Increased tau messenger RNA in Alzheimer's disease hippocampus.

The microtubule-associated protein tau is present in the pathologic hallmarks of Alzheimer's disease and its production and deposition have been implicated in the pathogenesis of the disease. We detected tau mRNA using in situ hybridization histochemistry in the hippocampus, visual cortex, and cerebellum, and compared its level in Alzheimer's disease with controls. The amount of tau mRNA also was determined as a ratio of total polyadenylated mRNA in each area. A significant and gene-specific increase in tau mRNA hybridization was found in hippocampal fields CA4 and CA3, with a similar trend in the dentate gyrus. In contrast, no change was found in the visual cortex or cerebellum in Alzheimer's disease. Increased hippocampal expression of tau mRNA also was present in cases of non-Alzheimer's dementia. Enhanced tau mRNA may be a marker of attempted plasticity involving the cytoskeleton in neuronal populations affected by various neurodegenerative disorders.

Neuropeptide Y expression in rat brain: effects of adrenalectomy.

Neuropeptide Y (NPY) messenger RNA was measured by hybridization of mRNA from the cortex, hippocampus, hypothalamus and striatum of rat brains. Adrenalectomized rats showed lowered level of NPY message in the striatum. A similar decline was found in the hypothalamus, while the cortex and hippocampus were unchanged. Levels of NPY message per unit total RNA were about the same for hypothalamus, cortex and striatum and about 50% less for hippocampus. Adrenalectomized rats that received replacement corticosterone had levels of NPY message that had returned to the levels found in rats receiving sham operation. Response elements consistent with our findings are reported in the NPY genomic sequence.

Sequence analysis, tissue distribution and regulation by cell depolarization, and second messengers of bovine secretogranin II (chromogranin C) mRNA.

Secretogranin II is a very acidic, tyrosine-sulfated protein found in secretory granules of cells belonging to the diffuse neuroendocrine system. It gained more general importance recently as a universal immunohistochemical marker for endocrine neoplasms. Sequence information was obtained from secretogranin II isolated from bovine anterior pituitaries, allowing the isolation of cDNA clones and deduction of its primary structure. Bovine secretogranin II is a 586-amino acid protein of 67,455 Da which is preceded by a signal peptide of 27 residues and contains 9 pairs of basic amino acids in its sequence which are used as potential cleavage sites for generation of physiologically active peptides. Moderately abundant mRNA levels were found in adrenal medulla, pituitary, hippocampus, and caudate. Secretogranin II message was absent from parathyroid gland, adrenal cortex, kidney, liver, and spleen. Depolarization of isolated chromaffin cells by various secretagogues significantly up-regulated secretogranin II mRNA levels by mechanisms distinct from those established for chromogranins and neuropeptides, components maintained along with secretogranin II in neuroendocrine storage vesicles.

Structural characterization of Y1 and Y2 receptors for neuropeptide Y and peptide YY by affinity cross-linking.

Pharmacological studies indicate that peptide YY (PYY) and neuropeptide Y interact with multiple binding sites, categorized as Y1 and Y2 subtypes. In order to identify and structurally characterize the Y1 and Y2 receptors we covalently cross-linked [125I-Tyr36]PYY to its receptors. The Y2 receptor in rat hippocampus and rabbit kidney membranes was affinity labeled using different homo- and heterobifunctional cross-linking reagents. Analysis by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and autoradiography resulted in a major labeled protein band of Mr = 50,000 in both hippocampal and kidney membranes, which was unaffected by reducing agents. The Y1 receptor was analyzed in membranes from the MC-IXC human neuroblastoma cell line. Autoradiography revealed two labeled bands at Mr = 70,000 and 45,000. As the intensity of the Mr = 45,000 band was reduced by protease inhibitors, it is likely that this band is a degradation product of the larger band. Labeling of these proteins was obtained only when N-5-azido-2-nitrobenzoyloxysuccinimide was employed for cross-linking followed by exposure to UV light. Labeling of the two cross-linked bands was unaffected by reducing agents. The binding of radiolabeled PYY and the intensity of the cross-linked bands, for both the Y1 and Y2 receptors, were inhibited similarly in a dose-dependent manner by increasing concentrations of unlabeled PYY. When exposed to agarose-coupled lectins, the detergent-solubilized Y1 receptor-hormone complex was completely adsorbed by wheat germ agglutinin and partially by ricin communis II. The cross-linked Y2 receptor was almost totally adsorbed by wheat germ agglutinin-agarose and partially adsorbed by concanavalin A. The adsorptions were in all cases blocked by the appropriate hapten sugar. These results indicate that the Y1 receptor is a glycoprotein with a Mr = 70,000 binding subunit, whereas the Y2 receptor is a glycoprotein with a Mr = 50,000 binding subunit. These results provide evidence that the Y1 and Y2 subtypes of neuropeptide Y and PYY receptors, previously characterized pharmacologically, are structurally distinct glycoproteins, not disulfide-linked to other subunits.

Alpha 3, alpha 5, and beta 4: three members of the rat neuronal nicotinic acetylcholine receptor-related gene family form a gene cluster.

We have identified two additional members of the neuronal nicotinic acetylcholine receptor (nAChR)-related gene family. cDNA clones for one new gene, designated alpha 5, were isolated from rat hippocampus and rat PC12 cell line cDNA libraries. The alpha 5 gene encodes a protein of 48,800 daltons (424 amino acids) which exhibits significant overall amino acid sequence identity with the previously cloned rat nAChR subunits alpha 1 (49%), alpha 2 (55%), alpha 3 (52%), and alpha 4 (49%). Features characteristic of other nAChR alpha-subunits are present such as conserved cysteine residues at positions 127, 141, 191, and 192, and four strongly hydrophobic domains. A second addition to the nAChR-related gene family, designated beta 4, is encoded in overlapping rat genomic clones lambda DD15 and lambda RG518A. The beta 4 gene, encoding a mature protein of 53,300 daltons (475 amino acids), consists of 6 exons and has a transcription unit length of approximately 18 kilobase pairs. The beta 4 gene encoded protein shows considerable amino acid sequence identity with nAChR beta 1 (43%), beta 2 (64%), and beta 3 (44%) subunits. Northern blots showed that, along with alpha 3 and beta 2, transcripts for both the alpha 5 and beta 4 genes are present in the PC12 cell line, while in situ hybridization experiments demonstrated expression of the alpha 5 and beta 4 genes in a small number of nuclei in the central nervous system. Finally, the genes that encode the beta 4, alpha 3, and alpha 5 proteins are transcribed with convergent polarities and form a tightly linked gene cluster spanning approximately 60 kilobase pairs.

Systemic administration of kainic acid produces elevations in TRH in rat central nervous system.

Several studies have suggested that the concentration of thyrotropin releasing hormone (TRH) in the central nervous system (CNS) is influenced by the level of CNS activation. Hibernation in the ground squirrel and estivation in the lungfish result in region-specific decreases in TRH concentrations. Repeated electroconvulsive shock (ECS) and amygdaloid kindling have been shown to result in elevations of TRH in limbic brain regions. In the present study, limbic seizures induced by systemic administration of kainic acid resulted in substantial increases in the TRH content of posterior cortex and of dorsal and ventral hippocampus, and in moderate elevations in anterior cortex, amygdala/piriform cortex and corpus striatum. Maximal elevations in TRH were observed 2-4 days after kainic acid administration, and by 14 days TRH levels were similar to control values, with the exception of the dorsal hippocampus, which exhibited more prolonged elevations in TRH levels. Prior exposure to limbic seizure activity attenuated the magnitude of TRH elevation in response to a second administration of kainic acid in the posterior cortex but in no other region. These results indicate that seizure-related processes or events influence TRH systems in the CNS. Neuronal populations involved in limbic seizure induced damage may be involved in the modulation of posterior cortical TRH levels.

Reserpine-induced decrease in type I and II corticosteroid receptors in neuronal and lymphoid tissues of adrenalectomized rats.

The effect of the biogenic amine depleting drug, reserpine, on the concentration of type II corticosteroid receptors (i.e., glucocorticoid receptors) in neuronal (hippocampus, frontal cortex, hypothalamus), lymphoid (circulating lymphocytes, spleen, thymus) and pituitary tissues as well as hippocampal type I (i.e., mineralocorticoid) receptors was examined in adrenal-intact and adrenalectomized (ADX) rats. Reserpine (2 mg/kg) or vehicle was administered to adrenal-intact rats for 2 consecutive days. Following the second injection rats were ADX and sacrificed 24 h later. Reserpine significantly decreased type I and II hippocampal receptors as well as type II receptors in frontal cortex, hypothalamus, lymphocytes and spleen. Since the reserpine-induced decreases in receptor content could be due to reserpine-induced elevations in circulating corticosterone levels, reserpine (2 mg/kg) or vehicle was administered to 1-day ADX rats which were then sacrificed 2 days later (i.e., 3 days post ADX). A 1-day ADX control group was also included. The 3-day ADX regimen produced significant or nearly significant increases in type II receptors in hippocampus, frontal cortex, hypothalamus, lymphocytes and spleen in vehicle-treated rats. Reserpine attenuated the ADX-induced upregulation of type II receptors in hippocampus, frontal cortex, lymphocytes and spleen, but had no effect on the ADX-induced upregulation of type II receptors in the hypothalamus. The ADX-induced increase in hippocampal type I receptors was not affected by reserpine treatment. In a final experiment, reserpine (2 mg/kg) or vehicle was administered immediately after ADX and rats were sacrificed 24 h later in order to assess the effect of reserpine on basal (i.e., nonupregulated) corticosteroid receptor levels in the absence of circulating corticosterone levels.(ABSTRACT TRUNCATED AT 250 WORDS)

Use of stable isotopes and gas chromatography-mass spectrometry in the study of different pools of neurotransmitter amino acids in brain slices.

A method was developed for simultaneous determination of endogenous and newly synthesized neurotransmitter amino acids (4-aminobutyric acid, glutamate and aspartate) and glutamine in brain in vitro. Brain slices were incubated in artificial cerebrospinal fluid in the presence of 13C-labeled precursors (glucose, pyruvate or acetate). After the incubation, the slices were homogenized in cold 80% ethanol and the supernatants were evaporated to dryness. The resultant residues were derivatized with N-methyl-N-(tert.-butyldimethylsilyl)trifluoroacetamide and analyzed by capillary gas chromatography-mass spectrometry in the electron-impact mode. N(O)-tert.-Butyldimethylsilyl derivatives of the naturally occurring amino acids, their 13C-enriched counterparts and deuterated internal standards were detected as their [M-57]+ fragments using selected-ion monitoring. The method was shown applicable to studying compartmentation of neurotransmitter amino acids.

Regional distribution of progesterone and 5 alpha-pregnane-3,20-dione in rat brain during progesterone-induced "anesthesia".

Progesterone and 5 alpha-pregnane-3,20-dione (5 alpha-DHP) concentrations were measured in seven brain areas and in plasma during "anesthesia" induced by progesterone (1-2 mg IV) in female rats. The highest levels of progesterone were detected in the striatum and hypothalamus (23.3 +/- 5.27 and 22.7 +/- 4.30 micrograms/g +/- SEM, respectively); these concentrations were approximately 1000 times higher than those during the post-ovulatory phase. Highest levels of 5 alpha-DHP were detected in the striatum and hippocampus (11.5 +/- 1.74 and 10.4 +/- 3.15 micrograms/g +/- SEM, respectively). The ratio of 5 alpha-DHP to progesterone was approximately 100 times higher in brain tissue than in plasma. We conclude that a conversion of progesterone to 5 alpha-DHP occurs in the brain during the course of progesterone-induced "anesthesia". This metabolic step may be an important contributory factor to the anesthetic potency of progesterone.

Hippocampal zinc thionein and pyridoxal phosphate modulate synaptic functions.

The hippocampus, a component of the limbic system, is a prominent subcortical structure, which not only contains high concentrations of zinc, but also exhibits regional variations in this essential element, with concentrations being highest in the hilar region and lowest in the fimbria. For example, the concentration of zinc in the mossy fiber axons has been estimated to approach 300-350 microM. Both zinc and pyridoxal phosphate (PLP) deficiency and excess have been reported to produce epileptiform seizures, which are blocked by gamma-aminobutyric acid (GABA). The proposed mechanism is that at physiological concentrations zinc stimulates the activity of the hippocampal pyridoxal kinase (50% stimulation at 1.7 x 10(-7) M), enhancing the formation of PLP, whereas in pharmacological doses zinc inhibits the activity of glutamate decarboxylase (GAD) directly (50% inhibition at 6.5 X 10(-4) M) by preventing the binding of PLP to HoloGAD. Furthermore, recent studies have shown that two forms of GAD are found in the rat brain. One form (GAD A) does not require PLP for maximal activity, while another form (GAD B) does. Furthermore, the ratio between GAD A and GAD B is nonuniform throughout brain areas, and the hippocampus contains twice as much GAD B (the PLP-requiring GAD) as GAD A. Although the hippocampus is a common target of exogenous neurotoxic agents, "free" zinc in greater than physiological concentrations should be considered an endogenous central neurotoxin. For example, iontophoretically applied zinc in the frontoparietal cortex enhances and prolongs the firing rate of neurons in urethane-anesthetized rat. In addition, zinc (50-500 microM) significantly depresses the paired-pulse potentation in the hippocampal CA3 subfield. Moreover, zinc selectively blocks the action of N-methyl-D-aspartate on cortical neurons and enhances the quisqualate receptor-mediated injury. Finally zinc competitively inhibits the calcium-dependent release of transmitter by inhibiting the entry of Ca2+ into the nerve terminals. Since zinc in a concentration of 300-350 microM could not possibly remain "unbound" in the hippocampus, we searched for and identified a metallothionein-like protein (MT) in the bovine hippocampus, which produces two isoforms on reverse-phase HPLC and lacks aromatic amino acids, but possesses metallomercaptide bonds. We believe that the hippocampal metallothionein, by donating zinc to an extensive number of zinc-activated, PLP-mediated biochemical reactions, modulates synaptic functions. Furthermore, by virtue of its inducibility, metallothionein binds additional amounts of zinc, maintains its steady-state concentration, prevents inhibition of an extensive number of sulfhydryl-containing enzymes and receptor sites, and hence averts metal-related neurotoxicity.

Physical evidence of the coupling of solubilized 5-HT1A binding sites with G regulatory proteins.

Previous investigations (El Mestikawy et al., J Neurochem 51: 1031-1040, 1988) have shown that 5-HT1A binding sites (R[5-HT1A]) solubilized by CHAPS from rat hippocampal membranes can be modulated by guanine nucleotides, as expected from their solubilization together with associated G regulatory proteins (G). Studies of the hydrodynamic properties of solubilized R[5-HT1A] have been presently carried out in order to assess in a more direct way the presence of R[5-HT1A]-G complexes in the soluble extract. Under control conditions, the sedimentation of a CHAPS extract from hippocampal membranes through a 5-30% sucrose gradient (200,000 g, 17 hr, 4 degrees) gave two maxima of [3H]8-OH-DPAT binding activity corresponding to sedimentation coefficients of 8.0 S and 10.0 S, respectively. Running the gradient in the presence of 1 microM GTP revealed a significant reduction of the 10.0 S peak, as expected from the loss of material (probably a G protein) normally associated with R[5-HT1A]. Conversely, attempts to prevent the dissociation of R[5-HT1A]-G by treatment of CHAPS soluble hippocampal extracts with the cross-linking reagent disuccinimidyl suberate (0.1 mM) resulted in a significant increase (+70%) in [3H]8-OH-DPAT binding activity associated with the appearance of a new sedimenting material with a higher coefficient (16.5 S). Furthermore, [3H]8-OH-DPAT binding became almost completely insensitive to guanine nucleotides as expected from the irreversible coupling by disuccinimidyl suberate of R[5-HT1A] with G protein(s). WGA-agarose chromatography of CHAPS soluble hippocampal extract supplemented with GTP allowed the physical separation of R[5-HT1A] from the bulk of G proteins, and a concomitant decrease of [3H]8-OH-DPAT high affinity binding capacity. Partial recovery of the latter could be achieved by reconstituting R[5-HT1A]-G complexes upon the addition of a mixture of pure bovine Gi + Go to G-deprived soluble extracts. Finally in vivo treatment with Pertussis toxin (5 micrograms intracerebroventricularly, 48 hr before killing) resulted in a significant reduction of the specific binding of [3H]8-OH-DPAT (-36%) to hippocampal membranes and corresponding CHAPS soluble extracts, and a marked decrease in the inhibitory effect of GppNHp. Accordingly the G protein associated with R[5-HT1A] belongs probably to the Gi or Go families.

[Role of noradrenaline of the dorsal hippocampus in the mechanism of self-stimulation reaction in rats]. / Rol' noradrenalina dorsal'nogo gippokampa v mekhanizme reaktsii samorazdrazheniia u krys.

In has been established that intrahippocampal bilateral injection of NA did not influence common frequency of lateral hypothalamic self stimulation. After the destruction of hippocampal NA - terminals of 6-OHDA increased the frequency of self stimulation and rearing. It is suggested that NA hippocampus inhibit the recall trace of the memory of sensory reinforcement stimuli in the course of stimulation "reward".

The effect of repeated administration of antidepressant drugs on the thyrotropin-releasing hormone (TRH) content of rat brain structures.

The present study investigated the effect of repeated treatment with the antidepressant drugs imipramine, amitryptyline, citalopram and mianserin (10 mg/kg PO, twice daily for 14 days) on levels of thyrotropin-releasing hormone (TRH) in several brain structures (cerebral cortex, amygdala + pyriform cortex, hippocampus, nucleus accumbens, striatum and hypothalamus) of the rat. Amitriptyline caused a marked increase in the TRH content in the striatum and nucleus accumbens. Citalopram and mianserin produced a smaller but significant increase in the TRH content in the striatum only, while imipramine did not significantly affect the TRH concentrations in any of the brain structures. None of the antidepressant drugs administered acutely significantly affected the TRH concentrations in the nucleus accumbens or the striatum. These results indicate that changes in brain TRH induced by antidepressant drugs are not related to their therapeutic activity.

Differential activation of the pituitary-adrenocortical axis after stress in the rat: use of two genetically selected lines (Roman low- and high-avoidance rats) as a model.

We have examined the activation of the pituitary-adrenal axis in two lines of rats, the Roman high (RHA)- and low (RLA)-avoidance rats known to be emotionally different. These rats are selected for rapid acquisition of a conditioned avoidance response (RHA) compared with failure to acquire this response (RLA). In this study the endocrine response (ACTH, corticosterone, aldosterone) of RLA and RHA rats to two types of stress was examined: exposure to open-field stress for 10 min (Op) or exposure to ether vapours for 3 min (E). Basal plasma ACTH concentrations were lower in RLA than in RHA rats (RLA: 110.8 +/- 24.5 ng/l; RHA: 252.7 +/- 60.8 ng/l, P less than 0.05) but the absolute values of ACTH reached after both types of stress were comparable between RLA and RHA rats. Plasma corticosterone and aldosterone under resting conditions were not different between RLA and RHA rats. Plasma corticosterone was higher in RLA following openfield stress (P less than 0.05) while no differences between RLA and RHA were observed after ether stress (RHA: basal = 66 +/- 14.nmol/l, Op = 384 +/- 55, E = 606 +/- 75; RLA: basal = 121 +/- 52, Op = 612 +/- 92, E = 698 +/- 89). Stress-induced increases in plasma aldosterone were higher in the RLA line after both types of stress (RHA: basal = 175 +/- 36 pmol/l, Op = 546 +/- 53, E = 563 +/- 47; RLA: basal = 272 +/- 64, Op = 1246 +/- 91, E = 863 +/- 72).

Selective lesions of granule cells by fluid injections into the dentate gyrus.

Granule cells were selectively lesioned by injections of fluid into the infragranular cleavage plane in the dentate gyrus. The granule cells were axotomized by the cavity created by the fluid and 6 days after the injection there were no granule cells at the injection site. The size of the granule cell loss could be altered by varying the volume and rate of the injection. The loss of granule cells led to a shrinkage of the molecular layer and to a reactive gliosis. The lesion also caused an increase in the density of AChE and Timm staining in the molecular layer above the lesion. Although the increased density of AChE and Timm staining may have been due in part to the shrinkage of the molecular layer, part was due to the growth of inputs in response to the loss of granule cells and/or to the axotomy of the input terminals. The changes seen in the molecular layer above the lesion site ended abruptly at the margins of the lesion and the adjacent molecular and granule cell layers appeared normal.

Characterization of the rat prodynorphin gene.

The structure of full-length rat prodynorphin cDNA and the corresponding gene has been determined. The 2400 base rat prodynorphin mRNA is encoded by four exons. Exons 1 and 2 encode the majority of the 5'-untranslated sequence, while exons 3 and 4 contain the translated region; the entire 3' -untranslated region is contained on exon 4 as well. RNase protection studies, in which a genomic DNA fragment was used to generate a cRNA hybridization probe, have determined the major transcriptional initiation site for both brain and testicular prodynorphin mRNA. Transient expression of transfected fusion genes containing the 5'-flanking DNA of the rat prodynorphin gene linked to the structural sequence of a reporter gene has been used to identify specific genomic DNA fragments from the prodynorphin gene locus which are capable of acting as transcriptional promoters. Multiple regions of genomic DNA appear to have transcriptional promoter activity when introduced into various eukaryotic cell lines.

Chromatographic analyses of the serotonin 5-HT1A receptor solubilized from the rat hippocampus.

Serotonin 5-HT1A receptors in rat hippocampal membranes were solubilized by 10 mM 3-[3-(cholamidopropyl)dimethylammonio]-1-propane sulfonate (CHAPS) and chromatographed on various gels in an attempt to design a relevant protocol for their (partial) purification. In particular, an affinity gel made of the 8-hydroxy-2-(di-n-propylamino)tetralin (8-OH-DPAT) derivative 8-methoxy-2-[(N-propyl, N-butylamino)amino]tetralin (8-MeO-N-PBAT) coupled to Affigel 202 was specially developed for this purpose. First, studies of the effects of various compounds (detergents, lipids, reducing agents, sugars, etc.) on the specific binding of [3H]8-OH-DPAT and on the rate of heat-induced inactivation of solubilized 5-HT1A sites led to a buffer composed of 50 mM Tris-HCl, 50 microM dithiothreitol, 1 mM CHAPS, 10% glycerol, 0.1 mM MnCl2, and 50 micrograms/ml of cholesteryl hemisuccinate, pH 7.4, ensuring a high degree of stability of solubilized 5-HT1A sites, compatible with chromatographic analyses for 2-4 days at 4 degrees C. Adsorption and subsequent elution of [3H]8-OH-DPAT specific binding sites were found with several chromatographic gels, including wheat germ agglutinin-agarose, phenyl-Sepharose, hydroxylapatite-Ultrogel, diethylaminoethyl (DEAE)-Sepharose, and DEAE-Sephacel. Similarly, 8-MeO-N-PBAT-Affigel 202 allowed the adsorption and subsequent elution (by 1 mM 5-HT) of active 5-HT1A binding sites solubilized from rat hippocampal membranes. The two-step chromatography using 8-MeO-N-PBAT-Affigel 202 followed by wheat germ agglutinin-agarose gave a fraction enriched (by at least 400-fold) in 5-HT1A sites. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis of this partially purified fraction revealed a major protein band with Mr close to 60,000.

Immunohistochemical evidence for reorganization of tau in the plaques and tangles in Alzheimer's disease.

Cytochemical and biochemical techniques have been used to assess the relationship of epitopes on the microtubule-associated protein, tau, to the cytoskeletal pathology of Alzheimer's disease. The main probes were Tau-1 and Alz-50, two monoclonal antibodies which recognize tau and a potentially related 68 kDa protein. Sequential treatment of tissue slices with combinations of the antibodies showed that each blocked the binding of the other to neurofibrillary tangles and neuritic plaques but not to normal axons. Western blot analysis of tau proteins isolated from Alzheimer's disease brains did not reveal such blocking patterns. The issue of steric hindrance affecting antibody binding in tissue sections was addressed by using Alz-50 in combination with Tau-2, another monoclonal antibody recognizing tau on blots and in Alzheimer's disease pathology. Neither antibody blocked the binding of the other to neurofibrillary tangles and neuritic plaques. These data suggest that the Alz-50 and Tau-1 epitopes are selectively organized in the tangles and plaques to be in close proximity which supports the hypothesis that in Alzheimer's disease pathology, tau is modified.

Multiple GABAA receptor alpha subunit mRNAs revealed by developmental and regional expression in rat, chicken and human brain.

GABAA receptor alpha subunit transcripts were detected by Northern analysis of rat, chicken and human brain mRNA using a series of 32P-labelled antisense RNA probes derived from human alpha 1 subunit cDNAs. These alpha subunit mRNAs differ in their distribution among various brain regions in the rat and at least one species is detected primarily in fetal brain. GABAA receptor alpha 1 subunit probes encoding the putative extracellular domain detect at least five alpha subunit transcripts in rat brain, whereas probes encoding the putative intracellular domain detect only two mRNAs. These data suggest the presence in brain of multiple GABAA receptor alpha subunits having homologous extracellular domains and whose expression is regionally and developmentally regulated. These alpha subunit transcripts may encode proteins that comprise GABAA isoreceptors differing in their pharmacological and physiological properties.