Resveratrol increases glutamate uptake and glutamine synthetase activity in C6 glioma cells

Resveratrol, a phytoalexin found mainly in grapes, is a promising natural product with anti-cancer and cardio-protective activities. Here, we investigated, in C6 glioma cells, the effect of resveratrol on some specific parameters of astrocyte activity (glutamate uptake, glutamine synthetase and secretion of S100B, a neurotrophic cytokine) commonly associated with the protective role of these cells. Cell proliferation was significantly decreased by 8% and 26%, following 24h of treatment with 100 and 250 microM resveratrol. Extracellular S100B increased after 48 h of resveratrol exposure. Short-term resveratrol exposure (from 1 to 100 microM) induced a linear increase in glutamate uptake (up to 50% at 100 microM resveratrol) and in glutamine synthetase activity. Changes in these glial activities can contribute to the protective role of astrocytes in brain injury conditions, reinforcing the putative use of this compound in the therapeutic arsenal against neurodegenerative diseases and ischemic disorders.     

Heat shock induction of a 65 kDa ATP—binding proteinase in rat C6 glioma cells

The 45,55,65 and 100kDa ATP-binding proteinases(ATP-BPases) of the heat-shocked (44℃ for 30 min,recovery for 12h) rat C6 glioma cells were purified by DEAE-ionexchange and ATP-affinity chromatography.Their molecular masses,isoelectric points (pI),pH-optima and other properties were analyzed by native proteinase gels.It was shown that the 65 kDa ATP-BPase is specifically induced by heat shock and not detectable in control cells.Its N-terminal 1-9amino acid sequence was determined by Edman degradation,but no homologies to other proteins in the protein data bases were found.30 and 31kDa proteinases can be cleaved from the 45,55 and 65 kDa proteinases to which they are linked.A possible relationship of the heat-induced 65 kDa ATP-BPase with the ATP-dependent proteinases (ATP-DPases) in prokaryotes and eukaryotes is discussed.     

Endogenous regulation of 2-deoxyglucose uptake in C6 glioma cells correlates with cytoskeleton-mediated changes of surface morphology

The cellular basis of the membrane-limited state of glucose utilization and the mechanism of the endogenous regulation of hexose uptake in dense monolayers of C6 glioma cells were investigated. In an earlier study, it was shown that at high rates of glucose transport and phosphorylation combined with the inhibition of glycolytic adenosine triphosphate (ATP) production by iodoacetate, an endogenous regulatory response occurred that resulted in rapid, periodic variations of the glucose uptake rates (Lange et al., 1982). Similar time-dependent periodic changes of uptake rates also occurred during incubation of C6 glioma cells with 2 mM 2-deoxyglucose (2-DG) without pretreatment of the cells with iodoacetate. These changes were accompanied by variations of the intracellular ATP content, by distinct alterations of the shape and arrangement of microvilli and lamellae (lamellipodia) on the cell surface, and by changes of the cytoskeletal F-actin content. Because the changes of 2-DG uptake rates occurred independent of the intracellular 2-DG concentration, the bulk of this 2-DG pool was assumed to be localized apart from the membranal transport sites. Downregulation of 2-DG uptake appeared to be triggered by a rapid decrease of a small pool of the cellular ATP involved in the phosphorylation of transported hexose. Scanning and transmission electron microscopic observations of cells fixed in different states of the endogenous uptake regulation supported the assumption that the interior of lamellae and microvilli may represent a small entrance compartment for transported hexoses in which occurred the observed close coupling between hexose transport and phosphorylation as well as the rapid variations of ATP content. Hexose uptake is supposed to be regulated by cytoskeleton-mediated changes of volume and diffusional accessibility of this compartment, modulating the degree of its metabolic coupling with the cytoplasmic main compartment.     

Purification and characterization of rat brain prostaglandin D synthetase

Prostaglandin D synthetase was purified 2,600-fold from rat brain to apparent homogeneity, as judged by polyacrylamide gel electrophoresis and ultracentrifugation. The purified enzyme was a monomeric protein with a molecular weight of 27,000 +/- 1,000. The pI value, sedimentation coefficient, and partial specific volume were 4.6, 4.1 s, and 0.73 ml/g, respectively. The enzyme was stable between pH 4 and 11 at the temperature lower than 25 degrees C and resistant to a heat treatment under alkaline conditions (pH 8-11). About 50% of the activity was detected after a heat treatment at 100 degrees C for 5 min at pH 10. However, the enzyme was readily inactivated by the isomerase reaction of prostaglandin H2 to prostaglandin D2. The enzyme required sulfhydryl compounds such as dithiothreitol, glutathione, beta-mercaptoethanol, cysteine, and cysteamine for the reaction, but stoichiometric oxidation of these sulfhydryl compounds was not observed. The optimum pH, Km value for prostaglandin H2, and the turnover number were 9.5, 14 microM, and 170 min-1, respectively. The antibody was raised against the purified enzyme in a rabbit, which showed only one positive band in immunoblotting after gel electrophoresis of crude extracts of the brain at the same position as that of the purified enzyme. More than 90% of the prostaglandin D synthetase activity in the brain was absorbed by an excess amount of the antibody, indicating that our preparation is a major component of the enzyme responsible for the biosynthesis of prostaglandin D2 in the brain.     

Serum catecholamines desensitize beta-adrenergic receptors of cultured C6 glioma cells

C6 glioma cells grown in medium containing fetal bovine serum have a decreased beta-adrenergic receptor number and beta-receptor-stimulated cyclic AMP accumulation as compared to cells grown in a serum-free, defined medium. The decreased number of receptors and decreased cAMP accumulation are attributable to a suppression of receptor binding and response by serum as opposed to increases produced by growth in the defined medium. Serum, when added to cells grown in the absence of serum, stimulated cellular cyclic AMP levels to 2-3 times basal levels. This direct stimulatory effect was blocked by incubation of the cells with the beta-adrenergic antagonist propranolol and was partially reversed by dialysis of the serum. In contrast, addition of serum to cells that have been grown with serum fails to stimulate cyclic AMP accumulation. The decrease in receptors following growth in serum can be mimicked by growing cells in serum-free medium in the presence of beta-adrenergic agonists such as isoproterenol or norepinephrine. Radioenzymatic assays indicate that fetal bovine serum contains approximately 0.3 nM norepinephrine and lower concentrations of epinephrine. It thus appears that growth of C6 cells in serum-containing media desensitizes the beta-adrenergic receptor/cyclic AMP system of these cells. This desensitized state appears to result primarily from the action of catecholamines present in serum. These data indicate that retained catecholamines are one component in serum that can modify expression of beta-adrenergic receptors and hormonal response of cultured glioma cells.     

Regulation of glutamine synthetase and glutaminase activities in cultured skeletal muscle cells

Glutamine is synthesized in skeletal muscle, released to the circulation, and transported to other tissues, where it may provide important substrate for gluconeogenesis, ammoniagenesis, and energy-yielding pathways. With the ultimate goal of delineating the factors that control glutamine production and release by skeletal muscle, we have studied the regulation of two key enzymes, glutamine synthetase and glutaminase, in the L6 line of rat skeletal muscle cells grown in monolayer culture. The cultured myotubes were found to have glutamine synthetase and phosphate-dependent glutaminase activities. Glutamine synthetase activity was increased following incubation (1) in glutamine-free medium (threefold); (2) in medium containing high glutamic acid concentrations (fourfold); and (3) in medium supplemented with dexamethasone (threefold). In each case the increase in glutamine synthetase activity required several hours to reach a maximum and was prevented by cycloheximide, suggesting that the change occurred through increased enzyme biosynthesis. No substances tested were found to affect glutaminase activity. We conclude that glutamine synthetase in cultured skeletal muscle is responsive to substrate, product, and hormonal regulation.     

Cobalt chloride-induced apoptosis and extracellular signal-regulated protein kinase 1/2 activation in rat C6 glioma cells

Brain ischemia brings about hypoxic insults. Hypoxia is one of the major pathological factors inducing neuronal injury and central nervous system infection. We studied the involvement of mitogen-activated protein (MAP) kinase in hypoxia-induced apoptosis using cobalt chloride in C6 glioma cells. In vitro cytotoxicity of cobalt chloride was tested by MTT assay. Its IC(50) value was 400 microM. The DNA fragment became evident after incubation of the cells with 300 microM cobalt chloride for 24 h. We also evidenced nuclear cleavage with morphological changes of the cells undergoing apoptosis with electron microscopy. Next, we examined the signal pathway of cobalt chloride-induced apoptosis in C6 cells. The activation of extracellular signal-regulated protein kinase 1/2 (ERK 1/2) started to increase at 1 h and was activated further at 6 h after treatment of 400 M cobalt chloride. In addition, pretreatment of PD98059 inhibited cobalt chloride-induced apoptotic cell morphology in Electron Microscopy. These results suggest that cobalt chloride is able to induce the apoptotic activity in C6 glioma cells, and its apoptotic mechanism may be associated with signal transduction via MAP kinase (ERK 1/2).     

Intracellular zinc distribution and transport in C6 rat glioma cells

In mammalian cells, the intracellular availability of zinc influences numerous crucial processes. Its distribution has previously been visualized with several fluorescent probes, but it was unclear how these probes are compartmentalized within the cell. Here, we show that in C6 cells the zinc-specific probe Zinquin is evenly distributed. Thus, the significantly lower level of fluorescence in the nucleus and a punctuate vesicular staining are real differences in the concentrations of zinc. Chemical perturbation of the steady state by releasing intracellular protein-bound zinc with the sulfhydryl-reactive N-ethylmaleimide (NEM) resulted in a vanadate sensitive transport of zinc out of the nucleus and into zincosomes. If the zinc-release was performed with the histidine-reactive diethylpyrocarbonate, sequestration was reduced compared to treatment with NEM, indicating the importance of histidine within membrane zinc transporters. Another major factor regulating the zinc homeostasis is ion export. As determined by atomic absorption spectroscopy, up to 50% of the cellular zinc was exported by a mechanism sensitive to lanthanum ions. We conclude that different concentrations of labile zinc exist in different cellular compartments, which are maintained by export and intracellular transport of zinc.     

Intracellular distribution of active and inactive transglutaminase in stimulated cultured C6 glioma cells

The cellular distribution of active and inactive transglutaminase (TGase) was studied in C6 glioma cells before and during stimulation by a serum-containing medium. The activity of the enzyme was determined in the soluble and insoluble fractions obtained by freezing and thawing the cells, followed by centrifugation at 12,000g for 5 min. In the soluble fractions, the activity of TGase decreased 2.5 h post-stimulation and increased after 5 and 8 h. In the corresponding insoluble fractions, no significant changes in the activity of the enzyme were noted up to 8 h after stimulating the cells with fresh medium. An immunological approach was next used to determine the quantity of TGase antigen during the stimulation of the cultured glioma cells. In the soluble fraction, the quantity of the antigen decreases significantly at 2.5, 5, and 8 h. In contrast, in the insoluble fraction, a significant increase in TGase antigen was detected 8 h after the addition of fresh medium. Cycloheximide completely inhibited the increase in the quantity of TGase antigen in the insoluble fraction, 8 h post-stimulation, while actinomycin D caused a partial inhibition. Trypsin, neuraminidase, or Sendai viruses increased the activity of TGase significantly, when added to nonstimulated cells. Trypsin had no effect on TGase activity when added to the cells 2 h after stimulation with a serum-containing medium. These findings suggest that an inactive form of the enzyme is present in the insoluble cellular fraction. A model has been proposed to explain the variations in TGase activity, its distribution and translocation during cellular stimulation.     

Biochemical and immunological characterization of rat spleen prostaglandin D synthetase

Rat spleen prostaglandin D synthetase (Christ-Hazelhof, E., and Nugteren, D. H. (1979) Biochim. Biophys. Acta 572, 43-51) is very similar to rat brain prostaglandin D synthetase (Urade, Y., Fujimoto, N., and Hayaishi O. (1985) J. Biol. Chem. 260, 12410-12415) as judged by their pI (4.7-5.2), Mr (26,000-27,000), and self-inactivation during the isomerase reaction from prostaglandin H2 to prostaglandin D2. However, the amino acid compositions of these two enzymes were quite different. Furthermore, the spleen enzyme was associated with the glutathione S-transferase activity, differing from the brain enzyme. The synthetase and transferase activities of the spleen enzyme showed almost identical pH and glutathione dependencies, the optimum pH = 8.0 and Km for glutathione = 300 microM. The Km values for prostaglandin H2 and 1-chloro-2,4-dinitrobenzene (a substrate for the transferase) were about 200 microM and 5 mM, respectively. The synthetase activity was dose-dependently inhibited by 1-chloro-2,4-dinitrobenzene (IC50: approximately 5 mM) and more strongly by nonsubstrate ligands, such as bilirubin and indocyanine green (IC50: 150 and 2 microM, respectively). Both the synthetase and transferase activities of the purified enzyme dose-dependently decreased and showed identical immunotitration curves by incubation with antibody against this enzyme, but remained unchanged when treated with antibody against the brain enzyme. The antibody specific for the spleen enzyme absorbed almost all of the synthetase activity and about 10% of the transferase activity in the spleen, but not the transferase activity in the liver, heart, and testis. These results show that the two types of prostaglandin D synthetase are similar but different enzymes and that the spleen enzyme is a unique glutathione S-transferase differing from other isozymes and their subunits reported previously.     

Selective induction of glial glutamate transporter GLT-1 by hypertonic stress in C6 glioma cells.

Glial glutamate transporter GLT-1 mRNA was selectively induced in C6 glioma cells exposed to hypertonic stress (HS), while the expression of two other subtypes, GLAST and EAAC1, was suppressed. HS increased phosphorylation of the MAPK family, ERK, p38 MAPK, and JNK. Treatment with a PKC inhibitor showed that phosphorylation of both p38 MAPK and JNK is PKC-dependent but ERK phosphorylation is independent. Inhibition of either ERK or p38 MAPK did not abolish GLT-1 mRNA induction. Inhibition of PKC also had no effect. These findings indicate that the induction of GLT-1 mRNA by HS is independent of the MAPK pathways. This is the first report that the expression of glial glutamate transporters is osmotically regulated.     

Exacerbated vulnerability to oxidative stress in astrocytic C6 glioma cells with stable overexpression of the glutamine transporter slc38a1

We have previously demonstrated the functional expression of glutamine (Gln) transporter (GlnT) believed to predominate in neurons for the neurotransmitter glutamate pool by rat neocortical astrocytes devoid of neuronal marker expression, with exacerbated vulnerability to oxidative stress after transient overexpression. To evaluate molecular mechanisms underlying the exacerbation, we established stable GlnT transfectants in rat astrocytic C6 glioma cells. In two different clones of stable transfectants with increased intracellular Gln levels, exposure to hydrogen peroxide (H(2)O(2)) and A23187, but not to tunicamycin or 2,4-dinitrophenol, led to significant exacerbation of the cytotoxicity compared to cells with empty vector (EV). Stable GlnT overexpression led to a significant increase in heme oxygenase-1 protein levels in a manner sensitive to H(2)O(2), whereas H(2)O(2) was significantly more effective in increasing NO(2) accumulation and reactive oxygen species (ROS) generation in stable GlnT transfectants than in EV cells. Moreover, exposure to A23187 led to a more effective increase in the generation of ROS in stable GlnT transfectants than in stable EV transfectants. These results suggest that GlnT may play a role in the mechanisms underlying the determination of cellular viability in astrocytes through modulation of intracellular ROS generation.     

Effects of medium glutamine,glutamate, and ammonia on glutamine synthetase activity in cultured mouse astroglial cells

Mouse astroglial cells were grown during the last week of culture in either glutamine-free or glutamine-containing medium. The addition of cortisol to the glutamine-containing medium resulted in a doubling of astroglial glutamine synthetase (GS) activity. Withdrawal of glutamine from the medium resulted in a 50% elevation of GS and addition of cortisol to such a medium resulted in a further increase in GS which was not additive to glutamine withdrawal. Both in glutamine-free and glutamine-containing medium, the addition of glutamate resulted in a depression of both basal and cortisol induced GS activity. The simultaneous addition of ammonia plus glutamate to the culture medium ameliorated the glutamate mediated depressive effects on cortisol induced but not basal GS activity. Glutamine withdrawal from the culture medium resulted in an astroglial protein deficit. The addition of ammonia to the medium considerably reduced this deficit and the addition of glutamate completely eliminated this protein deficit.     

An association between glutamine synthetase activity and adipocyte differentiation in cultured 3T3-L1 cells

Confluent 3T3-L1 Swiss mouse fibroblasts acquired morphological and biochemical characteristics of adipocytes when maintained in medium containing 10% calf serum and added insulin. Identical cultures maintained in the absence of added insulin did not differentiate into adipocytes. Incubation of confluent cultures for 48 h with 0.25 μm dexamethasone and 0.5 mm 1-methyl-3-isobutylxanthine yielded subsequent adipocyte differentiation when the culture medium contained 10% fetal calf serum. In contrast, differentiation did not occur when similarly treated cultures were maintained in medium containing 10% calf serum. The increase in glutamine synthetase which occurred during adipocyte differentiation was closely associated with an increased rate of triglyceride synthesis from acetate, with increased protein, and with increases in the activities of glycerol-3-P dehydrogenase and glucose-6-P dehydrogenase. Glutamine synthetase activity remained undetectable in insulin-treated confluent 3T3-C2 cells maintained under conditions which yielded high glutamine synthetase activity in 3T3-L1 cells. (3T3-C2 cells did not differentiate into adipocytes.) Glutamine accumulated in the culture medium of 3T3-L1 adipocytes, but it did not accumulate in the medium from identically treated 3T3-C2 cells. A half-maximal increase in glutamine synthetase specific activity occurred at a culture medium insulin concentration of 10 ng/ml. Neither adipocyte differentiation nor the rise in glutamine synthetase activity were substantially altered by maintaining confluent cultures in medium lacking added glutamine. Incubation of confluent 3T3-L1 cultures with 3 mml-methionine sulfone, a reversible inhibitor of glutamine synthetase, increased by two-fold both the activity and the cellular content of glutamine synthetase. Incubation of confluent 3T3-L1 cultures with 4 mml-glutamine and l-methionine-dl-sulfoximine, an irreversible inhibitor of glutamine synthetase activity, decreased glutamine synthetase activity to less than 5% of the activity in control cultures; however, neither cellular content of the enzyme nor synthesis rate of the enzyme were substantially altered. In the presence of added glutamine, neither methionine sulfone nor methionine sulfoximine had a significant effect on phenotypic adipocyte conversion. By contrast, when confluent cultures were incubated with methionine sulfoximine and no added glutamine, glutamine synthetase remained absent and there was no evidence of adipocyte conversion. Our data indicate (1) that added insulin is required for adipocyte differentiation of 3T3-L1 cells maintained in medium containing calf serum, (2) that glutamine synthetase activity increases during adipocyte conversion regardless of the culture conditions employed to achieve differentiation, and (3) that glutamine synthetase activity may be required for adipocyte differentiation when cultures are maintained in medium lacking added glutamine.     

Exfoliation of the beta-adrenergic receptor and the regulatory components of adenylate cyclase by cultured rat glioma C6 cells

Cultured rat glioma C6 cells exfoliate membrane vesicles which have been termed 'exosomes' into the culture medium. The exosomes contained both stimulatory and inhibitory GTP-binding components of adenylate cyclase (the stimulatory, Gs, and the inhibitory, Gi, regulatory components) and beta-adrenergic receptors but were devoid of adenylate cyclase activity. It was therefore apparent that the catalytic component of adenylate cyclase was either not exfoliated or was inactivated during the exfoliation process. The presence of Gs or Gi in the exosomes was detected by ADP ribosylation using [alpha-32P]NAD in the presence of cholera or pertussis toxins, respectively. The exosomal concentration of each of the two components was estimated to be about one fifth of that of the cell membrane when expressed on a per mg protein basis. Exosomal Gs was almost as active as the membrane-derived Gs in its ability to reconstitute NaF- and guanine nucleotide-stimulated adenylate cyclase activity in membranes of S49 cyc- cells, which lack a functional Gs. The ability of exosomal Gs to reconstitute isoproterenol-stimulated activity, however, was much lower than that of membrane Gs. The density of beta-adrenergic receptors in the exosomes was much less than that found in the membranes. Although the exosomal receptors bound the antagonist iodocyanopindolol with the same affinity as receptors from the cell membrane, the affinity for the agonist isoproterenol was 13- to 18-fold lower in the exosomes. In addition, this affinity was not modulated by GTP in the exosomes. Thus, exfoliated beta-adrenergic receptors seem to be impaired in their ability to couple to and activate Gs. This was directly tested by coupling the receptors to a foreign adenylate cyclase using membrane fusion. The fusates were then assayed for agonist-stimulated activity. While significant stimulation of the acceptor adenylate cyclase was obtained using C6 membrane receptors, the exosomal receptors were completely inactive. Thus during exfoliation, there appear to be changes in the components of the beta-adrenergic-sensitive adenylate cyclase that results in a nonfunctional system in the exosomes.