in synaptic transmission in rabbit retina. II. Inner plexiform layer.

Abstract: Intracellular recordings were obtained from amacrine and ganglion cells in the superfused, isolated retina-eyecup of the rabbit. The putative neurotransmitters aspartate, glutamate, and several of their analogues were added to the superfusate while the membrane potential and light-responsiveness of the retinal neurons were monitored. Both L-aspartate and L-glutamate displayed excitatory actions on the activity of the vast majority of amacrine and ganglion cells studied. However, these agents occasionally appeared to inhibit the responses of the inner retinal neurons by producing hyperpolarization of the membrane potential and blockage of the light-evoked responses. In either case, the effects of aspartate and glutamate were indistinguishable. The glutamate analogues kainate and quisqualate produced strong excitatory effects on the responses of amacrine and ganglion cells at concentrations some 200-fold less than those needed to obtain similar effects with aspartate or glutamate. The aspartate analogue, n-methyl DL-aspartate (NMDLA), also produced strong excitatory effects but was approximately three times less potent than kainate or quisqualate. On one occasion, we encountered a ganglion cell that was depolarized by kainate, but hyperpolarized by NMDLA. The glutamate antagonist alpha-methyl glutamate and the aspartate antagonist alpha-amino adipate effectively blocked the responses of amacrine and ganglion cells. However, on any one cell, one antagonist was always clearly more potent than the other. We examined the actions of the glutamate analogue 2-amino-4-phosphonobutyrate (APB) on the responses of inner retinal neurons and found that it selectively abolished all "on" activity in the inner retina. Together with our finding that APB selectively abolishes on-bipolar cell responses (see Ref. 6), these data support the hypothesis that on-bipolar cells subserve the "on" activity of amacrine and ganglion cells. Our data suggest that aspartate and glutamate are excitatory transmitters in the inner retina, possibly being released from bipolar cell axon terminals in the inner plexiform layer.

Neurotoxicity of L-glutamate and DL-threo-3-hydroxyaspartate in the rat striatum.

Abstract: : Destruction of the glutamatergic corticostriatal pathway potentiates the neurotoxic action of 1 μmol L-glutamate injected into the rat striatum, whereas the toxic effects of 10 nmol kainate are markedly attenuated. Injection of 170 nmol of the glutamate uptake inhibitor, DL-threo-3-hydroxyaspartate, into the intact striatum also causes neuronal degeneration, which is accompanied by a reduction in markers for cholinergic and GABAergic neurones. Prior removal of the corticostriatal pathway destroys the ability of DL-threo-3-hydroxyaspartate to cause lesions in the striatum. These results indicate that removal, or blockade, of uptake sites for glutamate increase the vulnerability of striatal neurones to the toxic effects of synaptically released glutamate.

Quisqualate and L-glutamate inhibit retinal horizontal-cell responses to kainate.

Abstract: Currents elicited by L-glutamate and the related agonists quisqualate and kainate were analyzed under voltage clamp in isolated goldfish horizontal cells, using the whole-cell recording configuration of the patch-clamp method [Hamill, O.P., Marty, A., Neher, E., Sakmann, B. & Sigworth, F. J. (1981) Pflugers Arch. 391, 85-100]. These currents resulted from an increase in cationic conductance and were indistinguishable from one another in terms of reversal potential (approximately equal to 0 mV) and apparent elementary conductance (2-3 pS). The power-density spectra of the noise increases produced by each agonist were fit by the sum of two Lorentzian curves having similar cutoff frequencies (tau 1 approximately equal to 5 msec, tau 2 approximately equal to 1 msec), but the relative power of these components were different for quisqualate and glutamate than for kainate. Moreover, the responses to high doses of either quisqualate or glutamate rapidly faded, whereas the responses to kainate did not. Finally, quisqualate and glutamate produced an inhibition of responses to kainate which appeared to be uncompetitive. Kainate, quisqualate, and in our preparation, glutamate appear to activate channels different than those activated by N-methyl-D-aspartate in other preparations. At least some of the effects of quisqualate and glutamate appear to be mediated by receptors bound by kainate.

Synaptic transmission at N‐methyl‐D‐aspartate receptors in the proximal retina of the mudpuppy.

Abstract: The effects of excitatory amino acid analogues and antagonists on retinal ganglion cells were studied using intracellular recording in the superfused mudpuppy eyecup preparation. Aspartate, glutamate, quisqualate (QA), kainate (KA) and N-methylaspartate (NMA) caused depolarization and decreased input resistance in all classes of ganglion cells. The order of sensitivity was QA greater than or equal to KA greater than NMA greater than aspartate greater than or equal to glutamate. All of these agonists were effective when transmitter release was blocked with 4 mM-Co2+ or Mn2+, indicating that they acted at receptor sites on the ganglion cells. At a concentration of 250 microM, 2-amino-5-phosphonovalerate (APV) blocked the responses of all ganglion cells to NMA, but not to QA or KA, indicating that NMA acts at different receptor sites from QA or KA. Responses to bath-applied aspartate and glutamate were reduced slightly or not at all in the presence of APV, indicating that they were acting mainly at non-NMDA (N-methyl-D-aspartate) receptors. In all ganglion cells 250 microM-APV strongly suppressed the sustained responses driven by the 'on'-pathway but not those driven by the 'off'-pathway. In most on-off ganglion cells the transient excitatory responses at 'light on' and 'light off' were not reduced by 500 microM-APV. APV-resistant transient excitatory responses were also present in some on-centre ganglion cells. APV did not block the transient inhibitory responses in any class of ganglion cells. At concentrations which blocked the sustained responses of ganglion cells, APV did not affect the sustained responses of bipolar cells, indicating that it acted at sites which were post-synaptic to bipolar cells. The simplest interpretation of these results is that the transmitter released by depolarizing bipolar cells acts at NMDA receptors on sustained depolarizing amacrine and ganglion cells. It may act at non-NMDA receptors at synapses which produce transient excitatory responses, but this could not be proved. The transmitter released by hyperpolarizing bipolar cells does not appear to act at NMDA receptors on any post-synaptic cells.

Blocking action of pentobarbital on receptors for excitatory amino acids in the guinea pig hippocampus.

Abstract: The actions of pentobarbital sodium (Pent) on receptors for glutamate (Glu) and related compounds were studied in thin sections of the guinea pig hippocampus. Depolarizations induced by Glu and quisqualate (Quis) in CA3 neurons were reduced in amplitude during iontophoretic administration of Pent. This action of Pent was not accompanied by any noticeable changes in membrane potential or neuron input resistance. Depolarizations induced by N-methyl-D-aspartate were less sensitive to Pent. The fast kainate (KA) response was as susceptible as the Glu response, whereas the slow KA response was unaffected by Pent in three quarters of the neurons examined. Pent suppressed the Glu response at lower concentrations than required to potentiate responses to gamma-amino butyric acid. Excitatory postsynaptic potentials (EPSPs) elicited by stimulation of mossy fibers were suppressed by Pent. The EPSPs were a little more resistant to Pent than were the Glu responses. These results indicate that Pent blocks receptors for excitatory amino acids in the hippocampus. Of the three different populations of the receptors, Quis receptors are the most sensitive to Pent and KA receptors are the least sensitive. The suppression of the EPSPs is in accordance with the notion that Glu is the transmitter released from mossy fibers.

Physiology of excitatory synaptic transmission in cultures of dissociated rat hippocampus.

Abstract: Cultures of dissociated rat hippocampal neurons were used to study the physiology and pharmacology of excitatory synaptic transmission. Rat hippocampal neurons depolarized when they were exposed to the excitatory transmitter candidates, glutamate (Glu) and aspartate (Asp), as well as to the pure excitatory amino acid agonists, N-methyl-D-aspartate (NMDA) and kainate (KA). Quisqualate (QUIS) produced responses in about two-thirds of these cells. Glu responses were much more effectively blocked by the excitatory amino acid antagonists cis-2,3-piperidine dicarboxylic acid (PDA) and gamma-D-glutamylglycine (DGG) than by D-2-amino-5-phosphonovaleric acid (APV) or D-alpha-aminoadipic acid (DAA). Asp depolarizations were depressed by all four antagonists. Monosynaptic excitatory postsynaptic potentials (EPSPs) were only decreased by PDA and DGG. Postsynaptic responses to both Glu and Asp were very voltage dependent, decreasing as the membrane potential was hyperpolarized up to 70 mV below resting levels. The EPSP, however, increased linearly in the hyperpolarized range. NMDA responses were also voltage dependent, while KA and QUIS responses behaved like EPSPs. DGG very effectively blocked KA, but not QUIS, depolarizations. APV, which only partially depressed Glu responses, markedly diminished their voltage sensitivity. These results all suggest that EPSPs in this preparation are produced by Glu acting at KA-type synaptic receptors. Exogenous Glu probably acts at both synaptic KA receptors and extrasynaptic NMDA receptors, which explains why it produces a voltage-dependent response different from the EPSP.

Inhibitors of high-affinity uptake augment depolarizations of hippocampal neurons induced by glutamate, kainate and related compounds.

Abstract: Actions of dihydrokainate (DHKA) and 3-hydroxy-DL-aspartate (HAsp), inhibitors of high-affinity uptake for L-glutamate (Glu), were studied in vitro in thin hippocampal slices of the guinea pig. The amplitude of the depolarizations induced by Glu and by L-aspartate (Asp) in CA3 neurons are markedly augmented by DHKA and HAsp. Depolarizations induced by D-homocysteate (DH) were unaffected by the inhibitors. In about half of the neurons, depolarizations induced by L-homocysteate (LH) and by quisqualate (Quis) were slightly augmented by the inhibitors. Fast responses to kainate (KA) were augmented by the inhibitors to a similar extent as were Glu responses whereas slow KA responses were insensitive to HAsp. HAsp was without effect on excitatory postsynaptic potentials elicited by stimulation of granular layer. These findings are in general agreement with the biochemical data on amino acid uptake processes and are also consistent with the slow time-courses of depolarizations induced by DH, LH and Quis. Augmentation of fast KA responses provides strong evidence for the hypothesis that an KA pulse causes a liberation of Glu and/or Asp from the tissue and the liberated amino acid(s) induces the fast KA response in neurons nearby.

Effect of excitatory amino acids on gamma‐aminobutyric acid release from frog horizontal cells.

Abstract: The effects of excitatory amino acids, analogues and K on [3H]gamma-aminobutyric acid [3H]GABA) release from horizontal cells of the isolated superfused frog retina were studied. Exposure of the retina to medium containing high concentrations (25-100 mM) of KCl increased the release of [3H]GABA to a maximum which was 40 times the spontaneous resting release. The K-evoked release of [3H]GABA was almost abolished in high-Mg/low-Ca medium. Glutamate, aspartate, kainate and quisqualate also stimulated the release of [3H]GABA from horizontal cells, the maximum evoked release being similar to that produced by KCl. The release of [3H]GABA evoked by glutamate, aspartate, kainate and quisqualate was abolished in high-Mg/low-Ca medium and by Na-free medium. The evoked releases of [3H]GABA were not reduced by tetrodotoxin. N-Methyl-D-aspartate (NMDA) at concentrations up to 10 mM had virtually no effect on [3H]GABA release from horizontal cells. In Mg-free medium, NMDA stimulated [3H]GABA release, but the maximum release was only 10% of that produced by other agonists. Mg-free medium did not significantly affect the evoked release of [3H]GABA by other agonists. NMDA apparently possessed affinity for the kainate receptor, because in normal medium it antagonized the effects of kainate but not glutamate, aspartate or quisqualate. The non-selective antagonist of excitatory amino acids, (+/-)-cis-2,3-piperidine dicarboxylic acid (PDA) antagonized the action of glutamate, aspartate, kainate and quisqualate on horizontal cell [3H]GABA release. D(-)-2-Amino-4-phosphonobutyrate (APB) and D-gamma-glutamylglycine (D-gamma-GG) antagonized the actions of kainate on horizontal cell [3H]GABA release at concentrations which had little affect on quisqualate-evoked responses. Approximate estimates of pA2 values (Schild, 1947) showed that the specificity and potency of the antagonists was low. Nevertheless, the retinal 'non-NMDA' receptors can probably be subdivided into kainate and quisqualate types. Glutamate diethylester (GDEE) did not affect the action of any agonist. We conclude that glutamate (and aspartate) probably stimulate the release of [3H]GABA from frog horizontal cells by activating receptors of the non-NMDA type. This activation may trigger the opening of tetrodotoxin-insensitive Na channels, resulting in the depolarization of the cell membrane and an increase in the conductance of voltage-sensitive Ca-channels. An influx of Ca ions would then trigger the release of [3H]GABA. Our results are not consistent with previous suggestions that GABA release from horizontal cells involves an outwardly directed transport process.