Negative allosteric modulators of AMPA-preferring receptors inhibit [3H]GABA release in rat striatum

Neurochemistry International 37 (2000) 33±45

www.elsevier.com/locate/neuint

Negative allosteric modulators of AMPA-preferring receptors inhibit [3H]GABA release in rat striatum Laszlo G. Harsing Jr*, Viola Csillik-Perczel, IstvaÂn Ling, SaÂndor SoÂlyom Institute for Drug Research Ltd, 47±49 Berlini ut, 1045 Budapest, Hungary Received 12 October 1999; accepted 16 December 1999

Abstract The e€ect of a-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA), a selective glutamate receptor agonist, on the release of previously incorporated [3H]GABA was examined in superfused striatal slices of the rat. The slices were loaded with [3H]GABA in the presence of b-alanine (1 mM) and superfused with Krebs-bicarbonate bu€er containing nipecotic acid (0.1 mM) and aminooxyacetic acid (0.1 mM) to inhibit GABA uptake and metabolism. AMPA (0.01 to 3 mM) increased basal [3H]GABA out¯ow and nipecotic acid potentiated this e€ect. The [3H]GABA releasing e€ect of AMPA was an external Ca2+dependent process in the absence but not in the presence of nipecotic acid. Cyclothiazide (0.03 mM), a positive modulator of AMPA receptors, failed to evoke [3H]GABA release by itself, but it dose-dependently potentiated the [3H]GABA releasing e€ect of AMPA. The AMPA (0.3 mM)-induced [3H]GABA release was antagonized by NBQX (0.01 mM) in a competitive fashion ( pA2 5.08). The negative modulator of AMPA receptors, GYKI-53784 (0.01 mM) reversed the AMPA-induced [3H]GABA release by a non-competitive manner ( pD2' 5.44). GYKI-53784 (0.01±0.1 mM) also decreased striatal [3H]GABA out¯ow on its own right, this e€ect was stereoselective and was not in¯uenced by concomitant administration of 0.03 mM cyclothiazide. GYKI-52466 (0.03±0.3 mM), another negative modulator at AMPA receptors, also inhibited basal [3H]GABA e‚ux whereas NBQX (0.1 mM) by itself was ine€ective in alteration of [3H]GABA out¯ow. The present data indicate that AMPA evokes GABA release from the vesicular pool in neostriatal GABAergic neurons. They also con®rm that multiple interactions may exist between the agonist binding sites and the positive and negative modulatory sites but no such interaction was detected between the positive and negative allosteric modulators. Since GYKI-53784, but not NBQX, inhibited [3H]GABA release by itself, AMPA receptors located on striatal GABAergic neurons may be in sensitized state and phasically controlled by endogenous glutamate. It is also postulated that these AMPA receptors are located extrasynaptically on GABAergic striatal neurons. 7 2000 Elsevier Science Ltd. All rights reserved. Keywords: AMPA receptor; 2, 3-Benzodiazepines; GYKI-53784; Cyclothiazide; [3H]GABA release; Neostriatum

1. Introduction Medium size g-aminobutyric acid(GABA)ergic spiny neurons in the neostriatum receive excitatory impulses from the cerebral cortex and the thalamus (Dube et al., 1988; Fonnum et al., 1981). Excitatory impulses are transmitted by glutamate released from nerve term* Corresponding author. Tel.: +36-1-399-3336; fax: +36-1-3993356. E-mail address: [email protected] (L.G. Harsing Jr).

inals establishing synaptic contact to the spines of the dendritic trees of striatal GABAergic projection neurons (Bolam and Bennett, 1995; Smith and Bolam, 1990). Glutamatergic excitation is mediated by postsynaptically located metabotropic and ionotropic glutamate receptors, the latter consist of N-methyl-Daspartate (NMDA), a-amino-3-hydroxy-5-methyl-4isoxazolepropionic acid (AMPA) and kainate types (Bleakman and Lodge, 1998; Pin and Duvoisin, 1995). AMPA receptors mediate fast glutamatergic neural impulses by opening receptor-linked ion channels permeable to Na+, although certain AMPA receptor sub-

0197-0186/00/$ - see front matter 7 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 1 9 7 - 0 1 8 6 ( 0 0 ) 0 0 0 0 5 - X

34

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

types might also be permeable to Ca2+ (Mayer and Westbrook, 1987; Pellegrini-Giampietro et al., 1997). AMPA receptor stimulation leads to membrane depolarization and concomitant neurotransmitter release, although the mechanism of AMPA-induced release is still question of debate (Galli et al., 1992; Garcia et al., 1995). AMPA-preferring receptors possess binding sites which are able to modulate receptor functions positively and negatively (Donevan and Rogawski, 1998). Cyclothiazide, which belongs to the benzothiadiazine class of diuretics, is widely used as a positive modulator since it decreases rapid AMPA receptor desensitization (Johansen et al., 1995; Partin et al., 1994; Vyklicky et al., 1991). On the contrary, a number of 2,3-benzodiazepines act as negative modulators of AMPA receptors (Tarnawa et al., 1992) and inhibit AMPA-induced current by a non-competitive fashion (Donevan and Rogawski, 1993; Paternain et al., 1995). GYKI-52466 was the ®rst compound shown to exhibit negative modulatory e€ect at AMPA receptors and since then several other molecules with similar structure were introduced (Tarnawa and Vizi, 1998; Vizi et al., 1996). Complex interactions may exist between the agonist binding site and the modulatory sites of AMPA receptors but the exact nature of these regulatory processes remained to be elucidated. Using electrophysiological or labelled neurotransmitter release techniques, interaction (Palmer and Lodge, 1993; Petitet et al., 1995; Zorumski et al., 1993) as well as lack of interaction (Desai et al., 1995; Okada et al., 1996; Yamada and Turetsky, 1996) between the positive and negative modulators of AMPA receptors have been demonstrated. The aim of the present experiments was to characterize glutamatergic and GABAergic interactions in the neostriatum and to assess the role of AMPA receptor-ion channel complex in the mechanism of GABA release. For this, GABAergic neurons in rat neostriatal slices were loaded with [3H]GABA in order to evoke labelled transmitter release by AMPAinduced depolarization. We found that although various interactions may exist among recognition sites at AMPA receptors, there is probably no such interaction between positive and negative modulatory sites. Our data also suggest that, depending on the rate of occupation by endogenous glutamate, AMPA receptors may be in sensitized or desensitized states which then ultimately determine their function in neurotransmitter release. 2. Experimental procedures 2.1. Preparation of striatal slices Male Sprague±Dawley rats weighing 200±250 g were

killed by decapitation and the brain was removed from the skull. Coronal slices approximately 350 mm thick were cut from the striatum using a McIlwain tissue chopper (The Mickie Laboratory Engineering Co., Gomshall, UK). The slices were collected into ice-cold Krebs-bicarbonate bu€er, pH 7.4 with the following composition in mM: NaCl 118, KCl, 4.7; CaCl2 1.25, NaH2PO4 1.2, MgCl2 1.2, NaHCO3 25, glucose 11.5. The Krebs-bicarbonate bu€er used throughout the experiments was continuously gassed with 5% CO2 in O 2. 2.2. Measurement of [3H]GABA e‚ux Striatal slices were incubated with [3H]GABA (2.5 mCi/ml) in oxygenated Krebs-bicarbonate bu€er for 30 min at 378C. b-Alanine (1 mM), an inhibitor of GABA uptake in glial cells (Iversen and Kelly, 1975), was present in the incubation bu€er. The tissues were then transferred into low-volume (300 ml) superfusion chambers (Experimetria, Inc., Budapest, Hungary) and superfused with aerated and preheated (378C) Krebsbicarbonate bu€er that contained the aminotransferase inhibitor aminooxyacetic acid (0.1 mM). Except where indicated otherwise, the GABA uptake inhibitor nipecotic acid was present in Krebs-bicarbonate bu€er in a concentration of 0.1 mM (Harsing and Zigmond, 1997). The ¯ow rate was kept at 1 ml/min by a Gilson multichannel peristaltic pump (type M312, Villiers-LeBel, France). The superfusate was discarded in the ®rst 60-min period of the experiment then 22 3-min fractions were collected by a Gilson fraction collector (type FC-2038, Middleton, WI, USA). The ionotropic glutamate receptor agonists (AMPA and NMDA) were added to the perfusion bu€er from fraction 8 for three consecutive 3-min periods. When the e€ect of an antagonist in the presence or absence of an agonist was studied, the antagonist was added to the superfusion bu€er from fraction 5 for six consecutive fractions. Cyclothiazide was added to Krebs-bicarbonate bu€er 30 min before starting fraction collection and was maintained throughout the experiments. At the end of superfusion, the tissues were collected from the superfusion chambers and homogenized in 0.8 ml Soluene-100 and an aliquot (100 ml) was processed for determination of tissue content of radioactivity. To determine the radioactivity released from the tissue, a sample (0.8 ml) of the superfusate was mixed with 4.2 ml of liquid scintillation reagent (Ultima Gold, Packard, Groningen, The Netherlands) and subjected to liquid scintillation spectrometry. Previously we separated the released radioactivity into [3H]GABA and [3H]metabolites on thin layer chromatography and found that, in these experimental conditions, 92±108% of the radioactivity released from striatal slices is in

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

the form of authentic [3H]GABA (Harsing and Zigmond, 1998). 2.3. Determination of [3H]GABA e‚ux from rat striatal slices The e‚ux of [3H]GABA was expressed as a fractional rate, i.e., as a percentage of the amount of radioactivity in the tissue at the time the release was determined (Harsing et al., 1992). A computer program (Quattro Pro V6.0) was used to estimate the fractional rate of tritium e‚ux. To estimate the AMPA-induced [3H]GABA over¯ow, the mean of the basal release determined before and after stimulation was subtracted from each sample and the evoked release represents the sum of the release in the drug exposed fractions. To calculate the inhibition, [3H]GABA e‚ux in each fraction was subtracted from the mean of the basal [3H]GABA out¯ow determined before and after the addition of AMPA antagonist and the di€erences were summed. The antagonist potency of NBQX, a competitive AMPA receptor antagonist, was calculated as pA2 by the method of Maura et al. (1985): pA2 ˆ log…E  =E ÿ 1† ÿ log B where E and E are the AMPA concentrations causing 0.5% of content increase in [3H]GABA release in the presence and absence of the antagonist and B is the concentration of the antagonist. The non-competitive antagonist potency of GYKI-53784 was calculated as pD2' by the method of Ariens and Van Rossum (1957): pD20 ˆ log…A =A ÿ 1† ÿ log B where A is the maximal e€ect of AMPA, A is the maximal e€ect of AMPA in the presence of the antagonist and B is the concentration of the antagonist. 2.4. Statistical analysis One way analysis of variance (ANOVA) followed by the Dunnett's multiple comparison test, the Student tstatistics for two-means and the paired t-test were used for statistical analysis of the data. The GraphPad Prism V2.0 software was used to calculate the slopes of concentration-release curves. The mean 2 SEM was calculated and the number of independent determinations (n ) is indicated. A level of probability (P ) less than 5% was considered signi®cant. 2.5. Materials The following drugs were used in this study: (S )-aamino-3-hydroxy-5-methyl-4-isoxazolepropionic acid ((S )AMPA), N-methyl-D-aspartic acid (NMDA),

35

Tocris-Cookson, Bristol, UK; cyclothiazide, 2,3-dihydro-6-nitro-7-sulphamoylbenzo(f)quinoxaline (NBQX), Research Biochemicals Inc., Natick, MA, USA; aminooxyacetic acid, b-alanine, (2)nipecotic acid, Sigma Chemicals Co., St Louis, MO, USA. GYKI-52466 (5(4-aminophenyl)-8-methyl-9H-1,3-dioxolo[4,5-h][2,3]benzodiazepine), GYKI-53784 (LY303070, (R )-5(4-aminophenyl)-7-methylcarbamoyl-8-methyl-8,9-dihydro-7H-1,3-dioxolo[4,5-h][2,3]-benzodiazepine), GYKI53785 (LY303071, (S )-5-(4-aminophenyl)-7-methylcarbamoyl-8-methyl-8,9-dihydro-7H-1,3-dioxolo[4,5-h][2,3]-benzodiazepine) and GYKI-54026 ((R,S )-5-(2aminophenyl)-7-methylcarbamoyl-8-methyl-8,9-dihydro-7H-1,3-dioxolo[4,5-h][2,3]-benzodiazepine) were synthesized in the Institute for Drug Research Ltd., Budapest, Hungary. g-[2,3-3H(N)]-aminobutyric acid ([3H]GABA), speci®c activity 1.5 TBq/mmol, was purchased from New England Nuclear Life Science Products, Boston, MA, USA. All other chemicals used were of analytical grade. 3. Results 3.1. Characterization of AMPA-induced [3H]GABA release In the absence of nipecotic acid, the basal [3H]GABA out¯ow approached a rate of 1.32 2 0.07 kBq/g in 3 min after a 60-min initial washout period (n = 6). In these experiments, the content of radioactivity in striatal tissue was found to be 527.06 2 71.56 kBq/g (n = 6). The calculated fractional [3H]GABA e‚ux at rest was 0.26 2 0.04% of content released in 3 min. When nipecotic acid was added to the superfusion bu€er in a concentration of 0.1 mM, the basal [3H]GABA out¯ow increased to 3.18 2 0.32 kBq/g in 3 min which corresponds to a fractional release of 0.66 2 0.06% of content released in 3 min, the di€erence was statistically signi®cant (P < 0.001, n = 6). Addition of nipecotic acid (0.1 mM) to the superfused tissue did not change the tissue content of radioactivity (439.49233.50 kBq/g, P > 0.20, n = 6). AMPA added to the superfusion bu€er in concentrations of 0.1±3 mM for three consecutive 3-min fractions, produced a concentration-dependent increase in [3H]GABA release (Fig. 1). The concentration of AMPA which increased basal fractional [3H]GABA out¯ow by 0.5% of content was 1.09 mM. The increase in [3H]GABA e‚ux was usually observed in four consecutive 3-min fractions then the release declined exponentially to the prestimulated value. The e€ect of 0.3 mM AMPA on [3H]GABA release measured in di€erent experimental conditions in superfused striatal slices is shown in Table 1. Addition of nipecotic acid (0.1 mM) to the superfusion bu€er

36

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

Fig. 1. E€ect of AMPA on [3H]GABA release from striatal slices of the rat (w). Cyclothiazide (0.03 mM) potentiated the stimulatory e€ect of AMPA on [3H]GABA release (*). The AMPA concentration which increased [3H]GABA over¯ow by 0.5% of content was 0.23 and 1.09 mM in the presence and absence of cyclothiazide. Slices from rat striatum were prepared, loaded with [3H]GABA and superfused with Krebs-bicarbonate bu€er containing aminooxyacetic acid and nipecotic acid (0.1 mM). AMPA (0.1±3 mM) was added to striatal slices from fractions 8 to 10 (9 min). Cyclothiazide was added to striatal slices 30 min before collection started and maintained throughout the experiment. Control [3H]GABA release was 0.0320.01% and 0.0420.01% of content in non-treated and cyclothiazide-treated groups, respectively (n = 4, P > 0.60). ANOVA followed by the Dunnett's multiple comparison test, F(4,15)=164.911, P < 0.01 and F(4,14)=68.553, P < 0.01 in AMPA- and AMPA plus cyclothiazide-treated groups, respectively. Each dose was compared to respective control value, indicates P < 0.05. Values shown are the mean2SEM of three to four experiments.

increased not only the basal [3H]GABA out¯ow but also the AMPA-evoked [3H]GABA over¯ow. The potentiation of AMPA-induced [3H]GABA release by nipecotic acid was also observed when Ca2+-free Krebs-bicarbonate bu€er was used for superfusion. Furthermore, removal of Ca2+ from the superfusion bu€er in the presence of nipecotic acid did not signi®cantly alter AMPA-evoked [3H]GABA release. In the absence of nipecotic acid, however, omission of Ca2+

from the superfusion bu€er led to a reduction in [3H]GABA release (Table 1). 3.2. Interaction between cyclothiazide and the agonist site on AMPA receptors The amount of [3H]GABA released by 1 mM of AMPA was identical when striatal slices were exposed to the agonist for 9 or 15 min (0.4720.03% and 0.52

Table 1 Characterization of AMPA depolarization-induced [3H]GABA release from rat striatal slicesa Conditions 1. 2. 3. 4.

Normal Ca2+ nipecotic acid-free Ca2+-free nipecotic acid-free Normal Ca2+ nipecotic acid Ca2+-free nipecotic acid

Stimulation

[3H]GABA release (% of content)

AMPA AMPA AMPA AMPA

0.1620.02 0.0720.02 0.3020.02 0.3020.01

Signi®cance (P ) 1:2 < 0.05 1:3 < 0.01 2:4 < 0.001 3:4 > 0.90

a Slices from rat striatum were prepared, loaded with [3H]GABA and superfused. AMPA (0.3 mM) was added to striatal slices from fractions 8 to 10 (9 min) to evoke [3H]GABA release, basal [3H]GABA release was 0.0420.01% of content released per 3 min (P < 0.01, n = 4). Normal concentration of Ca2+ in Krebs-bicarbonate bu€er was 1.25 mM, when Ca2+ was omitted, Na2EGTA was added to the bu€er (1 mM). Nipecotic acid was added to the superfusion bu€er in a concentration of 0.1 mM at the beginning of superfusion. Student t-statistics for two means was used for comparison. Values shown are the mean2SEM of four to eight experiments.

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

2 0.04% of content released, respectively (n = 4, P > 0.60). This ®nding indicates a rapid desensitization of AMPA receptors modulating [3H]GABA release. Inclusion of cyclothiazide (0.03 mM) during AMPA-induced depolarization resulted in a shift of the AMPA concentration-response curve to the left without apparently reaching plateau (Fig. 1). The concentration of AMPA which increased basal fractional [3H]GABA out¯ow by 0.5% of content in the presence of 0.03 mM cyclothiazide was 0.23 mM. Addition of cyclothiazide alone in a concentration of 0.03 mM did not alter the basal release of [3H]GABA from superfused striatal slices: the rate of e‚ux was 0.82 2 0.04 and 0.87 2 0.07% of content released in 3 min in the presence and absence of cyclothiazide (n = 3 and 4, P > 0.60). In another series of experiments, the e€ect of AMPA on [3H]GABA release was potentiated by various concentrations (0.03±0.1 mM) of cyclothiazide. Cyclothiazide, at all concentrations tested, elevated AMPA (0.3 mM)-induced release of [3H]GABA and this e€ect of cyclothiazide was concentration-dependent (Fig. 2). In contrast to AMPA-induced [3H]GABA release, cyclothiazide (0.03 mM) failed to potentiate NMDAevoked release. The NMDA (0.3 mM)-induced [3H]GABA release was 0.7420.13 in the presence and

37

0.69 2 0.07% of content in the absence of 0.03 mM cyclothiazide, respectively (P > 0.70 and n = 3±4). 3.3. Interaction between agonist and antagonist sites on AMPA receptors The concentration-e€ect curve of AMPA on striatal [3H]GABA release was also determined in the presence of NBQX (0.01 mM) and GYKI-53784 (0.01 mM). In this series of experiments, cyclothiazide (0.03 mM) was also added to provide a potentiated e€ect of AMPA on [3H]GABA release. Fig. 3 indicates that both NBQX and GYKI-53784 caused an inhibition of AMPA-induced [3H]GABA release in superfused striatal slices of the rat but the nature of antagonisms was di€erent. The inhibitory e€ect of NBQX was overcome by increasing concentration of AMPA. In addition, the concentration-e€ect curve for AMPA in the presence of NBQX was shifted to the right in a parallel manner. The slope of AMPA concentration-e€ect curve was 0.987 2 0.101 and that was 0.740 2 0.119 in the presence of 0.01 mM NBQX, not signi®cantly di€erent (P > 0.10, n = 15 and 16). The concentration of AMPA which increased basal fractional [3H]GABA out¯ow by 0.5% of content in the presence of 0.03 mM cyclothiazide was 0.23 mM and it was shifted by 0.01 mM NBQX to 0.51 mM (DR=2.21). The calculated pA2

Fig. 2. E€ect of cyclothiazide on AMPA-induced [3H]GABA release from striatal slices of the rat. Slices from rat striatum were prepared, loaded with [3H]GABA and superfused with Krebs-bicarbonate bu€er containing aminooxyacetic acid and nipecotic acid (0.1 mM). AMPA (0.3 mM) was added to striatal slices from fractions 8 to 10 (9 min) to evoke [3H]GABA release. Cyclothiazide (0±0.1 mM) was added 30 min before collection started and maintained throughout the experiment. Control [3H]GABA release was 0.04 2 0.01% of content, n = 4. ANOVA followed by the Dunnett's multiple comparison test, F(4,17)=36.360, P < 0.01, P < 0.05, P < 0.01 when compared to AMPA-treated group. Values shown are the mean2SEM of four to six experiments.

38

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

value for NBQX against AMPA on striatal [3H]GABA release was 5.08. On the contrary, the inhibitory e€ect of GYKI53784 (0.01 mM) on AMPA-induced [3H]GABA release was not overcome by increasing concentrations of AMPA (Fig. 3). Instead, a profound suppression of maximal AMPA responses was observed. The slopes of the AMPA concentration-e€ect curves were determined in the presence and absence of GYKI-53784: the slope for AMPA dose-response curve was 0.987 2 0.101 in the absence and it was 0.198 2 0.068 in the presence of 0.01 mM GYKI-53784, these slopes statistically di€ered (P < 0.001, n = 15 and 14). Therefore, the non-competitive antagonistic potency of GYKI53784 on striatal AMPA-induced [3H]GABA release was characterized with a pD2' value of 5.44. 3.4. The e€ects of negative modulators of AMPA receptors on the release of [3H]GABA in rat striatal slices Addition of GYKI-53784 (0.1 mM) to the superfusion bu€er for 18 min reduced the basal [3H]GABA

out¯ow from rat striatal slices (Fig. 4). The maximal inhibition of transmitter out¯ow developed gradually and this e€ect of GYKI-53784 was reversible as withdrawal of the drug from the superfusion bu€er resulted in a graduate increase of the transmitter out¯ow. The inhibition of [3H]GABA release by GYKI-53784 proved to be concentration-dependent when it was added in concentrations from 0.01 to 0.1 mM (Table 2). The (+)enantiomer, GYKI-53785 was found to be less e€ective in inhibition of [3H]GABA release from superfused striatal slices (Table 2). Similarly to GYKI-53784, GYKI-52466, another negative modulator of AMPA receptors, also reduced [3H]GABA over¯ow although its activity was less pronounced (Table 2). GYKI-54026, a 2,3-benzodiazepine compound, which exhibited low anity to AMPA receptors in spreading depression model on chick retina (Tarnawa and Vizi, 1998), was ine€ective in [3H]GABA release inhibition (Table 2). In contrast to the negative modulators of AMPA receptors, the competitive antagonist NBQX (0.1 mM) did not alter the release of [3H]GABA from superfused rat striatal slices (Table 2).

Fig. 3. E€ects of NBQX and GYKI-53784 on AMPA-induced [3H]GABA release from striatal slices of the rat. Slices from rat striatum were prepared, loaded with [3H]GABA and superfused with Krebs-bicarbonate bu€er containing aminooxyacetic acid and nipecotic acid (0.1 mM). AMPA (0.1±3 mM) was added to striatal slices from fractions 8 to 10 (9 min) to evoke [3H]GABA release (*). NBQX (0.01 mM, R) or GYKI53784 (0.01 mM, r) were added to the striatal slices 9 min before addition of AMPA for six consecutive fractions. Cyclothiazide (0.03 mM) was added 30 min before collection started and maintained throughout the experiment. Control [3H]GABA release was 0.04 2 0.01, 0.04 2 0.01 and 0.0220.01% of content for AMPA, AMPA plus NBQX and AMPA plus GYKI-53784-treated groups, respectively, n = 4. ANOVA followed by the Dunnett's test, F(4,14)=68.553, P < 0.01 for AMPA-, F(4,15)=26.424, P < 0.01 for AMPA plus NBQX- and F(4,13)=6.192, P < 0.01 for AMPA plus GYKI-53784-treated groups, respectively, P < 0.05, P < 0.01 when compared to AMPA-treated group. Values shown are the mean2SEM of three to four experiments.

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

39

Fig. 4. Inhibition of [3H]GABA release by GYKI-53784 (w) in superfused rat striatal slices, * indicates control experiments. Rat striatal slices were prepared, loaded with [3]GABA and superfused with Krebs-bicarbonate bu€er containing aminooxyacetic acid and nipecotic acid (0.1 mM). GYKI-53784 was added to the superfusion bu€er from fractions 5 to 10 (18 min) in a concentration of 0.1 mM. The average basal [3H]GABA release in fractions 1, 2 and 3 in control and GYKI-53784-treated groups were 0.6120.08 and 0.5320.03% of content (P > 0.50), respectively and were taken as 100%. Student t-statistics for two means, P < 0.05 when compared to the corresponding control values. Values shown are the mean2SEM of 12 in control experiments and of four in GYKI-53784-treated group.

3.5. Lack of interaction between cyclothiazide and 2,3benzodiazepine recognition sites on AMPA receptors When added to the perfusion bu€er, GYKI-53784 (0.1 mM) reduced [3H]GABA out¯ow from rat striatal slices and this inhibition was not a€ected by cyclothiaTable 2 The inhibitory e€ect of competitive and non-competitive AMPA antagonists on [3H]GABA release from rat striatal slicesa Compounds

Concentration (mM)

Inhibition of [3H]GABA release (% of content)

Control

± 0.01 0.03 0.10 0.10 0.03 0.10 0.30 0.10 0.10

0.0920.02 0.1720.03 0.3620.08 0.8220.06 0.3820.08 0.2620.02 0.5720.04 0.7320.07 0.2020.04 0.0920.01

GYKI-53784 GYKI-53785 GYKI-52466 GYKI-54026 NBQX

a Slices from rat striatum were prepared, loaded with [3H]GABA and superfused with Krebs-bicarbonate bu€er containing aminooxyacetic acid and nipecotic acid (0.1 mM). The AMPA receptor antagonists were added to the slices from fractions 5 to 10 (18 min) then washed out. ANOVA followed by the Dunnett's multiple comparison test, F(9,36)=17.106, P < 0.01, P < 0.05, P < 0.01 when compared to control. Values shown are the mean2SEM of three to eight experiments.

zide. The GYKI-53784-induced [3H]GABA release inhibition was 0.8220.06 in the absence and it was 0.74 2 0.08% of content in the presence of 0.03 mM cyclothiazide (P > 0.50, n = 4 and 8). In addition, cyclothiazide (0.03 mM) also failed to modify the inhibitory e€ect of 0.3 mM GYKI-52466: the inhibition of [3H]GABA release was 0.73 2 0.07 and 0.74 2 0.05% of content when GYKI-52466 and GYKI52466 plus cyclothiazide were added, respectively (P > 0.90, n = 4 and 4).

4. Discussion In the present study slices from rat striatum were incubated with [3H]GABA in the presence of b-alanine to block initial uptake into glial cells, superfused with a Krebs-bicarbonate bu€er containing aminooxyacetic acid to minimize the formation of GABA metabolites and AMPA was added to stimulate basal [3H]GABA out¯ow. Using this approach, we observed that: (1) AMPA induces GABA release from the vesicular pool; (2) cyclothiazide potentiates, NBQX and GYKI-53784 antagonize this e€ect by competitive and non-competitive fashion; (3) no interaction occurs between the positive and negative allosteric binding sites; and (4) AMPA receptors regulating GABA release in the striatum may be located extrasynaptically in sensitized

40

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

state and endogenous glutamate may regulate them in phasic manner. 4.1. Mechanism of AMPA depolarization-induced GABA release in striatum The selective glutamate receptor agonist AMPA dose-dependently enhanced the basal release of [3H]GABA from superfused striatal slices of the rat. This e€ect of AMPA may be explained by inducing depolarization of cell membrane as monovalent ions ¯ow through the receptor-channel complex upon AMPA receptor stimulation (Hollmann and Heinemann, 1994; Seeburg, 1993). The increase in Na+ in¯ux via AMPA channels secondarily leads to rise of intracellular Ca2+ concentration by opening of voltage-sensitive Ca2+ channels (Hoyt et al., 1995) and by direct Ca2+ entry through those AMPA receptors which are permeable for Ca2+ (Bettler and Mulle, 1995; Kardos, 1999; Pellegrini-Giampietro et al., 1997). The importance of the rise of [Ca2+]i concentration in AMPA response was shown by removing Ca2+ from the superfusion bu€er which led to decrease of AMPA-induced [3H]GABA release. These data strongly suggest that the AMPA-induced [3H]GABA release in neostriatum is an external Ca2+dependent, vesicle-originated exocytotical process (Garcia et al., 1995). The role of Ca2+ in the AMPAinduced GABA release is, however, controversial. Galli et al. (1992) reported that the e€ect of AMPA on GABA release is abolished in the absence of Ca2+. On the contrary, GABA e‚ux evoked by quisqualate, another agonist ligand for AMPA receptors, was completely Ca2+-independent from chick retina (Campochiaro et al., 1985) and in addition, quisqualate was reported to promote GABA release in Ca2+-free conditions from ®sh retina (Kato et al., 1985). Experiments involved the GABA transporter inhibitor nipecotic acid further con®rmed the vesicular origin of AMPA-induced GABA release. Nipecotic acid has higher anity to GABA transport system than does GABA itself and it is also a substrate for GABA transporter (Krogsgaard-Larsen, 1980; Storm-Mathisen et al., 1983). The fact, that nipecotic acid stimulates basal [3H]GABA release, indicates that an active GABA transporter may be operative even in superfused striatal slices (Bernath and Zigmond, 1988; Harsing and Zigmond, 1997). It is generally accepted that blockade of neurotransmitter reuptake by transporter inhibitors leads to increase the release originated from the vesicular pool whereas it usually reduces non-exocytotic e‚ux from the cytoplasmic pool (Bernath, 1992). Our observation that nipecotic acid potentiated AMPA-induced [3H]GABA release further supports that the pool which release is originated from is primarily vesicular and not cytoplasmic.

Although AMPA depolarization-induced GABA release was totally Ca2+-dependent when the GABA transporter was active, removal of Ca2+ from the superfusion bu€er did not in¯uence the AMPAinduced [3H]GABA release when the GABA carrier was blocked by nipecotic acid. The classical exocytosis mechanism for many neurotransmitters requires the presence of Ca2+, although external Ca2+-independent vesicular release has also been reported (Knight et al., 1989). It is likely that in the absence of Ca2+, Na+ is a putative mediator between membrane depolarization and transmitter secretion (cf. Adam-Vizi, 1992; Bernath, 1992; Vizi and Sperlagh, 1999). The role of Na+ was further supported by the fact that, in striatal cell cultures, GABA release was stimulated in the absence of Ca2+ by drugs known to increase intracellular Na+ concentration (Pin and Bockaert, 1989). Since GABA transporter acts as a symport transporting one zwitterionic GABA molecule with 2 Na+ ions and 1 Clÿ ion through the presynaptic membrane (Mager et al., 1998), addition of the GABA transporter substrate nipecotic acid may induce rise of [Na+]i and concomitant depolarization of neural membrane. As a consequence, increased inward ¯ux of Na+ in response to nipecotic acid could release Ca2+ from intracellular pools through a Na+/Ca2+ exchange system (Berride and Irvine, 1984; Miller, 1985; Scho€elmeer et al., 1988). An additional source of Ca2+ required for vesicular release of GABA can be the increase of inositol phospholipid hydrolysis after activation of AMPA receptors (Sugiyama et al., 1987; Weiss, 1988). A further source of rise in [Ca2+]i is those AMPA receptors lacking GluR2 subunit and are attributed to Ca2+ permeability (Hollmann et al., 1991). Thus, the AMPA-induced response may involve several intracellular mechanisms which may lead to increase intracellular free Ca2+ concentration and vesicle-originated GABA release may occur even in the absence of external Ca2+. 4.2. E€ects of positive and negative allosteric modulators on AMPA-induced [3H]GABA release in striatum AMPA receptor stimulation results in an increase of neurotransmitter release as it was shown in cell cultures, brain slices or in vivo microdialysis (Giovannini et al., 1998; Jin and Fredholm, 1994; Pin et al., 1989, Weiss, 1988). These experiments demonstrated that AMPA induces release of GABA (Galli et al., 1992; Garcia et al., 1995; Giovannini et al., 1995; Pin et al., 1989; Weiss et al., 1990), acetylcholine (Anderson et al., 1994; Giovannini et al., 1998; Jin, 1998), dopamine (Antonelli et al., 1997; Jin, 1998; Maione et al, 1995; Sakai et al., 1997), glutamate (Barnes et al., 1994; Chittajallu et al., 1996; Vizi and Kiss, 1998), norepinephrine (Cowen and Beart, 1998; Desai et al., 1995;

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

Schoepp et al., 1995) or serotonin (Tao et al., 1997; Whitton et al., 1994). The broad stimulatory e€ect of AMPA on neurotransmitter systems corresponds with a wild distribution of AMPA receptors in the CNS (Cotman et al., 1987). The [3H]GABA-releasing e€ect of AMPA observed from superfused striatal slices exhibited characteristics of receptor-ion channel desensitization as the evoked release of [3H]GABA was weak and gradually decreased even if the presence of AMPA was maintained for a longer time. When the [3H]GABA-releasing e€ect of AMPA was measured in the presence of low molar concentrations of cyclothiazide, a potentiation was found. This positive interaction is in accordance with the previously well characterized phenomenon that rapid desensitization of AMPA receptors can be e€ectively inhibited by the benzothiadiazine compound cyclothiazide (Zorumski et al., 1993). Although the potentiation by cyclothiazide of AMPA-induced release of various neurotransmitters was extensively studied (Barnes et al., 1994; Chittajallu et al., 1996; Cowen and Beartet, 1998; Jin, 1998; Maione et al., 1995; Patel and Croucher, 1997; Tao et al., 1997; Whitton et al., 1994), this interaction has not been demonstrated for the release of [3H]GABA. The increase of AMPA-induced [3H]GABA e‚ux observed in the presence of cyclothiazide may be due to receptor sensitization elicited by this drug. The sensitization of AMPA response by cyclothiazide was more pronounced at higher AMPA concentrations indicating by the di€erences in the evoked [3H]GABA release measured in the presence and absence of cyclothiazide. This ®nding may point on the importance of the rate of receptor occupation by the agonist in development of desensitization. In contrast to AMPA potentiation, cyclothiazide alone did not exert any e€ect on [3H]GABA release in rat striatal slices. In vitro and in vivo experiments indicate that positive modulators, on their own right, are able to in¯uence the release of some but not all neurotransmitters. Thus, it has been reported that cyclothiazide or diazoxide, another compound positively modulates AMPA response, are capable of increasing [3H]norepinephrine release from rat hippocampal slices (Cowen and Beart, 1998) or extracellular serotonin and dopamine concentrations in rat hippocampus in vivo (Maione et al., 1995; Whitton et al., 1994). Cyclothiazide was, however, without e€ect on [3H]glutamate release from rat hippocampal synaptosomes (Barnes et al., 1994) or on [3H]D-aspartate release from rat forebrain slices (Patel and Croucher, 1997). In addition, cyclothiazide also was without e€ect on [3H]norepinephrine release in rat hippocampal slices (Desai et al., 1995) and on [3H]arachidonic acid release measured from cultured mouse striatal neurons (Williams and Glowinski, 1996). Independently whether

41

cyclothiazide or diazoxide altered neurotransmitter release by themselves, positive modulators potentiated the neurotransmitter releasing e€ects of AMPA in these experiments. The fact, that positive modulators are able to induce transmitter release in some but not in all experimental conditions, suggests that certain AMPA receptors are in desensitized state while others are sensitized to the endogenous agonist glutamate. Since cyclothiazide in our experiments was without e€ect on [3H]GABA release in superfused striatal slices of the rat, AMPA receptors located on GABAergic neurons in this brain area may be in sensitized state. In experimental conditions in which AMPA receptors are sensitized and thus positive modulators do not alter neurotransmitter release by themselves, 2,3-benzodiazepines may shift AMPA receptors to desensitized state. This may explain our ®ndings that GYKI-53784 and GYKI-52466, compounds belong to negative allosteric modulators of AMPA receptors (cf. Tarnawa and Vizi, 1998), were capable of decreasing [3H]GABA release from superfused striatal slices. This e€ect of GYKI-53784 was stereoselective and its pharmacologically inactive analog possessing amino group in ortho position (GYKI-54026) failed to elicit inhibition of basal [3H]GABA out¯ow. Although the e€ect of GYKI-52466 in neurotransmitter release studies has been investigated previously in other studies as well, there appeared to be only a trend in reduction of transmitter out¯ow in the presence of this compound (Patel and Croucher, 1997). It may be worthwhile to point out that the lipophilic 2,3-benzodiazepines exhibit slow penetration speed into superfused brain slices (Tarnawa and Vizi, 1998), therefore the exposition time of tissues to these compounds may be critical for detection of their actions. Besides agonists and competitive antagonists which in¯uence AMPA receptor function acting on the same binding sites, ligands acting on allosteric modulatory sites also possess principle role in modulation of the receptor-ion channel complex. Positive and negative allosteric modulators exert opposite e€ects on AMPA receptors, although their actions may not necessarily be mediated by interactions with identical binding sites (Johansen et al., 1995; Kessler et al., 1996 but see Palmer and Lodge, 1993; Zorumski et al., 1993). Our ®nding, that the inhibitory e€ect of 2,3-benzodiazepines on [3H]GABA release was not a€ected by concomitant administration of cyclothiazide, further supports the contention that the non-competitive AMPA receptor antagonists act at a site distinct from that binds cyclothiazide. 4.3. Antagonism of AMPA depolarization-induced [3H]GABA release Both NBQX and GYKI-53784 antagonized the

42

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

stimulatory e€ect of AMPA on [3H]GABA release in superfused striatal slices of the rat, although the two antagonists acted di€erently. NBQX caused a parallel rightwards shift in the concentration-response curve of AMPA without changing the maximal e€ect. On the contrary, the inhibitory e€ect of GYKI-53784 on AMPA-induced [3H]GABA release was not overcome by increasing concentrations of the agonist and a profound suppression of maximal AMPA response was observed. Corresponding to the competitive and noncompetitive receptor antagonism, a pA2 value of 5.08 for NBQX and a pD2' value of 5.44 for GYKI-53784 against AMPA were calculated in the [3H]GABAreleasing experiments. These data indicate that the antagonistic activity of GYKI-53784 is about two-fold higher than that of NBQX in inhibiting AMPAmediated [3H]GABA release. The antagonistic potencies of NBQX and 2,3-benzodiazepines were compared in other studies as well. In a patch clamp study of hippocampal neurons, a six-fold greater activity of NBQX compared to GYKI-52466 against AMPA-evoked response was reported (Parsons et al., 1994). In addition, GYKI-53784 showed a 10-fold higher activity than GYKI-52466 in blocking AMPA receptors in chicken retina spreading depression model (Tarnawa and Vizi, 1998). It has been reported that the competitive AMPA receptor antagonists NBQX and CNQX decreased extracellular dopamine in hippocampus (Maione et al., 1995) and striatum (Sakai et al., 1997). In addition, CNQX also inhibited extracellular serotonin concentrations in rat hippocampus (Whitton et al., 1994). CNQX was, however, without e€ect on [3H]glutamate e‚ux from hippocampal synaptosomes (Barnes et al., 1994) or on [3H]dopamine release in rat striatal slices (Jin, 1998). NBQX was also without e€ect on [3H]Daspartate release from rat forebrain slices (Patel and Croucher, 1997) or on [3H]norepinephrine release in hippocampal preparation (Cowen and Beart, 1998). Furthermore, DNQX, another AMPA receptor antagonist, was found to be ine€ective on extracellular serotonin in raphe nuclei (Tao et al., 1997). Independently whether or not competitive AMPA antagonists altered neurotransmitter release on their own, these compounds inhibited the neurotransmitter-releasing e€ect of AMPA in the various reports cited above. The fact, that competitive AMPA antagonists inhibited neurotransmitter release in some but not in all experiments, suggests that certain AMPA receptors may be, while others are not, tonically in¯uenced by the endogenous agonist glutamate. It is interesting to point out that CNQX also blocked the e€ects of diazoxide when this positive modulator enhanced neurotransmitter release on its own right (Maione et al., 1995; Whitton et al., 1994). Therefore, we speculated that AMPA receptors, which are in desensitized state (i.e. positive modulators

are e€ective), are also tonically in¯uenced by the endogenous agonist glutamate. On the other hand, AMPA receptors in sensitized state (i. e. positive modulators are ine€ective) may be under the control of phasic glutamatergic tone. AMPA receptors, which stimulate [3H]GABA release from striatal slices in our experiments, may belong to the latter type. 4.4. The dual state of AMPA receptors and its possible functional signi®cance Although the great majority of GABAergic neurons in the neostriatum are projection neurons, GABAergic interneurons in a smaller number are also present (Bolam and Bennett, 1995). Both GABAergic interneurons and projection neurons possess AMPA receptors (Kita, 1996), however, the receptor subtypes they express may be di€erent (Kwok et al., 1997; Watkins and Olverman, 1987). Postsynaptic AMPA receptors regulating GABA release in the striatum may be located within glutamatergic synapses or extrasynaptically on neural membranes of GABAergic cells (cf. Vizi, 1984, 2000). Glutamate released from the nerve terminals reaches high concentrations within the synaptic gap and may exert tonic regulation on AMPA receptors. Competitive AMPA receptor antagonists suspend tonic regulation of synaptic AMPA receptors by displacing endogenous glutamate from the receptor-ion channel complex. Consequently, they exert opposite pharmacological e€ects. On the contrary, glutamate released from the nerve terminals is present only in low concentration in the extrasynaptic space and AMPA receptors located out of the synapse may express higher sensitivity to the endogenous agonist. Because of the low glutamate concentration in the extracellular space, AMPA receptors located by a distance from the synaptic cleft may be regulated in a more phasic fashion and competitive AMPA receptor antagonists do not in¯uence transmitter release by themselves. Whether glutamate is present in high or low concentrations at the vicinity of AMPA receptors also determines the actual sensitized or desensitized state of the receptors. Those located within the synapse are occupied at a higher rate by glutamate and receptor desensitization is induced. On the other hand, extrasynaptic AMPA receptors are only partially occupied by endogenous glutamate and the relatively low agonist concentration shifts the receptors to sensitized state. There may be a relation between the rate of sensitization of AMPA receptors and their role in regulation of neurotransmitter release. Positive modulators acting on desensitized AMPA receptors may shift them into sensitized state and thereby increase AMPA receptormediated neurotransmitter release (Maione et al., 1995; Whitton et al., 1994). On the contrary, negative

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

AMPA receptor modulators of the 2,3-benzodiazepine class may exert opposite e€ect by shifting AMPA receptors from sensitized to desensitized form and a concomitant inhibition of neurotransmitter release may occur. The molecular basis for the sensitized-desensitized state may be the ¯ip/¯op variants of AMPA receptors (Partin and Mayer, 1996). In conclusion, AMPA receptors may be located in the synapse or the extrasynaptic space mediating tonic or phasic regulation of neurotransmitter release. AMPA receptors involved in the tonic regulation of neurotransmitter release may be in desensitized state and positive modulators acting on these receptors induce neurotransmitter release. When glutamate concentration is low, phasic in¯uence on neurotransmitter release is more probable and AMPA receptors are shifted into sensitized state. 2,3-benzodiazepines, the negative modulators of AMPA receptors, may inhibit neurotransmitter release by acting on sensitized AMPA receptors.

Acknowledgements A preliminary report of these ®ndings was presented at the 28th Annual Meeting of the Society for Neuroscience held in Los Angeles, CA, USA (Harsing et al., 1998). This research was supported in part by the Research Council for Health Sciences, Hungarian Ministry of Health and Welfare (ETT-123/96/99) and the Hungarian Science Research Fund (T-025060). The authors wish to thank Ms Zsuzsa Major for her technical assistance.

References Adam-Vizi, V., 1992. External Ca2+-independent release of neurotransmitters. J. Neurochem. 58, 395±405. Anderson, J.J., Kuo, S., Chase, T.N., 1994. Endogenous excitatory amino acids tonically stimulate striatal acetylcholine release through NMDA but not AMPA receptors. Neurosci. Letters 176, 264±268. Antonelli, T., Govoni, B.M., Bianchi, C., Beani, L., 1997. Glutamate regulation of dopamine release in guinea pig striatal slices. Neurochem. Int. 30, 203±209. Ariens, E.J., Van Rossum, J.M., 1957. pDx, pAx pD'x values in the analysis of pharmacodynamics. Arch. int. Pharmacodyn. 110, 275±299. Barnes, J.M., Dev, K.K., Henley, J.M., 1994. Cyclothiazide unmasks AMPA-evoked stimulation of [3H]-L-glutamate release from rat hippocampal synaptosomes. Br. J. Pharmacol. 113, 339±341. Bernath, S., Zigmond, M.J., 1988. Characterization of [3H]GABA release from striatal slices: evidence for a calcium-independent process via the GABA uptake system. Neuroscience 27, 563±570. Bernath, S., 1992. Calcium-independent release of amino acid neurotransmitters: fact or artifact? Progress Neurobiol. 38, 57±91. Berride, M.J., Irvine, R.F., 1984. Inositol triphosphate: a novel sec-

43

ond messenger in cellular signal transduction. Nature 312, 315± 321. Bettler, B., Mulle, C., 1995. Review: Neurotransmitter receptors II. AMPA and kainate receptors. Neuropharmacology 34, 123±139. Bleakman, D., Lodge, D., 1998. Neuropharmacology of AMPA and kainate receptors. Neuropharmacology 37, 1187±1204. Bolam, J.P., Bennett, B.D. 1995. Microcircuitry of the neostriatum. In: Ariano, M.A., Surmeier, D.J. (Eds.), Molecular and Cellular Mechanisms of Neostriatal Function. Landes Company, Austin, TX, pp. 1±19. Campochiaro, P., Ferkany, J.W., Coyle, J.T., 1985. Excitatory amino acid analogs evoke release of endogenous amino acids and acetylcholine from chick retina in vitro. Vision Res. 25, 1375± 1386. Chittajallu, R., Vignes, M., Dev, K.K., Barnes, J.M., Collingridge, G.L., Henley, J.M., 1996. Regulation of glutamate release by presynaptic kainate receptors in the hippocampus. Nature 379, 78± 81. Cotman, C., Monaghan, D., Ottersen, O.P., Storm-Mathisen, J., 1987. Anatomical organization of excitatory amino acid receptors and their pathways. Trends Neurosci. 10, 273±279. Cowen, M.S., Beart, P.M., 1998. Cyclothiazide and AMPA receptor desensitization: analyses from studies of AMPA-induced release of [3H]-noradrenaline from hippocampal slices. Br. J. Pharmacol. 123, 473±480. Desai, M.A., Burnet, J.P., Ornstein, P.L., Schoepp, D.D., 1995. Cyclothiazide acts at a site on the a-amino-3-hydroxy-5-methyl-4isoxazole propionic acid receptor complex that does not recognize competitive or noncompetitive AMPA receptor antagonists. J. Pharm. Exp. Ther. 272, 38±43. Donevan, S.D., Rogawski, M.A., 1993. GYKI 52466, a 2,3-benzodiazepine, is a highly selective, noncompetitive antagonist of AMPA/kainate receptor responses. Neuron 10, 51±59. Donevan, S.D., Rogawski, M.A., 1998. Allosteric regulation of aamino-3-hydroxy-5-methyl-4-isoxazole-propionate receptors by thiocyanate and cyclothiazide at a common modulatory site distinct from that of 2,3-benzodiazepines. Neuroscience 87, 615±629. Dube, L., Smith, A.D., Bolam, J.P., 1988. Identi®cation of synaptic terminals of thalamic or cortical origin in contact with distinct medium size spiny neurons in the rat neostriatum. J. Comp. Neurol. 267, 455±471. Fonnum, F., Storm-Mathisen, J., Divac, I., 1981. Biochemical evidence for glutamate as neurotransmitter in corticostriatal and corticothalamic ®bers in rat brain. Neuroscience 6, 863±873. Galli, T., Desce, J., Martaud, F., Kemel, M.-L., Cheramy, A., Glowinski, J., 1992. Modulation of GABA release by a-amino-3hydroxy-5-methylisoxazole-4-propionate and N-methyl-D-aspartate receptors in matrix-enriched areas of the rat striatum. Neuroscience 50, 769±780. Garcia, Y., Ibarra, C., Ja€e, E.H., 1995. NMDA and non-NMDA receptor-mediated release of [3H]GABA from granule cell dendrites of rat olfactory bulb. J. Neurochem. 54, 662±669. Giovannini, M.G., Camilli, F., Mundula, A., Bianchi, L., Colivicchi, M.A., Pepeu, G., 1995. Di€erential regulation by N-methyl-Daspartate and non-N-methyl-D-aspartate receptors of acetylcholine release from the rat striatum in vivo. Neuroscience 65, 409±415. Giovannini, M.G., Rakovska, A., Della, Corte L., Bianchi, L., Pepeu, G., 1998. Activation of non-NMDA receptors stimulates acetylcholine and GABA release from dorsal hippocampus: a microdialysis study in the rat. Neuroscience Letters 243, 152±156. Harsing Jr, L.G., Sershen, H., Lajtha, A., 1992. Dopamine e‚ux from striatum after chronic nicotine: evidence for autoreceptor desensitization. J. Neurochem. 59, 48±54. Harsing, L.G., Zigmond, M.J., 1997. In¯uence of dopamine on GABA release in striatum: evidence for D1±D2 interactions and non-synaptic in¯uences. Neuroscience 77, 419±429.

44

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45

Harsing, L.G., Zigmond, M.J., 1998. Postsynaptic integration of cholinergic and dopaminergic signals on medium-sized GABAergic projection neurons in the neostriatum. Brain Res. Bull. 45, 607±613. Harsing Jr, L.G., Somogyi, P., Hamori, T., Solyom, S., 1998. Glutamate-GABA interaction in neostriatum: the role of AMPA receptors in the regulation of [3H]GABA release. Society for Neuroscience 24, 824.19. Hollmann, M., Hartley, M., Heinemann, S., 1991. Ca2+ permeability of KA-AMPA gated glutamate receptor channels depends on subunit composition. Science 252, 851±853. Hollmann, M., Heinemann, S., 1994. Cloned glutamate receptors. Ann. Rev. Neurosci. 17, 31±108. Hoyt, K.R., Raydev, S., Fattman, C.L., Reynolds, I.J., 1995. Cyclothiazide modulates AMPA receptor mediated increases in intracellular free Ca2+ and Mg2+ in cultured neurons from rat brain. J. Neurochem. 64, 2049±2056. Iversen, L.L., Kelly, J.S., 1975. Uptake and metabolism of g-aminobutyric acid by neurones and glial cells. Biochem. Pharmac. 24, 933±938. Jin, S., 1998. AMPA and kainate receptors di€erently mediate excitatory amino acid-induced dopamine and acetylcholine release from striatal slices. Neuropharmacology 36, 1503±1510. Jin, S., Fredholm, B.B., 1994. Role of NMDA, AMPA and kainate receptors in mediating glutamate and 4-AP-induced dopamine and acetylcholine release from rat striatal slices. Neuropharmacology 33, 1039±1048. Johansen, T.H., Chaudhary, A., Verdoorn, T.A., 1995. Interactions among GYKI-52466, cyclothiazide and aniracetam at recombinant AMPA and kainate receptors. Mol. Pharmacol. 48, 946± 955. Kardos, J., 1999. Recent advances in GABA research. Neurochem. Internat. 34, 353±358. Kato, S., Negishi, K., Teranishi, T., 1985. Dopamine inhibits calcium-independent [3H]g-aminobutyric acid release induced by kainate and high K+ in ®sh retina. J. Neurochem. 44, 893±899. Kessler, M., Arai, A., Quan, A., Lynch, G., 1996. E€ect of cyclothiazide on binding properties of AMPA-type glutamate receptors: lack of competition between cyclothiazide and GYKI 52466. Mol. Pharmacol. 49, 123±131. Kita, H., 1996. Glutamatergic and GABAergic postsynaptic responses of striatal spiny neurons to intrastriatal and cortical stimulation recorded in slice preparations. Neuroscience 70, 925± 940. Knight, D.E., von Grafenstein, H., Athayde, C.M., 1989. Calciumdependent and calcium-independent exocytosis. Trends Neurosci. 12, 451±458. Krogsgaard-Larsen, P., 1980. Inhibitors of the GABA uptake systems. Molec. Cell. Biochem. 31, 105±121. Kwok, K.H.H., Tse, Y.C., Wong, R.N.S., Yung, K.K.L., 1997. Cellular localization of GluR2/3 and GluR4 glutamate receptor subunits in neurons of the rat neostriatum. Brain Res. 778, 43± 55. Mager, S., Cao, Y., Lester, H.A., 1998. Measurement of transient currents from neurotransmitter transporters expressed in Xenopus oocytes. In: Amara, S.G. (Ed.), Methods in Enzymology, vol. 296. Academic Press, New York, pp. 551±578. Maione, S., Biggs, C.S., Rossi, F., Fowler, R.J., Whitton, P.S., 1995. a-Amino-3-hydroxy-5-methyl-4-isoxazolepropionate receptors modulate dopamine release in rat hippocampus and striatum. Neurosci. Letters 193, 181±184. Maura, G., Gemignani, A., Raiteri, M., 1985. Alpha-2 adrenoceptors in rat hypothalamus and cerebral cortex: functional evidence for pharmacologically distinct subpopulations. Eur. J. Pharmacol. 116, 335±339. Mayer, M.L., Westbrook, G.L., 1987. The physiology of excitatory

amino acids in the vertebrate nervous system. Prog. Neurobiol. 28, 197±276. Miller, D.J., 1985. The mitochondria and cellular calcium. Nature 313, 638. Okada, M., Kohara, A., Yamaguchi, T., 1996. Characterization of YM90 K, a selective and potent antagonist of AMPA receptors, in rat cortical mRNA-injected Xenopus oocytes. Eur. J. Pharmacol. 309, 299±306. Palmer, A.J., Lodge, D., 1993. Cyclothiazide reverses AMPA receptor antagonism of the 2,3-benzodiazepine, GYKI-53655. Eur. J. Pharmacol. 244, 193±194. Parsons, C.G., Grunner, R., Rozental, J., 1994. Comparative patch clamp studies on the kinetics and selectivity of glutamate receptor antagonism by 2,3-dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxaline (NBQX) and 1-(4-amino-phenyl)-4-methyl-7,8-methylendoxy-5H-2,3-benzodiazepine (GYKI-52466). Neuropharmacology 33, 589±604. Partin, K.M., Patneau, D.K., Mayer, M.L., 1994. Cyclothiazide differently modulates desensitization of a-amino-3-hydroxy-5methyl-4-isoxazolepropionic acid receptor splice variants. Mol. Pharmacol. 46, 129±138. Partin, K.M., Mayer, M.L., 1996. Negative allosteric modulation of wild-type and mutant AMPA receptors by GYKI-53655. Mol. Pharmacol. 49, 142±148. Patel, D.R., Croucher, M.J., 1997. Evidence for a role of presynaptic AMPA receptors in the control of neuronal glutamate release in the rat forebrain. Eur. J. Pharmacol. 332, 143±151. Paternain, A.V., Morales, M., Lerma, J., 1995. Selective antagonism of AMPA receptors unmasks kainate receptor-mediated responses in hippocampal neurons. Neuron 14, 185±189. Pellegrini-Giampietro, D.E., Gorter, J.A., Bennett, M.V.L., Zukin, R.S., 1997. The GluR2 (GluR-B) hypothesis: Ca2+-permeable AMPA receptors in neurological disorders. Trends Neurosci. 20, 464±470. Petitet, F., Blanchard, J.-C., Doble, A., 1995. E€ects of non-NMDA receptor modulators on [3H]dopamine release from rat mesencephalic cells in primary culture. J. Neurochem. 64, 1410±1412. Pin, J.-P., Bockaert, T., 1989. Two distinct mechanisms, di€erently a€ected by excitatory amino acids, trigger GABA release from fetal mouse striatal neurons in primary culture. J. Neurosci. 9, 648±656. Pin, J.-P., Van Vliet, B.J., Bockaert, T., 1989. Complex interaction between quisqualate and kainate receptors as revealed by measurement of GABA release from striatal neurons in primary culture. Eur. J. Pharmacol. 172, 81±91. Pin, J.-P., Duvoisin, R., 1995. The metabotropic glutamate receptors structure and functions. Neuropharmacology 34, 1±26. Sakai, K., Akiyama, K., Jashihara, K., Tsuchida, K., Ujika, H., Kuroda, S., Shohmori, T., 1997. AMPA receptors modulate dopamine release in the striatum, as measured by brain microdialysis. Neurochem. Int. 30, 329±336. Schoepp, D.D., Lodge, D., Blekman, D., Leander, J.D., Tizzano, J.P., Wright, R.A., et al., 1995. In vitro and in vivo antagonism of AMPA receptor activation by (3 S, 4aR,6R,8aR)6-[2-(1(2)H-tetrazole-5-yl)ethyl]decahydroisoquinoline-3-carboxylic acid. Neuropharmacology 34, 1159±1168. Scho€elmeer, A.N.M., Hogenboom, F., Mulder, A.H., 1988. Sodium dependent 3H noradrenaline release from rat neocortical slices in the absence of extracellular calcium: presynaptic modulation by m-opioid receptor and adenylate cyclase activation. NaunynSchmiedeberg's Arch. Pharmacol. 338, 548±552. Seeburg, P.H., 1993. The TINS/TiPS Lecture. The molecular biology of mammalian glutamate receptor channels. Trends Neurosci. 16, 359±365. Smith, A.D., Bolam, J.P., 1990. The neuronal network of the basal ganglia as revealed by the study of synaptic connections of identi®ed neurones. Trends Neurosci. 13, 259±265.

L.G. Harsing Jr et al. / Neurochemistry International 37 (2000) 33±45 Storm-Mathisen, J., Leknes, A.K., Bore, A.T., Vaaland, J.L., Edminson, P., Haug, F.-M.S., Otterson, O.P., 1983. First visualization of glutamate and neurones by immunocytochemistry. Nature 301, 517±520. Sugiyama, H., Ito, I., Hirono, C., 1987. A new type of glutamate receptor linked to inositol phospholipid metabolism. Nature 325, 531±533. Tarnawa, I., Molnar, P., Gal, L., Andrasi, F., 1992. Inhibition of hippocampal ®eld potentials by GYKI-52466 in vitro and in vivo. Acta. Physiol. Hung. 79, 163±169. Tarnawa, I., Vizi, E.S., 1998. 2,3-Benzodiazepine AMPA antagonists. Restorative Neurol. Neurosci. 13, 41±57. Tao, R., Ma, Z., Auerbach, S.B., 1997. In¯uence of AMPA/kainate receptors on extracellular 5-hydroxytryptamine in rat midbrain raphe and forebrain. Br. J. Pharmacol. 121, 1707±1715. Vizi, E.S., 1984. Non-Synaptic Interactions Between Neurons: Modulation of Neurochemical Transmission. Pharmacological and Clinical Aspects. Wiley, Chichester, New York. Vizi, E.S., Mike, A., Tarnawa, I., 1996. 2,3-Benzodiazepines (GYKI52466 and analogs): negative allosteric modulators of AMPA receptors. CNS Drug Rev. 2, 91±126. Vizi, E.S., Kiss, J.P., 1998. Neurochemistry and pharmacology of the major hippocampal transmitter systems: synaptic and nonsynaptic interactions. Hippocampus 8, 566±607. Vizi, E.S., Sperlagh, B., 1999. Separation of carrier mediated and vesicular release of GABA from rat brain slices. Neurochem. Internat. 34, 407±413. Vizi, E.S., 2000. Role of high-anity receptors and membrane trans-

45

porters in nonsynaptic communication and drug action in the CNS. Pharmacol. Reviews 55, in press. Vyklicky Jr, L., Patneau, D.K., Mayer, M.L., 1991. Modulation of excitatory synaptic transmission by drugs that reduce desensitization at AMPA/kainate receptors. Neuron 7, 971±984. Watkins, J.C., Olverman, H.J., 1987. Agonists and antagonists for excitatory amino acid receptors. Trends Neurosci. 10, 265± 280. Weiss, S., 1988. Excitatory amino acid-evoked release of g[3H]aminobutyric acid from striatal neurons in primary culture. J. Neurochem. 51, 435±441. Weiss, S., Kemp, D.E., Bauce, L., Tse, F.W.Y., 1990. Kainate receptors coupled to the evoked release of [3H]-g-aminobutyric acid from striatal neurons in primary culture: potentiation by lithium ions. Mol. Pharmacol. 38, 229±236. Whitton, P.S., Maione, S., Biggs, C.S., Fowler, L.J., 1994. Tonic desensitization of hippocampal a-amino-3-hydroxy-5-methyl-4isozazoleproprionic acid receptors regulates 5-hydroxytryptamine release in vivo. Neuroscience 63, 945±948. Williams, R.J., Glowinski, J., 1996. Cyclothiazide unmasks an AMPA-evoked release of arachidonic acid from cultured striatal neurones. J. Neurochem. 67, 1551±1558. Yamada, K.A., Turetsky, D.M., 1996. Allosteric interactions between cyclothiazide and AMPA/kainate receptor antagonists. Br. J. Pharmacol. 117, 1663±1672. Zorumski, C.H., Yamada, K.A., Price, M.T., Olney, J.W., 1993. A benzodiazepine recognition site associated with the non-NMDA glutamate receptor. Neuron 10, 61±67.