Mohammad H. Jalilian Tehrani

Chronic exposure of developing cortical neurons to GABA down-regulates GABA/benzodiazepine receptors and GABA-gated chloride currents.

Cultures of cerebral neurons were prepared from chick embryos, 8.5 days in ovo, and maintained in vitro. Following chronic exposure of these cells to GABA, the levels of [3H]flunitrazepam binding in situ and electrophysiological responsiveness to gamma-aminobutyric acid (GABA) was examined. Treatment with 100 microM GABA for 7 days reduced [3H]flunitrazepam binding in situ by 70 +/- 8% compared to untreated controls. The binding of [3H]N-methylscopolamine was unaffected by this treatment. The reduction in [3H]flunitrazepam binding was prevented by concomitant exposure of developing neurons to the GABA antagonist R 5135, suggesting that GABAA receptor occupancy is required. The loss of bezodiazepine receptors was dependent on the GABA concentration in the culture medium and a half-saturation (IC50) value of 11.2 +/- 3.7 microM was estimated. Whole-cell patch-clamp recordings were obtained to assess the functional properties of the labile receptor pool observed in the binding studies. Neurons cultured with 100 microM GABA for 7 days showed a 60-70% reduction in the peak current amplitudes observed in response to application of 10-100 microM GABA. However, the rate of rapid desensitization, quantified by measuring changes in input conductance, was unchanged by chronic GABA exposure, yielding decay time constants of 27.1 +/- 2.1 and 34.7 +/- 4.7 s for control and treated cells, respectively. The results are consistent with a GABA modulation of the GABAA/benzodiazepine receptor complex by means of down-regulation.

Agonist‐Dependent Internalization of γ‐Aminobutyric AcidA/Benzodiazepine Receptors in Chick Cortical Neurons

An impermeant benzodiazepine receptor ligand was prepared by derivatization of the aminobenzodiazepine 1012‐S with 4‐sulfophenylisothiocyanate. The resulting N‐sulfophenyl)‐thiocarbamoyl derivative of 1012‐S (SPTC‐1012S) was purified by reverse‐phase HPLC, and the predicted structure was verified by mass spectrometry. The apparent affinity of SPTC‐1012S (IC50= 9.8 ± 2.9 nM) for displacement of [3H] flunitrazepam from intact chick cortical neurons was similar to that of 1012‐S (IC50= 4.0 ±0.3 nM). However, at concentrations from 0.1 to 10 μM, 1012‐S was consistently more efficacious than SPTC‐1012S, a finding indicating that 6‐8% of the benzodiazepine receptor pool was not accessible to the impermeant compound. This inaccessible pool was eliminated by permeabilization of the cells with saponin or Triton ±‐100, a result suggesting that ∼7% of neuronal benzodiazepine receptors are intracellular. Acute treatment (1–4 h at 37°C) of neurons with 100μMγ‐aminobutyric acid (GABA) or 100 nM clonazepam had little effect on the level of [3H] flunitrazepam binding but increased the proportion of intracellular receptors by 61 and 74%, respectively, compared with untreated controls. Similar treatment with 1 mM GABA increased the level of intracellular sites by 154–176%. The effect of GABA on receptor internalization was blocked by cotreatment with the GABAA receptor antagonist R 5135. The results suggest that SPTC‐1012S can be used as a probe to study the internalization of the GABAA/benzodiazepine receptor complex under normal conditions or following acute or chronic treatment with agonists.

GABA down-regulates the GABA/benzodiazepine receptor complex in developing cerebral neurons

Neuronal cultures of the chick embryo cerebrum were used to study the chronic effects of gamma-aminobutyric acid (GABA) on the expression of the GABA/benzodiazepine receptor complex. A 7 day exposure of developing neurons to 100 microM GABA produced a 70% reduction in the level of [3H]flunitrazepam binding to intact cells, when compared to untreated controls. The reduction was due to a decrease in receptor density (Bmax) rather than the affinity. The same treatment also caused a 75% reduction in the rates of GABA-gated 36Cl- uptake by intact cells, without an effect on the basal (GABA-independent) flux. Eight days after removal of GABA from the medium of treated cultures, the neurons recovered [3H]flunitrazepam binding to levels corresponding to 74% of unexposed, age-matched controls. The results are consistent with a GABA-induced down-regulation of the GABA/benzodiazepine receptor.

Identification of GABAA/benzodiazepine receptors on clathrin-coated vesicles from rat brain

Abstract: To investigate the subcellular compartments that are involved in the endocytosis and intracellular trafficking of GABAA/benzodiazepine receptors, we have studied the distribution and properties of clonazepam‐displaceable binding of [3H]flunitrazepam to membrane fractions from rat brain. The microsomal fraction was subjected to density centrifugation and gel filtration to isolate clathrin‐coated vesicles. Homogeneity of the coated‐vesicle fraction was demonstrated by using electron microscopy and by analysis of clathrin subunits and clathrin light‐chain kinase. Vesicles exhibiting specific binding of [3H]flunitrazepam eluted from the sieving gel as a separate peak, which was coincident with that for coated vesicles. Scatchard analysis of equilibrium binding of [3H]flunitrazepam to coated vesicles yielded a KD value of 21 ± 4.7 nM and a Bmax value of 184 ± 28 fmol/mg. The KD value for coated vesicles was 12‐19‐fold that found with microsomal or crude synaptic membranes. This low‐affinity benzodiazepine receptor was not identified on any other subcellular fraction and thus appears to be a novel characteristic of coated vesicles. The Bmaxvalue for coated vesicles, expressed per milligram of protein, corresponded to 16 and 115% of that found for crude synaptic and microsomal membrane fractions, respectively. Because the trafficking of neurotransmitter receptors via clathrin‐coated vesicles is most likely to occur through endocytosis, the data suggest that an endocytotic pathway may be involved in the removal of GABAA/benzodiazepine receptors from the neuronal surfaces of the rat brain. This mechanism could play a role in receptor sequestration and down‐regulation that is produced by exposure to GABA and benzodiazepine agonists.

Ontogeny of the GABA receptor complex in chick brain: Studies in vivo and in vitro

The ontogeny of the gamma-aminobutyric acid-A (GABAA) receptor complex in the chick brain was studied by specific binding of [3H]muscimol, [3H]flunitrazepam (Flu) and [35S]t-butylbicyclophosphorothionate (TBPS) to isolated membranes. During development in ovo, the specific binding of muscimol and flunitrazepam increased from day 8 and reached 50% of adult levels of day 20, while a comparable level of TBPS binding was achieved by day 17. The increases in TBPS and Flu binding were reflected in Bmax rather than Kd changes. In embryonic brain, only a low-affinity site for muscimol (Kd = 23 nM) was observed while an additional high-affinity site (Kd = 0.4 nM), as well as the low-affinity site, was found in adult tissue. Similar studies were carried out with cultures of cerebral neurons prepared from 8-day embryos. The level of specific binding of muscimol, Flu and TBPS increased in culture, achieved one half of the maximum level by days 4-5, maximal levels by day 10 and decreased slowly thereafter. The maximal levels in culture corresponded, respectively, to 27%, 67% and 57% of these found in the 18-day embryo. The binding of Flu to membranes from neurons, embryos and adults was enhanced by addition of GABA while TBPS binding was inhibited. The EC50 and IC50 values for these effects corresponded to those for gating of chloride channels. These findings indicate a coordinated expression of receptors for GABA, benzodiazepines and convulsant/TBPS during neuronal maturation both in vivo and in vitro. The schedule for this postsynaptic ontogeny is very similar to that for presynaptic markers of GABAergic neurons (see companion paper).

Decreased expression of GABAA receptor α6 and β3 subunits in stargazer mutant mice: a possible role for brain-derived neurotrophic factor in the regulation of cerebellar GABAA receptor expression?

The cerebellar granule cells of the spontaneous recessive mutant mouse strain, stargazer (stg/stg), fail to express brain-derived neurotrophic factor mRNA. This deficit is exclusive to these neurons and is believed to underlie the motor irregularities displayed by stg/stg, though the molecular basis for their phenotype has still to be resolved. Brain-derived neurotrophic factor has been shown to play a role in the postnatal maturation of cerebellar granule cells. Differentiation of these neurons, postnatally, is characterised by a switch in their GABAA receptor subunit expression profile. Notably, the GABAA receptor alpha6 subunit, which is specific to these neurons, becomes detectable at postnatal days 10-14 (P10-14). To determine whether cerebellar GABAA receptor expression has been compromised in stg/stg mice, the expression levels of GABAA receptor alpha1, alpha6, beta2 and beta3 subunits were compared between stg/stg mice and the appropriate wild-type background strain, C57BL/6J (+/+). By quantitative immunoblotting, it was found that the expression of the alpha6 and beta3 subunits was 23+/-8% and 38+/-12% (mean+/-S.E.M., n=6) of control (+/+) levels, respectively. In contrast, the expression of the alpha1 and beta2 subunits was not significantly different from controls, being 116+/-11% and 87+/-24% (mean+/-S.E.M., n=6) of +/+ levels, respectively. Total specific [3H]Ro15-4513 binding activity detected in cerebellar membranes prepared from stg/stg was not significantly different from +/+ mice. However, the benzodiazepine agonist-insensitive subtype of [3H]Ro15-4513 binding activity, a pharmacological motif of alpha6 subunit-containing GABAA receptors, was lower in stg/stg mice relative to the +/+ strain which correlated with the lowered level of alpha6 subunit expression. Thus, we have identified an abnormality in the GABAA receptor profile of stg/stg mutant mice that might underpin its irregular phenotype.

Clathrin-coated vesicles from bovine brain contain uncoupled GABAA receptors.

Clathrin-coated vesicles are thought to be a vehicle for the sequestration of GABAA receptors. For coated vesicles from bovine cerebrum, we examined the binding properties of [3H]muscimol. a GABAA-specific agonist. [3H]flunitrazepam a benzodiazepine agonist, and [35S]t-butylbiocyclophosphorthionate (TBPS), a ligand for GABAA receptor channels. Under standard conditions, the binding level of [3H]muscimol, [3H]flunitrazepam, and [35S]TBPS to coated vesicles represented 12.3 +/- 1.8%, 7.9 +/- 1%, and 10.2 +/- 1.8%, respectively, of that in crude synaptic membranes. Coated vesicles showed a single [3H]flunitrazepam binding site with a KD value (12 nM) which was 9-fold that for synaptic membranes. The allosteric coupling between binding sites was measured by the addition of GABA to [3H]flunitrazepam and [35S]TBPS binding assays. For [3H]flunitrazepam binding to synaptic membranes, GABA gave an EC50 = 2.0 microM and at saturation (100 microM) an enhancement of 122%. This stimulation was completely blocked by the GABA antagonist SR95531. In contrast, neither GABA nor SR95531 had a significant effect on [3H]flunitrazepam binding to CCVs, indicating that the allosteric interaction between GABA and benzodiazepine binding sites is abolished. Likewise, GABA displaced nearly all of the [35S]TBPS binding to synaptic membranes but had no effect on binding to coated vesicles, indicating that coupling between the GABA binding sites and chloride channel is also impaired. Thus GABAA receptors appear to be uncoupled during normal intracellular trafficking via coated vesicles. The presence of major GABAA receptor subunits on these particles was verified by quantitative immunoblotting. Relative to the levels in synaptic membranes, CCVs contained 110 +/- 14% and 29.5 +/- 3.8%, respectively, of the immunoreactivity for GABAA receptor beta 2 and alpha 1 subunits. Thus, in comparison to GABAA receptors on synaptic membranes, those on CCVs have a reduced alpha 1/beta 2-subunit ratio. It may be suggested that a selective decline in the content of alpha 1 subunits in coated vesicles could in part account for GABAA receptor uncoupling.

Regulation of γ-aminobutyric acidB (GABAB) receptors in cerebral cortex during the estrous cycle

We examined binding of the GABAB receptor agonist baclofen to brain synaptic membranes as a function of the natural variations in gonadal steroids that occur during the estrous cycle of the adult rat. We found that the binding of baclofen to neocortical membranes varied systematically as a function of the estrous cycle, with the lowest binding occuring during the estrus stage. Binding to archicortical (hippocampal) and hypothalamic preparations also varied with the estrous cycle, except that the lowest level of binding in these latter cases occurred during the diestrus stage. The variation of [3H]baclofen binding during the estrous cycle was different with respect to the binding of [3H]muscimol, an agonist for GABAA receptors, and [3H]8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT), an agonist for serotonin 5-HT1A receptors that shares similar G proteins and effectors with GABAB receptors. Saturation binding studies of cortical GABAB receptors showed that apparent receptor density (Bmax) rather than affinity (Kd)best accountd for the change in binding during the estrous cycle in that Bmax), like total specific binding, was at a minimum during the estrus stage. The robust regulation of GABAB receptors in neocortex was unexpected and its functional significance is at present unknown. However, the correlation of the menstrual cycle with mood and other behavioral changes, and the correlations of the estrous and menstrual cycles with seizure susceptibility, may somehow depend upon hormonal regulation of transmitter systems such as the one we have observed here.

Interaction of t-Butylbicyclophosphorothionate with γ-Aminobutyric Acid-Gated Chloride Channels in Cultured Cerebral Neurons

Abstract: The role of t‐butylbicyclophosphorothionate (TBPS) as an antagonist of γ‐aminobutyric acid (GABA) was studied with primary cultures of neurons from the chick embryo cerebrum. The addition of GABA stimulated the uptake of 36Cl− by neurons and the dose dependence of this effect followed hyperbolic kinetics with a K0.5= 1.3 μM for GABA. TBPS proved to be a potent inhibitor of GABA‐dependent Cl− uptake (IC50= 0.30 μM). Analysis of the kinetics of this process revealed that TBPS is a noncompetitive inhibitor (Ki= 0.15 μM) with respect to GABA. Scatchard analysis of direct binding of [35S]TBPS to membranes isolated from neuronal cultures gave curvilinear plots. These could be resolved by nonlinear regression methods into two components with KD values of 3.1 nM and 270 nM. The TBPS binding constant for this lower affinity site agreed well with the IC50 and Ki values for inhibition of Cl− flux, suggesting that this site is physiologically relevant to GABA antagonism. GABA was a noncompetitive displacer of [35S]TBPS binding to the lower affinity site. The Ki value for this displacement by GABA (1.7 μM) was comparable to the value for GABA enhancement of Cl− flux. The binding of [35S]TBPS to its low‐affinity site on neuronal membranes was ninefold higher in the presence of Cl− than with gluconate, an impermeant anion. The rank order for anion stimulation of [35S]TBPS binding was Br−≧ SCN− > Cl−≧ NO−3 > I− > F− > gluconate. The EC50 value for Cl− enhancement of [35S]TBPS binding (160 mM) agreed well with the Km for Cl− influx via GABA‐gated channels (140 mM). These results indicate that TBPS acts as a GABA antagonist via direct blockade of neuronal Cl− channels. A minimum density of 6.5 × 104 chloride channels per neuron was obtained from TBPS binding at saturation.

Reduced function of γ-aminobutyric acidA receptors in tottering mouse brain: Role of cAMP-dependent protein kinase

Abstract The single-locus mutant mouse tottering ( tg ) displays spontaneous seizures that resemble those in human petit-mal epilepsy. In order to examine alterations in GABA A receptor function which could arise as a result of this mutation, the influx of 36 C1 − was determined using microsacs (membrane vesicles) isolated from the brain of tg/tg and coisogenic C57BL/6J ( + / + ) control mice. In microsacs from both tg/tg and + / + strains, the maximum level of 36 Cl − uptake induced by 50 μM GABA was observed during five seconds of incubation at 28°C. Compared to + / +, the GABA-dependent 36 Cl − uptake in tg/tg microsacs was significantly lower and faded rapidly during longer incubations. The levels of gated 36 C1 − uptake in tg/tg microsacs were 45 ± 6.3%, 65 ± 9.9%, and 33 ± 6.1% of control ( + / + ) values for 3-, 5-, and 10-s incubations, respectively. GABA A receptor-specific agonists (30 μM), muscimol, isoguvacine and THIP (4,5,6,7-tetrahydroisoazolo-[5,4-c]pyridin-3-ol) induced 36 C1 − influx in the order muscimol > GABA > isoguvacine > THIP. This order was similar for both strains, but the agonist-dependent influx was always significantly lower in tg/tg compared to + / +. Treatment of the microsacs with 10 μ M H-89, a membrane-permeant inhibitor of the cAMP-dependent protein kinase (protein kinase A, PKA), was without effect on GABA-gated 36 Cl − uptake in + / +, but increased the gated uptake in tg/tg microsacs by 44 ± 16%. PKA was assayed using [ γ - 32 ]ATP and kemptide as the substrate. Triton ×-100 (0.1%) increased both the basal and 8-Br-CAMP dependent PKA activity in microsacs by 3–4 four fold, showing that most of the enzyme was intravesicular. In the presence of Triton, the basal activity of PKA in the tg/tg preparations was twice that of + / +, while the strain difference was no longer apparent in assays containing 8-Br-cAMP. The data suggest that an abnormal elevation of protein kinase A activity in tottering mouse brain contributes to an impairment of GABA A receptor function. It is suggested that the resulting loss of inhibition could play a role in induction of the seizures which characterize the mutant phenotype.