Functional coupling of gamma-aminobutyric acid (GABA)A and benzodiazepine type II receptors: analysis using purified GABA/benzodiazepine receptor complex from bovine cerebral cortex

Possible functional coupling between gamma-aminobutyric acid (GABA) and benzodiazepine receptors was examined using a purified GABA/benzodiazepine receptor complex. The purified receptor complex was obtained by 1012-S-acetamide adipic hydrazide Sepharose 4B affinity column chromatography, following the solubilization of synaptic membrane from the bovine cerebral cortex with Nonidet P-40. The binding of [3H]GABA to the purified GABA receptor was displaced significantly by muscimol and bicuculline, GABAA receptor agonists and antagonists, respectively, but not by baclofen, a GABAB receptor agonist. On the other hand, the binding of [3H]flunitrazepam to the purified benzodiazepine receptor was found to be displaced by microM ranges of CL 218,872, which is known to be sensitive to the benzodiazepine type II receptor. Furthermore, it was found that the binding of [3H]muscimol to these purified GABAA receptors was enhanced by benzodiazepines, while the binding of [3H]flunitrazepam to these benzodiazepine type II receptors was increased by GABA receptor agonists. These enhancements by GABA agonists and benzodiazepine agonists were found to be blocked by bicuculline and a benzodiazepine receptor antagonist, Ro15-1788, respectively. These results strongly suggest that the purified receptor may consist of GABAA and benzodiazepine type II receptors and possess a functional coupling of these sites, as shown in cerebral synaptic membranes.     

Functionally relevant gamma-aminobutyric acidA receptors: equivalence between receptor affinity (Kd) and potency (EC50)?

Many neurotransmitter receptors bind agonists with high affinity (Kd in the nanomolar range), whereas micromolar concentrations of the same agonists are required to elicit a functional effect. We have identified low affinity agonist binding sites for the gamma-amino-butyric acidA (GABAA) receptor-chloride channel under conditions normally used in 36Cl- uptake assays (a measure of receptor function). The GABAA agonist [3H]muscimol bound to a population of receptors with a Kd (2 microM) similar to its EC50 value for 36Cl- uptake. Binding was inhibited by the GABA agonist 4,5,6,7-tetrahydroisoxazolo[5,4-c]pyridin-3-ol and by the GABA antagonist bicuculline methiodide. A reduction in the number of [3H]muscimol binding sites (Bmax) by a thiol-modifying reagent produced a corresponding decrease in the Emax for muscimol. The benzodiazepine diazepam enhanced the potency of muscimol in ion flux experiments but did not alter the affinity of [3H]muscimol binding sites. We propose that benzodiazepines enhance GABAergic function by increasing receptor-ion channel coupling, rather than by increasing GABAA receptor affinity. These studies suggest that the study of physiologically relevant (low affinity) binding sites is necessary when examining regulation of receptors by cellular processes, drugs, and disease.     

Gamma-aminobutyric acid enhancement of potassium-stimulated release of [3H]norepinephrine by multiple mechanisms in rat cortical slices

Gamma-Aminobutyric acid (GABA) and GABAA agonists enhance stimulated release of [3H]norepinephrine [( 3H]NA) in several regions of the rat brain. In this study, the mechanisms by which GABA and GABAergic agonists augment potassium-stimulated release of [3H]NA from rat frontal cortical slices were examined. GABA enhanced potassium-stimulated [3H]NA release, but did not alter release of [3H]NA evoked by the calcium ionophore A23187, 10(-5) M, either in the presence or the absence of extracellular calcium. The effect of GABA on potassium-stimulated [3H]NA release was apparently reduced by the GABAA antagonist bicuculline methiodide, 10(-4) M, and by the selective inhibitor of GABA uptake SKF 89976A, 10(-5) M, but was abolished only when bicuculline methiodide and SKF 89976A were present in combination. The GABAA agonist muscimol enhanced potassium-stimulated release of [3H]NA in a manner similar to GABA. In addition, nipecotic acid, a substrate for GABA uptake, enhanced potassium-stimulated [3H]NA release. Thus, GABA appears to enhance potassium-stimulated [3H]NA release by acting upon both GABA uptake and GABAA receptors. The GABAA receptors involved in this effect may be a subtype of GABAA receptors since they are not modulated by benzodiazepines. These results support the involvement of the GABA uptake carrier and the GABAA receptor in mediating the enhancement by GABA of potassium-stimulated [3H]NA release in the cortex of the rat.     

Histamine H(3) receptor-mediated inhibition of depolarization-induced, dopamine D(1) receptor-dependent release of [(3)H]-gamma-aminobutryic acid from rat striatal slices

1. A study was made of the regulation of [(3)H]-gamma-aminobutyric acid ([(3)H]-GABA) release from slices of rat striatum by endogenous dopamine and exogenous histamine and a histamine H(3)-agonist. Depolarization-induced release of [(3)H]-GABA was Ca(2+)-dependent and was increased in the presence of the dopamine D(2) receptor family antagonist, sulpiride (10 microM). The sulpiride-potentiated release of [(3)H]-GABA was strongly inhibited by the dopamine D(1) receptor family antagonist, SCH 23390 (1 microM). Neither antagonist altered basal release. 2. The 15 mM K(+)-induced release of [(3)H]-GABA in the presence of sulpiride was inhibited by 100 microM histamine (mean inhibition 78+/-3%) and by the histamine H(3) receptor-selective agonist, immepip, 1 microM (mean inhibition 81+/-5%). The IC(50) values for histamine and immepip were 1.3+/-0.2 microM and 16+/-2 nM, respectively. The inhibitory effects of histamine and immepip were reversed by the H(3) receptor antagonist, thioperamide, 1 microM. 3. The inhibition of 15 mM K(+)-induced [(3)H]-GABA release by immepip was reversed by the H(3) receptor antagonist, clobenpropit, K(d) 0.11+/-0.04 nM. Clobenpropit alone had no effect on basal or stimulated release of [(3)H]-GABA. 4. Elevated K(+) caused little release of [(3)H]-GABA from striatal slices from reserpinized rats, unless the D(1) partial agonist, R(+)-SKF 38393, 1 microM, was also present. The stimulated release in the presence of SKF 38393 was reduced by 1 microM immepip to the level obtained in the absence of SKF 38393. 5. These observations demonstrate that histamine H(3) receptor activation strongly inhibits the dopamine D(1) receptor-dependent release of [(3)H]-GABA from rat striatum; primarily through an interaction at the terminals of GABA neurones.     

Modulation of calcium uptake and D-aspartate release by GABAB receptors in cultured cerebellar granule cells

(-)Baclofen, a GABAB receptor agonist, and GABA attenuated by 60% the high K+-evoked 45Ca2+ uptake into cultured cerebellar granule cells with an EC50 of 110 +/- 18 nM and 2.4 +/- 0.2 microM, respectively. The attenuation by baclofen of 45Ca2+ uptake was stereospecific and the effect of GABA was unaffected by bicuculline. Moreover, muscimol, a GABAA receptor agonist did not affect the K+-evoked 45Ca2+ uptake. (-)Baclofen and GABA also decreased the K+-evoked and calcium-dependent release of preloaded [3H]D-aspartate from granule cells; however, their potency and efficacy appeared to be less than those for inhibiting the 45Ca2+ uptake. (+)Baclofen and muscimol failed to change this K+-evoked release. The release of [3H]D-aspartate induced by the calcium ionophore A23187 was unaffected by (-)-baclofen. The K+-evoked release of [3H]D-aspartate was effectively inhibited by nimodipine, a voltage sensitive calcium channel blocker. The results suggest that GABAB receptor in cultured cerebellar granule cells plays a crucial role in modulating the uptake of calcium and release of the excitatory transmitter. Moreover, these two effects mediated by GABAB receptor activation may be casually related.     

Ligand-dependent effects of ethanol and diethylether at brain benzodiazepine receptors.

The GABAA receptor chloride channel complex interacts with various categories of sedatives, including the benzodiazepines, and possibly ethanol and volatile general anesthetics. Thus, specific binding of tritiated derivatives of a benzodiazepine antagonist, flumazenil, and an agonist, flunitrazepam, to rat brain membrane fragments was monitored at equilibrium in the presence and absence of anesthetizing concentrations of ethanol and diethylether. Ethanol produced a concentration-dependent inhibition of [3H]flumazenil binding, which was not reversed by the GABAA receptor competitive antagonist bicuculline, but had no effect on [3H]flunitrazepam binding. Both ethanol and diethylether decreased the affinity of the benzodiazepine site for [3H]flumazenil. These data indicate that ethanol and diethylether have GABA-independent effects at the benzodiazepine sites of the GABAA receptor. These findings are inconsistent with a two-state functional model of the benzodiazepine site and, instead, support a model containing a specific, antagonist-favored conformation.     

Intrahippocampal injections of benzodiazepine and muscimol impair working memory but not reference memory of rats in the three-panel runway task.

In a three-panel runway task, the benzodiazepine chlordiazepoxide at 3.2 and 10 mg/kg i.p. significantly increased the number of errors (attempts to pass through two incorrect panels of the three panel-gates at four choice points) in a test of working memory, but it had no effect on errors in a test of reference memory. This effect of 10 mg/kg chlordiazepoxide on working memory was blocked by the benzodiazepine receptor antagonist flumazenil at 10 mg/kg. Intrahippocampal injection of chlordiazepoxide at 10 and 32 micrograms/side significantly increased the number of working memory errors. This effect of intrahippocampal chlordiazepoxide (32 micrograms/side) was attenuated not only by flumazenil at 10 mg/kg but also by the gamma-aminobutyric acid (GABA)A receptor antagonist bicuculline at 3.2 mg/kg. Intrahippocampal injection of the GABAA receptor agonist muscimol at 100 and 320 ng/side also significantly increased working memory errors. Neither chlordiazepoxide nor muscimol affected the number of reference memory errors when injected into the hippocampus at doses up to 32 micrograms/side or 320 ng/side, respectively. These results suggest that activation of the GABAA/benzodiazepine receptor complex in the hippocampus impairs working memory, but does not affect reference memory.     

Studies on [3H]GABA and endogenous GABA release in rat cerebral cortex suggest the presence of autoreceptors of the GABAB type

The presence of autoreceptors for gamma-aminobutyric acid (GABA) in the CNS was reinvestigated using rat cortex synaptosomes prelabeled with [3H]GABA and exposed to GABA by superfusion in the presence of a new GABA uptake inhibitor, N-(4,4-diphenyl-3-butenyl)-nipecotic acid (SK&F 89976A). This compound itself did not increase the basal or the depolarization-evoked release of [3H]GABA. GABA reduced in a concentration-dependent way the release of [3H]GABA evoked by 15 mM K+. The effect was not antagonized by bicuculline, picrotoxin or by the new GABAA antagonist SR 95531. The GABAA agonist muscimol did not affect [3H]GABA release. This was reduced by (-)baclofen (but not by the (+) isomer) and the concentration-inhibition curve of (-)baclofen was superimposable on to that of GABA. Also the K+-evoked release of endogenous GABA was stereoselectively and concentration dependently inhibited by the (-) enantiomer of baclofen. It is concluded that the release of GABA from rat cortical nerve endings may be inhibited through the activation of autoreceptors which appear to belong to the GABAB type.     

Release-regulating autoreceptors of the GABAB-type in human cerebral cortex.

1. The depolarization-evoked release of gamma-aminobutyric acid (GABA) and its modulation mediated by autoreceptors were investigated in superfused synaptosomes prepared from fresh human cerebral cortex. 2. The release of [3H]-GABA provoked by 15 mM K+ from human cortex nerve endings was almost totally (85%) calcium-dependent. 3. In the presence of the GABA uptake inhibitor SK&F 89976A (N-(4,4-diphenyl-3-butenyl)-nipecotic acid), added to prevent carrier-mediated homoexchange, GABA (1-10 microM) decreased in a concentration-dependent manner the K+-evoked release of [3H]-GABA. The effect of GABA was mimicked by the GABAB receptor agonist (-)-baclofen (1-100 microM) but not by the GABAA receptor agonist muscimol (1-100 microM). Moreover, the GABA-induced inhibition of [3H]-GABA release was not affected by two GABAA receptor antagonists, bicuculline or SR 95531 (2-(3'-carbethoxy-2'-propenyl)-3-amino-6-paramethoxy-phenyl-pyr idazinium bromide). 4. (-)-Baclofen also inhibited the depolarization-evoked release of endogenous GABA from human cortical synaptosomes. 5. It is concluded that GABA autoreceptors regulating the release of both newly taken up and endogenous GABA are present in human brain and appear to belong to the GABAB subtype.     

GABAB autoreceptors in rat cortex synaptosomes: response under different depolarizing and ionic conditions

Rat cerebral cortex synaptosomes prelabeled with [3H]gamma-aminobutyric acid [( 3H]GABA) were exposed in superfusion to various concentrations of KCl (9-50 mM). The evoked release of [3H]GABA reached a plateau at about 35 mM KCl. The K+-induced release was Ca2+-dependent, particularly at the lowest K+ concentrations. The GABAB agonist (-)-baclofen concentration dependently inhibited the release of [3H]GABA evoked by K+; this effect decreased with increasing K+ concentration and disappeared at 35 mM KCl. The GABAA agonist muscimol (1-100 microM) was totally ineffective to inhibit the release of [3H]GABA. Veratrine (1-30 microM) induced the release of [3H]GABA and the effect was tetrodotoxin-sensitive. (-)-Baclofen, but not muscimol, decreased the veratrine-induced [3H]GABA release; the GABAB agonist was particularly effective in presence of low concentrations of veratrine (1-3 microM) but the effect disappeared when 30 microM of the alkaloid was used. The inhibitory effect of (-)-baclofen on the release of [3H]GABA evoked by 15 mM KCl was dependent on the concentration of Ca2+: the effect increased as the concentration of Ca2+ was raised, reaching a plateau at 0.6 mM Ca2+. Exogenous GABA, in presence of the GABA uptake blocker SK & F 89976A, inhibited the release of [3H]GABA evoked by K+; this effect was antagonized by phaclofen. The data support the idea that terminal GABA autoreceptors in the rat cerebral cortex are of the GABAB type.     

3-Chloro,4-methoxyfendiline is a potent GABA(B) receptor potentiator in rat neocortical slices

Using grease-gap recording from rat neocortical slices, the GABA(B) receptor agonist baclofen elicited reversible and concentration-dependent hyperpolarizing responses (EC50=18+/-2.3 microM). The hyperpolarizations were antagonised by the GABA(B) receptor antagonist Sch 50911 [(+)-(S)-5,5-dimethylmorpholinyl-2-acetic acid). (+)-N-1-(3-chloro-4-methoxyphenyl)ethyl-3,3-diphenylpropylamine (3-chloro,4-methoxyfendiline; 3-Cl,4-MeO-fendiline) reversibly potentiated baclofen-induced hyperpolarizing responses, which were reduced by Sch 50911, producing leftward shifts of the baclofen concentration-response curves, with a marked increase in the maximal hyperpolarization (EC50=2+/-0.5 microM). In slices preincubated with either [3H]GABA or [3H]glutamic acid, 3-Cl,4-MeO-fendiline (1 microM) potentiated the inhibitory effect of baclofen (2 microM) on the electrically evoked release of [3H]GABA and had a similar effect on the release of [3H]glutamic acid at a concentration of 0.5 microM, without affecting the basal release. These effects were blocked by Sch 50911 (10 microM). Our findings suggest that 3-Cl,4-MeO-fendiline is a potent potentiator of pre- and postsynaptic GABA(B) receptor-mediated functions.     

GABA induces down-regulation of the benzodiazepine-GABA receptor complex in the rat cultured neurons

Cultured neurons from embryonic rat brain display central type benzodiazepine receptors characterized by high-affinity binding of [3H]flunitrazepam which is allosterically enhanced in the presence of gamma-aminobutyric acid (GABA). A 48 h treatment of the cultured neurons with 1 microM diazepam, 0.1 microM clonazepam or 0.1 microM beta-carboline ester derivatives did not change either Bmax or KD values of the [3H]flunitrazepam specific binding. A 48 h incubation in the presence of GABA (1 mM) or muscimol (0.1 mM) induced a 30% decrease of the Bmax value of [3H]flunitrazepam specific binding without change of the KD value. The down-regulation was dependent on GABA concentrations and temperature, and was partially inhibited by bicuculline but not by the benzodiazepine antagonist Ro 15-1788. The other subunits of the benzodiazepine-GABA-chloride channel receptor complex also seemed to be down-regulated by GABA since there was a decrease of the specific binding of [3H]muscimol and [35S]t-butylbicyclophosphorothionate (TBPS) to the GABAA and chloride channel sites respectively. The GABA-induced down-regulation of the GABA-benzodiazepine receptor seems to be selective since the specific binding of ligands to other receptors was not affected. Our results suggests that activation of the low-affinity GABA subunit which is involved in cellular electrophysiological responses, induced the receptor down-regulation.     

Benzodiazepine interactions with GABAA receptors on chick ciliary ganglion neurons

gamma-Aminobutyric acidA (GABAA) receptors on chick ciliary ganglion neurons can be modulated by benzodiazepines and identified by radiolabeled benzodiazepine binding. Enhancement of submaximal GABA responses by benzodiazepines was demonstrated using a multibarrel pipette to construct complete benzodiazepine dose-response curves for single cells in culture. EC50 values of 22 +/- 5 nM, 1.1 +/- 0.3 microM, and 4.6 +/- 0.5 microM were obtained for flunitrazepam, clonazepam, and chlordiazepoxide, respectively. Chlordiazepoxide shifted the GABA dose-response curve to lower GABA concentrations without increasing the maximal response to GABA, demonstrating that benzodiazepines enhance the GABA response by increasing the receptor affinity for GABA. The imidazodiazepine Ro15-1788 potentiated the GABA response with an EC50 of 250 +/- 70 nM, and Ro5-4864 (chlorodiazepam) partially blocked the GABA response both in the presence and absence of chlordiazepoxide. Scatchard analysis of data from binding studies with [3H]flunitrazepam to ganglion membrane homogenates was consistent with the presence of a single class of high affinity sites with a KD of 34 +/- 6 nM and a Bmax of 145 +/- 26 fmol/mg of protein. Several lines of evidence indicated that the sites were associated with GABAA receptors. The KD of [3H]flunitrazepam binding was similar to the EC50 for flunitrazepam modulation of the GABA response. The level of [3H]flunitrazepam binding was enhanced approximately 50% over control levels by GABA. The binding was decreased both by clonazepam and by Ro5-4864 at concentrations similar to those required for the compounds to modulate the GABA response. These studies demonstrate that ciliary ganglion GABAA receptors are similar in major respects to GABAA receptors in the central nervous system but may differ in minor pharmacological properties.     

Ro 15-4513 binding to GABAA receptors: subunit composition determines ligand efficacy.

The bidirectional modulation of ligand binding to benzodiazepine receptors (BzR) by GABA (the "GABA shift") has been widely used to predict ligand efficacy. The present study examined the effects of GABA and muscimol on [3H]Ro 15-4513 binding to "diazepam-insensitive" (DI) and "diazepam-sensitive" (DS) BzR. Neither GABA nor muscimol significantly altered [3H]Ro 15-4513 binding to DI in cerebellum, while both compounds inhibit [3H]Ro 15-4513 binding to cerebellar DS in a concentration-dependent fashion. The maximum reductions in [3H]Ro 15-4513 binding to cerebral cortical and hippocampal membranes elicited by GABA were comparable to those obtained in cerebellar DS, but significantly less than obtained with the full inverse agonist [3H]3-carbomethoxy-beta-carboline. The qualitatively different effect of GABAmimetics on [3H]Ro 15-4513 binding to DS and DI is not species specific since identical effects were obtained in rat and mouse brain. Based on previously established criteria, Ro 15-4513 can be classified as a "GABA-neutral" (antagonist) ligand at DI and "GABA negative" (inverse agonist) at other BzR. These findings suggest that GABAA receptor subunit composition determines not only ligand affinity but also ligand efficacy.     

Chronic agonist exposure induces down-regulation and allosteric uncoupling of the gamma-aminobutyric acid/benzodiazepine receptor complex

Using [3H]flunitrazepam as a probe for the benzodiazepine-sensitive modulator site located on the gamma-aminobutyric acid (GABA)A receptor complex, we have investigated the cellular regulation of the GABAA receptor in neuronal cultures derived from embryonic chick brain. Treatment of cultures with 1 mM GABA for 48 hr causes a reversible 35% decrease in the number of [3H]flunitrazepam binding sites with no change in affinity. The EC50 for chronic GABA-induced down-regulation is 94 microM and the half-time is 25 hr. The effect of GABA is blocked by SR-95531, a GABAA receptor antagonist, and mimicked by muscimol but not baclofen. Consistent with the decrease in [3H]flunitrazepam binding, chronic GABA exposure causes a 43% decrease in the binding of [35S]t-butylbicyclophosphorothionate, a ligand for the receptor-associated chloride ionophore. In addition to chronic GABA-induced down-regulation, allosteric interactions between GABA and benzodiazepine recognition sites are uncoupled by 34%. The half-time and pharmacology for chronic GABA-induced uncoupling is indistinguishable from that for GABA-induced down-regulation, consistent with the hypothesis that the action of GABA at a common site induces both down-regulation and uncoupling.     

GABA(B) receptor activation inhibits dopamine D1 receptor-mediated facilitation of [(3)H]GABA release in substantia nigra pars reticulata

GABA(B) receptors inhibit and dopamine D1 receptors stimulate the release of GABA from striatal terminals in the pars reticulata of the substantia nigra. Here we have studied the interaction between both classes of receptors by exploring the effect of GABA(B) receptors upon the stimulation of depolarization-induced [(3)H]GABA release induced by the activation of D1 receptors in slices of the pars reticulata of the rat substantia nigra. The activation of GABA(B) receptors with baclofen (100 microM) inhibited by 48+/-8% the evoked [(3)H]GABA release in normal slices but did not modify the release in slices from reserpine-treated rats, indicating that the inhibition was dependent on endogenous dopamine. The inhibitory effect of baclofen was also abolished by the D1 receptor antagonist SCH 23390 (1 microM), indicating a D1 receptor-dependence of the baclofen inhibition. Baclofen dose-dependently inhibited (IC(50)=3.6 microM) the stimulation of release induced by the D1 agonist SKF 38393 (1 microM). Baclofen also blocked the stimulation of release induced by forskolin but not that induced by 8-Br-cAMP, indicating that the inhibitory effect was exerted before cAMP synthesis. N-ethylmaleimide (NEM), a selective inactivator of PTX-sensitive G-proteins, abolished the baclofen inhibition of the SKF 38393-induced stimulation of the release without affecting the stimulation induced by the D1 agonist, suggesting that the baclofen effect was mediated by Galpha(i/o) proteins. These results might have relevance in the control motor disorders associated with D1 receptor supersensitivity.     

Glycine stimulates [3H]noradrenaline release by activating a strychnine-sensitive receptor present in rat hippocampus

Rat hippocampus slices were prelabeled with [3H]noradrenaline ([3H]NA) and depolarized by superfusion with KCl. The release evoked by 12 mM K+ was totally calcium-dependent and more than 90% tetrodotoxin (TTX)-sensitive. Glycine (0.1-1 mM) increased the K(+)-evoked [3H]NA overflow in a concentration-dependent manner. The effect of 1 mM glycine reached 300%. Strychnine (0.3 microM) shifted to the right the concentration-response curve for glycine. The effect of glycine (0.1 or 1 mM) was totally abolished by 3 microM strychnine but was unaffected by the GABAA receptor antagonist, bicuculline (10 microM), or by 100 microM of 1-hydroxy-3-aminopyrrolidone-2 (HA-966), a proposed antagonist of glycine at the strychnine-insensitive site located on the N-methyl-D-aspartate (NMDA) receptor. The effect of glycine was mimicked by L-serine, although less potently; the release of [3H]NA was enhanced by 200% in presence of 3 mM L-serine. At this concentration D-serine was ineffective. Strychnine shifted to the right the concentration-response curve for L-serine. Glycine (1 mM) had only a minor effect (less than 20% potentiation) on the release of [3H]NA evoked by 12 mM KCl in hippocampal synaptosomes. While the effect of glycine in slices was increased by decreasing the depolarizing concentration of K+ (about 500% potentiation at 9 mM K+), the response of synaptosomes remained minimal, even in presence of 9 mM KCl. Hippocampal synaptosomes prelabeled with [3H]glycine released the radiolabeled amino acid when exposed to superfusion with 12 mM KCl. The release of [3H]glycine was more than 75% calcium-dependent. The results suggest that the release of NA in rat hippocampus may be enhanced by glycine through the activation of a strychnine-sensitive receptor. This receptor does not seem to be located on noradrenergic terminals.     

Both systemic and local administration of benzodiazepine agonists inhibit the in vivo release of 5-HT from ventral hippocampus

The effects of benzodiazepine- and GABAA-receptor agonists and antagonists on the release and metabolism of 5-HT were measured in the ventral hippocampus of freely moving rats using microdialysis. Systemic injections of the benzodiazepine agonists, flurazepam and diazepam reduced the levels of 5-HT while the partial inverse agonist, FG 7142 (N-methyl-beta-carboline-3-carboxamide), had no effect. Local perfusion of flurazepam through the dialysis probe also decreased the release of 5-HT in the ventral hippocampus, an effect which was completely blocked by the benzodiazepine antagonist, Ro15-1788 (flumazenil). Local application of the GABAA agonist muscimol had no effect on the release of 5-HT, while the antagonist picrotoxin, administered locally, caused a 4-fold enhancement of release of 5-HT. Picrotoxin also resulted in a complete block of the inhibitory effect of flurazepam on release of 5-HT. None of these drugs caused significant changes in the levels of the metabolite of 5-HT, 5-hydroxyindoleacetic acid (5-HIAA) in the ventral hippocampus. These results suggest that the inhibitory effect of flurazepam on the release of 5-HT is mediated by benzodiazepine/GABAA receptors in the hippocampus and that GABA exerts a tonic inhibitory effect on the release of 5-HT in the region of the brain.     

Molecular mechanisms of benzodiazepine-induced down-regulation of GABAA receptor alpha 1 subunit protein in rat cerebellar granule cells.

1. Chronic benzodiazepine treatment of rat cerebellar granule cells induced a transient down-regulation of the gamma-aminobutyric acidA (GABAA) receptor alpha 1 subunit protein, that was dose-dependent (1 nM-1 microM) and prevented by the benzodiazepine antagonist flumazenil (1 microM). After 2 days of treatment with 1 microM flunitrazepam the alpha 1 subunit protein was reduced by 41% compared to untreated cells, which returned to, and remained at, control cell levels from 4-12 days of treatment. Chronic flunitrazepam treatment did not significantly alter the GABAA receptor alpha 6 subunit protein over the 2-12 day period. 2. GABA treatment for 2 days down-regulates the alpha 1 subunit protein in a dose-dependent (10 microM-1 mM) manner that was prevented by the selective GABAA receptor antagonist bicuculline (10 microM). At 10 microM and 1 mM GABA the reduction in alpha 1 subunit expression compared to controls was 31% and 66%, respectively. 3. The flunitrazepam-induced decrease in alpha 1 subunit protein is independent of GABA, which suggests that it involves a mechanism distinct from the GABA-dependent action of benzodiazepines on GABAA receptor channel activity. 4. Simultaneous treatment with flunitrazepam and GABA did not produce an additive down-regulation of alpha 1 subunit protein, but produced an effect of the same magnitude as that of flunitrazepam alone. This down-regulation induced by the combination of flunitrazepam and GABA was inhibited by flumazenil (78%), but unaffected by bicuculline. 5. The flunitrazepam-induced down-regulation of alpha 1 subunit protein at 2 days was completely reversed by the protein kinase inhibitor staurosporine (0.3 microM). 6. This study has shown that both flunitrazepam and GABA treatment, via their respective binding sites, caused a reduction in the expression of the GABAA receptor alpha 1 subunit protein; an effect mediated through the same neurochemical mechanism. The results also imply that the benzodiazepine effect is independent of GABA, and that the benzodiazepine and GABA sites may not be equally coupled to the down-regulation process, with the benzodiazepine site being the more dominant. The biochemical mechanism underlying the benzodiazepine-mediated down-regulation of the alpha 1 subunit protein seems to involve the activity of staurosporine-sensitive protein kinases.     

Human brain somatostatin release from isolated cortical nerve endings and its modulation through GABAB receptors.

1. The release of somatostatin-like immunoreactivity (SRIF-LI) in the human brain was studied in synaptosomal preparations from fresh neocortical specimens obtained from patients undergoing neurosurgery to remove deeply sited tumours. 2. The basal outflow of SRIF-LI from superfused synaptosomes was increased about 3 fold during exposure to a depolarizing medium containing 15 mM KCl. The K(+)-evoked overflow of SRIF-LI was almost totally dependent on the presence of Ca2+ in the superfusion medium. 3. The GABAB receptor agonist, (-)-baclofen (0.3 - 100 microM), inhibited the overflow of SRIF-LI in a concentration-dependent manner (EC50 = 1.84 +/- 0.20 microM; maximal effect: about 50%). The novel GABAB receptor ligand, 3-aminopropyl(difluoromethyl)phosphinic acid (CGP 47656) mimicked (-)-baclofen in inhibiting the SRIF-LI overflow (EC50 = 3.06 +/- 0.52 microM; maximal effect: about 50%), whereas the GABAA receptor agonist, muscimol, was ineffective up to 100 microM. 4. The inhibition by 10 microM (-)-baclofen of the K(+)-evoked SRIF-LI overflow was concentration-dependently prevented by two selective GABAB receptor antagonists, 3-amino-propyl (diethoxymethyl)-phosphinic acid (CGP 35348) (IC50 = 24.40 +/- 2.52 microM) and [3-[[(3,4-dichlorophenyl) methyl]amino]propyl] (diethoxymethyl) phosphinic acid (CGP 52432) (IC50 = 0.06 +/- 0.005 microM). 5. The inhibition of SRIF-LI overflow caused by 10 microM CGP 47656 was abolished by 1 microM CGP 52432. 6. When human synaptosomes were labelled with [3H]-GABA and depolarized in superfusion with 15 mM KCl, the inhibition by 10 microM (-)-baclofen of the depolarization-evoked [3H]-GABA overflow was largely prevented by 10 microM CGP 47656 which therefore behaved as an autoreceptor antagonist. 7. In conclusion: (a) the characteristics of SRIF-LI release from synaptosomal preparations of human neocortex are compatible with a neuronal origin; (b) the nerve terminals releasing the neuropeptide possess inhibitory receptors of the GABAB type; (c) these receptors differ pharmacologically from the GABAB autoreceptors present on human neocortex nerve terminals since the latter have been shown to be CGP 35348-insensitive but can be blocked by CGP 47656.