Oxidative stress induced by ascorbate causes neuronal damage in an in vitro system

Brain Research 895 (2001) 66–72 www.elsevier.com / locate / bres

Research report

Oxidative stress induced by ascorbate causes neuronal damage in an in vitro system Jin H. Song, Seon H. Shin*, Gregory M. Ross Department of Physiology, Botterell Hall, Queen’ s University, Kingston, Ontario, Canada, K7 L 3 N6 Accepted 19 December 2000

Abstract Of particular physiological interest, ascorbate, the ionized form of ascorbic acid, possesses strong reducing properties. However, it has been shown to induce oxidative stress and lead to apoptosis under certain experimental conditions. Ascorbate in the brain is released during hypoxia, including stroke, and is subsequently oxidized in plasma. The oxidized product (dehydroascorbate) is transported into neurons via a glucose transporter (GLUT) during a reperfusion period. The dehydroascorbate taken up by cells is reduced to ascorbate by both enzymatic and non-enzymatic processes, and the ascorbate is stored in cells. This reduction process causes an oxidative stress, due to coupling of redox reactions, which can induce cellular damage and trigger apoptosis. Ascorbate treatment decreased cellular glutathione (GSH) content, and increased the rates of lipid peroxide production in rat cortical slices. Wortmannin, a specific inhibitor of phosphatidylinositol (PI)-3-kinase (a key enzyme in GLUT translocation), prevented the ascorbate induced-decrease of GSH content, and suppressed ascorbate-induced lipid peroxide production. However, wortmannin was ineffective in reducing hydrogen peroxide (H 2 O 2 )induced oxidative stress. The oxidative stress caused ceramide accumulation, which was proportionally changed with lipid peroxides when the cortical slices were treated with ascorbate. These differential effects support the hypothesis that GLUT efficiently transports the dehydroascorbate into neurons, causing oxidative stress. Crown Copyright  2001 Published by Elsevier Science B.V. Theme: Disorders of the nervous system Topic: Neurotoxicity Keywords: Apoptosis; Ceramide; Dehydroascorbate; Oxidative stress; Stroke

1. Introduction Ascorbate (vitamin C) concentrations in the brain (1–2 mM) are much higher than plasma concentrations (24–85 mM) [21,31,35,41], suggesting that it is actively taken up by a transport mechanism and plays important physiological roles. However, only a limited number of functions are known, and including a cofactor in catecholamine biosynthesis [14], facilitation of neurotransmitter release [4,23,33] and modulation of ligand binding [24,45]. Ascorbate is a well-known reducing agent, and is easily oxidized to dehydroascorbate by molecular oxygen (O 2 ) in solution. Ascorbate is also known to act as pro-oxidant [16,17], but the mechanism of ascorbate-induced oxidative *Corresponding author. Tel.: 11-613-533-2802; fax: 11-613-5336880. E-mail address: [email protected] (S.H. Shin). 0006-8993 / 01 / $ – see front matter PII: S0006-8993( 01 )02029-7

action and apoptosis is not well established. Our studies have previously shown the oxidant action of ascorbate in PC12 cells [43] and in brain slices [44]. Clarification of the mechanism carries profound physiological and pathological significance, since ascorbate recycling may be involved in oxidative damage of neurons associated with stroke. The normal physiological levels of ascorbate in the brain cannot be maintained during hypoxia [28,29]; the released ascorbate will likely be oxidized in plasma, and the oxidized product (dehydroascorbate) is known to be transported into cells via a glucose transporter (GLUT) [34,51,53]. Reduction processes of cytosolic dehydroascorbate (by dehydroascorbate reductase or by chemical reaction) cause oxidation of vital cellular components [3,29,40], and the oxidative stress may trigger neuronal apoptosis [7,10,37]. Ceramide is the breakdown product of sphingomyelin (SPM) generated by the activation of a sphingomyelinase,

Crown Copyright  2001 Published by Elsevier Science B.V.

J.H. Song et al. / Brain Research 895 (2001) 66 – 72

and is a lipid second messenger which is involved in cellular stress and apoptosis [2,15,36]. The oxidative stress caused by physiologic and pharmacologic agents leads to ceramide accumulation which initiates apoptosis [8]. Apoptosis is inhibited by scavengers of reactive oxygen species such as glutathione (GSH) and N-acetylcystein, a thiol antioxidant [25]. In this study, we showed that oxidative stress induced by ascorbate involves the activation of SPM-ceramide signaling. These events may play a crucial role in neuronal death associated with pathological conditions such as stroke.

2. Materials and methods Male Sprague–Dawley rats (Charles River Canada, Montreal) weighing 275–300 g were housed with illumination for 14 h daily (06:00–20:00 h). Purnia lab chow and tap water was supplied ad libitum. The rats were killed by decapitation and brains were collected. Cortices were sliced (,1 mm) after the brains were wetted with DMEM medium (Gibco, Green Island, NY) containing 15% horse serum and 2.5% fetal calf serum (culture medium). The cerebral cortical slices were suspended in culture medium (400 mg wet tissue / 4 ml culture medium) and the cortical slices were treated by replacing fresh culture medium containing pharmacological agents. Petri dishes containing the brain slices suspended in culture medium were incubated at 378C under a water-saturated atmosphere of 5% CO 2 –95% air. After treatment for a 2-h period, the cortical slices were washed with phosphate buffered saline (PBS; 4 ml) containing 0.002% butylated hydroxy toluene (BHT; Sigma), transferred to 10 ml test tubes and then centrifuged (10003g for 10 min). The washing processes were repeated once and sample pellets were resuspended in PBS and then homogenized to make 10% homogenates (w / v) in the presence of 4.4 mM BHT. The homogenate was used to quantify the amounts of lipid peroxide using thiobarbituric acid reactive substances (TBARS) assay [5]. Briefly, samples were mixed with 1 ml of 0.67% thiobarbituric acid (TBA; Sigma) and 0.5 ml 20% trichloroacetic acid (Sigma) in test tubes (133100 mm) with a marble on the top and the mixtures were incubated in a boiling water bath for 20 min. After cooling the tubes on ice, the reaction mixture was centrifuged at 30003g for 10 min and absorbance of the supernatant was read at 532 nm. The concentrations of TBARS were calculated using tetraethoxy propane (Sigma) as a reference standard. Protein content was determined spectrophotometrically using a Bio-Rad protein determination kit. The quantities of TBARS were expressed in terms of amount (nmol) per 100 mg protein, and presented in percentage to compensate for variations of absolute weights of lipid oxides among different experiments. GSH content was measured using a method described by Hissin and Hilf [19]. Cortical slices (100 mg) were washed

67

with 0.5 ml of 25% metaphosphoric acid (Na 2 HPO 3 ) and 2 ml of 0.1 M sodium phosphate–0.005 M EDTA buffer (pH 8.0). After centrifugation at 10,0003g for 30 min, the supernatants were incubated with o-phthalaldehyde (OPT) for GSH determination. The GSH content was determined by a fluorometry with excitation and emission wave lengths of 350 and 420 nm, respectively. Data were calculated on the basis of GSH calibration curves. Tissue ceramide level was measured by diacylglycerol kinase assay [49]. Total lipids were extracted by Folch partition [13], dried with a stream of nitrogen, and incubated in 20 ml of aqueous 7.5% n-octyl-b-D-glucopyranoside (w / v), 5 mM cardiolipin, 1 mM diethylenetriamine pentaacetic acid, pH 6.6 (OG / CL buffer) and E. coli diacylglycerol kinase in the presence of r-[ 32 P]ATP at room temperature for 1 h. Labeled lipids were separated by thin layer chromatography (TLC) in chloroform / acetone / methanol / acetic acid / water (10:4:3:2:1, v / v). The autoradiogram was analyzed by an electronic autoradiography (Instant Imager, Packard), and the ceramide content was determined with a standard curve 0–1.7 nmol of ceramides. The ceramide content was normalized with total phospholipid, determined according to the method described by Van Veldhoven and Mannaerts [48].

3. Results Increasing concentrations of ascorbate enhanced lipid peroxide in a dose dependent manner (Fig. 1A). Exposure of cortical slices to 1 mM ascorbate significantly increased amount of lipid peroxide from 100.063.8% (control value) to 190.3623.3% (P,0.001) (Fig. 1B), wortmannin (10 mM) treatment did not change quantities of lipid peroxide (112.066.5) (Fig. 1B). However, a treatment group of ascorbate (1 mM) plus wortmannin (10 mM) contained less lipid peroxide (116.9612.0%, P,0.01) than ascorbate alone (190.3623.3%) (Fig. 1B). Ascorbate is readily oxidized to dehydroascorbate in culture medium, and the dehydroascorbate is known to efficiently transport into the lymphoblast by a glucose transporter (GLUT) [17–19]. Wortmannin inhibits phosphatidylinositol (PI)-3-kinase [1], which induces migration of GLUT to plasma membrane [20]. GSH alone (500 mM) did not change basal lipid peroxide content (105.464.1%) (Fig. 1B). When 500 mM GSH was introduced to the cortical slices in the presence of 1 mM ascorbate, lipid peroxide content decreased to 119.267.3% (ascorbate1GSH) from 190.3623.3% (ascorbate alone, P,0.01) (Fig. 1B). GSH inhibits oxidative stress either by preventing oxidation of ascorbate in medium, or by scavenging dehydroascorbate in the cytosol. The effect of rotenone on oxidative stress was tested to illustrate whether GSH suppresses oxidation of ascorbate in medium or scavenges dehydroascorbate in the cells. A low concentration of rotenone (10 mM) was ineffective in

68

J.H. Song et al. / Brain Research 895 (2001) 66 – 72

Fig. 2. The effects of rotenone (10, 30 mM) on lipid peroxide production with or without glutathione (GSH 0.5 mM, 1 mM) in cortical slices. Lipid peroxides were quantified by measuring thiobarbituric acid reactive substances (TBARS). Quantities are expressed as percentage of control values. Each vertical bar represents mean6S.E.M. (n56). *P,0.05 vs. control.

Fig. 1. Dose–response relationship between ascorbate (AA) and lipid peroxide production in cerebral cortical slices (A). Effects of 500 mM glutathione (GSH) and 10 mM wortmannin (WT) on the oxidative stress induced by AA (1 mM) (B). Lipid peroxides were quantified by measuring thiobarbituric acid reactive substances (TBARS). Quantities are expressed as percentage of control values. Each vertical bar represents mean6S.E.M. (n56). *P,0.01.

producing oxidative stress (99.466.2%) under the present experimental conditions, but a higher concentration (30 mM) significantly elevated lipid peroxide concentration to 129.364.7% (Fig. 2). The elevated lipid peroxide concentration was not decreased by GSH treatment (30 mM rotenone11 mM GSH; 129.863.5%) indicating that either GSH does not penetrate brain cells, or that GSH penetration into neurons is insignificant (Fig. 2). Therefore, a possibility of scavenging cytosolic dehydroascorbate is excluded, and thus GSH likely acts to suppress oxidation of ascorbate in culture medium. As expected, hydrogen peroxide (100 mM) treatment increased quantities of lipid peroxide (157.464.5%) and GSH neutralized the effect of hydrogen peroxide on lipid peroxide production (103.5613.1%) (Fig. 3). However, when the cortical slices were treated with wortmannin in

the presence of hydrogen peroxide, quantities of hydrogen peroxide-induced lipid peroxide (157.464.5%) were not decreased (156.2611.5%) (Fig. 3). The differential effects of wortmannin on ascorbate – (Fig. 1) and hydrogen peroxide – (Fig. 3) induced lipid peroxide suggest that wortmannin is not a general suppressor of oxidative stress, but a specific inhibitor of ascorbate-induced oxidative stress. In order to further support the concept that ascorbate generates oxidative stress, GSH content was quantified. Endogenous GSH levels were reduced in a dose-dependent manner in cortical slices treated with different concen-

Fig. 3. The effects of hydrogen peroxide (H 2 O 2 ) (100 mM), 100 mM H 2 O 2 plus glutathione (H 2 O 2 1GSH), and 100 mM H 2 O 2 plus 10 mM wortmannin (H 2 O 2 1WT) on lipid peroxide production in cerebral cortex slices. Lipid peroxides were quantified by measuring thiobarbituric acid reactive substances (TBARS). Quantities are expressed as percentage of control values. Each vertical bar represents mean6S.E.M. (n56). *P, 0.01.

J.H. Song et al. / Brain Research 895 (2001) 66 – 72

trations of ascorbate (Fig. 4A). Ascorbate (1 mM) treatment significantly decreased GSH content from 100.063.4% (control group) to 43.664.1% (Fig. 4B). Treatment with wortmannin plus ascorbate (1 mM) did not decrease GSH content (109.963.2%) (Fig. 4B). This observation shows that the wortmannin antagonizes ascorbate-induced oxidative stress and suggests that it reduces transport capacity of dehydroascorbate into the tissue. Changes of ceramide levels were examined to determine if ascorbate-induced oxidative stress activates apoptosis in the cerebral cortical tissue. The ceramide level in the presence of 1 mM ascorbate was significantly increased to 199.6614.3% (P,0.001) (Fig. 5). Exposure of cortical slices to GSH (500 mM) in the presence of ascorbate (1 mM), decreased ceramide levels to 131.862.3% (P,0.05) while wortmannin (10 mM) plus ascorbate further de-

69

Fig. 5. Effects of 500 mM glutathione (GSH), 10 mM wortmannin (WT), 1 mM ascorbate (AA), 1 mM ascorbate plus 500 mM GSH (AA1GSH) and 1 mM ascorbate plus 10 mM wortmannin (AA1WT) on ceramide formation in cerebral cortex. Quantities of ceramide are expressed as percentage of control (100%). Each vertical bar represents means6S.E.M. (n56). *P,0.01.

creased ceramide levels to 83.166.9% (P,0.001) from the ascorbate treated group (199.6614.3%) (Fig. 5). When cortical slices were treated with 100 mM hydrogen peroxide for a 2-h period, ceramide levels were significantly increased to 156.364.6% (P,0.05 vs. control). However, GSH (500 mM) plus hydrogen peroxide (100 mM) showed lower levels of ceramide (115.665.9%, P,0.05) than hydrogen peroxide-treated group (15664.6%) (Fig. 6) indicating that GSH inactivated the hydrogen peroxide effect on ceramide production. Wortmannin (10 mM) treatment in the presence of hydrogen peroxide (100 mM) did not change ceramide activities (155.663.9%) from the hydrogen peroxide-treated group (156.364.6%) (Fig. 6).

Fig. 4. Dose–response relationships between ascorbate (AA) and endogenous glutathione (GSH) levels in cerebral cortical slices (A). Effects of 1 mM ascorbate (AA), 1 mM ascorbate plus 10 mM wortmannin (AA1WT) on GSH levels (B). Quantities are expressed as percentage of control values. Each vertical bar represents mean6S.E.M. (n56). **P,0.001.

Fig. 6. Effects of 100 mM hydrogen peroxide (H 2 O 2 ), 100 mM H 2 O 2 plus 500 mM glutathione (H 2 O 2 1GSH) and 100 mM H 2 O 2 plus 10 mM wortmannin (H 2 O 2 1WT) on ceramide formation in cerebral cortex. Quantities of ceramide are expressed as percentage of control (100%). Each vertical bar represents means6S.E.M. (n56). *P,0.05.

70

J.H. Song et al. / Brain Research 895 (2001) 66 – 72

4. Discussion Ascorbate is a well-known reducing agent, which will easily donate electrons, becoming an oxidized compound. The mechanism of an oxidative action induced by ascorbate is difficult to envisage, because the molecule is unable to accept electrons. It has previously been suggested that an ascorbate-iron interaction is responsible for the initiation of lipid peroxidation [17,32], commonly believed to be mediated through the Fenton reaction. However, this reaction has been demonstrated only with cell homogenates in protein-free medium. The ascorbate-induced oxidative stress in live neuronal cells is not caused by interaction between ascorbate and transition metal ions [44]. We have proposed an alternate model to elucidate ascorbate-induced oxidative action [43,44]. Dehydroascorbate, an oxidized form of ascorbate, is carried into cells and reduction of the cytosolic dehydroascorbate to ascorbate causes oxidative stress [44]. The dehydroascorbate is known to be efficiently transported into cells by a GLUT [50]. Two other mechanisms are involved in the transport of ascorbate into cells: Na 1 -dependent ascorbate transporters (SVCT1 and SVCT2) [46] and glutamate–ascorbate heteroexchange [30,38]. These two mechanisms would not generate oxidative stress, since ascorbate is transported into cells. Therefore, only a ‘dehydroascorbate transporter’ (GLUT) will be involved in oxidative stress generated by ascorbate. The processes of decapitation and preparing cortical slices cause hypoxia due to the discontinuation of blood supply, and slice incubation will resemble in someways a reperfusion period. Many chemicals in brain cells, including ascorbate, are released during a period of hypoxia, and then the released ascorbate will be oxidized to dehydroascorbate by molecular oxygen in plasma and taken up during a reperfusion period as dehydroascorbate. The halflife of ascorbate is relatively short (t 1 / 2 55–50 min) in solution [11,39]. Plasma contains sufficient quantities of free oxygen (Pa O 2 9562 Torr) [52] to oxidize ascorbate to dehydroascorbate. Plasma dehydroascorbate concentration is approximately 100 mM [54]. The dehydroascorbate in cells is rapidly reduced to ascorbate by the action of dehydroascorbate reductase [3,29,40] and non-enzymatic chemical reactions. The reduction process generates an oxidative stress, which decreases quantities of cellular reducing agents such as GSH (Fig. 4), and increases quantities of oxidized cellular components such as lipid peroxide (Fig. 1). The oxidative stress is known to trigger apoptosis [7,10,37] causing neuronal degeneration in the central nervous system. Rapid uptake of an oxidizing agent such as dehydroascorbate damages vital cellular components. However, oxidative stress can be completely scavenged by enzymes such as dismutase and catalase [6,22], and by endogenous reducing agents such as GSH when the uptake rates are slow or the

total amount of oxidizing agents are small. A high uptake rate of dehydroascorbate is a critical factor to generate oxidative stress. We have attempted to suppress rates of dehydroascorbate uptake by inhibiting ‘dehydroascorbate transportation system’ with wortmannin treatment. Wortmannin inhibits activity of PI-3-kinase [26,47]. PI-3-kinase plays many different roles in signal transduction and is a key enzyme of a signal cascade downstream of insulin stimulation of its receptor. Insulin translocates GLUT from vesicular storage compartment to plasma membrane via PI-3-kinase [20]. Wortmannin inhibits the function of GLUT by suppressing GLUT migration to the plasma membrane by suppressing PI-3 kinase [20,56]. Production of ascorbate-induced lipid peroxide and of ceramide were prevented by wortmannin in our test system (Figs. 1 and 5). The decreased amounts of lipid peroxide and ceramide in a treatment group of ascorbate plus wortmannin support the concept that GLUT mediates dehydroascorbate transportation into cells. The wortmannin effect on hydrogen peroxide-induced lipid peroxide production was tested in order to eliminate a possibility that wortmannin may be a non-specific general inhibitory agent on lipid peroxide production. Wortmannin was ineffective against hydrogen peroxide-induced lipid peroxide production (Fig. 3). The differential effects of wortmannin between ascorbate- and hydrogen peroxideinduced lipid peroxide production (Figs. 1 and 3) show that wortmannin’s inhibitory action is not a non-specific general phenomenon, but due to the inhibitory action on dehydroascorbate uptake via a ‘dehydroascorbate transporter’ (GLUT). GSH is an important antioxidant in biological systems, scavenging reactive oxygen species such as hydroxyl radical, hydrogen peroxides and singlet oxygen [43,12]. Thus, GSH can protect brain tissues from oxidative damage. GSH in medium inhibited ascorbate-induced oxidative stress in cerebral cortical slices (Fig. 1). GSH suppresses oxidative stress either (1) by preventing ascorbate’s oxidation to dehydroascorbate in culture medium and / or (2) by penetrating into neurons and scavenging ROS. (1) Suppression of the oxidation in the medium results in less dehydroascorbate available to be taken up into the neurons. (2) It is known that GSH does not easily diffuse into cells, but there is a possibility that GSH may leak into the cells due to a high concentration in medium. We tried to resolve a question whether GSH’s inhibitory action on ascorbate-induced oxidative stress is due to inhibition of ascorbate oxidation extracellularly, or due to protection of critical cellular elements against oxidation intracellularly. The oxidative action of hydrogen peroxide was completely inhibited by GSH (Fig. 3), which is most likely caused by inactivating hydrogen peroxide in medium. In addition, we examined whether GSH inhibits oxidative stress via an extracellular or cytoplasmic mechanism. We used rotenone (an inhibitor of Complex I in

J.H. Song et al. / Brain Research 895 (2001) 66 – 72

mitochondria) [18,42,55] to determine if cytoplasmic oxidative stress could be inhibited by extracellular GSH. Only intracellular GSH will suppress rotenone-induced oxidative damage as oxidative stress is generated within mitochondria. Under these conditions, exogenous GSH treatment did not have any effect on the rotenone-induced oxidative stress (Fig. 2). Therefore, GSH does not penetrate the neurons, or the amount of GSH diffused into the neuron is insignificant and thus unable to scavenge a substantial amount of oxidative potential. We, therefore, concluded that the protective effect of the GSH on ascorbate-induced oxidative stress is caused by prevention of ascorbate oxidation to dehydroascorbate since GSH cannot diffuse into the neurons. The reduction process of cytosolic dehydroascorbate causes production of lipid peroxide, but ascorbate in the neurons (or taken up via Na 1 -dependent mechanism [50] and glutamate–ascorbate heteroexchange [30,38]) does not generate oxidative stress since it cannot take an extra electron from cellular components. The uptake rate through the Na 1 -dependent mechanism is much lower (approximately 10 times less) than the one via ‘dehydroascorbate transporter’ (GLUT) [50]. Ingestion of large amounts of ascorbate is unrelated to oxidative stress, but rapid uptake of dehydroascorbate into cells will generate oxidative stress. Ceramide is an established member in the cascade of reactions of apoptosis [2,15,36] and is generated mainly from de novo synthesis via ceramide synthase. It may also be produced from the hydrolysis of sphingomyelin via the activation of acidic or neutral sphingomyelinase [27]. Ascorbate generates oxidative stress via dehydroascorbate, and the oxidative stress activates ceramide in neuronal cells [8]. Changes of ceramide levels parallel quantities of lipid peroxide in the neurons in several different experimental conditions (Figs. 4, 5 and 6). The close relationship between ascorbate-induced oxidative stress and ceramide levels supports the notion that ascorbateinduced oxidative stress triggers apoptosis in the neurons. Several recent studies have suggested that ceramide is activated by oxidative stress and hypoxia–ischemia in neuronal cells [15,9]. Our studies have shown that ascorbate reduces endogenous GSH levels with their oxidative potential corresponding to that of ceramide formation. Taken together with the observations that enhanced GSH biosynthesis by N-acetylcystein is able to inhibit sphingomyelinase activity and thereby reduce ceramide formation, it is possible that depletion of GSH by ascorbate may trigger that activation of sphingomyelinase resulting in ceramide production and induction of apoptosis. In summary, observations reported here support the hypothesis that ascorbate is oxidized to dehydroascorbate by oxygen (O 2 ) in solution (culture medium or plasma), and the dehydroascorbate is transported into the neurons by a ‘dehydroascorbate transporter’ (GLUT) where it is

71

reduced back to the ascorbate generating oxidative stress. The oxidative stress, in turn, triggers apoptosis.

Acknowledgements These studies are supported by the Heart and Stroke Foundation of Ontario (NA-4474 to SHS and NA-3718 to GMR).

References [1] A. Arcaro, M.P. Wymann, Wortmannin is a potent phosphatidylinositol 3-kinase inhibitor: the role of phosphatidylinositol 3,4,5-trisphosphate in neutrophil responses, Biochem. J. 296 (1993) 297–301. [2] T. Ariga, W.D. Jarvis, R.K. Yu, Role of sphingolipid-mediated cell death in neurodegenerative diseases, J. Lipid Res. 39 (1998) 1–16. [3] A.M. Bode, C.R. Yavarow, D.A. Fry, T. Vargas, Enzymatic basis for altered ascorbic acid and dehydroascorbic acid levels in diabetes, Biochem. Biophys. Res. Commun. 191 (1993) 1347–1353. [4] P.L. Brick, A.C. Howlett, M.C. Beinfeld, Synthesis and release of vasoactive intestinal polypeptide (VIP) by mouse neuroblastoma cells: modulation by cyclic nucleotides and ascorbic acid, Peptides 6 (1985) 1075–1078. [5] J.A. Buege, S.D. Aust, Microsomal lipid peroxidation, Methods Enzymol. 52 (1978) 302–310. [6] M.V. Clement, A. Ponton, S. Pervaiz, Apoptosis induced by hydrogen peroxide is mediated by decreased superoxide anion concentration and reduction of intracellular milieu, FEBS Lett. 440 (1998) 13–18. [7] S. Clutton, The importance of oxidative stress in apoptosis, Br. Med. Bull. 53 (1997) 662–668. [8] M. Degli Esposti, H. McLennan, Mitochondria and cells produce reactive oxygen species in virtual anaerobiosis: relevance to ceramide-induced apoptosis, FEBS Lett. 430 (1998) 338–342. [9] M.D. Esposti, I. Hatzinisiriou, H. McLennan, S. Ralph, Bcl-2 and mitochondrial oxygen radicals. New approaches with reactive oxygen species-sensitive probes, J. Biol. Chem. 274 (1999) 29831– 29837. [10] J.M. Esteve, J. Mompo, J. Garcia de la Asuncion, J. Sastre, M. Asensi, J. Boix, J.R. Vina, J. Vina, F.V. Pallardo, Oxidative damage to mitochondrial DNA and glutathione oxidation in apoptosis: studies in vivo and in vitro, FASEB J. 13 (1999) 1055–1064. [11] J. Feng, A.H. Melcher, D.M. Brunette, H.K. Moe, Determination of L-ascorbic acid levels in culture medium: concentrations in commercial media and maintenance of levels under conditions of organ culture, In Vitro 13 (1977) 91–99. [12] J.C. Fernandez-Checa, C. Garcia-Ruiz, A. Colell, A. Morales, M. Mari, M. Miranda, E. Ardite, Oxidative stress: role of mitochondria and protection by glutathione, Biofactors 8 (1998) 7–11. [13] J. Folch, M. Lees, G.H.S. Stanley, A simple method for the isolation and purification of total lipides from animal tissues, J. Biol. Chem. 226 (1957) 497–509. [14] S. Friedman, S. Kaufman, 3,4-dihydroxyphenylethylamine beta-hydroxylase. Physical properties, copper content, and role of copper in the catalytic activity, J. Biol. Chem. 240 (1965) 4763–4773. [15] T. Goldkorn, N. Balaban, M. Shannon, V. Chea, K. Matsukuma, D. Gilchrist, H. Wang, C. Chan, H 2 O 2 acts on cellular membranes to generate ceramide signaling and initiate apoptosis in tracheobronchial epithelial cells, J. Cell Sci. 111 (1998) 3209–3220. [16] B. Halliwell, C.H. Foyer, Ascorbic acid, metal ions and the superoxide radical, Biochem. J. 155 (1976) 697–700.

72

J.H. Song et al. / Brain Research 895 (2001) 66 – 72

[17] R.E. Heikkila, F.S. Cabbat, Ascorbate-induced lipid peroxidation and inhibition of [ 3 H]spiroperidol binding in neostriatal membrane preparations, J. Neurochem. 41 (1983) 1384–1392. [18] M. Higuchi, R.J. Proske, E.T. Yeh, Inhibition of mitochondrial respiratory chain complex I by TNF results in cytochrome C release, membrane permeability transition, and apoptosis, Oncogene 17 (1998) 2515–2524. [19] P.J. Hissin, R. Hilf, A fluorometric method for determination of oxidized and reduced glutathione in tissues, Anal. Biochem. 74 (1976) 14–26. [20] G.D. Holman, S.W. Cushman, Subcellular localization and trafficking of the GLUT4 glucose transporter isoform in insulin-responsive cells, Bioessays 16 (1994) 753–759. [21] D. Hornig, Distribution of ascorbic acid, metabolites and analogues in man and animals, Ann. N Y Acad. Sci. 258 (1975) 103–118. [22] J.N. Keller, M.S. Kindy, F.W. Holtsberg, D.K. St Clair, H.C. Yen, A. Germeyerm, S.M. Steiner, A.J. Bruce-Keller, J.B. Hutchins, M.P. Mattson, Mitochondrial manganese superoxide dismutase prevents neural apoptosis and reduces ischemic brain injury: suppression of peroxynitrite production, lipid peroxidation, and mitochondrial dysfunction, J. Neurosci. 18 (1998) 687–697. [23] C.H. Kuo, F. Hata, H. Yoshida, A. Yamatodani, H. Wada, Effect of ascorbic acid on release of acetylcholine from synaptic vesicles prepared from different species of animals and release of noradrenaline from synaptic vesicles of rat brain, Life Sci. 24 (1979) 911–915. [24] F.M. Leslie, C.E. Dunlap, B.M. Cox, Ascorbate decreases ligand binding to neurotransmitter receptors, J. Neurochem. 34 (1980) 219–221. [25] B. Liu, N. Andrieu-Abadie, T. Levade, P. Zhang, L.M. Obeid, Y.A. Hannun, Glutathione regulation of neutral sphingomyelinase in tumor necrosis factor-alpha-induced cell death, J. Biol. Chem. 273 (1998) 11313–11320. [26] D. Malide, S.W. Cushman, Morphological effects of wortmannin on the endosomal system and GLUT4-containing compartments in rat adipose cells, J. Cell Sci. 110 (1997) 2795–2806. [27] S. Mathias, L.A. Pena, R.N. Kolesnick, Signal transduction of stress via ceramide, Biochem. J. 335 (1998) 465–480. [28] Y. Matsuo, T. Kihara, M. Ikeda, M. Ninomiya, H. Onodera, K. Kogure, Role of neutrophils in radical production during ischemia and reperfusion of the rat brain: effect of neutrophil depletion on extracellular ascorbyl radical formation, J. Cereb. Blood Flow Metab. 15 (1995) 941–947. [29] J.M. May, C.E. Cobb, S. Mendiratta, K.E. Hill, R.F. Burk, Reduction of the ascorbyl free radical to ascorbate by thioredoxin reductase, J. Biol. Chem. 273 (1998) 23039–23045. [30] M. Miele, M.G. Boutelle, M. Fillenz, The physiologically induced release of ascorbate in rat brain is dependent on impulse traffic, calcium influx and glutamate uptake, Neuroscience 62 (1994) 87– 91. [31] K. Milby, A. Oke, R.N. Adams, Detailed mapping of ascorbate distribution in rat brain, Neurosci. Lett. 28 (1982) 15–20. [32] D.M. Miller, S.D. Aust, Studies of ascorbate-dependent, iron-catalyzed lipid peroxidation, Arch. Biochem. Biophys. 271 (1989) 113–119. [33] B.T. Miller, T.J. Cicero, Ascorbic acid enhances the release of luteinizing hormone-releasing hormone from the mediobasal hypothalamus in vitro, Life Sci. 39 (1986) 2447–2454. [34] F.C. Ngkeekwong, L. L Ng, Two distinct uptake mechanisms for ascorbate and dehydroascorbate in human lymphoblasts and their interaction with glucose, Biochem. J. 324 (1998) 225–230. [35] A.F. Oke, L. May, R.N. Adams, Ascorbic acid distribution patterns in human brain: a comparison with nonhuman mammalian species, Ann. N Y Acad. Sci. 498 (1987) 1–12. [36] A. Quillet-Mary, J.P. Jaffrezou, V. Mansat, C. Bordier, J. Naval, G. Laurent, Implication of mitochondrial hydrogen peroxide generation in ceramide-induced apoptosis, J. Biol. Chem. 272 (1997) 21388– 21395.

[37] R.R. Ratan, J.M. Baraban, Apoptotic death in an in vitro model of neuronal oxidative stress, Clin. Exp. Pharmacol. Physiol. 22 (1995) 309–310. [38] G.V. Rebec, R.C. Pierce, A vitamin as neuromodulator: ascorbate release into the extracellular fluid of the brain regulates dopaminergic and glutamatergic transmission, Prog. Neurobiol. 43 (1994) 537–565. [39] H. Sakagami, K. Satoh, H. Ohata, H. Takahashi, H. Yoshida, M. Iida, N. Kuribayashi, T. Sakagami, K. Momose, M. Takeda, Relationship between ascorbyl radical intensity and apoptosis-inducing activity, Anticancer Res. 16 (1996) 2635–2644. [40] I. Savini, I. D’Angelo, M. Annicchiarico-Petruzzelli, L. Bellincampi, G. Melino, L. Avigliano, Ascorbic acid recycling in N-myc amplified human neuroblastoma cells, Anticancer Res. 18 (1998) 819– 822. [41] J.O. Schenk, E. Miller, R. Gaddis, R.N. Adams, Homeostatic control of ascorbate concentration in CNS extracellular fluid, Brain Res. 253 (1982) 353–356. [42] T.A. Seaton, J.M. Cooper, A.H. Schapira, Free radical scavengers protect dopaminergic cell lines from apoptosis induced by complex I inhibitors, Brain Res. 777 (1997) 110–118. [43] F. Si, G.M. Ross, S.H. Shin, Glutathione protects PC12 cells from ascorbate- and dopamine-induced apoptosis, Exp. Brain Res. 123 (1998) 263–268. [44] J.H. Song, S.H. Shin, G.M. Ross, Prooxidant effects of ascorbate in rat brain slices, J. Neurosci. Res. 58 (1999) 328–336. [45] R.D. Todd, P.A. Bauer, Ascorbate modulates 5-[3H]hydroxytryptamine binding to central 5-HT3 sites in bovine frontal cortex, J. Neurochem. 50 (1988) 1505–1512. [46] H. Tsukaguchi, T. Tokui, B. Mackenzie, U.V. Berger, X.Z. Chen, Y. Wang, R.F. Brubaker, M.A. Hediger, A family of mammalian Na 1 -dependent L-ascorbic acid transporters, Nature 399 (1999) 70–75. [47] A.M. Valverde, P. Navarro, T. Teruel, R. Conejo, M. Benito, M. Lorenzo, Insulin and insulin-like growth factor I up-regulate GLUT4 gene expression in fetal brown adipocytes, in a phosphoinositide 3-kinase-dependent manner, Biochem. J. 337 (1999) 397–405. [48] P.P. Van Veldhoven, G.P. Mannaerts, Inorganic and organic phosphate measurements in the nanomolar range, Anal. Biochem. 161 (1987) 45–48. [49] P.P. Van Veldhoven, W.R. Bishop, R.M. Bell, Enzymatic quantification of sphingosine in the picomole range in cultured cells, Anal. Biochem. 183 (1989) 177–189. [50] J.C. Vera, C.I. Rivas, F.V. Velasquez, R.H. Zhang, I.I. Concha, D.W. Golde, Resolution of the facilitated transport of dehydroascorbic acid from its intracellular accumulation as ascorbic acid, J. Biol. Chem. 270 (1995) 23706–23712. [51] J.C. Vera, C.I. Rivas, J. Fischbarg, D.W. Golde, Mammalian facilitative hexose transporters mediate the transport of dehydroascorbic acid, Nature 364 (1993) 79–82. [52] J.K. Walker, B.L. Lawson, D.B. Jennings, Breath timing, volume and drive to breathe in conscious rats: comparative aspects, Respir. Physiol. 107 (1997) 241–250. [53] R.W. Welch, Y. Wang Jr., A. Crossman, J.B. Park, K.L. Kirk, M. Levine, Accumulation of vitamin C (ascorbate) and its oxidized metabolite dehydroascorbic acid occurs by separate mechanisms, J. Biol. Chem. 270 (1995) 12584–12592. [54] J.C. Will, T. Byers, Does diabetes mellitus increase the requirement for vitamin C?, Nutr. Rev. 54 (1996) 193–202. [55] E.J. Wolvetang, K.L. Johnson, K. Krauer, S.J. Ralph, A.W. Linnane, Mitochondrial respiratory chain inhibitors induce apoptosis, FEBS Lett. 339 (1994) 40–44. [56] J. Yang, J.F. Clarke, C.J. Ester, P.W. Young, M. Kasuga, G.D. Holman, Phosphatidylinositol 3-kinase acts at an intracellular membrane site to enhance GLUT4 exocytosis in 3T3-L1 cells, Biochem. J. 313 (1996) 125–131.