Glycolysis and Glutamate Accumulation into Synaptic Vesicles
2003; Elsevier BV; Volume: 278; Issue: 8 Linguagem: Inglês
10.1074/jbc.m211617200
ISSN1083-351X
AutoresAtsushi Ikemoto, D G Bole, Tetsufumi Ueda,
Tópico(s)Lipid Membrane Structure and Behavior
ResumoGlucose is the major source of brain energy and is essential for maintaining normal brain and neuronal function. Hypoglycemia causes impaired synaptic transmission. This occurs even before significant reduction in global cellular ATP concentration, and relationships among glycolysis, ATP supply, and synaptic transmission are not well understood. We demonstrate that the glycolytic enzymes glyceraldehyde phosphate dehydrogenase (GAPDH) and 3-phosphoglycerate kinase (3-PGK) are enriched in synaptic vesicles, forming a functional complex, and that synaptic vesicles are capable of accumulating the excitatory neurotransmitter glutamate by harnessing ATP produced by vesicle-bound GAPDH/3-PGK at the expense of their substrates. The GAPDH inhibitor iodoacetate suppressed GAPDH/3-PGK-dependent, but not exogenous ATP-dependent, [3H]glutamate uptake into isolated synaptic vesicles. It also decreased vesicular [3H]glutamate content in the nerve ending preparation synaptosome; this decrease was reflected in reduction of depolarization-induced [3H]glutamate release. In contrast, oligomycin, a mitochondrial ATP synthase inhibitor, had minimal effect on any of these parameters. ADP at concentrations above 0.1 mminhibited vesicular glutamate and dissipated membrane potential. This suggests that the coupled GAPDH/3-PGK system, which converts ADP to ATP, ensures maximal glutamate accumulation into presynaptic vesicles. Together, these observations provide insight into the essential nature of glycolysis in sustaining normal synaptic transmission. Glucose is the major source of brain energy and is essential for maintaining normal brain and neuronal function. Hypoglycemia causes impaired synaptic transmission. This occurs even before significant reduction in global cellular ATP concentration, and relationships among glycolysis, ATP supply, and synaptic transmission are not well understood. We demonstrate that the glycolytic enzymes glyceraldehyde phosphate dehydrogenase (GAPDH) and 3-phosphoglycerate kinase (3-PGK) are enriched in synaptic vesicles, forming a functional complex, and that synaptic vesicles are capable of accumulating the excitatory neurotransmitter glutamate by harnessing ATP produced by vesicle-bound GAPDH/3-PGK at the expense of their substrates. The GAPDH inhibitor iodoacetate suppressed GAPDH/3-PGK-dependent, but not exogenous ATP-dependent, [3H]glutamate uptake into isolated synaptic vesicles. It also decreased vesicular [3H]glutamate content in the nerve ending preparation synaptosome; this decrease was reflected in reduction of depolarization-induced [3H]glutamate release. In contrast, oligomycin, a mitochondrial ATP synthase inhibitor, had minimal effect on any of these parameters. ADP at concentrations above 0.1 mminhibited vesicular glutamate and dissipated membrane potential. This suggests that the coupled GAPDH/3-PGK system, which converts ADP to ATP, ensures maximal glutamate accumulation into presynaptic vesicles. Together, these observations provide insight into the essential nature of glycolysis in sustaining normal synaptic transmission. Glycolysis plays a vital role in maintaining normal brain function. Glucose is known to serve as the major substrate for cerebral energy under normal conditions (1Siesjo B.K. Brain Energy Metabolism. John Wiley & Sons, Inc., New York1978: 101-130Google Scholar). Recent evidence suggests a direct correlation between glucose utilization and cognitive function (2McNay E.C. Fries T.M. Gold P.E. Proc. Natl. Acad. Sci. U. S. A. 2000; 97: 2881-2885Google Scholar). Reduction of glucose levels results in pathophysiological states and abnormal electrophysiological activity; however, this occurs long before significant alteration in tissue ATP levels is detected (3Lewis L.D. Ljunggren B. Ratcheson R.A. Siesjo B.K. J. Neurochem. 1974; 23: 673-679Google Scholar, 4Dirks B. Hanke H. Krieglstein J. Stock R. Wickop G. J. Neurochem. 1980; 35: 311-317Google Scholar, 5Ghajar J.B.G. Plum F. Duffy T.E. J. Neurochem. 1982; 38: 397-409Google Scholar, 6Bachelard H.S. Cox D.W.G. Drower J. J. Physiol. 1984; 352: 91-102Google Scholar, 7Fleck M.W. Henze D.A. Barrionuevo G. Palmer A.M. J. Neurosci. 1993; 13: 3944-3955Google Scholar). Substitution of pyruvate for glucose does not support normal evoked neuronal activity, although tissue ATP level returns to normal (8Cox D.W.G. Bachelard H.S. Brain Res. 1982; 239: 527-534Google Scholar, 9Cox D.W.G. Morris P.G. Feeney J. Bachelard H.S. Biochem. J. 1983; 212: 365-370Google Scholar, 10Kanatani T. Mizuno K. Okada Y. Experientia. 1995; 51: 213-216Google Scholar). Abnormal synaptic transmission caused by hypoglycemia occurs in part if not entirely by a presynaptic mechanism (7Fleck M.W. Henze D.A. Barrionuevo G. Palmer A.M. J. Neurosci. 1993; 13: 3944-3955Google Scholar, 11Shoji S. Synapse. 1992; 12: 322-332Google Scholar, 12Spuler A. Endres W. Grafe P. Exp. Neurol. 1988; 100: 248-252Google Scholar). Fleck et al. (7Fleck M.W. Henze D.A. Barrionuevo G. Palmer A.M. J. Neurosci. 1993; 13: 3944-3955Google Scholar) have shown that substantial reduction of extracellular glucose results in a decrease in stimulus-evoked Glu release, with no changes in ATP levels. These studies together suggest that glycolysis or glycolytic intermediate(s) are necessary for normal synaptic transmission independent of global cellular ATP levels. In an attempt to reveal the underlying mechanism of hypoglycemia-induced aberrant synaptic transmission, we previously explored the possibility that glycolytic intermediates could modify proteins localized in the nerve ending (13Ueda T. Plagens D.G. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 1229-1233Google Scholar, 14Morino H. Fischer-Bovenkerk C. Kish P.E. Ueda T. J. Neurochem. 1991; 56: 1049-1057Google Scholar). 3-Phosphoglycerate (3-PG) 1The abbreviations used are: 3-PG, 3-phosphoglycerate; t-ACPD, trans-1-aminocyclopentane-1,3-dicarboxylic acid; 4-AP, 4-aminopyridine; 1, 3-BPG, 1,3-bisphosphoglycerate; DHAP, dihydroxyacetone phosphate; FCCP, carbonyl cyanidep-(trifluoromethoxy)-phenylhydrazone; GAP, glyceraldehyde-3-phosphate; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; 3-PGK, 3-phosphoglycerate kinase; PGM, monophosphoglycerate mutase; ACSF, artificial cerebrospinal fluid 1The abbreviations used are: 3-PG, 3-phosphoglycerate; t-ACPD, trans-1-aminocyclopentane-1,3-dicarboxylic acid; 4-AP, 4-aminopyridine; 1, 3-BPG, 1,3-bisphosphoglycerate; DHAP, dihydroxyacetone phosphate; FCCP, carbonyl cyanidep-(trifluoromethoxy)-phenylhydrazone; GAP, glyceraldehyde-3-phosphate; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; 3-PGK, 3-phosphoglycerate kinase; PGM, monophosphoglycerate mutase; ACSF, artificial cerebrospinal fluid was demonstrated to stimulate phosphorylation of 155- and 72-kDa proteins. The latter was identified as glucose-1,6-bisphosphate synthetase, and 1,3-bisphosphoglycerate (1,3-BPG) was found to serve as the direct substrate for phosphorylation of this enzyme, by donating 1-phosphate. Both of these phosphorylated proteins are enriched in the synaptosomal (nerve ending preparation) as well as cell body soluble fractions, but the significance of these modifications in synaptic transmission remains unclear. In this paper, we show that the glycolytic intermediate 1,3-BPG forms an acyl-enzyme intermediate with vesicle-bound glyceraldehyde phosphate dehydrogenase (GAPDH), that vesicle-bound GAPDH exists in a complex with 3-phosphoglycerate kinase (3-PGK), and that activation of vesicle-associated GAPDH and 3-PGK is sufficient to support vesicular uptake of Glu. Glutamate is now recognized as the major excitatory neurotransmitter responsible for triggering neuronal firing, in the vertebrate central nervous system (15Watkins J.C. Evans R.H. Annu. Rev. Pharmacol. Toxicol. 1981; 21: 165-204Google Scholar, 16Cotman C.W. Foster A. Lanthorn T. DiChiara G. Gessa G.L. Glutamate as a Neurotransmitter. Raven Press, New York1981: 1-27Google Scholar, 17Fonnum F. J. Neurochem. 1984; 42: 1-11Google Scholar, 18Ueda T. Roberts P.J. Storm-Mathisen J. Bradford H.F. Excitatory Amino Acids. Macmillan, London, UK1986: 173-195Google Scholar, 19Collingridge G.L. Bliss T.V.P. Trends Neurosci. 1987; 10: 288-293Google Scholar, 20Cotman C.W. Monaghan D.T. Ganong A.H. Annu. Rev. Neurosci. 1988; 11: 61-80Google Scholar, 21Nicholls D.G. J. Neurochem. 1989; 52: 331-341Google Scholar, 22Maycox P.R. Hell J.W. Jahn R. Trends Neurosci. 1990; 13: 83-87Google Scholar, 23Özkan E.D. Ueda T. Jpn. J. Pharmacol. 1998; 77: 1-10Google Scholar). As such, proper Glu synaptic transmission is not only essential for basic neuronal communication but also is involved in learning and memory formation (19Collingridge G.L. Bliss T.V.P. Trends Neurosci. 1987; 10: 288-293Google Scholar, 20Cotman C.W. Monaghan D.T. Ganong A.H. Annu. Rev. Neurosci. 1988; 11: 61-80Google Scholar). Glutamate accumulation into synaptic vesicles in the nerve terminal is an initial crucial step in Glu transmission (18Ueda T. Roberts P.J. Storm-Mathisen J. Bradford H.F. Excitatory Amino Acids. Macmillan, London, UK1986: 173-195Google Scholar, 22Maycox P.R. Hell J.W. Jahn R. Trends Neurosci. 1990; 13: 83-87Google Scholar, 23Özkan E.D. Ueda T. Jpn. J. Pharmacol. 1998; 77: 1-10Google Scholar, 24Reimer R.J. Fremeau Jr., R.T. Bellocchio E.E. Edwards R.H. Curr. Opin. Cell Biol. 2001; 13: 417-421Google Scholar, 25Takamori S. Rhee J.S. Rosenmund C. Jahn R. Nature. 2000; 407: 189-194Google Scholar, 26Otis T.S. Neuron. 2001; 29: 11-14Google Scholar). This process requires ATP to generate an electrochemical gradient, which is the driving force for Glu uptake into synaptic vesicles (27Naito S. Ueda T. J. Biol. Chem. 1983; 258: 696-699Google Scholar, 28Naito S. Ueda T. J. Neurochem. 1985; 44: 99-109Google Scholar, 29Maycox P.R. Deckwerth T. Hell J.W. Jahn R. J. Biol. Chem. 1988; 263: 15423-15428Google Scholar, 30Hell J.W. Maycox P.R. Jahn R. J. Biol. Chem. 1990; 265: 2111-2117Google Scholar, 31Tabb J.S. Ueda T. J. Neurosci. 1991; 11: 1822-1828Google Scholar, 32Tabb J.S. Kish P.E. Van Dyke R. Ueda T. J. Biol. Chem. 1992; 267: 15412-15418Google Scholar, 33Wolosker H. Reis M. Assreuy J. de Meis L. J. Neurochem. 1996; 66: 1943-1948Google Scholar, 34Bellocchio E.E. Reimer R.J. Fremeau Jr., R.T. Edwards R.H. Science. 2000; 289: 957-960Google Scholar). We present evidence that glycolytically produced ATP, in particular that produced by GAPDH and 3-PGK, but not mitochondria-derived ATP, is harnessed for accumulation of Glu into synaptic vesicles in synaptosomes; Glu transported into synaptic vesicles in this manner is released upon depolarization. These findings could provide an explanation for hypoglycemia-induced aberrant synaptic transmission and insight into the essential nature of glycolysis in normal synaptic transmission. [γ-32P]ATP (6,000 Ci/mmol) was obtained from PerkinElmer Life Sciences.l-[G-3H]Glutamic acid (42.0 Ci/mmol) was purchased from Amersham Biosciences. The affinity-purified polyclonal antibodies (rabbit IgG) specific to recombinant monophosphoglycerate mutase (PGM) type B was kindly provided by Oriental Yeast Co., Ltd. (Tokyo, Japan). Anti-GAPDH monoclonal antibody (mouse IgG) and anti-3-PGK polyclonal antibody (rabbit IgG) were purchased from U.S. Biological (Swampscott, MA) and Accurate Chemical & Science Co. (Westbury, NY), respectively. Glycolytic enzymes and all other chemicals were purchased from Sigma-Aldrich unless mentioned elsewhere. Synaptic vesicles were prepared from bovine cerebrum through the discontinuous sucrose gradient procedure as described previously (35Özkan E.D. Lee F.S. Ueda T. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 4137-4142Google Scholar). The subcellular fractions of bovine cerebrum were prepared as described previously (36Ueda T. Greengard P. Berzins K. Cohen R.S. Blomberg F. Grab D.J. Siekevitz P. J. Cell Biol. 1979; 83: 308-319Google Scholar). Synaptosomes were prepared from cerebra of male Sprague-Dawley rats (150–200 g) and purified through the Percoll gradient centrifugation step, as described by Dunkley et al. (37Dunkley P.R. Heath J.W. Harrison S.M. Jarvie P.E. Glenfield P.J. Rostas A.P. Brain Res. 1988; 441: 59-71Google Scholar). Protein concentration was determined by the method of Bradford (38Bradford M.M. Anal. Biochem. 1976; 72: 248-254Google Scholar) with a Coomassie protein assay reagent kit (Pierce) with bovine serum albumin as standard protein. [32P]Dihydroxyacetone phosphate (DHAP) was prepared by phosphorylation of dihydroxyacetone by glycerol kinase with [γ-32P]ATP in a mixture containing 5 mm Tris-HCl (pH 7.0), 40 μmMgSO4, 10 mm dihydroxyacetone, 20 μm [γ-32P]ATP (6,000 Ci/mmol), and 0.5 units of glycerol kinase (Bacillus stearothermophilus). The reaction was performed in a volume of 40 μl at 37 °C for 20 min. [3-32P]1,3-BPG was prepared from [32P]DHAP by conversion to [32P]GAP with triose-phosphate isomerase, followed by a GAPDH reaction in the presence of the lactate dehydrogenase-coupled NAD-regenerating system. The reaction mixture (400 μl) contained 12.5 mm triethanolamine (pH 8.0), 0.2 mm EDTA, 2 mm NAD, 2 mm sodium pyruvate, 2 mm KH2PO4, 50 μm DHAP, 3.2 units of GAPDH (rabbit muscle), 20 units of triose-phosphate isomerase (rabbit muscle), 10 units of lactate dehydrogenase (rabbit muscle), and 40 μl of the [32P]DHAP reaction mixture. The entire mixture was incubated at 25 °C for 5 min, filtered to remove the enzymes by an Amicon Centricon-10 concentrator at 4 °C, and put onto DEAE-cellulose (Whatman DE32, 1.2 × 2.2 cm) previously equilibrated with 10 mm glycylglycine, pH 7.4. Elution was carried out with stepwise increases in the NaCl concentration: 50, 75, 100, 125, 150, and 200 mm. DHAP, GAP, and inorganic phosphate (Pi) were eluted with 75 mm NaCl in the same buffer. 3-PG and 1,3-BPG were eluted with 125 and 150 mm NaCl, respectively, in the same buffer. All the compounds thus prepared were stored at −80 °C until use. Radioactive compounds were analyzed by high pressure liquid chromatography on a Whatman Partisil 10 SAX WCS column (4.6 × 250 mm), comparing their retention times with those of nonradioactive authentic standards monitored at 214 nm. The column was equilibrated with 0.4 m sodium phosphate buffer (pH 3.2), and glycolytic intermediates and nucleotides were eluted isocratically, as described previously (14Morino H. Fischer-Bovenkerk C. Kish P.E. Ueda T. J. Neurochem. 1991; 56: 1049-1057Google Scholar). Retention times for glycolytic intermediates were 4.1 min for DHAP and GAP, 5.3 min for Pi, 5.9 min for 2-phosphoglycerate and 3-PG, 6.0 min for ADP, 7.4 min for phosphoenolpyruvate, 15.4 min for 1,3-BPG, and 30.0 min for ATP. The synaptic vesicle fraction (30 μg of protein) was preincubated at 37 °C for 30 s in 27 μl of 5 mm Tris-maleate (pH 7.4). The reaction was initiated by the addition of 3 μl of [3-32P]1,3-BPG (240 Ci/mmol) to a final concentration of 140 nm and allowed to continue for 10 s. For electrophoretic protein separation under neutral pH conditions, the reaction was terminated by the addition of 10 μl of buffer containing 4% SDS, 30% sucrose, 40 mm Tris-HCl (pH 7.0), 4 mm EDTA, 160 mm 2-mercaptoethanol, and 50 μg/ml bromphenol blue; aliquots (25 μl) were subjected to polyacrylamide gel (1% SDS and 5.6% acrylamide) without stacking gel according to the method of Fairbanks et al. (39Fairbanks G. Steck S.L. Wallach D.F.H. Biochemistry. 1971; 10: 2606-2617Google Scholar). For electrophoretic protein separation under alkaline pH conditions (standard SDS-PAGE), the reaction was terminated by the addition of 10 μl of SDS sample buffer containing 4% SDS, 40% glycerol, 0.2m Tris-HCl (pH 6.8), 20% 2-mercaptoethanol, and 50 μg/ml bromphenol blue; aliquots (25 μl) were subjected to polyacrylamide gel (0.1% SDS and 12% acrylamide) according to the method of Laemmli (40Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar), except for omission of sample boiling. Autoradiography was carried out as described previously (41Ueda T. Maeno H. Greengard P. J. Biol. Chem. 1973; 248: 8295-8305Google Scholar) and analyzed using an image analyzer (Bio-Rad Gel Doc 2000). Antibodies (10 μg) were absorbed onto immobilized protein G (0.1 ml as 50% slurry) and chemically cross-linked, using the Seize X mammalian immunoprecipitation kit (Pierce). The synaptic vesicle fraction (50 μg of protein) was stirred in 0.2 ml of buffer containing 0.32 m sucrose, 4 mm Tris-maleate (pH 7.4), and 0.2 m NaCl for 5 min at 4 °C and centrifuged at 200,000 g maxfor 1 h at 4 °C. When 32P-labeled protein was detected, the synaptic vesicle fraction was incubated at 37 °C for 10 s with the same buffer containing 140 nm (240 Ci/mmol) [3-32P]1,3-BPG. The supernatant (180 μl) was subjected to immunoprecipitation with immobilized antibody, according to the manufacturer's protocol. Aliquots (20 μl) were subjected to SDS-PAGE, except for omission of sample boiling, according to Fairbankset al. (39Fairbanks G. Steck S.L. Wallach D.F.H. Biochemistry. 1971; 10: 2606-2617Google Scholar) or Laemmli (40Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar), followed by Western blot analysis or autoradiography, as appropriate. For analysis of subcellular fractions, 30 μg of protein were subjected to standard SDS-PAGE (40Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar). In NaCl solubilization experiments, the synaptic vesicle fraction (50 μg of protein) was stirred in 0.2 ml of buffer containing 0.32m sucrose, 4 mm Tris-maleate (pH 7.4), and various concentrations of NaCl for 5 min at 4 °C and then centrifuged at 200,000 × g maxfor 1 h. The pellet was dissolved in 30 μl of SDS sample buffer. The protein in the supernatant (180 μl) was precipitated with 15% trichloroacetic acid and dissolved in 30 μl of SDS sample buffer. Aliquots (25 μl) of the 200,000 ×g max pellet and supernatant fractions were subjected to standard SDS-PAGE. For analysis of GAPDH and 3-PGK binding to synaptic vesicles, synaptic vesicles (1 mg of protein) were washed twice by 20 ml of buffer containing 0.32 m sucrose, 4 mm Tris-maleate (pH 7.4), and 0.8 m NaCl; bound NaCl was then removed by washing twice with 20 ml of the same buffer without NaCl. The washed synaptic vesicles (40 μg of protein) were incubated at 37 °C for 10 min in 0.1 ml of buffer in the absence or presence of 2 μg of purified GAPDH or 3-PGK. The synaptic vesicles were pelleted by centrifugation at 200,000 ×g max for 1 h at 4 °C and washed twice with the buffer. The pellet was suspended in 80 μl of SDS sample buffer, and an aliquot (25 μl) was subjected to standard SDS-PAGE. Proteins were electrotransferred onto the Immobilon-P polyvinylidene difluoride membrane (Millipore Corp.) using a semidry transfer apparatus (Bio-Rad Trans-Blot SD). The membrane was treated with 5% nonfat dry milk in a solution containing 50 mm Tris-HCl (pH 7.4), 0.5 m NaCl, and 0.1% Tween 20 (TBS-T) for 1 h and then incubated for 2 h at room temperature with anti-GAPDH monoclonal antibody at a 1:50 dilution or anti-3-PGK polyclonal antibodies at 1:500 dilution, followed by incubation with alkaline phosphatase-conjugated goat anti-mouse IgG or anti-rabbit IgG, respectively, at room temperature for 1.5 h. Unbound antibodies were washed out with TBS-T. 5-Bromo-4-chloro-3-indolyl phosphate and nitro blue tetrazolium (Bio-Rad) were used as substrates for color development and analyzed using an image analyzer (Bio-Rad Gel Doc 2000). Vesicular glutamate uptake was measured by the filtration-based assay using Whatman GF/C filters, as described previously (27Naito S. Ueda T. J. Biol. Chem. 1983; 258: 696-699Google Scholar, 28Naito S. Ueda T. J. Neurochem. 1985; 44: 99-109Google Scholar), with minor modifications. In the standard assay, aliquots (10 μg of protein) of bovine synaptic vesicles were incubated at 30 °C for 10 min with 100 μm [3H]Glu (a specific activity of 7.4 GBq/mmol was obtained by the addition of unlabeled Glu to [3H]Glu) in 0.1 ml of an incubation medium (pH 7.4) containing 20 mm Hepes-KOH, 0.25 m sucrose, 4 mm MgSO4, 4 mm KCl, and 2 mml-aspartic acid in the absence or presence of 2 mm ATP (pH adjusted to 7.4 by the addition of Tris-base). Prior to incubation, preincubation without ATP and [3H]Glu was carried out for 1 min at 30 °C. When inhibitor effect was tested, test agents were added at the start of the preincubation period. In some experiments, ATP was replaced with a mixture of 2 mm GAP, 2 mm Pi, 2 mm NAD, and 0.1 mm ADP (pH adjusted to 7.4 by the addition of Tris-base) as a source of a Glu uptake activator. [3H]Glu content in synaptic vesicles within the synaptosome was assayed as described previously (42Bole D.G. Hirata K. Ueda T. Neurosci. Lett. 2002; 322: 17-20Google Scholar, 43Ogita K. Hirata K. Bole D.G. Yoshida S. Tamura Y. Leckenby A.M. Ueda T. J. Neurochem. 2001; 77: 34-42Google Scholar) with minor modifications. Synaptosomes (100 μg of protein) were suspended in 0.1 ml of oxygenated (95% O2, 5% CO2) Krebs-Ringer buffer containing 150 mmNaCl, 2.4 mm KCl, 1.2 mmNa2HPO4, 1.2 mm CaCl2, 1.2 mm MgSO4, 5 mm Hepes-Tris (pH 7.4), and 10 mm glucose and preincubated at 37 °C for 10 min in the absence or presence of iodoacetate, oligomycin (Calbiochem), and pyruvate at indicated concentrations. After 3 μCi of [3H]Glu (42.0 Ci/mmol) were added to the medium, synaptosomes were incubated for an additional 10 min. Aliquots (10 μl) were removed and filtered on Whatman GF/C filters to determine the total amount of [3H]Glu taken up by the synaptosomes, and the rest were immediately frozen on dry ice. For vesicular [3H]Glu content determination, the frozen synaptosomes (90 μl) were thawed by adding 1.5 ml of ice-cold hypotonic solution containing 6 mm Tris-maleate (pH 8.1) and 2 mmaspartate and incubated for 20 min at 0 °C. Aliquots (1 ml) were filtered on Whatman GF/C filters, and radioactivity retained on filters was determined in a Beckman LS 6500 scintillation spectrophotometer. Glutamate release from synaptosomes was assayed using the superfusion technique described previously (43Ogita K. Hirata K. Bole D.G. Yoshida S. Tamura Y. Leckenby A.M. Ueda T. J. Neurochem. 2001; 77: 34-42Google Scholar), with minor modifications. Synaptosomes (200 μg of protein) were suspended in 1.5 ml of oxygenated (95% O2/5% CO2) artificial cerebrospinal fluid (ACSF) containing 124 mm NaCl, 5 mm KCl, 2 mm MgSO4, 1.25 mmNaH2PO4, 22 mm NaHCO3, and 10 mmd-glucose and preincubated with or without 300 μm iodoacetate or 2 μmoligomycin at 37 °C for 10 min. After 3.5 μCi of [3H]Glu (42.0 Ci/mmol) were added to the medium, synaptosomes were incubated for an additional 10 min. An aliquot (1.0 ml) of [3H]Glu-loaded synaptosomal suspension was layered onto a cellulose-acetate membrane filter (pore size 0.45 μm) placed in a superfusion chamber. The synaptosomes were superfused (0.5 ml/min) with ACSF for 60 min before application of 50 μm4-aminopyridine (4-AP) plus 2 mm CaCl2 to trigger depolarization of the synaptosomal membrane. In some control experiments, synaptosomes were superfused with ACSF containing 300 μm iodoacetate or 2 μm oligomycin for the first 20 min of the superfusion period. This period was the same as the sum of the synaptosomal preincubation and [3H]Glu-loading periods. These procedures were all carried out at 37 °C. Fractions were collected every 10 s for 3 min, from 30 s prior to 4-AP application. The amount of [3H]Glu released into each fraction was expressed as a percentage of the total [3H]Glu taken up into synaptosomes at the end of 10 min of [3H]Glu loading. The total [3H]Glu loaded into synaptosomes was calculated from the amount of [3H]Glu in 0.1 ml of loaded synaptosomes, determined by the same filtration method used for the vesicular Glu uptake assay. GAPDH activity was measured using 20 μg of protein of synaptic vesicles in a reaction mixture (1 ml) containing 0.1 m Tris-HCl (pH 8.5), 1.7 mmsodium arsenate, 20 mm sodium fluoride, 1 mmNAD, 1 mm GAP, and 5 mmKH2PO4 (44Berenski L.M. Kim C.J. Jung C.Y. J. Biol. Chem. 1990; 265: 15449-15454Google Scholar). Activity was calculated from the rate of increase in NADH formation by monitoring absorbance at 340 nm at 25 °C. ATP levels in the synaptic vesicle and synaptosome suspensions were determined under the same conditions described for vesicular Glu uptake and vesicular Glu content assays, respectively, by the luciferin/luciferase method, using an ATP bioluminescent assay kit (Sigma-Aldrich), according to the manufacturer's protocol. Generation of the membrane potential across the synaptic vesicle membrane was monitored by ATP-induced fluorescence quenching of the membrane potential-sensitive dye oxonol V (Molecular Probes, Inc., Eugene, OR) using a Fluorolog III fluorospectrophotometer (Horiba Jobin Yvon Co., Ltd., Tokyo, Japan) as described previously (32Tabb J.S. Kish P.E. Van Dyke R. Ueda T. J. Biol. Chem. 1992; 267: 15412-15418Google Scholar, 43Ogita K. Hirata K. Bole D.G. Yoshida S. Tamura Y. Leckenby A.M. Ueda T. J. Neurochem. 2001; 77: 34-42Google Scholar). Vesicular ATPase activity was assayed by determining free inorganic phosphate liberated upon incubation of synaptic vesicles (30 μg of protein) with ATP, according to the method of Lanzetta et al. (45Lanzetta P.A. Alvarez L.J. Reinach P.S. Candia O.A. Analyt. Biochem. 1979; 100: 95-97Google Scholar), with minor modifications as described previously (43Ogita K. Hirata K. Bole D.G. Yoshida S. Tamura Y. Leckenby A.M. Ueda T. J. Neurochem. 2001; 77: 34-42Google Scholar). Purified synaptic vesicles were allowed to react with [3-32P]1,3-BPG for 10 s and subjected to SDS-PAGE at either neutral (Fig. 1 A) or alkaline (Fig. 1 B) pH (39Fairbanks G. Steck S.L. Wallach D.F.H. Biochemistry. 1971; 10: 2606-2617Google Scholar, 40Laemmli U.K. Nature. 1970; 227: 680-685Google Scholar). Electrophoresis at neutral pH revealed incorporation of a radioactive moiety of [3-32P]1,3-BPG into vesicular proteins ofM r = 37,000 and 29,000. In contrast, electrophoresis at alkaline pH revealed only the 29-kDa protein, which appears to retain more of the 32P label than when electrophoresis was conducted at neutral pH. These results indicate that both the 29- and 37-kDa vesicular proteins can incorporate a32P-containing moiety; however, the stability of the labeled proteins differs depending on pH. Incorporation of the radioactive moiety into the 37-kDa protein was stimulated by Mg2+ but not affected by 3-PG. Labeling of the 29-kDa protein was inhibited by Mg2+ and completely blocked by 3-PG. The linkage of the 32P-containing moiety to the 37-kDa protein was labile in the presence of 0.2 mhydroxylamine (data not shown). These observations suggested that the 37- and 29-kDa proteins might be GAPDH and PGM, respectively. This was confirmed by immunoprecipitation with antibodies directed against these proteins (Fig. 1 C). Thus, labeling of the 37-kDa GAPDH with [3-32P]1,3-BPG probably occurs by formation of a thioester bond between a cysteine residue and the 3-phosphoglyceroyl moiety of 1,3-BPG; thioester bonds are known to be labile at alkaline pH. Labeling of the 29-kDa PGM probably represents phosphorylation of a histidine residue that is labile at either neutral or acidic pH. In order to determine the subcellular distribution of GAPDH and PGM, we examined the amount of [3-32P]1,3-BPG radioactivity incorporated into these two proteins present in synaptosomal cytosol, perikaryal cytosol, microsome, synaptic vesicle, and plasma membrane fractions (Fig. 1 D). For each subcellular fraction, an equivalent amount of protein was incubated with [3-32P]1,3-BPG. Labeled GAPDH was found in all subcellular fractions but was enriched in the synaptic vesicle fraction. In contrast, PGM was most enriched in the synaptosomal cytosol. Subcellular fractions were also subjected to Western blotting with anti-GAPDH antibody. GAPDH was found in the greatest concentration in the synaptic vesicle fraction (Fig. 1 E), in accord with the results obtained with the labeling method. Analysis of subcellular fractions indicates that GAPDH not only occurs in the cytosol but also can bind to various types of membranes. To determine the nature of the interaction of GAPDH with synaptic vesicles, synaptic vesicles were treated with various concentrations of NaCl (Fig. 1 F). GAPDH was dissociated from synaptic vesicles with increasing salt concentrations. At 0.8 m NaCl, only about 10% of GAPDH was found to remain associated with synaptic vesicle membranes. At physiologic ionic strength, GAPDH is associated with synaptic vesicle membranes, although it does also occur in the cytosol fractions. In the glycolytic pathway, GAPDH is known to exist in a complex with 3-PGK (46Srivastava D.K. Bernhard S.A. Curr. Top. Cell. Reg. 1986; 28: 1-68Google Scholar). The activities of GAPDH and 3-PGK are energetically coupled, utilizing GAP, NAD, Pi, and ADP, to yield 3-PG, ATP, NADH, and H+ (46Srivastava D.K. Bernhard S.A. Curr. Top. Cell. Reg. 1986; 28: 1-68Google Scholar, 47Weber J.P. Bernhard S.A. Biochemistry. 1982; 21: 4184-4194Google Scholar). Because GAPDH is enriched in synaptic vesicles, we considered the possibility that 3-PGK may also preferentially localize to synaptic vesicle membranes. Western blots were conducted on subcellular fractions using anti-3-PGK antibody (Fig. 2 A). Like GAPDH, 3-PGK was found enriched in the synaptic vesicle fraction. Immunoreactive 3-PGK was also dissociated from synaptic vesicles by increasing concentrations of NaCl in a manner similar to that observed for GAPDH (Fig. 2 B). To determine whether GAPDH and 3-PGK exist in a complex on synaptic vesicles, we performed experiments to determine whether GAPDH and 3-PGK could be co-immunoprecipitated. Synaptic vesicle salt extracts were incubated with anti-3-PGK antibody, and the resulting immunoprecipitates were subjected to Western blotting with anti-GAPDH antibody (Fig. 2 C). Immunoreactive GAPDH was detected in the anti-3-PGK antibody
Referência(s)