Effect of Denervation on a Cholinergic-Specific Ganglioside Antigen (Chol-1) Present in Torpedo Electromotor Presynaptic Plasma Membranes

The presence of Chol-1, an antigen identified in the plasma membrane of cholinergic electromotor nerve terminals of Torpedo marmorata, was investigated in Torpedo electric organ after 3, 6, and 9 weeks' denervation. Denervation was monitored by the cessation of stimulus-evoked discharge potentials, by the reduction in nerve terminals seen morphologically, and by the decrease in ACh and ChAT contents. The content of ganglioside-bound sialic acid did not show any appreciable change with time. Some modification of ganglioside pattern on TLC was observed after 9 weeks' denervation. The presence of Chol-1 after denervation was assayed by its activity in inhibiting the selective complement-induced lysis of the cholinergic subpopulation of guinea pig cortical synaptosome which is mediated by the anti-Chol-1 antiserum. Denervation did not affect Chol-1 immunoreactivity although it did alter the distribution of the immunoreactivity among gangliosides. The possible significance of the results is discussed.     

Specific lysis of GABAergic synaptosomes by an antiserum to glutamate decarboxylase

An antiserum to pure glutamate decarboxylase (GAD) when incubated with rat cortical synaptosomes in the presence of complement caused release of 33-53% of lactate dehydrogenase (LDH) and 22-41% of total GAD. In addition most of the gamma-aminobutyrate (GABA) present was released. Anti-GAD antiserum alone, or complement alone, were without action. The antiserum plus complement had no effect on noradrenaline or choline uptake, and did not release choline acetylase (ChAT). Anti-ChAT serum plus complement released 30-37% of ChAT and 10-13% of LDH. It prevented choline uptake. This serum did not produce GAD release or prevent GABA, choline or noradrenaline uptake. When cortical synaptosomes were exposed to both antisera plus complement, their actions were strictly additive. The data indicate specific lysis of GABAergic and cholinergic synaptosomal sub-populations.     

Confirmation of the Cholinergic Specificity of the Chol-1 Gangliosides in Mammalian Brain Using Affinity-Purified Antisera and Lesions Affecting the Cholinergic Input to the Hippocampus

An antiserum raised to Torpedo electromotor synaptosomal membranes (anti-TSM antiserum) induces a cholinergic-specific immune lysis of mammalian brain synaptosomes and recognizes a group of minor gangliosides appeared, therefore, to be specific to the cholinergic neuron and were designated Chol-1. To confirm the cholinergic specificity of the Chol-1 gangliosidic antigens, we have shown that not only does a mammalian ganglioside fraction that is enriched with respect to the Chol-1 gangliosides inhibit the cholinergic-specific immune lysis induced by the anti-TSM antiserum, but also it can be used to affinity-purify a subpopulation of immunoglobulins from the anti-TSM antiserum that also induce a cholinergic-specific lysis. Furthermore, we have demonstrated that fimbrial lesions, which cause a massive degeneration of cholinergic terminals in the ipsilateral hippocampus, lead to a loss of the Chol-1 gangliosides concomitant with that shown by choline acetyltransferase activity and that lesions to the entorhinal cortex, which cause a loss of mainly glutamergic synapses in the ipsilateral dentate gyrus leading to cholinergic sprouting from adjacent hippocampal areas and an increase in cholinergic markers in the dentate gyrus, produce concomitant increases in choline acetyltransferase activity and Chol-1 content. These results provide strong evidence in favour of the cholinergic specificity of the Chol-1 gangliosides.     

Effects of Nucleus Basalis Lesions on the Muscarinic and Nicotinic Modulation of [3H] Acetylcholine Release in the Rat Cerebral Cortex

Presynaptic muscarinic and nicotinic receptors in the cerebral cortex reportedly inhibit and increase acetylcholine (ACh) release, respectively. In this study, we investigated whether these receptors reside on cholinergic nerve terminals projecting to the cerebral cortex from the nucleus basalis magnocellularis (nbm). Adult male rats received unilateral infusions of ibotenic acid (5 micrograms/1 microliter) in the nbm. Two weeks later, cerebral cortical cholinergic markers (choline acetyltransferase activity, high-affinity choline uptake, and coupled ACh synthesis) were significantly reduced in synaptosomes prepared from the lesioned hemispheres compared to contralateral controls. The depolarization-induced release of [3H]ACh from these synaptosomes was also reduced in the lesioned hemispheres, reflecting the reduced synthesis of transmitter. However, the nbm lesions had no effect on the inhibition of release induced by 100 microM oxotremorine. Synaptosomal [3H]ACh release was not altered by nicotine or the nicotinic agonists anabaseine and 2-(3-pyridyl)-1,4,5,6-tetrahydropyrimidine. Nicotine (10-100 microM) did increase [3H]ACh release in control and lesioned hemispheres in cortical minces, but to a similar extent. These results suggest that neither muscarinic nor nicotinic receptors modulating ACh release reside on nbm-cholinergic terminals.     

Cholinergic-specific glycoconjugates

Cholinergic nerve terminals utilize glycoconjugates in several ways, as surface markers and as structural components of the synaptic vesicles present within them. The surface markers have been discovered immunochemically: antibodies raised against them are able specifically to sensitize the cholinergic subpopulation of mammalian brain synaptosomes to complement-mediated lysis. One such group of antigens (Chol-1) have been identified as a novel series of minor gangliosides having in common a sialylatedN-acetylgalactosamine residue. These gangliosides may constitute the major gangliosides at cholinergic terminals. A second surface antigen (Chol-2) is thought to be a protein with an epitope in common with aTorpedo electric organ ganglioside. Cholinergic synaptic vesicles are rich in a proteoglycan which appears to assist in the sequestration of acetylcholine within the vesicle and to stabilize the vesicle membrane during cycles of exocytosis and recovery. It may be the cholinergic equivalent of the chromogranins.Abbreviations AP affinity purified - ATPase adenosine 5-triphosphate phosphohydrolase - cer ceramido - ChAT choline acetyltransferase - Chol-1, –2 cholinergic-specific antigens - DA dopamine - DOG deoxyglucose - ELISA enzyme-linked immunosorption assay - EOD electric organ discharge - FAB fast atomic bombardment - GABA -aminobutyrate - GAG glycosaminoglycan - gal galactosyl - gaINAc N-acetylgalactosaminyl - glc glucosyl - Glu glutamate - 5-HT 5-hydroxytryptamine - LDH lactate dehydrogenase - NA noradrenaline - NGF nerve growth factor - S, S-S mono-, disialyl - SPM synaptosomal plasma membrane - TH tyrosine hydroxylase - TLC thin-layer chromatography - TSM Torpedo electromotor synaptosomal membrane - VIP vasoactive intestinal polypeptide - VPG vesicle proteoglycan Special issue dedicated to Dr. Leon Wolfe.     

Specific immunolysis of serotonergic nerve terminals using an antiserum against tryptophan hydroxylase

An antiserum to tryptophan hydroxylase purified from whole rat brain when incubated with rat striatal synaptosomes in the presence of complement caused release of 18% of LDH, 20% loss of potassium and 60% loss of tryptophan hydroxylase. Uptake of 5-HT was reduced by 60%. Anti-tryptophan hydroxylase alone, or complement alone were without action. The antiserum plus complement had no effect on DA uptake and did not release TH or GAD. These results suggest selective lysis of serotonergic nerve terminals had occurred. The antiserum plus complement reduced choline uptake by 45%. However, this did not seem due to lysis of cholinergic terminals, as ChAT was not released.     

Biochemical evidence that acetylcholine release from cholinergic nerve terminals is mostly vesicular

The nature of the intraterminal compartments from which acetylcholine (ACh) is released following presynaptic stimulation was investigated. This was pursued by examining the effects of the anticholinergic drug 2-(4-phenylpiperidino)cyclohexanol (AH5183) on the release of newly synthesized [3H]ACh and of endogenous ACh from purified cholinergic nerve terminals (synaptosomes) which were isolated from the electric organs of Torpedo. Preincubation of the synaptosomes, with AH5183 (1-10 microM), does not affect either the intraterminal synthesis of [3H]ACh or the uptake of its precursors, but results in a marked inhibition (85%) of the release of the newly synthesized [3H]ACh. However, when AH5183 is added following the accumulation of [3H]ACh in the nerve terminals, it does not affect [3H]ACh release. AH5183 also has no effect on the release of preformed endogenous ACh. These findings, together with the previous in vitro demonstrations that AH5183 is a potent inhibitor of ACh uptake into isolated cholinergic vesicles, suggest that most of the synaptosomal ACh is secreted by a vesicular mechanism.     

Vasoactive Intestinal Peptide Increases Acetylcholine Synthesis by Rat Hippocampal Slices

The purpose of this study was to determine whether vasoactive intestinal peptide (VIP) might have a presynaptic modulatory effect at cholinergic terminals in the rat hippocampal formation. The exposure of rat hippocampal slices to VIP increased [3H]acetylcholine ([3H]ACh) synthesis from the precursor [3H]choline when tissue was incubated in normal or in high K+ medium; the maximal effect was apparent at 10(-8) M VIP and 10(-7) M VIP, respectively. Also, 10(-7) M VIP increased the activity of choline acetyltransferase (ChAT) in a hippocampal homogenate system. The increased synthesis by hippocampal slices was not the result of a VIP-induced alteration in either the basal release of ACh or the uptake of choline via the high-affinity uptake system. The increase in ACh synthesis induced by VIP in hippocampal slices was not associated with either adenylate cyclase or protein kinase C second messenger systems. There was no correlation between the effect of VIP on cyclic AMP production with that on ACh synthesis; also, forskolin, an activator of adenylate cyclase that increased cyclic AMP production 3.5-fold, did not mimic the effect of VIP on ACh synthesis. Similarly, there was no effect of the protein kinase C activator, phorbol myristate acetate, on ACh synthesis in hippocampal slices. However, the effect of VIP to increase ACh synthesis was not evident in the absence of extracellular calcium, suggesting that the effect of VIP is mediated by a calcium-requiring mechanism. The results suggest that, in the rat hippocampus, VIP has a presynaptic action at cholinergic terminals that results in enhanced synthesis of ACh, possibly by an action that alters ChAT activity.     

Transfection analysis of functional roles of complexin I and II in the exocytosis of two different types of secretory vesicles

Classical neurotransmitters such as gamma-aminobutyric acid and glutamate are released from synaptic nerve terminals by exocytosis of synaptic vesicles. PC12 cells also have SSVs capable of storing acetylcholine (ACh). A novel method to examine the effect of transient transfection of any gene of interest on the exocytosis of SSVs was developed. The transfection of choline acetyltransferase (ChAT) into PC12 cells which have lost ACh synthesizing activity resulted in the accumulation of a substantial amount of ACh. Synthesized ACh was released in Ca(2+)-dependent manner. Release was thought to occur by an exocytosis of SSVs because: (1) release was abolished by treating the cells with vesamicol, a specific inhibitor of the vesicular ACh transporter (VAChT) localizing specifically in SSVs; and (2) the release was further increased by cotransfecting rat VAChT with the ChAT. By means of this method, we showed that overexpression of complexin I or II with ChAT markedly suppressed high-K(+)-dependent ACh release of SSVs.     

Presynaptic Distribution of the Cholinergic-Specific Antigen Chol-1 and 5''-Nucleotidase in Rat Brain, as Determined by Complement-Mediated Release of Neurotransmitters

Nerve terminals prepared from rat cortex and hippocampus were loaded with seven radioactive putative neurotransmitters (serotonin, noradrenaline, dopamine, gamma-aminobutyric acid, aspartate, glutamate, and taurine). The release of these transmitters, choline acetyltransferase, 3,4-dihydroxyphenylalanine decarboxylase, enolase, and lactate dehydrogenase was monitored during complement-mediated lysis. Three antisera were used: anti-5'-nucleotidase, anti-Chol-1, and anti-rat cerebrum. Anti-5'-nucleotidase serum did not cause the release of any labelled transmitter or of any of the enzymes studied. Anti-Chol-1 serum released choline acetyltransferase and small amounts of enolase and lactate dehydrogenase. Anti-rat cerebrum caused the release of all seven transmitters, choline acetyltransferase, and small amounts of the other three enzymes. It was concluded that 5'-nucleotidase was not present on any of the terminals studied, and that Chol-1 is only present on cholinergic terminals.     

Synaptosomal "Membrane-Bound" Choline Acetyltransferase Is Most Sensitive to Inhibition by Choline Mustard

The objectives of the present study were to validate the presence of cytoplasmic and membrane-associated pools of choline acetyltransferase (ChAT) in rat brain synaptosomes, and to evaluate inhibition of these different forms of the enzyme by the nitrogen mustard analogue of choline, choline mustard aziridinium ion (ChM Az). The relative distribution of ChAT and lactate dehydrogenase (LDH) was followed in subfractions of synaptosomes to establish whether ChAT activity associated with salt-washed presynaptic membranes represents membrane-bound protein rather than cytosolic enzyme trapped within undisrupted synaptosomes or revesiculated membrane fragments. The percentage of total synaptosomal ChAT activity (14%) recovered in the final membrane pellet always exceeded that of LDH (6%), lending support to the hypothesis that much of the ChAT associated with the membranes was a membrane bound form of the enzyme. Incubation of purified synaptosomes with ChM Az led to irreversible inhibition of ChAT activity; this loss of enzyme activity could not be accounted for by lysis of nerve terminals during incubation in the presence of the mustard analogue. Subfractionation of the ChM Az-treated nerve terminals revealed that the membrane-bound form of ChAT was inhibited to the greatest extent, followed by the ionically membrane-associated enzyme, with the activity of the water-solubilized enzyme not differing significantly from control. Preparation of the synaptosomal ChAT subfractions from untreated nerve terminals prior to incubation with varying concentrations of ChM Az or naphthylvinylpyridine revealed that under these conditions water-solubilized, ionically membrane-associated, and detergent-solubilized membrane-bound pools of ChAT were not differentially inhibited by either compound.(ABSTRACT TRUNCATED AT 250 WORDS)     

Direct evidence that release-stimulating alpha7* nicotinic cholinergic receptors are localized on human and rat brain glutamatergic axon terminals

The existence on glutamatergic nerve endings of nicotinic acetylcholine receptors (nAChRs) mediating enhancement of glutamate release has often been suggested but not demonstrated directly. Here, we study the effects of nAChR agonists on [3 H]-d-aspartate ([3 H]-d-ASP) release from synaptosomes superfused in conditions known to prevent indirect effects. Nicotinic receptor agonists, while unable to modify the basal [3 H]-d-ASP release from human neocortex or rat striatal synaptosomes, enhanced the Ca2+ -dependent exocytotic release evoked by K+ (12 mm) depolarization. Their rank order of potency were anatoxin-a > epibatidine > nicotine > ACh (+ atropine). The anatoxin-a effect, both in human and rat synaptosomes, was antagonized by mecamylamine, alpha-bungarotoxin or methyllycaconitine. The basal release of [3 H]ACh from human cortical synaptosomes was increased by (-)-nicotine (EC50 = 1.16 +/- 0.33 microm) or by ACh plus atropine (EC50 = 2.0 +/- 0.04 microm). The effect of ACh plus atropine was insensitive to alpha-bungarotoxin, methyllycaconitine or alpha-conotoxin MII, whereas it was totally antagonized by mecamylamine or dihydro-beta-erythroidine. To conclude, glutamatergic axon terminals in human neocortex and in rat striatum possess alpha7* nicotinic heteroreceptors mediating enhancement of glutamate release. Release-enhancing cholinergic autoreceptors in human neocortex are nAChRs with a pharmacological profile compatible with the alpha4beta2 subunit combination.     

Synaptic Function of Cholinergic-Specific Chol-1α Ganglioside

The function of a cholinergic-specific ganglioside, Chol-1alpha, was investigated. The release of acetylcholine from synaptosomes was inhibited by anti-Chol-1alpha monoclonal antibody but not by monoclonal antibodies against other brain gangliosides tested. Chol-1alpha ganglioside stimulated the high-affinity choline uptake by synaptosomes and consequently enhanced acetylcholine synthesis, resulting in an increased release of acetylcholine from synaptosomes. The memory and learning abilities of rats given anti-Chol-1alpha antibody were remarkably suppressed. These in vitro and in vivo studies suggest that Chol-1alpha ganglioside plays a pivotal role in cholinergic synaptic transmission and participates in cognitive function.     

Phenotype dependent differential effects of interleukin-1beta and amyloid-beta on viability and cholinergic phenotype of T17 neuroblastoma cells

Amyloid-beta accumulation in brains of Alzheimer's disease (AD) victims is accompanied by glial inflammatory reactions and preferential loss of cholinergic neurons. Therefore, the aim of this study was to find out whether proinflamatory cytokine interleukin 1beta (IL1beta) modifies effects of amyloid-beta (Abeta) on viability and cholinergic phenotype of septum derived T17 cholinergic neuroblastoma cells. In nondifferentiated T17 cells (NC) Abeta(25-35) (1 microg/ml) caused no changes in choline acetyltransferase (ChAT) activity, acetylcholine (ACh) release, subcellular distribution of acetyl-CoA, but doubled content of trypan blue positive cells. IL1beta (10 ng/ml) increased ACh release (125%) but did not change other parameters of NC. In the presence of Abeta IL1beta also increased ChAT activity (47%), ACh release (100%) but had no effect on acetyl-CoA distribution and cell viability. Differentiation with retinoic acid and dibutyryl cyclic AMP caused over two-fold increase of ChAT activity and ACh content, four-fold increase of ACh release and about 50% decrease of acetyl-CoA level in the mitochondria. In differentiated cells (DC), Abeta decreased ChAT activity (31%), ACh release (47%) and content of acetyl-CoA (80%) in cell cytoplasmic compartment, whereas IL1beta elevated ChAT activity (54%) and ACh release (32%). IL1beta totally reversed Abeta-evoked inhibition of ChAT activity and ACh release and restored control level of cytoplasmic acetyl-CoA but increased fraction of nonviable cells to 25%. Thus, IL1beta could compensate Abeta-evoked cholinergic deficits through the restoration of adequate expression of ChAT and provision of acetyl-CoA to cytoplasmic compartment in cholinergic neurons that survive under such pathologic conditions. These data indicate that IL1beta possess independent cholinotrophic and cholinotoxic activities that may modify Abeta effects on cholinergic neurons.     

Cetiedil, a Drug That Inhibits Acetylcholine Release in Torpedo Electric Organ

The effects of cetiedil, a vasodilatator substance with reported anticholinergic properties, were examined on cholinergic presynaptic functions at the nerve electroplaque junction of Torpedo marmorata using either synaptosomes or slices of intact tissue. Cetiedil abolished the calcium-dependent release of acetylcholine (ACh) triggered by depolarization or by addition of A23187 ionophore, a finding localizing the site of action downstream from the calcium entry step. In addition, a direct effect on the release process itself was indicated by the observation that cetiedil blocks the release of ACh mediated by a recently isolated presynaptic membrane protein, the mediatophore, reconstituted into ACh-containing proteoliposomes. In all three preparations, ACh release was inhibited by cetiedil with a Ki of 5-8 microM. Under the conditions used in these release experiments, the synthesis of ACh and its compartmentation within the nerve terminals were not modified. However, the drug was able to reduce high-affinity choline uptake and vesicular ACh incorporation when it was given together with the radioactive precursor, a result showing that cetiedil has a broad inhibitory action on cholinergic uptake processes.     

Putative Cholinergic-Specific Gangliosides in Guinea Pig Forebrain

The nature of the cholinergic-specific antigen Chol-1 recognized by an antiserum raised against Torpedo cholinergic electromotor synaptosomal plasma membranes was investigated in guinea pig forebrain to establish whether it has a gangliosidic nature in guinea pig as in Torpedo. Gangliosides extracted from guinea pig forebrain and extensively purified to eliminate peptide contaminants were effective in inhibiting the selective lysis of the cholinergic subpopulation of cortical synaptosomes induced by the antiserum. Neuraminidase, protease, alkali, and heat treatment did not impair the inhibitory activity of gangliosides. Whereas the antiserum recognized many gangliosides from Torpedo electric organ, the immunostaining of guinea pig forebrain gangliosides separated on TLC showed only two immunopositive bands migrating close to GT1b and GQ. After affinity purification on Torpedo electric organ gangliosides the activity of the antiserum in inducing complement-mediated lysis was increased and it still recognized the two ganglioside bands on TLC. These results strongly suggest the existence of two polysialogangliosides bearing antigenic determinants specific for the cholinergic neurons.     

Regional Selectivity of a γ-Aminobutyric Acid-Induced [3H]Acetylcholine Release Sensitive to Inhibitors of γ-Aminobutyric Acid Uptake

The effects of gamma-aminobutyric acid (GABA) on the release of [3H]acetylcholine ([3H]ACh) were studied in synaptosomes prepared from rat hippocampus, cerebral cortex, hypothalamus, and striatum and prelabelled with [3H]choline. When synaptosomes were exposed in superfusion to exogenous GABA (0.01-0.3 mM) the basal release of newly synthesized [3H]ACh was increased in a concentration-dependent way in hippocampus, cortex, and hypothalamus nerve endings. In contrast, the release of [3H]ACh was not significantly affected by GABA in striatal synaptosomes. The effect of GABA was not antagonized significantly by bicuculline or picrotoxin. Muscimol caused only a slight not significant increase of [3H]ACh release when tested at 0.3 mM whereas, at this concentration, (-)-baclofen was totally inactive. The GABA-induced release of [3H]ACh was counteracted by SKF 89976A, SKF 100561, and SKF 100330A, three strong and selective GABA uptake inhibitors. The data suggest that, in selective areas of the rat brain, GABA causes release of [3H]ACh following penetration into cholinergic nerve terminals through a GABA transport system.     

Depolarization-induced release of ATP from cholinergic synaptosomes is not blocked by botulinum toxin type A

We report here the effects of Botulinum Toxin type A on the release of ATP and Acetylcholine from Torpedo electric organ synaptosomes. Our results show that Botulinum Toxin type A inhibits specifically the K⁺-induced release of Acetylcholine from synaptosomes without affecting the release of ATP. Membrane potential and calcium uptake into cholinergic nerve terminals are not modified after Botulinum Toxin poisoning. It is suggested that either most of the ATP released during the depolarization of the cholinergic synaptosomes does not originate from cholinergic synaptic vesicles or that there are two populations of synaptic vesicles, Acetylcholine-enriched synaptic vesicles and ATP-enriched synaptic vesicles. However, the possibility that the ACh and ATP released could come from different intrasynaptosomal compartments cannot be excluded.     

Aging Decreases the Sensitivity of Rat Cortical Synaptosomes to Calcium Ionophore-Induced Acetylcholine Release

The capacity of calcium ions to trigger acetylcholine release was studied in cerebral cortical synaptosomes from adult (6-month-old) and senescent (24-month-old) rats, using a calcium ionophore, A23187, that bypasses voltage-sensitive calcium channels. The potency but not the efficacy of the A23187 was reduced with respect to releasing acetylcholine (ACh) in the aged animals. There was no age-related difference in the synthesis of ACh or potency of the ionophore with respect to increasing 45calcium uptake. These results suggest that aging reduces the sensitivity of cerebral cortical nerve terminals to calcium-triggered ACh-release.     

Changes in cortical acetyl-CoA metabolism after selective basal forebrain cholinergic degeneration by 192IgG-saporin

The aim of the present study was to reveal whether reduced cortical cholinergic input affects the acetyl-CoA metabolism in cholinoceptive cortical target regions which may play a causative role for the deficits in cerebral glucose metabolism observed in Alzheimer's disease. The effect of cortical cholinergic denervation produced by a single intracerebroventricular application of the cholinergic immunotoxin 192IgG-saporin, on activities of pyruvate dehydrogenase and adenosine triphosphate (ATP)-citrate lyase as well as on the level of synaptoplasmic and mitochondrial acetyl-CoA and acetylcholine release in cortical target regions was studied. Cholinergic lesion produced 83%, 72% and 32% decreases in the activities of choline acetyltransferase, acetylcholinesterase and ATP-citrate lyase in nerve terminals isolated from rat brain cortex, respectively, but no change in pyruvate dehydrogenase activity. Spontaneous and Ca2+-evoked acetylcholine release from synaptosomes was inhibited by 76% and 73%, respectively, following immunolesion. The lesion-induced 39% decrease of acetyl-CoA level in synaptosomal mitochondria was accompanied by 74% increase in synaptoplasmic fraction. Levels of acetyl-CoA and CoASH assayed in fraction of whole brain mitochondria from lesioned cortex were 61% and 48%, respectively, higher as compared to controls. The data suggest a preferential localization of ATP-citrate lyase in cholinergic nerve terminals, where it may contribute to the transport of acetyl-CoA from the mitochondrial to the cytoplasmic compartment. They provide evidence on differential distribution of acetyl-CoA in subcellular compartments of cholinergic and non-cholinergic nerve terminals. There are also indications that cholinergic activity affects acetyl-CoA level and its intracellular distribution in glial and other non-cholinergic cortical cells.