DT-n-PROPYLACETATE AND GABA DEGRADATION. PREFERENTIAL INHIBITION OF SUCCINIC SEMIALDEHYDE DEHYDROGENASE AND INDIRECT INHIBITION OF GABA-TRANSAMINASE

The kinetic constants for 4-aminobutyrate: 2-oxoglutarate aminotransferase (GABA-trans-aminase) and succinate-semialdehyde: NAD⁺ oxidoreductase (SSA-DH) have been determined using rat brain homogenate. The Michaelis constants for GABA-T at saturated substrate concentrations were calculated to be K_gaba= 1.5 mM, K_2-OG= 0.25 mM, K_GLU= 620 μM, and K_SSA= 87 μm. The V_max for the reaction using GABA and 2-oxoglutarate (2-OG) as substrates (forward reaction) was found to be 35.2 μmol/ These results indicate that MOPEG is a measure for changes in central noradrenaline turnover and that drugs affect MOPEG in the brain and spinal cord similarly. Fractional rate constants of MOPEG in the rat brain and spinal cord were estimated with the exponential decline curves produced by treatment with pargyline. Turnover rates of 193 pmol/gh and 167 pmol/g These results indicate that MOPEG is a measure for changes in central noradrenaline turnover and that drugs affect MOPEG in the brain and spinal cord similarly. Fractional rate constants of MOPEG in the rat brain and spinal cord were estimated with the exponential decline curves produced by treatment with pargyline. Turnover rates of 193 pmol/g/h and 167 pmol/g/h in the brain and spinal cord respectively were calculated. The kinetics of GABA-T have been shown to be consistent with a Ping Pong Bi Bi mechanism. Substrate inhibition of the forward reaction, through formation of a dead-end complex, was found to occur with 2-OG (K_i 3.3 mM), whereas GABA was found to be a product inhibitor of the reverse reaction (K_i= 0.6 mM). Using the appropriate Haldane relationship, a K_eq of 0.04 for GGBA-T was found, indicating that the reaction was strongly biased towards GABA. For SSA-DH, the K_m of SSA was determined (9.1 μM) and the V_max was 27.5 μmol/ These results indicate that MOPEG is a measure for changes in central noradrenaline turnover and that drugs affect MOPEG in the brain and spinal cord similarly. Fractional rate constants of MOPEG in the rat brain and spinal cord were estimated with the exponential decline curves produced by treatment with pargyline. Turnover rates of 193 pmol/g/h and 167 pmol/g These results indicate that MOPEG is a measure for changes in central noradrenaline turnover and that drugs affect MOPEG in the brain and spinal cord similarly. Fractional rate constants of MOPEG in the rat brain and spinal cord were estimated with the exponential decline curves produced by treatment with pargyline. Turnover rates of 193 pmol/g/h and 167 pmol/g/h in the brain and spinal cord respectively were calculated. h. The effect of di-n-propylacetate (DPA) on both GABA-T and SSA-DH was measured. DPA inhibited SSA-DH competitively with respect to SSA, giving a K_i of 0.5 mM. GABA-T was only slightly inhibited. The K_i of DPA for the forward reaction was 23.2 mM with respect to GABA, which was 40-50 times higher than that for SSA-DH. For the reverse reaction the K_i of DPA was found to be nearly the same (15.2 mM with respect to Glu and 22.9 mM with respect to SSA). These results suggest that GABA accumulation in the brain, after administration of DPA in vivo, is caused by SSA-DH inhibition. Two mechanisms are indicated by the data. (1) The higher level of SSA, which results from inhibition of SSA-DH, initiates the reverse reaction of GABA-T, thus increasing the level of GABA via conversion of SSA. (2) The degradation of GABA is inhibited by SSA, since SSA has a strong inhibitory effect on the forward reaction, as calculated from the present data.     

SOME EFFECTS OF MONOAMINE OXIDASE INHIBITORS ON THE METABOLISM OF γ-AMINOBUTYRIC ACID IN RAT BRAIN

(1) The inhibitor of γ-aminobutyrate transaminase (GABA-T), amino-oxyacetic acid (AOAA), drastically reduced the activity of GABA-T to 30 per cent of the control value, with a corresponding increase of brain GABA, but had no effect on the activity of glutamate decarboxylase (GAD). (2) The monoamine oxidase (MAO) inhibitors phenelzine, phenylpropylhydrazine and phenylvalerylhydrazine, lowered GABA-T activity to 58, 49 and 48 per cent, respectively; this was associated with a marked elevation of brain GABA. (3) The action of phenelzine and phenylpropylhydrazine in vivo and in vitro could be abolished by pre-treatment of the tissue with the structurally related MAO inhibitors phenylisopropylhydrazine and trans-2-phenylcyclopropylamine. These had no action on the GABA system in vivo, either on the GABA content or on the GABA-T activity. These latter drugs, however, were unable to influence the effects of AOAA either on GABA or on GABA-T. (4) The possible mechanism of action on GABA and the enzyme activities of the GABA system is discussed.     

Interactions of Di-n-Propylacetate, Gabaculine, and Aminooxyacetic Acid: Anticonvulsant Activity and the γ-Aminobutyrate System

Abstract: Di-n-propylacetate (DPA), aminooxyacetic acid (AOAA), and gabaculine were administered alone or in combination to Swiss mice. Six hours after administration of the drugs the anticonvulsant action (against isonicotinic acid hydrazide-induced seizures) of AOAA and DPA combined was less than that of AOAA alone. The cause of this phenomenon appeared to be an interaction between DPA and AOAA with respect to inhibition of GABA-T activity, resulting in a long-term diminished inhibition by AOAA, which in turn led to a lessening of the AOAA-induced elevation in the GABA content of nerve endings (synaptosomes). An excellent correlation was observed between the delay in onset of seizures and the elevation of synaptosomal GABA content.     

gamma-Vinyl GABA (4-amino-hex-5-enoic acid), a new selective irreversible inhibitor of GABA-T: effects on brain GABA metabolism in mice

Abstract— γ-Vinyl GABA (4-amino-hex-5-enoic acid, RMI 71754) is a catalytic inhibitor of GABA-T in vitro. When given by a peripheral route to mice, it crosses the blood-brain barrier and induces a long-lasting, dose-dependent, irreversible inhibition of brain GABA transaminase (GABA-T). Glutamate decarboxylase (GAD) is only slightly affected even at the highest doses used. γ -Vinyl GABA has little or no effect on brain succinate semialdehyde dehydrogenase, aspartate transaminase and alanine transaminase activities. GABA-T inhibition is accompanied by a sustained dose-dependent increase of brain GABA concentration. From the rate of accumulation of GABA it was estimated that GABA turnover in brain was at least 6.5 μmol/g/h. Based on recovery of enzyme activity the half-life of GABA-T was found to be 3.4 days, that of GAD was estimated to be about 2.4 days. γ -Vinyl GABA should be valuable for manipulations of brain GABA metabolism.     

Functioning of the γ-Aminobutyrate Pathway in Wheat Seedlings Affected by Osmotic Stress

γ-Aminobutyrate (GABA) was the only amino acid out of three amino acid intermediates of GABA shunt that increased significantly after 28 h from the beginning of osmotic stress induced by 20 % polyethylene glycol 6000 in wheat seedlings. At the same time specific activities of glutamate decarboxylase (GAD) and GABA aminotransferase (GABA-T) two enzymes of GABA pathway did not change as compared with the control plants. The response of two GABA-T activities (with pyruvate or 2-oxoglutarate as amino acid acceptor) to aminooxyacetate, 3-chloro-L-alanine and p-hydroxymercuribenzoate prompted us to suggest that at least two isoforms of GABA-T showing different substrate specificity do exist in wheat leaves. This revised version was published online in July 2006 with corrections to the Cover Date.     

Molecular cloning and differential expression of an γ-aminobutyrate transaminase gene, OsGABA-T, in rice (Oryza sativa) leaves infected with blast fungus

γ-Aminobutyrate transaminase (GABA-T) catalyzes the conversion of GABA to succinic semialdehyde. Using differential display PCR and cDNA library screening, a full-length GABA-T cDNA (OsGABA-T) was isolated from rice (Oryza sativa) leaves infected with an incompatible race of Magnaporthe grisea. The deduced amino acid sequence comprises 483 amino acid residues and shares 85–69% identity with GABA-T sequences from other plants. OsGABA-T expression is induced by blast fungus infection, mechanical wounding and ultraviolet radiation in rice leaves and is not detected in normal rice organs. This gene is also induced by defense signal molecules such as salicylic acid and abscisic acid, but not by jasmonic acid. Our data suggest that OsGABA-T (GABA shunt) may play a role in restricting the levels of cell death during the host–pathogen interaction.     

Kinetic Studies on the Inhibition of GABA-T by γ-Vinyl GABA and Taurine

γ-Aminobutyric acid transaminase (GABA-T, EC 2.6.1.19) is a pyridoxal phosphate (PLP) dependent enzyme that catalyzes the degradation of γ-aminobutyric acid. The kinetics of this reaction are studied in vitro, both in the absence, and in the presence of two inhibitors: γ-vinyl GABA (4-aminohex-5-enoic acid), and a natural product, taurine (ethylamine-2-sulfonic acid). A kinetic model that describes the transamination process is proposed. GABA-T from Pseudomonas fluorescens is inhibited by γ-vinyl GABA and taurine at concentrations of 51.0 and 78.5?mM. Both inhibitors show competitive inhibition behavior when GABA is the substrate and the inhibition constant (K_i) values for γ-vinyl GABA and taurine were found to be 26±3?mM and 68±7?mM respectively. The transamination process of α-ketoglutarate was not affected by the presence of γ-vinyl GABA, whereas, taurine was a noncompetitive inhibitor of GABA-T when α-ketoglutarate was the substrate. The inhibition dissociation constant (K_ii) for this system was found to be 96±10?mM. The Michaelis-Menten constant (K_m) in the absence of inhibition, was found to be 0.79±0.11?mM, and 0.47±0.10?mM for GABA and α-ketoglutarate respectively.     

Effect of Inhibitors of GABA Transaminase on the Synthesis, Binding, Uptake and Metabolism of GABA

Abstract: Four catalytic inhibitors of GABA aminotransferase (gabaculine, γ-acetylenic GABA, γ-vinyl GABA, ethanolamine O -sulphate) as well as aminooxyacetic acid and valproate were studied for effects on neurochemical assays for GABA synthesis, receptor binding, uptake and metabolism in mouse and rat brain preparations. Gabaculine did not interfere with GABA synthesis as reflected by the activity of glutamate decarboxylase (GAD), it was only a weak inhibitor (IC₅₀= 0.94 mM) of GABA receptor binding sites but was a moderately potent inhibitor of GABA uptake (IC₅₀= 81 μM) and very potent (IC₅₀= 1.8 μM) with respect to inhibition of the GABA-metabolizing enzyme GABA aminotransferase (GABA-T). γ-Acetylenic GABA was a weak inhibitor of GAD and GABA binding (IC₅₀ > 1 mM), but virtually equipotent to inhibit uptake and metabolism of GABA (IC₅₀ 560 and 150 μM, respectively). This was very similar to γ-vinyl GABA, except that this drug did not decrease GAD activity. Ethanolamine O -sulphate was found to show virtually no inhibition of GAD and GABA uptake, but was a fairly potent inhibitor of GABA binding (IC₅₀= 67 μM) and in this respect, 500 times more potent than as an inhibitor of GABA-T. Aminooxyacetic acid was a powerful inhibitor of both GAD and GABA-T (IC₅₀ 14 and 2.7 μM, respectively), but had very little affinity to receptor and uptake sites for GABA. Valproate showed no effects on GABA neurochemical assays which could be related to anticonvulsant action. The present results suggest that the anticonvulsant properties of the four catalytic inhibitors of GABA-T tested are at least in part mediated through a direct influence on GABA receptors and uptake sites.     

The cell death factor,cell wall elicitor of rice blast fungus (Magnaporthe grisea) causes metabolic alterations including GABA shunt in rice cultured cells

An elicitor derived from the cell wall of rice blast fungus (Magnaporthe grisea) causes cell death in suspension cultured cells of rice (Oryza sativa L.). To elucidate the role of M. grisea elicitor on metabolic pathway of rice cells, we performed metabolite profiling using capillary electrophoresis-mass spectrometry (CE/MS). Treatment with M. grisea elicitor increased the amounts of antioxidants and free amino acids and decreased the amount of metabolites in the tricarboxylic acid (TCA) cycle. Lower ATP concentration caused aberrant energy charge, concurrently with reduced amount of NAD(P)H in elicitor treated cells. Among free amino acids detected in this study, the level of gamma-aminobutyric acid (GABA) increased. GABA is metabolized through a bypass pathway of the TCA cycle called GABA shunt, which is composed of glutamate decarboxylase (GAD), GABA transaminase (GABA-T) and succinic semialdehyde dehydrogenase (SSADH). While M. grisea elicitor negligibly affected GAD and SSADH, GABA-T activity significantly decreased. The decrease in GABA-T activity was recovered by NADPH oxidase inhibitor, which prevents cell death induced by M. grisea elicitor. Thus, GABA accumulation observed in rice cells under elicitor stress is partly associated with GABA-T activity.Key words: metabolome, Magnaporthe grisea, capillary electrophoresis, mass spectrometry, gamma-aminobutyric acid, GABA transaminase, Oryza sativa     

Effects of some anticonvulsant drugs on brain GABA level and GAD and GABA-T activities

The effect of anticonvulsant drugs was examined on brain GABA levels and GAD and GABA-T activities. The level of GABA was increased by the treatment with diphenylhydantoin. The drug had no effect on GABA-T activity, whereas GAD activity was inhibited. Carbamazepine increased the GABA level but did not effect GAD and GABA-T activities. Diazepam had no effect on GABA level and GAD activity, whereas it caused a slight inhibition of GABA-T activity. Phenobarbital administration decreased GABA level only at the higher concentration. Clonazepam effected only GAD activity. Some anticonvulsant drugs generally increase brain GABA level; however the lack of correlation with an effect on the GAD and GABA-T activities indicate that other factors than metabolism, such as membrane transport processes, are involved in the mechanism of action of anticonvulsant drugs.     

Effect of Inhibitors of GABA Aminotransferase on the Metabolism of GABA in Brain Tissue and Synaptosomal Fractions

Abstract: Five inhibitors of the GABA degrading enzyme GABA-aminotransferase (GABA-T), viz., gabaculine, γ-acetylenic GABA, γ-vinyl GABA, ethanolamine O -sulphate, and aminooxyacetic acid, as well as GABA itself and the antiepileptic sodium vdproate were administered to mice in doses equieffective to raise the electroconvulsive threshold by 30 V. The animals were killed at the time of maximal anticonvulsant effect of the respective drugs and GABA, GABA-T and glutamate decarboxylase (GAD) were determined in whole brain and synaptosomes, respectively. The synaptosomal fraction was prepared from brain by conventional ultracentrifugation procedures. All drugs studied brought about significant increases in both whole brain and synaptosomal GABA concentrations, and, except GABA itself, inhibited the activity of GABA-T. Furthermore, all drugs, except GABA and γ-acetylenic GABA, activated GAD in the synaptosomal fraction. This was most pronounced with ethanolamine O -sulphate, which induced a twofold activation of this enzyme but exerted only a weak inhibitory effect on GABA-T. The results suggest that activation of GAD is an important factor in the mechanism by which several inhibitors of GABA-T and also valproate increase GABA concentrations in nerve terminals, at least in the relatively non-toxic doses as used in this study.     

Transaminase encoded by NCgl2515 gene of Corynebacterium glutamicum ATCC13032 is involved in γ-aminobutyric acid decomposition

Corynebacterium glutamicum that expresses an exogenous l-glutamate decarboxylase (GAD) gene can synthesize γ-aminobutyric acid (GABA). GABA is decomposed to succinic semialdehyde (SSA) by GABA transaminase (GABA-T) and to succinate thereafter by SSA dehydrogenase (SSADH). However, deletion of the gabT gene encoding GABA-T could not prevent GABA from decomposing at neutral pH. In this study, an additional transaminase gene, NCgl2515, was deleted in a gabT-deleted GAD strain, and GABA fermentation in this gabT NCgl2515-deleted GAD strain was investigated. GABA concentration remained at 22.5–24.0 g/L when pH was maintained at 7.5–8.0, demonstrating that GABA decomposition was reduced. Activity assay indicated that unlike GabT, which exhibits high GABA-T activity (1.34 ± 0.06 U/mg) and utilizes only α-ketoglutarate as amino acceptor, the purified NCgl2515 protein exhibits very low GABA-T activity (approximately 0.03 U/mg) only when coupled with the SSADH, GabD, but can utilize both α-ketoglutarate and pyruvate as amino acceptor. The optimum pH for coupled NCgl2515–GabD was 8.0, similar to that of GabT (7.8). Therefore, NCgl2515 has weak GABA-T activity and is involved in GABA decomposition in C. glutamicum. Deletion of gabT and NCgl2515 could effectively reduce GABA decomposition at neutral pH.     

Anticonvulsant Activity of Muscimol and γ-Aminobutyric Acid Against Pyridoxal Phosphate-Induced Epileptic Seizures

The intracerebroventricular injection of pyridoxal phosphate (PLP, 0.125-1.25 μmol/rat) causes epileptic seizures (4 min → 1 min) that are preventable or reversible by GABA (1 μmol/rat), by muscimol (O.025 μmol/rat), or by diazepam (1.75 μmol/rat). At the peak of PLP-induced convulsions, the activities of GAD and GABA-T in 14 regions of rat brain remained unaltered, whereas the concentrations of PLP remained elevated. The PLP-induced convulsion was blocked by DABA (10 μmol/rat) but was not altered by β-alanine (50 μmol/rat). The previous in vitro studies have shown that PLP increases the uptake of [³H]GABA into synaptosomes and inhibits the binding of [³H]GABA to synaptic membranes. These data suggest that PLP-induced convulsion is due to reduced availability of GABA to its recognition sites, rather than to alteration in the activity of GABA metabolizing enzymes, or unavailability of PLP as a coenzyme for GAD and GABA-T. Since the duration of PLP-induced epileptic seizures is short and can be prevented by GABA agonists, PLP may be used as a tool to study the nature of GABA-mediated neuroinhibition and the properties of GABA receptor sites.     

Glucose inhibits GABA release by pancreatic beta-cells through an increase in GABA shunt activity

GABA is the major inhibitory neurotransmitter in the nervous system. It is also released by the insulin-producing beta-cells, providing them with a potential paracrine regulator. Because glucose was found to inhibit GABA release, we investigated whether extracellular GABA can serve as a marker for glucose-induced mitochondrial activity and thus for the functional state of beta-cells. GABA release by rat and human beta-cells was shown to reflect net GABA production, varying with the functional state of the cells. Net GABA production is the result of GABA formation through glutamate decarboxylase (GAD) and GABA catabolism involving a GABA-transferase (GABA-T)-mediated shunt to the TCA cycle. GABA-T exhibits K(m) values for GABA (1.25 mM) and for alpha-ketoglutarate (alpha-KG; 0.49 mM) that are, respectively, similar to and lower than those in brain. The GABA-T inhibitor gamma-vinyl GABA was used to assess the relative contribution of GABA formation and catabolism to net production and release. The nutrient status of the beta-cells was found to regulate both processes. Glutamine dose-dependently increased GAD-mediated formation of GABA, whereas glucose metabolism shunts part of this GABA to mitochondrial catabolism, involving alpha-KG-induced activation of GABA-T. In absence of extracellular glutamine, glucose also contributed to GABA formation through aminotransferase generation of glutamate from alpha-KG; this stimulatory effect increased GABA release only when GABA-T activity was suppressed. We conclude that GABA release from beta-cells is regulated by glutamine and glucose. Glucose inhibits glutamine-driven GABA formation and release through increasing GABA-T shunt activity by its cellular metabolism. Our data indicate that GABA release by beta-cells can be used to monitor their metabolic responsiveness to glucose irrespective of their insulin-secretory activity.     

Bicuculline and its Effect on γ-Amino Butyric Acid Transaminase in the Guinea-pig Brain Stem

IT has been claimed that the inhibitory effect of γ-amino butyric acid (GABA) is antagonized post-synaptically by the alkaloid bicuculline^1–3, although others⁴ have been unable to demonstrate this in feline cortical neurones. When applied through a microtap, GABA has a profound inhibitory effect on many cochlear nucleus⁵ and other brain stem neurones⁶ and the brain stem is also rich in GABA-transaminase (GABA-T) (ref. 7 and our unpublished results), the enzyme catalysing the initial step in the degradation of GABA. Although there is little doubt that GABA-T is largely a mitochondrial enzyme^8,9, there is considerable activity in purified nerve ending fractions. The data of Salganicoff and de Robertis¹⁰ indicate that two iso-enzymes of GABA-T exist, one in free mitochondria, the other in nerve ending mitochondria.     

Stereoselective uptake of the GABA-transaminase inhibitors gamma-vinyl gaba and gamma-acetylenic GABA into neurons and astrocytes

The cellular uptake of the GABA-transaminase inhibitors gamma-vinyl GABA (GVG) and gamma-acetylenic GABA (GAG) was studied in cultured neurons and astrocytes. By the use of the individual enantiomersR- andS-GVG andR- andS-GAG it could be shown that in both cell types only theS-enantiomers could be actively transported. Comparing neurons and astrocytes only neurons exhibited a high affinity uptake system forS-GVG (K _m 78.2±20.3 M;V _max 0.71±0.06 nmol · min^–1 · mg^–1 cell protein). In case ofS-GAG it could not be established with certainty whether the neuronal uptake was of the high affinity type. Both GVG and GAG were studied as inhibitors of GABA uptake into neurons and astrocytes.S-GVG andS-GAG were found to be weak inhibitors of GABA uptake suggesting thatS-GVG is not transported by the GABA carrier in neurons. The finding of a much more efficient uptake ofS-GVG into neurons than into astrocytes is in line with the previous observation that neuronal GABA-T is more sensitive than astrocytic GABA-T toS-GVG.     

The alteration of γ-aminobutyric acid-transaminase expression in the gerbil hippocampus induced by seizure

It is well established that GABA degradation may play a key role in epileptogenesis. However, whether or not the expression of GABA-transaminase (GABA-T), which catalyzes GABA degradation and participates in the neuronal metabolism via GABA shunt, changes chronologically after on-set of seizure remains to be clarified. To identify the change of GABA-T expression in seizure, GABA-T expression in the gerbil hippocampus, associated with different sequelae of spontaneous seizures, was investigated. The distribution pattern of GABA-T immunoreactive neurons in the hippocampus between the seizure-resistant and pre-seizure group of seizure sensitive gerbils was similar. Interestingly, at 30 min postictal, the enhancement of GABA-T immunoreactivity in the perikarya was apparently observed. This contrasted with the decline in GABA-T immunoreactivity in the granular and pyramidal layer. At 12–24 h postictal, GABA-T immunoreactivity in the hilar neurons had declined significantly. However, the GABA-T immunoreactivity in the granular layer increased. These findings suggest that in the gerbil, the alteration in GABA-T expressions may play an important role in the self-recovery mechanism from seizure attack via both GABA degradation and regulation of neuronal metabolism.     

Regulatory interrelations between GABA and polyamines. I. Brain GABA levels and polyamine metabolism

Elevation of brain GABA levels by GABA-T inhibition is accompanied by a decrease ofS-adenosylmethionine decarboxylase activity. This is followed by an increase of ornithine decarboxylase activity and a severalfold increase of brain putrescine levels. Spermidine and spermine levels are not significantly affected under these conditions. These unexpected findings support a regulatory interaction between GABA and polyamine metabolism.     

Comparative studies on the GABA-transaminase immunoreactivity in rat and gerbil brains.

Gamma-aminobutyric acid (GABA) is the most important inhibitory neurotransmitter in the central nervous system (CNS). Degradation of GABA in the CNS is catalyzed by the action of GABA transaminase (GABA-T). However, the neuroanatomical characteristics of GABA-T in the gerbil, which is a useful experimental animal in neuroscience, are still unknown. Therefore, we performed a comparative analysis of the distribution of GABA-T in rat and gerbil brains using immunohistochemistry. GABA-T immunoreactive neurons were observed in the regions which contained GABAergic neurons of both animals: corpus striatum; substantia nigra, pars reticulata; septal nucleus; and accumbens nucleus. GABA-T + neurons were restricted to layers III and V in the rat. Unlike the rat GABA-T + neurons were observed in layers II, III, and V of the gerbil cerebral cortex. These results suggest that the expression of GABA-T in the gerbil brain may be similar to that in the rat brain, except in the cerebral cortex.     

Regulation of GABA Level in Rat Brain Synaptosomes: Fluxes Through Enzymes of the GABA Shunt and Effects of Glutamate, Calcium, and Ketone Bodies

Abstract: Stable isotopes were used to measure both the rate of GABA formation by glutamic acid decarboxylase (GAD) and the rate of utilization by GABA-transaminase (GABA-T). The initial rate of GABA accumulation, determined with either [2-¹⁵N]glutamine or [²H₅]glutamine as precursor, was 0.3–0.4 nmol/min/mg of protein. Addition of the calcium ionophore A23187 enhanced GAD activity, whereas changes in levels of inorganic phosphate and H⁺ were without influence. Flux through GABA-T (GABA → glutamate), measured with [¹⁵N]GABA as precursor, was 0.82 nmol/min/mg of protein, whereas the reamination of succinic acid semialdehyde (reverse flux through GABA-T) was almost sixfold faster, 4.8 nmol/min/mg of protein. The rate of GABA metabolism via the tricarboxylic acid cycle was very slow, with the upper limit on flux being 0.03 nmol/min/mg of protein. Addition of either acetoacetate or β-hydroxybutyrate raised the internal content of glutamate and reduced that of aspartate; the GABA concentration and the rate of its formation increased. It is concluded that in synaptosomes (a) GABA-T is a primary factor in regulating the turnover of GABA, (b) a major regulator of GAD activity is the concentration of internal calcium, (c) GAD in nerve endings may not be saturated with its substrate, glutamate, and the concentration of the latter is a determinant of flux through this pathway, and (d) levels of ketone bodies increase, and maintain at a higher value, the synaptosomal content of GABA, a phenomenon that may contribute to the beneficial effect of a ketogenic diet in the treatment of epilepsy.