Evidence for a role of a vicinal dithiol in the transport of gamma-butyrobetaine in Agrobacterium sp

An Agrobacterium sp. isolated from soil is able to use gamma-butyrobetaine as its sole source of carbon and nitrogen. The involvement of thiol groups for active transport of gamma-butyrobetaine was investigated by use of the thiol alkylating reagent N-ethylmaleimide (NEM) and the dithiol specific reagent phenylarsine oxide (PAO). Both reagents strongly inhibited gamma-butyrobetaine uptake, but also induced the release of the accumulated substrate, suggesting that the transport system either contains a dithiol-dependent protein or that a small thiol-containing molecule is implicated in the uptake phenomenon.     

INHIBITION OF THE UPTAKE OF GABA AND RELATED AMINO ACIDS IN RAT BRAIN SLICES BY THE OPTICAL ISOMERS OF NIPECOTIC ACID

R(-)-Nipecotic acid was a more potent inhibitor than the S(+)-isomer of the uptake of GABA, (+)-nipecotic acid, and β-alanine in rat brain slices. (-)-Nipecotic acid was an order of magnitude more potent as an inhibitor of GABA uptake than as an inhibitor of β-alanine uptake, whereas the (+)-isomer was less selective. (–)-Nipecotic acid was a weak inhibitor of L-proline uptake and of rat brain acetylcholinesterase activity. Kinetic studies showed that both isomers of nipecotic acid were competitive inhibitors of GABA uptake when added at the same time as GABA, but non-competitive inhibitors when preincubated with the tissue for 15 min before addition of GABA. The apparent slope inhibition constants, which were not influenced by preincubation, indicated that (–)-nipecotic acid has an affinity for the carrier some 5 times higher than that for (+)-nipecotic acid. (–)-Nipecotic acid stimulated the release of preloaded radioactive GABA from rat brain slices. These observations indicate that (–)-nipecotic acid is a substrate-competitive inhibitor of GABA which combines with the GABA carrier and is taken up. (?)-Nipecotic acid and (+)-2,4-diaminobutyric acid, on the basis of their absolute structures and inhibition kinetics, are proposed to interact in a similar way with the GABA transport system.     

Unmasking of an essential thiol during function of the membrane-bound enzyme II of the phosphenolpyruvate beta-glucoside phosphotransferase system of Escherichia coli.

beta-Glucoside transport by phosphoenolpyruvate-hexose phosphotransferase system in Escherichia coli is inactivated in vivo by thiol reagents. This inactivation is strongly enhanced by the presence of transported substrates. In a system reconstituted from soluble and membrane-bound components, only the particulate component, the membrane-bound enzyme IIbgl appeared as the target of N-ethylmaleimide inaction. The same feature was found in the case of methyl-alpha-D-glucoside uptake via enzyme IIglc. It is shown that the sensitizing effect of substrates is specific and not generalized, methyl-alpha-D-glucoside only sensitizes enzyme IIglc and p-nitrophenyl-beta-D-glucoside only sensitizes enzyme IIbgl towards N-ethylmaleimide inactivation. The inactivation of enzyme IIbgl by thiol reagents is also promoted in vivo by fluoride inhibition of phosphoenolpyruvate synthesis. In toluene-treated bacteria, the presence of phosphoenolpyruvate protects against inactivation by thiol reagents of p-nitrophenyl-beta-D-glucoside phosphorylation. Both results suggest that the inactivator resistent form of enzyme IIbgl is an energized form of the enzyme.     

Taurine and GABA release from mouse cerebral cortex slices: Effects of structural analogues and drugs

The effects of structural analogues, excitatory amino acids and certain drugs on spontaneous and potassium-stimulated exogenous taurine and GABA release were investigated in mouse cerebral cortex slices using a superfusion system. Spontaneous efflux of both amino acids was rather slow but could be enhanced by their uptake inhibitors. Taurine efflux was facilitated by exogenous taurine, hypotaurine, -alanine and GABA, whereas GABA, nipecotic acid and homotaurine effectively enhanced GABA release. The stimulatory potency of the analogues closely corresponded to their ability to inhibit taurine and GABA uptake, respectively, indicating that these efflux processes could be mediated by the carriers operating outwards. Glutamate induced GABA release, whereas taurine efflux was potentiated by aspartate, glutamate, cysteate, homocysteate and kainate. The centrally acting drugs, including GABA agonists and antagonists, as well as the proposed taurine antagonist TAG (6-aminomethyl-3-methyl-4H-1,2,4-benzothiadiazine-1,1-dioxide), had no marked effects on spontaneous taurine and GABA release. Potassium ions stimulated dosedependently both taurine and GABA release from the slices, the responses of taurine being strikingly slow but sustained. Exogenous GABA and nipecotic acid accelerated the potassium-stimulated GABA release, whereas picrotoxin and bicuculline were ineffective. The potassium-stimulated taurine release was unaltered or suppressed by exogenous taurine and analogues, differing in this respect from GABA release. The apparent magnitude of the depolarization-induced GABA release is thus influenced by the function of membrane transport sites, but the same conclusion cannot be drawn with regard to taurine. Haloperidol and imipramine were able to affect the evoked release of both taurine and GABA.     

Carnitine transport in rat heart slices: I. The action of thiol reagents on the acetylcarnitine/carnitine exchange

Rat heart slices show a permeability barrier that can be crossed by carnitine but not by sucrose and inulin. The integrity of thiol groups of heart cell membrane is essential for the uptake of carnitine. N-ethylmaleimide inhibits the transport into heart slices which is insensitive to Mersalyl. On the contrary both N-ethylmaleimide and Mersalyl inhibit acetyl carnitine/carnitine exchange. The amount of thiol groups titrated by the above reagents are related to the extent of exchange inhibition.     

Transport of L-carnitine in isolated cerebral cortex neurons.

The accumulation of carnitine was measured in cerebral cortex neurons isolated from adult rat brain. This process was found to be lowered by 40% after preincubation with ouabain and with SH-group reagents (N-ethylmaleimide and mersalyl). The initial velocity of carnitine transport was found to be inhibited by 4-aminobutyrate (GABA) in a competitive way (Ki = 20.9 +/- 2.4 mM). However, of various inhibitors of GABA transporters, only nipecotic acid and very high concentrations of 1-[2-([(diphenylmethylene)amino]oxy)ethyl]-1,2,5,6-tetrahydro-3-pyridinecarboxylic acid hydrochloride (NO-711) acid decreased carnitine accumulation while betaine, taurine and beta-alanine had no effect. The GABA transporters expressed in Xenopus laevis oocytes did not transport carnitine. Moreover, carnitine was not observed to diminish the accumulation of GABA in cerebral cortex neurons, which further excluded a possible involvement of the GABA transporter GAT1 in the process of carnitine accumulation, despite the expression of this protein in the cells under study. The absence of carnitine transporter OCTN2 in rat cerebral cortex neurons (K. A. Na?ecz, D. Dymna, J. E. Mroczkowska, A. Bro?r, S. Bro?r, M. J. Na?ecz and R. Cecchelli, unpublished results), together with the insensitivity of carnitine accumulation towards betaines, implies that a novel transporting protein is present in these cells.     

A Highly Active ATP-Insensitive K+ Import Pathway in Plant Mitochondria

We describe here a regulated and highly active K+ uptake pathway in potato (Solanum tuberosum), tomato (Lycopersicon esculentum), and maize (Zea mays) mitochondria. K+ transport was not inhibited by ATP, NADH, or thiol reagents, which regulate ATP-sensitive K+ channels previously described in plant and mammalian mitochondria. However, K+ uptake was completely prevented by quinine, a broad spectrum K+ channel inhibitor. Increased K+ uptake in plants leads to mitochondrial swelling, respiratory stimulation, heat release, and the prevention of reactive oxygen species formation. This newly described ATP-insensitive K+ import pathway is potentially involved in metabolism regulation and prevention of oxidative stress.     

POSTNATAL ALTERATIONS IN EFFECTS OF POTASSIUM ON UPTAKE AND RELEASE OF GLUTAMATE AND GABA IN RAT BRAIN CORTEX SLICES

The uptake and release of glutamate and of GABA, as well as the effect of high potassium concentrations (35 or 80 mM) hereupon, were studied by aid of ¹⁴C-labelled amino acids in brain cortex slices from rats of different ages between birth and adulthood. Both the extent of the uptake (i.e. the tissue/medium ratio of ¹⁴C at, or close to, equilibrium) and the rate of uptake (i.e. the tissue/ medium ratio of ¹⁴C after short (5 min) incubation periods) increased with age. Differences were, however, found between glutamate and GABA, and the extent of the GABA uptake had a distinct maximum during the second postnatal week. At all ages, high concentrations of potassium caused a decrease in the rate of GABA uptake but were without effect on the rate with which glutamate was taken up. The release of the two amino acids occurred with approximately the same half-time (50 min) in slices from animals of at least 14 days of age. Before that time the release of glutamate was somewhat faster, whereas that of GABA was much slower, especially during the first postnatal week (half-time 90 min). The ontogenetic alterations in the effect of excess potassium were complex and varied both between the two potassium concentrations used and between the two amino acids. The results are thus compatible with the existence of different transport systems for the two amino acids, They also suggest that glutamate may exert other functions in addition to its role as a putative transmitter.     

Effect of Homo-β-Proline and Other Heterocyclic GAB A Analogues on GABA Uptake in Neurons and Astroglial Cells and on GABA Receptor Binding

Abstract: Two groups of GABA (γ-aminobutyric acid) analogues, one comprising derivatives of β-proline and the other compounds structurally related to nipecotic acid, were investigated as potential inhibitors of high-affinity GABA transport in neurons and glial cells, as well as displacers of GABA receptor binding. In addition to cis -4-hydroxynipecotic acid, which is known as a potent inhibitor of GABA uptake, homo-β-proline was the only compound which proved to be a potent inhibitor of glial as well as neuronal GABA uptake. IC₅₀ values for GABA uptake into glial cells and brain cortex "prisms" were 20 and 75 μM, respectively, and the IC₅₀ value obtained for GABA uptake into cultured neurons was 10 μM. A kinetic analysis of the action of homo-β-proline on GABA uptake into cultured astrocytes and neurons showed that this compound acts as a competitive inhibitor of GABA uptake in both cell types. From the apparent K _m values, K _i values for homo-β-proline of 16 and 6 μM could be calculated for glial and neuronal uptake, respectively. This mechanism of action strongly suggests that homo-β-proline interacts with the GABA carriers. Furthermore, homo-β-proline also displaced GABA from its receptor with an IC₅₀ value of 0.3 μM. The cis -4-hydroxynipecotic acid analogues, cis- and trans-4-mercaptonipecotic acid, had no inhibitory effect on glial or neuronal GABA uptake. Other SH reagents, PCMB, NEM and DTNB, were shown to be relatively weak inhibitors of GABA uptake into cultured astrocytes, suggesting that SH groups are not directly involved in the interaction between GABA and its transport carrier.     

Peroxidation induced changes in synaptosomal transport of dopamine and gamma-aminobutyric acid

The effects of iron-dependent peroxidation on respiration and neurotransmitter transport of brain nerve endings has been studied. Rat brain synaptosomes were peroxidized by exposure to an ADP-Fe/ascorbate system and the protective effect of added Se, Cd, or Zn was investigated with regard to dopamine and gamma-aminobutyric acid (GABA) transport. Peroxidation impaired the respiration of synaptosomes by about 20% and caused a marked increase in dopamine uptake; but in contrast, peroxidation induced a large decrease in synaptosomal uptake of GABA. The increased dopamine transport into synaptosomes was partially prevented by the presence of Zn, Se, or Cd. The presence of Zn, Cd, or Se, in order of decreasing effectiveness, also slowed down ADP-Fe/ascorbate mediated peroxidation of synaptosomes. Peroxidation caused a significant inhibition of veratridine-dependent release of both dopamine and GABA from synaptosomes, but the KCl-dependent release of these neurotransmitters was not effected by peroxidation. These results implicate that peroxidation damage of nerve endings may lead to large changes in neurotransmitter transport thus resulting in an alteration in the function of the central nervous system.     

Glycine is taken up through GLYT1 and GLYT2 transporters into mouse spinal cord axon terminals and causes vesicular and carrier-mediated release of its proposed co-transmitter GABA

Glycine and GABA are likely co-transmitters in the spinal cord. Their possible interactions in presynaptic terminals have, however, not been investigated. We studied the effects of glycine on GABA release using superfused mouse spinal cord synaptosomes. Glycine concentration dependently elicited [(3)H]GABA release which was insensitive to strychnine or 5,7-dichlorokynurenic acid, but was Na(+) dependent and sensitive to the glycine uptake blocker glycyldodecylamide. The glycine effect was external Ca(2+) independent, but was reduced when intraterminal Ca(2+) was chelated with 1,2-bis-(2-aminophenoxy)ethane-N,N,N',N'-tetracetic acid or depleted with thapsigargin, or when vesicular storage was impaired with bafilomycin. Glycine-induced [(3)H]GABA release was prevented, in part, by blocking GABA transport. The glycine effect was halved by sarcosine, a GLYT1 substrate/inhibitor, or by amoxapine, a GLYT2 blocker, and abolished by a mixture of the two. The sensitivity to sarcosine, used as a transporter inhibitor or substrate, persisted in synaptosomes prelabelled with [(3)H]GABA in the presence of beta-alanine, excluding major gliasome involvement. To conclude, in mice spinal cord, transporters for glycine (both GLYT1 and GLYT2) and for GABA coexist on the same axon terminals. Activation of the glycine transporters elicits GABA release, partly by internal Ca(2+)-dependent exocytosis and partly by transporter reversal.     

Protective effect of alpha-tocopherol on the Ca2+-transport system of the sarcoplasmic reticulum membranes in hypercholesterolemia

The effect of antioxidant--alpha-tocopherol--on Ca2+-transporting system in sarcoplasmic reticulum (SR) of the rabbit skeletal muscles was studied in hypercholesterolemia (HC). alpha-tocopherol administration to animals with HC produced a break on the curve of temperature dependence of Ca-ATPase activity at about 20 degrees C, that disappeared in HC, increased the rate of "rapid" SH-group binding by thiol reagents, and normalized the level of unsaturated fatty acids in SR membranes without altering phospholipid content. It is suggested that the damage of Ca-ATPase in HC is mainly due to activation of lipid peroxidation.     

Aspartate aminotransferase from chicken heart cytosol. Characterization of SH-groups]

Homogeneous aspartate aminotransferase has been prepared from chicken heart cytosol. The purification procedure includes fractionation with NH4-sulfate and with ethanol, chromatography on ion-exchange cellulose DE-32 and on hydroxylapatite. Crystallization of the enyme is described. The enzyme was shown to contain 4 SH-groups per protein subunit of molecular weight 50 000. Two of the SH-groups are fully buried, they can be blocked with thiol reagents only upon denaturation of the protein. One exposed SH-group is readily modified at alkaline pH by iodoacetamide, N-ethymaleimide or tetranitromethane, without any inhibition of enzymic activity; this group readily reacts also with 5,5,-ditthiobis (2-nitrobenzoate) and p-mercuribenzoate. One SH-group is semi-buried: it is inaccessible to the above-mentioned reagents at pH 8, but can be blocked by p-mercuribenzoate at pH about 5. Blocking with p-mercuribenzoate of two SH-groups-the exposed and the semi-buried one-lowers enzymic activity to 70% of the initial value. Syncatalytic modication of a SH-group observed in aspartate aminotransferase from pig heart cytosol does not occur in chicken enzyme.     

Induced Release of γ-Aminobutyric Acid by a Carrier-Mediated, High-Affinity Uptake of L-Glutamate in Cultured Chick Retina Cells

[3H]gamma-aminobutyric acid (GABA) was taken up by cultured embryonic retina cells during the initial stages of cell differentiation. The accumulated GABA was released in the bathing medium and a transient increase in the efflux of GABA was observed when cultures were pulse-stimulated (2 min) with 0.1 mM L-glutamate but not with D-glutamate. The EC50 for L-glutamate to evoke [3H]GABA release was approximately 15 microM. This value is close to the Km for high-affinity uptake of L-glutamate by retina cells. When Na+ ions were replaced by Li+ ions, L-glutamate-induced release of GABA was abolished. Moreover, L-[14C]glutamate uptake by retina cells was significantly reduced when NaCl was replaced by LiCl in the incubation medium. L-Glutamate elicited release of GABA was Ca2+ independent, and was observed when Ca2+ was replaced by Co2+ or when Mg2+ ions were increased to 10 mM concentration. D-Aspartate, which is taken up by the same high-affinity uptake mechanism as L-glutamate, induced an increase in [3H]GABA efflux comparable to L-glutamate. The addition of unlabeled GABA to the medium also promoted the release of accumulated [3H]GABA. However, GABA was twofold less effective than L-glutamate in eliciting [3H]GABA release. The addition of both GABA and L-glutamate to the incubation medium indicated that [3H]GABA efflux due to L-glutamate and GABA was additive. L-Aspartate also promoted an increase in the efflux of [3H]GABA accumulated by retina cells. However, L-aspartate effect was significantly decreased in the absence of Ca2+ or when Na+ ions were replaced by Li+. Our results indicate that at least three releasable pools of GABA are present in the chick embryo retina cells: (a) a GABA-promoted GABA release-homoexchange, (b) a Ca2+-dependent L-aspartate-promoted release, and (c) a Ca2+-independent, Na+-dependent L-glutamate-evoked release. In addition, our data strongly suggest that the L-glutamate-promoted GABA release is due to a process of exchange of L-glutamate with GABA, which may play a fundamental role in the fine control of the excitability of local circuits in the retina.     

Phenylarsine oxide is able to dissipate synaptic vesicle acidic pool

Phenylarsine oxide (PAO) has a number of targets in the neurons, one of them is exocytotic process. In this study, we have focused on the mechanisms of phenylarsine oxide action on Ca(2+)-dependent and Ca(2+)-independent neurotransmitter release from rat brain synaptosomes. We investigated the influence of phenylarsine oxide on: (i) l-[(14)C]glutamate and [(3)H]GABA release and uptake; (ii) plasma membrane potential using a potential-sensitive fluorescent probe rhodamine 6G; (iii) exo/endocytotic process using a pH-sensitive fluorescent probe acridine orange (AO). It has been found that phenylarsine oxide induced deacidification of synaptic vesicles. This effect was completely abolished by preliminary treatment of synaptosomes with a protonophore FCCP indicating that both reagents injured a proton electrochemical gradient. Dissipation of the proton gradient by low concentrations of phenylarsine oxide (not exceed 1 microM) did not prevent KCl-triggered exocytotic response, but essentially modified endocytotic one. At higher concentrations of phenylarsine oxide (up to 10 microM), the proton gradient dissipation was intensified and the exocytotic response was fully abolished. The reagent did not change plasma membrane potential, but depolarized mitochondria. It also caused potent inhibition of the Ca(2+)-stimulated l-[(14)C]glutamate and [(3)H]GABA release and increase the Ca(2+)-independent release of l-[(14)C]glutamate, but not of [(3)H]GABA. Disulfide-reducing reagents (dithiothreitol and beta-mercaptoethanol) completely prevented phenylarsine oxide-evoked injuries. They could also restore the initial levels of the mitochondrial potential, the exocytotic response to KCl and the release and uptake of neurotransmitters. Our data provide the evidence that phenylarsine oxide causes dissipation of synaptic vesicle acidic pool resulting in the reduction of vesicle filling and as consequence in attenuation of Ca(2+)-stimulated neurotransmitter release.     

Possible reasons for the failure of glutamine to influence GABA release in rat hippocampal slices; Effect of nipecotic acid and methionine sulfoximine

Although labelled glutamine is readily incorporated into labelled releasable GABA, it has been shown recently that high concentrations (0.1–0.5 mM) glutamine do not increase the release of GABA from brain slices, while greatly enhancing that of glutamate. Two possible reasons for this discrepancy were investigated: (a) That released GABA, in contrast to glutamate is not freshly synthesized but derives from GABA taken up by terminals. The possibility was made unlikely by the present finding which showed that even in the presence of the uptake inhibitor nipecotic acid, glutamine failed to enhance GABA release. (b) That glutamine is transported into GABA-ergic terminals by a high-affinity transport system which is saturated even at low glutamine concentrations obtained without adding glutamine to the superfusion fluid. However, when glutamine efflux was further reduced by prolonging depolarization with 50 mM K⁺ and by pretreatment with the glutamine synthetase inhibitor methionine sulfoximine, GABA release was depressed only very little and this decrease was related to the duration of depolarization and not to extracellular glutamine levels. These results can be reconciled with the ready incorporation of labelled glutamine into releasable GABA by assuming that GABA originates from a glutamate pool to which both glutamine and glucose contribute. The formation of releasable GABA however, is not governed by the supply of glutamate in this pool but by the activity of the rate-limiting enzyme glutamate decarboxylase.     

Evidence for net uptake of GABA into mouse astrocytes in primary cultures—Its sodium dependence and potassium independence

Content of GABA was measured in cultured, normal astrocytes (from the brain cortex of newborn mice) together with the effect of nonradioactive GABA on the efflux of labeled GABA from cells previously loaded with [¹⁴C]GABA. An increase of external GABA concentration from 0 to 25 M evoked a rise of the GABA content in the cells to a level which was approximately 50 times that of the incubation medium. Neither 200 nor 2000 M nonradioactive GABA had any effect on the rate of release of radioactivity from cells loaded with [¹⁴C]GABA. Both the high tissue/medium ratio and the lack of a GABA-induced enhancement of the release of radioactivity indicate that the previously observed high-affinity uptake of GABA in cultured astrocytes represents a net uptake and not a homoexchange with endogenous GABA. This uptake is sodium dependent but was found to be unaffected in potassium-free media; the quantitative correlation between GABA transport and sodium transport differed from that reported for synaptosomes.     

Preferential glutamine uptake in rat brain synaptic mitochondria

Glutamine uptake has been studied in purified rat brain mitochondria of synaptic or non-synaptic origin. It was taken up by an active saturable transport mechanism, with an affinity two-times higher in synaptic than in non-synaptic mitochondria (Km = 0.45 and 0.94 mM, respectively). Vmax of uptake was 7-times higher in synaptic mitochondria (Vmax = 9.2 and 1.3 nmol/min per mg protein, respectively). Glutamine transport was found to be inhibited by L-glutamate (IC50 = 0.64 mM) as well as thiol reagents (mersalyl, N-ethylmaleimide). It is suggested that differential uptake of glutamine in mitochondria of synaptic or non-synaptic origin may be a major mechanism in the regulation of the synthesis of the neurotransmitter glutamate.     

Mutual interactions in the transport of taurine,hypotaurine, and GABA in brain slices

The mutual interactions and the effects of GABA on the saturable transport components of taurine and hypotaurine were investigated with mouse brain slices. The low-affinity taurine transport was competitively inhibited by both hypotaurine and GABA. Hypotaurine did not alter the kinetic parameters of high-affinity taurine uptake, whereas there occurred some stimulation with GABA, possibly by heteroexchange. Taurine had no significant effects on high-affinity hypotaurine uptake, whereas the low-affinity component was reduced by both taurine and GABA, GABA strongly interfered with the high-affinity hypotaurine uptake, being the preferred substrate in simultaneous uptake experiments. The results confirm that taurine, hypotaurine, and GABA are transported into brain slices by only one two-component system with affinities highest for GABA and lowest for taurine.     

HYPEROSMOLALITY-INDUCED GABA RELEASE FROM RAT BRAIN SLICES: STUDIES OF CALCIUM DEPENDENCY AND SOURCES OF RELEASE

Abstract— The effects of hyperosmolal superfusion upon the release of preloaded, radio-labeled GABA has been studied, using both first cortical and first pontine brain slices. GABA release was stimulated with either hyperosmolal Na⁺ or sucrose superfusion in cortical slices. This stimulated release of radio-labeled GABA was partially Ca²⁺-dependent in cortical slices. When barium ions replaced Ca²⁺ in hyperosmolal medium, a similar effect was seen. High concentration of magnesium in Ca²⁺ -free hyperosmolal medium did not induce stimulation. The increased release of α-aminoisobutyric acid (AIBA), a non-metabolized amino acid induced by hyperosmolality, was not Ca²⁺-dependent.
GABA release was also stimulated with hyperosmolal sucrose superfusion in pontine slices. The effect of pre-treatment of cortical and pontine slices with β-alanine or L-2,4-diaminobutyric acid (DABA) was used to study the source of exogenous GABA release induced by hyperosmolality. In cortical slices, β-alanine blocked the hyperosmolal release of GABA and also slightly inhibited GABA uptake. DABA did not change hyperosmolal GABA release, although it inhibited GABA uptake. In pontine slices, both DABA and β-alanine inhibited GABA uptake, but were unable to inhibit the hyperosmolal release of GABA.
The data suggest that hyperosmolality causes increased release of GABA from neurons, analogous to that seen with K⁺-depolarization. AIBA, unlike GABA, is released from brain cells as a non-Ca²⁺ -dependent response to osmotic equilibration. The observation that pre-treatment with β-alanine inhibits the hyperosmolal release of GABA suggests that hyperosmolality alters glial cell function.