Immunocytochemical localization of the α6 subunit of the γ-aminobutyric acidA receptor in the rat nervous system

The localization in the rat central nervous system and retina of the α₆ subunit peptide of the γ-aminobutyric acid (GABA_A) receptor has been studied by light microscopy immunocytochemistry with a specific anti-α₆ antibody. The α₆ subunit was present in the granule cells of the cerebellum, the granule cells of the dorsal cochlear nucleus, axons of the olfactory nerve including the glomerular endings, layer II of the dorsal horn of the spinal cord, and in the retinal synaptic layers, particularly the inner plexiform layer. Thus, contrary to the general belief, the α₆ subunit is not exclusively localized in the granule cells of the cerebellum. It is also expressed in some sensory neurons and other neurons involved in the initial processing of sensory information. © 1996 Wiley-Liss, Inc.     

Chronic pentobarbital administration alters γ-aminobutyric acidA receptor α6-subunit mRNA levels and diazepam-insensitive [3H]Ro15-4513 binding

In order to study the chronic effects of pentobarbital, a positive GABA_A receptor modulator, on the inverse agonist binding of the benzodiazepine site, binding of [³H]Ro15-4513 and levels of GABA_A receptor α₆-subunit mRNA were investigated in the brains of pentobarbital-tolerant/dependent animals, using receptor autoradiography and in situ hybridization histochemistry in consecutive brain sections. Pentobarbital was administered to rats either 60 mg/kg, i.p., once, for acute treatment, or 300 μg/10μl/h i.c.v. continuously for 6 days via osmotic minipumps to render rats tolerant to pentobarbital. Rats assigned to the dependent group were sacrificed 24 h after discontinuance of pentobarbital infusion, while those assigned to the tolerant group were sacrificed at the end of infusion. The α₆ subunit mRNA was increased in the tolerant group only. Diazepam-insensitive [³H]Ro15-4513 binding was increased in the cerebellar granule layer of pentobarbital-tolerant and -dependent rats. No alterations in these parameters were observed in acutely treated animals. These data suggest that chronic pentobarbital treatment induced expression of α₆-subunit mRNA. This was in contrast to α₁- and γ₂-subunit mRNA, which in tolerant animals are unchanged, but for which withdrawal triggers a surge in levels. Because the α₆-subunit is a major component of the diazepam-insensitive [³H]Ro15-4513 binding site, the increased diazepam-insensitive [³H]Ro15-4513 binding implied de novo synthesis of the receptor subunit protein. © 1996 Wiley-Liss, Inc.     

Temporal changes in extracellular acetylcholine and CA1 pyramidal cells in gerbil hippocampus following transient cerebral ischemia

Temporal changes in cholinergic functions following transient cerebral ischemia (10 min) were studied in the hippocampus of awake unrestrained gerbils using in vivo microdialysis. These data were compared with the results for temporal change in the area of each CA1 cell soma, measured with a microcomputer imaging device. KCl-induced release of acetylcholine (ACh) tended to be lower within 1 day after recirculation, and was significantly lower on the 4th, 7th and 14th days. Atropine-induced release of ACh gradually decreased over the test period. In histological estimation, no differences were observed within the 1st day, but a significant decrease of the area of CA1 cell soma was observed from the 4th to 14th days. Moreover, ischemia over 2 min decreased KCl- and atropine-induced ACh release on the 14th day without significant changes of hippocampal CA1 pyramidal cell. From these results, it is clear that ischemia produced dysfunction of hippocampal cholinergic neurons, and that dysfunction of the hippocampal cholinergic system following transient ischemia precedes pyramidal cell damage in the hippocampal CA1 subfield.     

Effect of ischemic preconditioning on antioxidant status in the gerbil hippocampal CA1 region after transient forebrain ischemia

Ischemic preconditioning (IPC) is a condition of sublethal transient global ischemia and exhibits neuro-protective effects against subsequent lethal ischemic insult. We, in this study, examined the neuroprotective effects of IPC and its effects on immunoreactive changes of antioxidant enzymes including superoxide dismutase (SOD) 1 and SOD2, catalase (CAT) and glutathione peroxidase (GPX) in the gerbil hippocampal CA1 region after transient forebrain ischemia. Pyramidal neurons of the stratum pyramidale (SP) in the hippocampal CA1 region of animals died 5 days after lethal transient ischemia without IPC (8.6%(ratio of remanent neurons) of the sham-operated group);however, IPC prevented the pyramidal neurons from subsequent lethal ischemic injury (92.3%(ratio of remanent neurons) of the sham-operated group). SOD1, SOD2, CAT and GPX immunoreactivities in the sham-operated animals were easily detected in pyramidal neurons in the stratum pyramidale (SP) of the hippocampal CA1 region, while all of these immunoreac-tivities were rarely detected in the stratum pyramidale at 5 days after lethal transient ischemia without IPC. Meanwhile, their immunoreactivities in the sham-operated animals with IPC were similar to (SOD1, SOD2 and CAT) or higher (GPX) than those in the sham-operated animals without IPC. Furthermore, their immunoreactivities in the stratum pyramidale of the ischemia-operated animals with IPC were steadily maintained after lethal ischemia/reperfusion. Results of western blot analysis for SOD1, SOD2, CAT and GPX were similar to immunohistochemical data. In conclusion, IPC maintained or increased the expression of antioxidant enzymes in the stratum pyramidale of the hippocampal CA1 region after subsequent lethal transient forebrain ischemia and IPC exhibited neuroprotective effects in the hippocampal CA1 region against transient forebrain ischemia.     

Ethyl β-carboline-3-car☐ylate antagonizes the action of GABA and benzodiazepines in the hippocampus

In urethane-anaesthetized rats, the β-carboline derivative βCCE (0.3–1.0 mg/kg i.v.) excited hippocampal pyramidal cells which were inhibited by GABA (applied iontophoretically) and benzodiazepines (applied iontophoretically or intravenously). While benzodiazepines facilitated the action of GABA, the effects of GABA and benzodiazepines were antagonized by βCCE. This electrophysiological study supports the behavioural observations that βCCE is a benzodiazepine receptor antagonist.     

Immunocytochemical localization of the β2 subunit of the gamma-aminobutyric acidA receptor in the rat brain

An antiserum to the β₂ subunit of the rat gamma-aminobutyric acid (GABA_A) receptor was prepared by immunizing a rabbit with a fusion protein expressed in bacteria. The fusion protein had the large, intracellular loop expanding between the putative M3 and M4 transmembrane domains of the β₂ subunit fused to staphylococcal protein A (SPA). The antiserum immunoprecipitated both the solubilized and the affinity-purified GABA_A receptors. The anti-β₂ antibodies were affinity purified on immobilized β₂ intracellular loop peptide. The antibodies recognized a 55–57 kDa peptide in immunoblots of either crude membranes from rat cerebral cortex or affinity-purified GABA_A receptors from bovine cerebral cortex. Immunocytochemistry with the affinity-purified antibody has revealed for the first time the localization of the β₂ subunit in the rat brain. A comparative study of the regional and cellular immunoreactivities of the affinity-purified anti-β₂ antibody and the monoclonal antibody 62-3G1 (which recognizes both β₂ and β₃ subunits) is presented. The procedure described for generating and preparing specific anti-β₂ subunit antibodies that are valuable for immunocytochemistry could be extended to other GABA_A receptor subunits. © 1994 Wiley-Liss, Inc.     

Spatial and temporal alterations in the GABA shunt in the gerbil hippocampus following transient ischemia

In the present study, we have identified the alteration in the expressions of GABA shunt-associated enzymes and the GABA transporter in order to determine the relationship between the neuronal damage and GABA metabolism following ischemia. At 30 min post-ischemia, the immunoreactivities of the glutamic acid decarboxylase (GAD) isoforms were markedly elevated in the CA1 region, as compared with the sham operated group. At 3-12 h post-ischemia, their immunoreactivities recovered at the sham level. These patterns were similarly observed up to 12 h following ischemia insult. However, the intensity of GAD67 was markedly increased at 24 h post-ischemic insult. The temporal changes in GABA transporter 1 (GAT-1) expressions were similar to that of GAD67, but not GAD65, expression, at least prior to 12 h after ischemic insults. GAT-1 immunoreactivity was significantly elevated in the CA1 region posterior to 12 h post-ischemia. Both succinic semialdehyde dehydrogenase (SSADH) and succinic semialdehyde reductase (SSAR) immunoreactivities were not altered in GABAergic neurons following ischemia. In contrast, in pyramidal cells, both SSADH and SSAR immunoreactivities showed chronological alterations in the CA1 region. Thus, our findings suggest that the differential alterations of GABA metabolism may be one of the important factors in neuronal damages induced by ischemia.     

Transport of fragmented DNA in apical dendrites of gerbil CA1 pyramidal neurons following transient forebrain ischemia

Transport of fragmented DNA in apical dendrites of the CA1 pyramidal neurons of gerbil hippocampus is observed in the apoptotic process following transient forebrain ischemia. The time-course of specific DNA fragmentation was examined after the ischemic insult by in situ nick-end-labeling method and fluorescence detection technique by DAPI. Although the role of the fragmented DNA movement is unclear, the transport mechanism of fragmented DNA is still active in the late phase of apoptotic process.     

Synaptic target selectivity and input of GABAergic basket and bistratified interneurons in the CA1 area of the rat hippocampus

To assess the position of interneurons in the hippocampal network, fast spiking cells were recorded intracellularly in vitro and filled with biocytin. Sixteen non-principal cells were selected on the basis of 1) cell bodies located in the pyramidal layer and in the middle of the slice, 2) extensive labeling of their axons, and 3) a branching pattern of the axon indicating that they were not axo-axonic cells. Examination of their efferent synapses (n = 400) demonstrated that the cells made synapses on cell bodies, dendritic shafts, spines, and axon initial segments (AIS). Statistical analysis of the distribution of different postsynaptic elements, together with published data (n = 288) for 12 similar cells, showed that the interneurons were heterogeneous with regard to the frequency of synapses given to different parts of pyramidal cells. When the cells were grouped according to whether they had less or more than 40% somatic synaptic targets, each population appeared homogeneous. The population (n = 19) innervating a high proportion of somata (53 ± 10%, SD) corresponds to basket cells. They also form synapses with proximal dendrites (44 ± 12%) and rarely with AISs and spines. One well-filled basket cell had 8,859 boutons within the slice, covering an area of 0.331 mm² of pyramidal layer tangentially and containing 7,150 pyramidal cells, 933 (13%) of which were calculated to be innervated, assuming that each pyramidal cell received nine to ten synapses. It was extrapolated that the intact axon probably had about 10,800 boutons innervating 1,140 pyramids. The proportion of innervated pyramidal cells decreased from 28% in the middle to 4% at the edge of the axonal field. The other group of neurons, the bistratified cells (n = 9), showed a preference for dendritic shafts (79 ± 8%) and spines (17 ± 8%) as synaptic targets, rarely terminating on somata (4 ± 8%). Their axonal field was significantly larger (1,250 ± 180 μm) in the medio-lateral direction than that of basket cells (760 ± 130 μm). The axon terminals of bistratified cells were smaller than those of basket cells. Furthermore, in contrast to bistratified cells, basket cells had a significant proportion of dendrites in stratum lacunosum-moleculare suggesting a direct entorhinal input. The results define two distinct types of GABAergic neuron innervating pyramidal cells in a spatially segregated manner and predict different functions for the two inputs. The perisomatic termination of basket cells is suited for the synchronization of a subset of pyramidal cells that they select from the population within their axonal field, whereas the termination of bistratified cells in conjunction with Schaffer collateral/commissural terminals may govern the timing of CA3 input and/or voltage-dependent conductances in the dendrites. © 1996 Wiley-Liss, Inc.     

GABAAβ2–3 immunoreactive cells in the developing chick retina

Gama-aminobutyric acid (GABA) is the major inhibitory neurotransmitter in the central nervous system (CNS). It has been shown that GABA is an important factor for CNS maturation and that its functions are mainly mediated by GABA(A) receptors. Thus, in order to fully comprehend the role of GABA during development, it is essential to establish the developmental features of the catalytic subunits (beta) of GABA(A) receptor. Here, we determine the ontogenesis and neurogenesis of cells expressing beta2-3 subunits of GABA(A) receptor (GABA(Abeta2-3)) in the chick retina. In the ontogenetic experiments, only the immunohistochemistry for GABA(Abeta2-3) approach was employed. For neurogenesis a double-labeling method (autoradiography and immunohistochemistry) was applied. [H(3)]-thymidine was injected into eggs (2-11 days) and the embryos were sacrificed at embryonic day 19 (E19). GABA(Abeta2-3) immunohistochemistry was processed and then autoradiography was performed. We used a cumulative counting method to quantify the autoradiographic grains. The ontogenesis study revealed that at E9, GABA(Abeta2-3) immunoreactivity was restricted to the inner plexiform layer and the first cell bodies immunoreactive to GABA(Abeta2-3) were seen at E14. Thereafter, the number of cell bodies and the intensity of GABA(Abeta2-3) immunoreactivity increased until the adult pattern was established. The neurogenesis study showed that cells that will express GABA(Abeta2-3) were generated between E6 and E9. In addition, from E7 to E9 the rate of neurogenesis of GABA(Abeta2-3) immunoreactive cells quickly increases. Therefore, the detection of GABA(Abeta2-3) occurred only after the end of generation period of this cell population.     

Up-regulation of the peripheral-type benzodiazepine receptor expression and [(3)H]PK11195 binding in gerbil hippocampus after transient forebrain ischemia

In mammalian CNS, the peripheral-type benzodiazepine receptor (PTBR) is localized on the outer mitochondrial membrane within the astrocytes and microglia. The main function of PTBR is to transport cholesterol across the mitochondrial membrane to the site of neurosteroid biosynthesis. The present study evaluated the changes in the PTBR density, gene expression and immunoreactivity in gerbil hippocampus as a function of reperfusion time after transient forebrain ischemia. Between 3 to 7 days of reperfusion, there was a significant increase in the maximal binding site density (B(max)) of the PTBR antagonist [(3)H]PK11195 (by 94-156%; P < 0.01) and PTBR mRNA levels (by 1.8- to 2.9-fold; P < 0.01). At 7 days of reperfusion, in the hippocampal CA1 (the brain region manifesting selective neuronal death), PTBR immunoreactivity increased significantly. Increased PTBR expression after transient forebrain ischemia may lead to increased neurosteroid biosynthesis, and thus may play a role in the ischemic pathophysiology.     

Is γ-aminobutyric acid the neurotransmitter of barnacle photoreceptors?

The hypothesis that γ-aminobutyric acid (GABA) is the neurotransmitter ofbarnacle photoreceptors was tested by studying the effect of GABA on the membrane of the cell directly postsynaptic to the photoreceptor, by testing the ability of GABA antagonists to block transmission at this synapse, and by estimating the free GABA content of the photoreceptor. The results of these experiments suggest that GABA is not the photoreceptor's neurotransmitter.     

Effect of transient forebrain ischemia on superoxide dismutases in gerbil hippocampus

Substantial generation of oxygen-derived free radicals has been implicated in pathophysiology of ischemic brain damage. Immunoreactive mitochondrial manganese and cytosolic copper-zinc superoxide dismutases, initial and essential enzymes to scavenge superoxide radical anions, increased in the gerbil hippocampal neurons after transient forebrain ischemia. Neuronal cells responded to oxidative stress in ischemia and induced the protective mechanism to increase superoxide dismutases.     

α7-Nicotinic receptor expression and the anatomical organization of hippocampal interneurons

C3H and DBA/2 mice differ in their hippocampal inhibitory function, as measured by the inhibitory gating of pyramidal neuron response to repeated auditory stimulation. This functional difference appears to be related to differences in expression of the α7 nicotinic cholinergic receptor, which may be generally expressed by interneurons. This study examines the relationship between genetic variation in α7 receptor subunit expression and GABAergic interneuron distribution in various regions and layers of the hippocampus in the two mouse strains. Subpopulations of hippocampal interneurons in both mouse strains were found to bind [¹²⁵I]α-bungarotoxin. However, the distribution of the [¹²⁵I]α-bungarotoxin-positive hippocampal interneurons was significantly different between C3H and DBA/2 mice. In region CA1, and to a lesser extent in region CA3, DBA/2 mice had increased numbers of [¹²⁵I]α-bungarotoxin-positive neurons in stratum lacunosum-moleculare and decreased numbers in stratum oriens. Similar differences in GABAergic neuron distribution were observed in region CA1 in the two strains. C3H/DBA/2 F1 animals were backcrossed to the C3H parental strain for six generations, with selection for either the DBA/2 or C3H allelic variant of the α7 receptor gene. The distribution of [¹²⁵I]α-bungarotoxin labeling closely resembled the DBA/2 parental phenotype in animals retaining the DBA/2 allele of the α7 gene. These data suggest that the α7 receptor gene locus may influence the anatomical organization of at least a subset of hippocampal interneurons by an as yet unidentified mechanism. This difference in interneuron anatomy may also contribute to functional differences in inhibitory sensory gating between the two strains.     

Immunohistochemistry of γ-aminobutyric acid in the cat nucleus tractus solitarius

Using a new antibody directed solely against the γ-aminobutyric acid (GABA) molecule, distribution of GABA was studied in the nucleus tractus solitarii of the cat. Both immunoreactive puncta and cell bodies had a homogenous distribution within the nucleus. The one exception was in the parvocellular subdivision where very little immunoreactive puncta, but numerous immunoreactive cell bodies, were found. Results of this investigation provide immunohistochemical evidence of GABA's localization in an autonomic nucleus involved in cardiovascular regulation.     

Rapid decline of GABAA receptor subunit mrna expression in hippocampus following transient cerebral ischemia in the gerbil

Inhibitory neurotransmission may play an important role in neuronal degeneration following transient cerebral ischemia. We studied the effect of transient forebrain ischemia on the GABA_A receptor system in the gerbil hippocampus. Gerbils were subjected to 5 minutes of bilateral carotid occlusion and were sacrificed at various times over 4 days following reperfusion. There was a substantial loss of pyramidal cells in the CA1 area of the hippocampus, granule cell layer of the dentate gyrus, and ventroposterior medial and ventroposterior lateral nuclei of the thalamus at any time following ischemia. Examination of brain slices by in situ hybridization histochemistry revealed that a change in expression of the GABA_A receptor α₁ and β₂ subunit mRNAs occurred in two phases following onset of reperfusion. The early phase (rapid) occurred within the first 4 hours following reperfusion. The expression of mRNAs significantly decreased (up to 25%) within 1 hour after occlusion in CA1 and CA3 pyramidal cell layers of the hippocampus and in the granule cell layer of the dentate gyrus. The expression of the mRNAs in these regions continued to decrease for 4 hours (up to 43%). In the second phase, which began between 4 and 12 hours following reperfusion, mRNA expression started to return to control levels in CA3 hippocampus and in the dentate. However, expression of both mRNAs continued to decline slowly in the CA1 pyramidal cell layer (up to 85%) over the next 3 days, concomitantly with degeneration of the CA1 pyramidal cells. Expression of mRNAs in the ventroposterior medial or ventroposterior lateral nuclei of the thalamus was similar to control values. To determine if a change in GABA_A receptor distribution paralleled changes in receptor subunit mRNA expression, we also measured the binding of [³⁵S]t-butylbicylophosphorothionate to GABA_A receptor chloride channels. The t-butylibicyclophosphorothionate [³⁵S] binding decreased between 1 and 4 days after reperfusion in the dendritic fields of CA1 pyramidal cells (strata oriens, radiatum, and lacunosum-moleculare) but not in the pyramidal cell body layer. These results indicate that expression of GABA_A receptor subunit mRNAs decrease well before CA1 pyramidal cell degeneration and loss of GABA_A receptors. At present, it is not clear if an early loss of mRNA expression after an ischemic insult leads to a functional defect in GABA_A receptors. If so, a loss of GABA neurotransmission may contribute to the development of neuronal degeneration following cerebral ischemia. The maintenance of normal GABA neurotransmission in surviving cells may explain their resistance to ischemia-induced neuronal death.     

DNA fragmentation in the CA2 sector of gerbil hippocampus following transient forebrain ischemia

It has been reported that following transient forebrain ischemia in the gerbil, "delayed neuronal death" and "reactive change" occur in hippocampal CA1 and CA2 sectors, respectively. In the present study, using the gerbil transient forebrain ischemia model, we examined brain sections after various recirculation periods and demonstrated, employing the in situ nick-end labeling (TUNEL) method, a nuclear DNA fragmentation in the damaged CA2 neurons.     

Multiple neurotrophic signals converge in surviving CA1 neurons of the gerbil hippocampus following transient forebrain ischemia

Delayed cell death involving the CA1 area of the hippocampus was produced following 5 minutes of transient forebrain ischemia in gerbils. Cell death mainly affected CA1 pyramidal neurons, whereas parvalbumin-immunoreactive (parv-ir) cells were spared. Synaptophysin immunoreactivity was observed in the strata oriens and radiatum of CA1 for months, although immunoreactivity decreased in gerbils surviving 1 year post-ischemia. Golgi studies disclosed a few pyramidal neurons with dendrites, variably covered with dendritic spines, in the CA1 area of 1-year surviving gerbils. In the normal gerbil, the majority of CA1 neurons expressed brain-derived neurotrophic factor (BDNF), tyrosine protein kinase C (TrkC), fibroblast growth factor receptor 1 (Flg), transforming growth factor-alpha (TGF-alpha), and epidermal growth factor-receptor (EGF-R), but only a minority of cells were tyrosine protein kinase B (TrkB)-immunoreactive. Marked reduction in the number of BDNF-, TrkC-, Flg-, TGF-alpha-, and EGF-R-ir cells was observed in CA1 from 24 hours to 1 year after ischemia. In contrast, TrkB-ir cells survived the ischemic insult. Double-labeling immunohistochemistry disclosed that about 90% of surviving BDNF-ir and 85% of TrkB-ir neurons co-localized parvalbumin in the CA1 area. In control gerbils, only about 5% of BDNF-ir cells in CA1 co-expressed TrkB. However, TrkB co-localized in about 95% of surviving BDNF-ir neurons in CA1 in ischemic gerbils. In addition, parvalbumin was co-expressed in about 90% of TrkC-, 95% Flg-, and 85% EGF-R-ir surviving neurons in the stratum pyramidale of CA1. Finally, basic fibroblast growth factor (bFGF) was expressed by reactive astrocytes from day 4 onwards. These data show that the subpopulation of TrkB-/parv-ir neurons in CA1 survive the ischemic episode and that multiple neurotrophic signals converge in surviving neurons of the gerbil hippocampus following transient forebrain ischemia. J. Comp. Neurol. 394:416–430, 1998. © 1998 Wiley-Liss, Inc.     

Chronic ethanol treatment reduces inhibition in CA1 of the rat hippocampus

The effect of chronic ethanol exposure on inhibition in the rat hippocampal slice was investigated using paired-pulse stimulation techniques with stimulation in stratum radiatum or stratum oriens of CA1. Experimental animals were fed ethanol in a liquid diet for 20 weeks and were withdrawn for at least 8 weeks prior to electrophysiological recording. Prior ethanol treatment had no effect on basic input-output relationships for the extracellular population spike. Ethanol treatment significantly reduced the recurrent inhibition produced by antidromic stimulation in a manner dependent upon stimulus intensity. In addition, with orthodromic paired-pulse stimulation of either stratum radiatum or oriens, a trend toward an augmentation of the facilitation of population spike amplitude was observed, suggesting that feedforward inhibition may also be reduced. These results are similar to those found with treatments that reduce inhibition. Therefore, we conclude that chronic ethanol exposure produces an enduring disruption of inhibitory neuronal function in the rat hippocampus.     

Purification of γ-aminobutyric acid (GABA) receptor from rat brain by affinity column chromatography using a new benzodiazepine, 1012-S, as an immobilized ligand

The purification of γ-aminobutyric acid (GABA) and benzodiazepine receptors from the rat brain was employed by the affinity column using a new benzodiazepine, 1012-S, as immobilized ligand. The 1012-S has a aliphatic primary amino group and exhibited an extremely high potency for displacing [³H]flunitrazepam binding to solubilized benzodiazepine receptor preparation (IC₅₀ = 6.0 × 10⁻¹¹M). This benzodiazepine affinity gel retained almost all of the solubilized GABA receptors from synaptic membranes applied to the column, and 25.6% of the receptor was eluted bio-specifically following the application of 1 mM 1012-S. The highest purification fold thus obtained was 4576 (specific activity: 0.99 nmol/mg protein). Furthermore, the successive application of 1–2 M NaSCN also resulted the elution of a highly enriched GABA receptor (specific activity: 0.41 nmol/mg protein; purification fold: 1889). SDS-polyacrylamide gel electrophoretic profiles of the bio-specifically eluted fraction with 1012-S showed the existence of two major bands having the molecular weights of approximately 48,500 and 54,500 in which the former band was selectively photoaffinity-labelled with [³H]flunitrazepam. On the other hand, it was found that the non-specifically eluted fraction with NaSCN contained 4 additional minor bands having molecular weights of 41,000 to 51,000. These results indicate that GABA receptor of the rat brain is coupled, at least in part, with benzodiazepine receptor and is readily purified by the use of highly specific benzodiazepine affinity gel, 1012-S-acetamide adipic hydrazide Sepharose 4B. The present results also suggest that the purified GABA/benzodiazepine receptor complex may contain two different kinds of subunits having the molecular weights of 48,000 and 54,500, in which the former subunit may possess benzodiazepine binding sites.