Radiation Inactivation Studies of the Benzodiazepine/γ-Aminobutyric Acid/Chloride Ionophore Receptor Complex

Radiation inactivation was used to estimate the molecular weight of the benzodiazepine (BZ), gamma-aminobutyric acid (GABA), and associated chloride ionophore (picrotoxinin/barbiturate) binding sites in frozen membranes prepared from rat forebrain. The target size of the BZ recognition site (as defined by the binding of the agonists [3H]diazepam and [3H]flunitrazepam, the antagonists [3H]Ro 15-1788 and [3H]CGS 8216, and the inverse agonist [3H]ethyl-beta-carboline-3-carboxylate) averaged 51,000 +/- 2,000 daltons. The presence or absence of GABA during irradiation had no effect on the target size of the BZ recognition site. The apparent molecular weight of the GABA binding site labelled with [3H]muscimol was identical to the BZ receptor when determined under identical assay conditions. However the target size of the picrotoxinin/barbiturate binding site labelled with the cage convulsant [35S]t-butylbicyclophosphorothionate was about threefold larger (138,000 daltons). The effects of lyophilization on BZ receptor binding activity and target size analysis were also determined. A decrease in the number of BZ binding sites (Bmax) was observed in the nonirradiated, lyophilized membranes compared with frozen membranes. Lyophilization of membranes prior to irradiation at -135 degrees C or 30 degrees C resulted in a 53 and 151% increase, respectively, in the molecular weight (target size) estimates of the BZ recognition site when compared with frozen membrane preparations. Two enzymes were also added to the membrane preparations for subsequent target size analysis. In lyophilized preparations irradiated at 30 degrees C, the target size for beta-galactosidase was also increased 71% when compared with frozen membrane preparations. In contrast, the target size for glucose-6-phosphate dehydrogenase was not altered by lyophilization.(ABSTRACT TRUNCATED AT 250 WORDS)     

Purification of the γ-Aminobutyric AcidA/Benzodiazepine Receptor Complex by Immunoaffinity Chromatography

The bovine gamma-aminobutyric acidA/benzodiazepine receptor complex has been purified by a novel immunoaffinity chromatography method on immobilized monoclonal antibody 62-3G1. Immunopurification of the complex was achieved in a single step with an improved yield over affinity chromatography on the benzodiazepine Ro 7-1986/1. High-resolution sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) of the immunoaffinity-purified receptor revealed three major peptide bands of 51,000, 55,000, and 57,000 Mr which were also present in the Ro 7-1986/1 affinity-purified receptor. Peptide mapping, immunoblotting with subunit specific antibodies, and photoaffinity labeling with [3H]flunitrazepam and [3H]muscimol have been used for the identification of receptor subunits, including several which comigrated in a single band in SDS-PAGE.     

Taurine Modulation of the Benzodiazepine-γ-Aminobutyric Acid Receptor Complex in Brain Membranes

In unwashed brain membranes taurine produced an inhibition of [3H]flunitrazepam [( 3H]FNZ) binding with IC50 ranging between 31.5 and 11.9 microM; the IC20 varied between 18 and 26 nM. This inhibitory effect was of a mixed type, with a reduction in Bmax and an increase in KD. Various precursors and metabolites of taurine have a less inhibitory effect. Taurine also has little inhibitory effect (IC50 above 500 microM) on the binding of [3H]ethyl-beta-carboline-3-carboxylate. In extensively washed membranes, 10(-5) M taurine produces a 16-21% increase in the binding of [3H]FNZ while 10(-5) M gamma-aminobutyric acid (GABA) increases it between 31 and 42%. However, if 10(-5) M GABA plus 10(-5) M taurine is included in the assay there is a dramatic inhibitory effect. Taurine causes an inhibition of the GABAergic enhancement of [3H]FNZ binding with an IC50 between 7.3 and 7.8 microM. Binding experiments with [3H]taurine done under different conditions failed to detect a Na+-independent and specific [3H]taurine receptor. These results suggest that endogenous taurine, the second most abundant free amino acid in brain, may play an important modulatory role in the GABA-benzodiazepine receptor complex.     

γ-Aminobutyric Acid: Temperature Dependence in Benzodiazepine Receptor Binding

The effects of muscimol and/or incubation temperature on the inhibition of [3H]flunitrazepam receptor binding by benzodiazepine receptor ligands were investigated. At 0 degree C muscimol decreased the Ki values for some ligands as displacers of [3H]flunitrazepam binding to brain-specific sites while increasing or having no effect on the Ki values for other ligands. The Ki values for some ligands are higher at 37 degrees C than at 0 degree C but are reduced by muscimol at both 0 degrees and 37 degrees C. In contrast, the ligands whose Ki values are increased by muscimol either decreased or did not alter the Ki values at 37 degrees C as compared to those at 0 degree C. Incubation of membranes at 37 degrees C for 30 min accelerated gamma-aminobutyric acid (GABA) release by 221% over that at 0 degree C. These results indicate that changes in incubation temperature alter benzodiazepine receptor affinity for ligands via GABA.     

Characterization with Antibodies of the γ-Aminobutyric AcidA/Benzodiazepine Receptor Complex During Development of the Rat Brain

The postnatal development of the gamma-aminobutyric acidA/benzodiazepine receptor (GABAR/BZDR) complex of the rat brain has been investigated using the monoclonal antibody 62-3G1 and the polyclonal rabbit antiserum A, specific for the 57,000 and 51,000 Mr receptor subunits, respectively. Both GABAR and BZDR binding activities co-precipitated during all postnatal ages. Adult rats showed a main 51,000 Mr[3H]flunitrazepam photoaffinity-labeled peptide, whereas newborn rats showed several photolabeled peptides of higher Mr. All the photolabeled peptides could be immunoprecipitated with each antibody regardless of the age of the rats. These results suggest that the physical coupling between the GABAR and the BZDR is already present in newborn animals and it is maintained afterwards during development. Glycosidase and peptidase treatments of the immunoprecipitated GABAR/BZDR complex indicated that all the [3H]flunitrazepam-photolabeled subunits are different peptides, although they seem to conserve a high degree of homology. In addition to the age-dependent heterogeneity, the results also suggest that for each age, there is heterogeneity in the subunit composition of the GABAR/BZDR complex.     

An N-Protected γ-Aminobutyric Acid Dipeptide with Anticonvulsant Action

Abstract: N -Pivaloyl-leucyl–γ-aminobutyric acid (PLG) is a synthetic dipeptide with a partition coefficient of 1.67 in an ethyl acetate/water system that partially inhibits the synaptosomal uptake and activates the release of [U- ¹⁴C]-γ-aminobutyric acid ([U-¹⁴C]GABA). The displacement of GAB A from crude synaptic membranes by PLG occurs with an IC₅₀ of 10⁻⁵ M . The compound has the capacity to cross the blood-brain barrier and increase central GABA levels. Its ED₅₀ on cardiazol-induced convulsions is 60-65 mg/kg. PLG is resistant to hydrolysis by chymotrypsin and partially inhibits the proteolytic activity of trypsin.     

γ-Aminobutyric Acid Receptor Complex of Insect CNS: Characterization of a Benzodiazepine Binding Site

The specific binding of [N-methyl-3H]flunitrazepam ([3H]FNZP) to a membrane fraction from the supraoesophageal ganglion of the locust (Schistocerca gregaria) has been measured. The ligand binds reversibly with a KD of 47 nM. The binding is Ca2+-dependent, a property not found for the equivalent binding site in vertebrate brain. The pharmacological characteristics of the locust binding site show similarities to both central and peripheral benzodiazepine receptors in mammals. Thus binding is enhanced by gamma-aminobutyric acid (GABA), a feature of mammalian central receptors, whereas the ligand Ro 5-4864 was more effective in displacing [3H]FNZP than was clonazepam, which is the pattern seen in mammalian peripheral receptors. The locust benzodiazepine binding site was photoaffinity-labelled by [3H]FNZP, and two major proteins of Mr 45K and 59K were specifically labelled. In parallel experiments with rat brain membranes a single major protein of Mr 49K was labelled, a finding in keeping with many reports in the literature. We suggest that the FNZP binding site described here is part of the GABA receptor complex of locust ganglia. The insect receptor appears to have the same general organization as its mammalian counterpart but differs significantly in its detailed properties.     

Antibodies to the Human γ2 Subunit of the γ-Aminobutyric AcidA/Benzodiazepine Receptor

Abstract: A γ-aminobutyric acid_A (GABA_A) receptor (GABA_AR) γ₂ subunit (short form) was cloned from an adult human cerebral cortex cDNA library in bacteriophage λgt11. The 261-bp intracellular loop (IL) located between M3 and M4 was amplified using the polymerase chain reaction and inserted into the expression vectors λgt11 and pGEX-3X. Both γ-galactosidase (LacZ) and glutathione-S-transferase (GST) fusion proteins containing the γ₂IL were purified, and a rabbit antibody to the LacZ–γ₂IL was made. The antibody reacted with the γ₂IL of both LacZ and GST fusion proteins and immunoprecipitated the GABA_AR/ benzodiazepine receptor (GABA_AR/BZDR) from bovine and rat brain. The antibody reacted in affinity-purified GABA_AR/BZDR immunoblots with a wide peptide band of 44,000–49,000 M_r. Immunoprecipitation studies with the anti-γ₂IL antibody suggest that in the cerebral cortex, 87% of the GABA_ARs with high affinity for benzodiazepines and 70% of the GABA_ARs with high affinity for muscimol contain at least a γ subunit, probably a γ₂. These results indicate that there are [³H]muscimol binding GABA_ARs that do not bind [³H]flunitrazepam with high affinity. Immunoprecipitations with this and other anti-GABA_AR/BZDR antibodies indicate that the most abundant combination of GABA_AR subunits in the cerebral cortex involves α₁, γ₂ (or other γ), and β₂ and/or β₃ subunits. These subunits coexist in >60% of the GABA_AR/BZDRs in the cerebral cortex. The results also show that a considerable proportion (20–25%) of the cerebellar GABA_AR/BZDRs is clonazepam insensitive. At least 74% of these cerebellar receptors, which likely contain α₆, also contain γ₂ (or other γ) subunit(s). The α₁ and β₂ or β₃ subunits are also frequently associated with γ₂ (or other γ) and α₆ in these cerebellar receptors.     

Calcium-Independent γ-Aminobutyric Acid Release from Growth Cones: Role of γ-Aminobutyric Acid Transport

Neuronal growth cones isolated in bulk from neonatal rat forebrain have uptake and K(+)-stimulated release mechanisms for gamma-aminobutyric acid (GABA). Up to and including postnatal day 5, the K(+)-stimulated release of [3H]GABA and endogenous GABA is Ca2+ independent. At these ages, isolated growth cones neither contain synaptic vesicles nor stain for synaptic vesicle antigens. Here we examined the possibility that the release mechanism underlying Ca2(+)-independent GABA release from isolated growth cones is by reversal of the plasma membrane GABA transporter. The effects of two GABA transporter inhibitors, nipecotic acid and an analogue of nipecotic acid, SKF 89976-A, on K(+)-stimulated release of [3H]GABA from superfused growth cones were examined. Nipecotic acid both stimulated basal [3H]GABA release and enhanced K(+)-stimulated release of [3H]GABA, which indicates that this agent can stimulate GABA release and is, therefore, not a useful inhibitor with which to test the role of the GABA transporter in K(+)-stimulated GABA release from growth cones. In contrast, SKF 89976-A profoundly depressed both basal and K(+)-stimulated [3H]GABA release. This occurred at similar concentrations at which uptake was blocked. These observations provide evidence for a major role of the GABA transporter in GABA release from neuronal growth cones.     

γ-Aminobutyric Acid and Benzodiazepine Receptor Changes Induced by Unilateral 6-Hydroxydopamine Lesions of the Medial Forebrain Bundle

Quantitative autoradiography was used to ascertain alterations in [3H]muscimol, [3H]flunitrazepam (FLU), [3H]naloxone, [3H]D-alanine-D-leucine-enkephalin (DADL), and [3H]spiroperidol binding in basal ganglia 1 week, 4 weeks, and 5 months after unilateral 6-hydroxydopamine lesions of the medial forebrain bundle (MFB) in the rat. At 1 and 4 weeks following lesions, [3H]spiroperidol binding increased 33% in striatum. At 5 months, [3H]spiroperidol was only nonsignificantly increased above control. At 1 week, [3H]muscimol binding decreased 39% in ipsilateral globus pallidus (GP), but increased 41% and 11% in entopeduncular nucleus (EPN) and substantia nigra pars reticulata (SNr), respectively. At 4 weeks, [3H]muscimol binding was reduced 19% in striatum and 44% in GP and remained enhanced by 32% in both EPN and SNr. These changes in [3H]muscimol binding persisted at 5 months. [3H]FLU binding was altered in the same direction as [3H]muscimol binding; however, changes were slower in onset and became significant (and remained so) only at 4 weeks after lesions. Decreases in [3H]naloxone and [3H]DADL binding were seen in striatum, GP, EPN, and SNr. Scatchard analyses revealed that only receptor numbers were altered. This study provides biochemical evidence for differential regulation of striatal GABAergic output to GP and EPN/SNr.     

Heterogeneity of γ-Aminobutyric Acid/Benzodiazepine/β-Carboline Receptor Complex in Rat Spinal Cord

The properties of muscimol, beta-carboline (BC), and benzodiazepine (BZD) binding to crude synaptic membranes were studied in the spinal cord and cerebellum of rats. In cerebellar membranes, the density of high-affinity [3H]muscimol and [3H]6,7-dimethoxy-4-ethyl-beta-carboline ([3H]BCCM) binding sites is almost identical to that of [3H]flunitrazepam ([3H]FLU) or [3H]flumazenil (Ro 15-1788; ethyl-8-fluoro-5,6-dihydro-5-methyl-6-oxo-4H-imidazo[1,5-a] [1-4]benzodiazepine-3-carboxylate). In contrast to the cerebellum, the number of muscimol and BC binding sites in rat spinal cord is approximately 20-25% of the number of FLU or flumazenil binding sites. Moreover, in spinal cord membranes, BC recognition site ligands displace [3H]-flumazenil bound to those sites, with low affinity and a Hill slope significantly less than 1; the potency of the different BCs in displacing [3H]flumazenil is 20-50-fold lower in the spinal cord than in the cerebellum. [3H]Flumazenil is not displaced from spinal cord membranes by the peripheral BZD ligand Ro 5-4864 (4'-chlorodiazepam), whereas it is displaced with low affinity and a Hill slope of less than 1 (nH = 0.4) by CL 218,872 (3-methyl-6-(3-trifluoromethylphenyl)-1,2,4-triazolol[4,3-b] pyridazine). These data suggest that a large number of BZD binding sites in spinal cord (approximately 80%) are of the central-type, BZD2 subclass, whereas the BZD binding sites in cerebellum are predominantly of the central-type, BZD1 subclass.(ABSTRACT TRUNCATED AT 250 WORDS)     

Existence of Sites for Anions and Divalent Cations in the Solubilized γ-Aminobutyric Acid/Benzodiazepine Receptor Complex

This study evaluated the ability of gamma-aminobutyric acid (GABA), baclofen, monovalent anions, divalent cations, and various combinations thereof to protect solubilized benzodiazepine (BZ) receptors of types 1 and 2, when contained together on the complex, against heat inactivation. Neither anions, cations, nor GABA alone provided significant protection of solubilized BZ receptors against heat, but inclusion of monovalent anions or divalent cations together with 500 microM GABA did afford protection. Monovalent anions combined with GABA (500 microM) provided 50% to full protection. Divalent cations, such as CaCl2 (2.5 mM) or MgCl2 (2.5 mM) in the presence of GABA (500 microM) yielded 45% and 24% protection, respectively. Other divalent cations tested (Zn2+, Hg2+, Co2+, and Ni2+) were poor protectors, even when combined with GABA. Monovalent anions (200 mM NaCl) and divalent cations (5 mM CaCl2) when tested together provided no protection. Similarly, baclofen (the GABA-B agonist) provided no protection, either alone or together with anions or divalent cations. These results indicate that the independent but interacting recognition sites of GABA, BZ, anions, and divalent cations, previously detected in the membrane-bound state, are retained in the solubilized state.     

Effects of Sodium and Bicarbonate Ions on γ-Aminobutyric Acid Receptor Binding in Synaptic Membranes of Rat Brain

Crude synaptic membranes treated with Triton X-100 (TX) bound gamma-aminobutyric acid (GABA) to two classes of receptor site in Na+-free 10 mM-Tris-sulfate buffer (pH 7.4), but to only a single class of receptor site in 10 mM Tris-sulfate buffer (pH 7.4), containing 150 mM-NaCl. The high-affinity receptor site in TX membranes was specifically masked in the presence of Na+. However, TX membranes incubated in Krebs-Ringer bicarbonate solution (pH 7.4) bound GABA to two classes of receptor site despite the presence of Na+. It was found that addition of bicarbonate ions to the Na+-containing 10 mM-Tris-sulfate buffer (pH 7.4) could restore that high-affinity class of GABA receptors, rendering both classes detectable. This finding suggests that both Na+ and HCO-3 may have a regulatory function on GABA binding to the receptor.     

Evidence for Functionally Distinct Subclasses of γ-Aminobutyric Acid Receptors in Rabbit Retina

Abstract: γ-Aminobutyric acid (GABA) is a major inhibitory neurotransmitter in the mammalian retina, where it serves many roles in establishing complex response characteristics of ganglion cells. We now provide biochemical and physiological evidence that at least three subclasses of GABA receptors (A1, A2, and B) contribute to different types of synaptic integration. Receptor binding studies indicate that approximately three-fourths of the total number of [³H]GABA binding sites in retina are displaced by the GABA_A receptor antagonist, bicuculline, whereas one-fourth are displaced by the GABA-B receptor agonist, baclofen. GABA_A receptors can be described by a three-site binding model with K_D values of 19 n M , 122 n M , and 5.7 μ M . Benzodiazepines and barbiturates potentiate binding to the GABA_A site, which suggests that significant numbers of GABA_A receptors are coupled to regulatory sites for these compounds and thus are classified as GABA_A1 receptors. The response to pentobarbital appears to involve a conversion of low-affinity sites to higher-affinity sites, and is reflected in changes in the densities of sites at different affinities. Functional studies were used to establish which of the different receptor subclasses regulate release from cholinergic amacrine cells. Our results show that GABA suppresses light-evoked [³H]acetylcholine release via GABA_A2 receptors not coupled to a benzodiazepine or barbiturate regulatory site, and enhances release via GABA_B receptors. GABA_A1 sites do not appear to control acetylcholine release in rabbit retina.     

γ-Aminobutyric AcidA Receptor Function Is Desensitised in Rat Cultured Cerebellar Granule Cells Following Chronic Flunitrazepam Treatment

Abstract: This study examined γ-aminobutyric acid_A (GABA_A) receptor function in cultured rat cerebellar granule cells by using microphysiometry following chronic flunitrazepam exposure, and correlated the findings with the α1 and β2/3 subunit protein expression and [³H]muscimol binding after the same treatment paradigm. Flunitrazepam treatment reduced ( p < 0.05) the maximal GABA-stimulated increase in extracellular acidification rate ( E _max) (16.5 ± 1.2% and 11.3 ± 1.0%, 2-day control and treated cells, respectively; 17.4 ± 1.0% and 9.9 ± 0.7%, 7-day control and treated cells, respectively; best-fit E _max± SEM, n = 7), without affecting the GABA concentration required to elicit 50% of maximal response (EC₅₀) (1.2 ± 1.7 and 2.3 ± 1.8 µ M , 2-day control and treated cells, respectively; 1.7 ± 1.5 and 1.5 ± 1.5 µ M , 7-day control and treated cells, respectively; best-fit EC₅₀± SEM, n = 7). Flunitrazepam exposure also abolished the flunitrazepam potentiation of the GABA response, caused a transient reduction of the GABA_A receptor α1 and β2/3 subunit proteins over the initial 2 days, but did not alter [³H]muscimol binding compared with vehicle-treated cells. The results suggest that changes in GABA_A receptor subunit protein expression, rather than loss of [³H]muscimol binding sites, underlie the chronic flunitrazepam-mediated desensitisation of GABA_A receptor function.     

Effect of Ethanol on γ-Aminobutyric Acid and Glycine Receptor-Coupled Cl− Fluxes in Rat Brain Synaptoneurosomes

Chloride fluxes in synaptoneurosomes in response to additions of gamma-aminobutyric acid, glycine, and ethanol were measured using a chloride-sensitive fluorescent probe 6-methoxy-N-(3-sulfopropyl)quinolinium (SPQ). The Cl- gradient was directed outward by bathing cells in a medium low in Cl- concentration. The synaptoneurosomes responded to both gamma-aminobutyric acid and glycine by outflow of Cl- ions, as judged from an increase in SPQ fluorescence. These effects were inhibited by picrotoxin and strychnine, respectively. Ethanol also produced an outflow of Cl- ions from the synaptoneurosomes. Both picrotoxin and strychnine inhibited this effect. When the antagonists were used together, the inhibiting effect was additive. These results indicate that ethanol affects both gamma-aminobutyric acid and glycine receptor-linked chloride fluxes in the rat brain.     

Distribution of Glycine, γ-Aminobutyric Acid, Glutamate Decarboxylase, and γ-Aminobutyric Acid Transaminase in Rabbit and Mudpuppy Retinas

Abstract: The distributions of glycine, γ-aminobutyric acid (GABA), glutamate decarboxylase (EC 4.1.1.15), and GABA transaminase (EC 2.6.1.19) were determined in rabbit and mudpuppy retinas. In both species, peak levels of the amino acids and the enzymes occurred in the inner plexiform layer. Glutamate decarboxylase was almost entirely confined to the inner plexiform layer. Determinations were also made of the GABA content of 107 individual putative amacrine cell somas from mudpuppy retina. About 30% of those somas were found to have high endogenous GABA levels.     

Effect of Nipecotic Acid, a γ-Aminobutyric Acid Transport Inhibitor, on the Turnover and Release of γ-Aminobutyric Acid in Rat Cortical Slices

Abstract: To see the effect of a γ-aminobutyric acid GABA uptake inhibitor on the efflux and content of endogenous and labeled GABA, rat cortical slices were first labeled with [³H]GABA and then superfused in the absence or presence of 1 mM nipecotic acid. Endogenous GABA released or remaining in the slices was measured with high performance liquid chromatography, which was also used to separate [³H]GABA from its metabolites. In the presence of 3 mM K⁺, nipecotic acid released both endogenous and [³H]GABA, with a specific activity four to five times as high as that present in the slices. The release of labeled metabolite(s) of [³H]GABA was also increased by nipecotic acid. The release of endogenous GABA evoked by 50 mM K⁺ was enhanced fourfold by nipecotic acid but that of [³H]GABA was only doubled when expressed as fractional release. In a medium containing no Ca²⁺ and 10 mM Mg²⁺, the release evoked by 50 mMK⁺ was nearly suppressed in either the absence or the presence of nipecotic acid. In the absence of nipecotic acid electrical stimulation (bursts of 64 Hz) was ineffective in evoking release of either endogenous or [³H]GABA, but in the presence of nipecotic acid it increased the efflux of endogenous GABA threefold, while having much less effect on that of [³H]GABA. Tetrodotoxin (TTX) abolished the effect of electrical stimulation. Both high K⁺ and electrical stimulation increased the amount of endogenous GABA remaining in the slices, and this increase was reduced by omission of Ca²⁺ or by TTX. The results suggest that uptake of GABA released through depolarization is of major importance in removing GABA from extracellular spaces, but the enhancement of spontaneous release by nipecotic acid may involve intracellular heteroexchange. Depolarization in the presence of Ca²⁺ leads to an increased synthesis of GABA, in excess of its release, but the role of this excess GABA remains to be established.     

Pyrrolines as Prodrugs of γ-Aminobutyric Acid Analogues

Abstract: Δ'-Pyrroline, S-methyl-Δ'-pyrroline, and 5.5-dimethyl-Δ'-pyrroline have been identified as substances metabolized to γ-aminobutyric acid (GABA), 4-aminopentanoic acid (raethyl GABA), and 4-amino-4-methylpen-tanoic acid (dimethyl GABA), respectively. An enzyme system residing in the soluble fraction of rabbit liver catalyzes the conversion of Δ'-pyrroline to GABA and its lactam, 2-pyrrolidinone. Acetaldehyde, allopurinol, and cyanide inhibited the reaction. Incubation of deuterium-labeled Δ'-pyrroline with mouse brain homogenates produced deuterated GABA. Mouse liver 10,000 g supernatant and mouse brain homogenates converted S-methyl-Δ'-pyrroline to methyl GABA, and 5,5-dimethyl-Δ'-pyrroline to dimethyl GABA. Four hours after intraperitoneal injection of 5-methyl-Δ'-pyrroline (200 mg/kg), methyl GABA was detected in mouse brain (0.27 μ-mol/g). Dimethyl GABA (1.21 μmol/g) was determined in mouse brain 30 min after intraperitoneal administration of 5.5-dimethyl-Δ'-pyrroline (200 mg/kg). Neither methyl GABA nor dimethyl GABA penetrated into the central nervous system when administered in the periphery. The present studies suggest that pyrrolines may represent a chemical class of brain-penetrating precursors of pharmacologically active analogues of GABA.