Arachidonic acid promotes glutamate-induced cell death associated with necrosis by 12- lipoxygenase activation in glioma cells

Life Sciences 80 (2007) 1856 – 1864 www.elsevier.com/locate/lifescie

Arachidonic acid promotes glutamate-induced cell death associated with necrosis by 12- lipoxygenase activation in glioma cells Yoshihiro Higuchi a,⁎, Hideji Tanii b , Yoshiki Koriyama c , Yuji Mizukami d , Tanihiro Yoshimoto a a

Department of Molecular Pharmacology, Kanazawa University Graduate School of Medical Science, 13–1 Takara-machi, Kanazawa, Ishikawa 920–8640, Japan b Department of Hygene, Graduate School of Medical Science, Kanazawa University, Kanazawa 920–8640, Japan c Department of Molecular Neurobiology, Graduate School of Medical Science, Kanazawa University, Kanazawa 920–8640, Japan d Department of Radiological Technology, Graduate School of Medical Science, Kanazawa University, Kanazawa 920–8640, Japan Received 8 November 2006; accepted 15 February 2007

Abstract Glutamate induced glutathione (GSH) depletion in C6 rat glioma cells, which resulted in cell death. This cell death seemed to be apoptosis through accumulation of reactive oxygen species (ROS) or hydroperoxides representing cytochrome c release from mitochondria and internucleosomal DNA fragmentation. A significant increase of 12-lipoxygenase enzyme activity was observed in the presence of arachidonic acid (AA) under GSH depletion induced by glutamate. AA promoted the glutamate-induced cell death, which reduced caspase-3 activity and diminished internucleosomal DNA fragmentation. Furthermore, AA reduced intracellular NAD, ATP and membrane potentials, which indicated dysfunction of the mitochondrial membrane. Protease inhibitors such as N-α-tosyl-L-phenylalanine chloromethyl ketone (TPCK) and 3, 4dichloroisocumarin (DCI) but no Ac-DEVD, a caspase inhibitor, suppressed the glutamate-induced cell death. AA reduced the inhibitory effect of TPCK and DCI on the glutamate-induced cell death. These results suggest that AA promotes cell death by inducing necrosis from caspase-3independent apoptosis. This might occur through lipid peroxidation initiated by ROS or lipid hydroperoxides generated during GSH depletion in C6 cells. © 2007 Published by Elsevier Inc. Keywords: Apoptosis; Arachidonic acid; Glutamate; GSH depletion; 12-Lipoxigenase; Necrosis; Reactive oxygen species

Introduction It has been proposed that glutamate exerts cytotoxic action through inhibition of cystine uptake, which leads to a marked decrease in cellular glutathione (GSH) levels, exposing the cells to oxidative stress (Murphy et al., 1989). Receptor-mediated effects of excitatory amino acids such as glutamate, kainate and N-methyl-D-aspartate (NMDA) (Campagne et al., 1995) have been well characterized. Cell death caused by glutamate-induced GSH depletion in some neuronal tissues and cell lines are distinct from amino acid excitotoxicity (Cyle and Puttfarcken,

⁎ Corresponding author. Tel.: +81 76 265 2186; fax: +81 76 234 4227. E-mail address: [email protected] (Y. Higuchi). 0024-3205/$ - see front matter © 2007 Published by Elsevier Inc. doi:10.1016/j.lfs.2007.02.031

1993). Intracellular GSH depletion has been reported to induce apoptosis in other cell lines (Pereicra and Oliveira, 1997). Thus, characterization of the type of cell death initiated by glutamate could further define the molecular mechanisms underlying this form of neuronal cell death and reveal sites for pharmacological intervention. The GSH and GSH peroxidase system is the primary defense mechanism for peroxide removal from the brain against the effects of reactive oxygen species (ROS) damage, which is involved in neuropathological disorders (Simonian and Coyle, 1996). When cells are stimulated, phospholipase A2 (PLA2) is expressed in many cell types and appears to be a crucial enzyme that is involved in the selective release of AA from phospholipids (Murakami et al., 1999). AA is an important cellular signaling mediator, the precursor of eicosanoids and a critical component of cellular membranes. AA metabolism is catalyzed by two

Y. Higuchi et al. / Life Sciences 80 (2007) 1856–1864

major groups of enzyme, lipoxygenases including 5-lipoxygenase, 12-lipoxygenase and 15-lipoxygenase, and cyclooxygenases. AA is metabolized to its hydroperoxides such as hydroperoxy eicosatetraenoic acids (HPETE) under GSH depletion (Spiteller, 1996). Twelove-lipoxygenase plays a significant role in tumor cell survival and apoptosis (Tang et al., 1996). Twelve-lipoxygenase is expressed in neurons and glia such as oligodendrocytes and astrocytes (Ishyama et al., 1993). Therefore, it is important to evaluate not only the role of AA but also whether AA metabolites are involved in glutamate-induced cell death. Apoptosis and necrosis are two distinct forms of cell death that have profoundly different implications for the surrounding tissues. Apoptosis is characterized by cell shrinkage, chromatin condensation, caspase activation, and fragmentation of DNA at internucleosomal linker sites giving rise to discrete bands of multiples of 180–200 base pairs (Carson and Ribeiro, 1993). In contrast, necrosis is a passive process, typified by cell and organelle swelling with spillage of the intracellular contents into the extracellular milieu. Necrosis is an uncontrolled event resulting from loss of homeostasis and cell contents are dispersed, which may then have adverse effects on neighboring tissue (Swartz et al., 1993). In this study, we investigated the effect of AA on the apoptotic or necrotic events through activation of 12-lipoxygenase induced by glutamate in C6 rat glioma cells.

1857

Materials and methods Materials Anti-apopain (Ac-DEVD-CHO), baicalein, ETYA, indomethacin, sodium N-lauroyl sarcosine, nordihydroguaiaretic acid (NDGA), sodium glutamate monohydrate, 3.4-dichloroisocumarin (DCI), N-α-tosyl-L-phenylalanine chloromethyl ketone (TPCK) and α-tocopherol (α-Toc) were purchased from Sigma Chemicals (St. Louis, USA). Four-sulfamoyl-7fluoro-2,1,3-benzoxadiazole (ABD-F) was purchased from Wako Chemicals (Osaka, Japan). Arachidonic acid (AA) was purchased from Biomol Research Lab)Plymouth, USA). Dulbecco's modified Eagle's medium (DMEM) and fetal calf serum were purchased from Gibco (BRL, Rockville, USA). We purchased Na251CrO4 (300 mCi/mg chromium) and [3 H]tetraphenylphosphonium bromide (TPP+, 29.0 Ci/mmol) from Amersham (Tokyo, Japan). We purchased 2′,7′-dichlorodihydro fluorescein diacetate bisester (DCFH-DA) from Molecular Probes (Eugene, USA). Alpha-Toc was dissolved in dimethyl sulfoxide and AA was dissolved in 99.5% ethanol. Sodium glutamate monohydrate (glutamate) was dissolved in phosphate-buffered saline (PBS) and other agents were dissolved in distilled water. Antibodies against caspase-3 and cytochrome c were purchased from Santa Cruz Biotech. (Santa Cruz, USA). Anti-12-ipoxygenase antibody was made by recombinant 12-

Fig. 1. Effect of AA on depletion of intracellular GSH and on cytolysis caused by glutamate. C6 cells (2 × 106) were treated with 10 mM glutamate alone (closed circle), 10 mM glutamate + 50 μM AA (open circle), 10 mM glutamate + 50 μM AA + 100 μM α-Toc (closed square), 10 mM glutamate + 50 μM AA + 1 mM NAC (open squqre), 10 mM glutamate + 50 μM AA + 10 μM NDGA (closed triangle), or 50 μM AA alone (open triangle). GSH levels in the cells (A) and 51Cr activity released in the medium (B) were determined after incubation for the indicated periods. Values are the mean of 4 independent experiments. (C) and (D), cells were treated with (closed column) or without (open column) 10 mM glutamate in the presence of AA at the indicated concentrations. GSH levels (C) at 6 h and 51Cr activity released in the medium (D) at 9 h after the incubation in C6 cells were determined. Values are the mean ± SD of 4 independent experiments.

1858

Y. Higuchi et al. / Life Sciences 80 (2007) 1856–1864

at 60 °C for 5 min and was extracted twice with 400 μl of ethyl acetate after being chilled. The GSH in the mixture was analyzed as follows. A Shimadzu solvent delivery system (model LC-9A HPLC, Kyoto) was used for the HPLC analysis. Fluorescence detection of the ABD-F derivative of GSH was carried out using a fluorescence HPLC monitor (Shimadzu Model RF-535) at an excitation wavelength of 380 nm and emission wavelength of 510 nm. Agarose gel electrophoresis

Fig. 2. Chromosomal DNA fragmentation in C6 cells treated with glutamate. AGE analysis of C6 cells treated with 10 mM glutamate in the absence (lanes 2–3) or presence (lanes 4–7) of AA at the indicated concentrations, 50 μM AA + 10 μM NDGA (lane 6), or 50 μM AA + 100 μM α-Toc (lanes 7) for 12 h or 24 h. Lane M indicates size marker DNAs.

Agarose gel electrophoresis (AGE) was carried out according to the method described previously (Higuchi, 2002). Briefly, harvested C6 cells were embedded in 1% low melting agarose and hardened. The cells in the agarose were lysed in digesting solution that consisted of TE buffer [10 mM Tris–HCl (pH 8.6) and 0.5 M EDTA], 1% SDS and 1 mg/ml proteinase K, and were incubated at 50 °C overnight. The digestion outside of the gel was diluted with distilled water, extracted twice with phenol–chloroform (1:1, v/v) and precipitated with 70% ethanol. The precipitate was dried, dissolved in TE buffer [10 mM Tris–HCl (pH 7.4), 1 mM EDTA] and treated with

ipoxygenase and provided by Dr. Kawajiri of Kanazawa University Graduate School of Medical Science. Cell culture and Cytotoxic assay C6 rat glioma cells were grown in Dulbecco's modified Eagle's medium (DF-5) supplemented with 5% fetal calf serum containing 25 mM sodium bicarbonate, 100 U/ml of penicillin G and 60 μg/ml of kanamycin. C6 cells were usually seeded into 60 mm plastic culture dishes and grown in 5 ml of DF-5 at 37 °C in a humidified atmosphere containing 5% CO2. The 51Cr release assay was carried out according to the method described previously (Higuchi, 1996). Briefly, C6 cells were labeled with Na251CrO4 at a final concentration of 10 μCi/ 106 cells in 500 μl of DF-5 for 60 min at 37 °C. After being washed several times in chilled Hank's balanced salt solution (HBSS), the 51Cr-labeled cells were treated with glutamate under the various conditions specified for the times indicated in the figure legends at 37 °C in humidified air containing 5% CO2. Specific 51Cr release, which indicates cell death, was calculated by the formula described previously (Higuchi, 1996). Determination of GSH GSH levels were quantified by high performance liquid chromatography (HPLC) using the sulfhydryl group specific fluorigenic agent, ABD-F, for thiol, as described by Toyooka et al. (Toyook et al., 1988). The C6 glioma cell pellet was suspended in 50 μl of 5% trichloroacetic acid solution containing 5 mM EDTA and then the resultant suspension was allowed to stand for 10 min on ice. The acid extract was centrifuged at 12,000 ×g for 10 min and 10 μl of the supernatant was mixed with 190 μl of ABD-F solution (4.6 mM 4(Aminosulphonyl)-7-fluoro-2,1,3-benzoxadiazole, 100 mM sodium borate, pH 9.3, 1 mM EDTA). The mixture was incubated

Fig. 3. Effect of AA on ROS production and lipid peroxidation induced by glutamate. C6 cells were treated with or without 10 mM glutamate in the presence or absence of 50 μM AA, 100 μM α-Toc, or 10 μM NDGA for 6 h (open column) or 12 h (closed column). After treatment, the amount of DCFHreactive substances (DCF, A) and TBARS (MDA, B) in the cells were determined. DCF amount was expressed as a percentage of the control. Values are the mean ± SD of four independent experiments. ⁎ and ⁎ ⁎ indicate P b 0.01 and P b 0.05, respectively.

Y. Higuchi et al. / Life Sciences 80 (2007) 1856–1864

1859

Fig. 4. Twelve-lipoxygenase activity and effect of lipoxygenase inhibitors on cytotoxicity in C6 cells treated with glutamate. C6 cells were treated with or without 10 mM glutamate in the presence or absence of 50 μM AA, 100 μM α-Toc, or 10 μM NDGA for 10 h. After treatment, the activity (A) and the western blot amount (B) of 12-lipoxygenase were determined. Densitometrical analysis of 12–lipoxygenase was expressed as percent of the control. (C) C6 cells were treated with 10 mM glutamate in the presence (closed column) or absence (open column) of 50 μM AA with 10 μM NDGA, 50 μM baicalein, 10 μM ETYA or 50 μM indomethacin for 12 h. Specific 51Cr release (%) in the cells was then analyzed. Values are the mean ± SD of three independent experiments. ⁎ and ⁎ ⁎ indicate P b 0.01 and P b 0.05, respectively.

RNase A. DNA fragments obtained by this method were loaded into a 1.5% (w/v) agarose horizontal gel. Electrophoresis was performed in TBE buffer (pH 8.0) at 15 V/cm. The gel was visualized by staining with ethidium bromide and was photographed.

with aspectrophotometer. The amount of TBARS was calculated according to the molar absorption coefficient of TBA-reacted malon dialdehyde (MDA), ϵ = 1.56 × 105 M− 1 cm− 1 at 535 nm.

Determination of ROS or lipid hydroperoxides and lipid peroxidation

12-Lipoxygenase was measured by the cell lysis assay of Kawajiri et al. (Kawajiri et al., 1997). Briefly, cell lysate was prepared by sonicating the cells in 25 mM Tris–HCl (pH 7.4) under appropriate conditions. The enzyme reaction was initiated by the addition of 50 μM AA. After 15 min at 30 °C, the mixture was stopped by addition of 50 mM citric acid and then the reaction mixture was extracted with an appropriate volume of diethyl ether. The extract in the ether phase was dried by purging with nitrogen gas and analyzed for 12s-hydroxyeicosa-tetraenoic acid (12s-HETE) by reverse phase HPLC using a Tosoh ODS-120Ts column (4.6 mm × 25 cm).

Oxidative products such as hydrogen peroxide and lipid hydroperoxides were determined using DCFH-DA, which produces green fluorescence by oxidation after deacetylation in a cell (Royall and Ischiropoulos, 1993). Ten μM DCFH-DA was added to the culture of C6 glioma cells with the appropriate volume of HBSS, and the cells were further incubated at 37 °C for 30 min. The cells were harvested, washed with HBSS and resuspended in PBS. The cell suspension was measured at 485 nm excitation and 535 nm emission using a fluorescence spectrophotometer. Hydroperoxide accumulation in the cells was calculated from a DCF standard curve. Lipid peroxidation was quantified by determining 2-thiobarbituric acid reactive substance (TBARS) formation according to the method described by Buege and Aust (Buege and Aust, 1978). Harvested C6 cells (0.5–1 mg protein, corresponding to approximately 5 × 10 6 cells) were suspended in PBS. 2Thiobarbituric acid (TBA) reagent consisting of 0.375% TBA, 15% trichloroacetic acid, and then 0.25 N HCl was added to the cell suspension. The mixture of cell suspension and TBA reagent was heated at 100 °C for 20 min, chilled quickly and centrifuged at 1500 ×g for 10 min. The supernatant was measured at 535 nm

Lipoxygenase assay

Caspase-3 assay The harvested cell pellet (3 × 106 cells) was resuspended in the cell lysis buffer provided by MBL (Nagoya, Japan) that consisted of 10 mM Hepes-KOH (pH 7.4), 2 mM EDTA, 1% Nonident P-40 and 1 mM phenylmethylsulfonyl fluoride, and kept on ice for 10 min. The cell lysate was centrifuged at 12,000 ×g for 2 min and the supernatant was mixed with 2× reaction buffer consisting of 10 mM dithiothreitol, 20% glycerol and 50 mM Hepes-KOH (pH 7.4). The sample was mixed with DEVD-pNA (Asp-Glu-Val-Asp-p-nitroanilide)

1860

Y. Higuchi et al. / Life Sciences 80 (2007) 1856–1864

Western blot analysis Western blot analysis was performed on 40 μg whole-cell extracts. The cell extract was prepared by lysing cells in a buffer consisted of 1% Nonident P-40, 0.1% sodium dodecyl sulfate (SDS), 50 mM Tris–HCl (pH 8.0), 50 mM NaCl, 0.05% deoxycholate and some protease inhibitors. Proteins were sizefractionated by SDS-PAGE on a 10% polyacryl amide gel and transferred onto a polyvinylidene difluoride membrane (0.45 μm, Millipore) by wet electroblotting. The blotted membrane was blocked in TPBS (PBS supplemented with 0.1% Tween 20) that contained 5% non-fat dry milk, and then was probed with the polyclonal rabbit antibodies, followed by horseradish peroxidase-conjugated secondary goat anti-rabbit

Fig. 5. Effect of AA on cytochrome c release into cytosol and caspase-3 activity and the effect of specific inhibitors for caspase-3 on cell death induced by glutamate. C6 cells were treated with or without 10 mM glutamate in the presence or absence of 50 μM AA with or without 100 μM α-Toc or 10 μM NDGA for 12 h (open column) or 24 h (closed column). After treatment, the amount (A, for 12 h) of caspase-3 and cytochrome c and the enzyme activity (B) of caspase-3 were determined and the activity was expressed as optical density (OD) at 405 nm. (C) C6 cells were treated with 10 mM glutamate in the presence (closed column) or absence (open column) of 50 μM AA with 20 μM Ac-DEVD-CHO, 10 μM TPCK or 10 μM DCI for 12 h. Specific 51Cr release (%) for cytolysis in the treated cells was determined. Values are the mean ± SD of four independent experiments.

substrate incubated at 37 °C for 2 h, and the p-nitroanilide that was produced in the mixture was measured at 405 nm with a spectrophotometer.

Fig. 6. Effect of AA on the decrease in NAD and ATP levels and in mitochondrial membrane potentials of cells treated with glutamate. C6 cells were treated with or without 10 mM glutamate in the presence or absence of 50 μM AA, with 100 μM α-Toc or 10 μM NDGA. Cellular NAD (A) levels, ATP levels (B) and TPP+ amounts (C) were analyzed at 6 h (open column) or 12 h (closed column) of the treatment. Values are the mean±SD of three independent experiments.

Y. Higuchi et al. / Life Sciences 80 (2007) 1856–1864

IgG antibody. Antibody binding protein bands were detected by using enhanced chemiluminescence reagent (ECL, Amersham, Tokyo, Japan) and X-ray film.

1861

eluent B consisted of 60% eluent A and 40% methanol. UV absorbance was monitored at 260 nm. Determination of cell membrane integrity

Determination of mitochondrial membrane potentials Mitochondrial membrane potentials were assessed by the retention of TPP+ (Shimizu et al., 1996). Briefly, cells grown in a dish were loaded with [3 H]-TPP+ (0.025 μCi/ml) for 60 min, washed twice with medium and treated for the indicated time. After washing with PBS several times, the cells were trapped on a Millipore cellulose membrane (HA, Bedford, MA). 3 Hactivity on the membrane was measured with a scintillation counter. To correct for the nonspecific binding of [3 H]-TPP + , the value from 0.1% Triton-X100-treated cells for 5 min was subtracted from each value. Determination of intracellular nucleotides The prepared cell pellet (2 × 10 6 ) was suspended in appropriate volume of 0.5 M HClO4 and then vortexed for 5 min. The suspension was centrifuged at 12,000 ×g for 10 min and the supernatant was neutralized with 2 M KOH in 0.66 M potassium phosphate buffer (pH 7.5). The precipitates were removed by centrifugation at 12,000 ×g for 5 min, and the supernatant was applied to an HPLC separating system that consisted of a UV detector and equipped with a Tosoh ODS80T column (4.6 × 250 mm, Tokyo, Japan). The gradient protocol for separation of NAD and ATP was performed as follows: eluent A consisted of 0.1 M KH2PO4 (pH 5.95) and

Fig. 7. Effect of AA on the membrane integrity in C6 cells treated with glutamate. C6 cells were treated with 10 mM glutamate in the presence or absence of 50 μM AA with 100 μM α-Toc or 10 μM NDGA for 6 h (open column) or 12 h (closed column). After incubation, PI uptake of these sample cells was determined and expressed as fluorescent intensity/mg protein of the cells. Values are the mean ± SD of four independent experiments.

Propidium iodide (PI) was used to assess the membrane integrity of C6 (Bevensee et al., 1995). The cells were harvested, washed in PBS, resuspended in DMEM (FCS free medium) containing 15 μM PI and incubated at 37 °C. The cells treated with PI were washed in PBS several times and resuspended in appropriate volume of cell lysis buffer. The PI amount in the cell suspension was measured with a fluorometer at 485 nm for excitation and at 590 nm for emission wavelength using an analysis system with Fluoro Scan Ancent FL. Statistical analysis Results were analyzed using the Student's t test. A difference was considered significant when P b 0.05. Results AA promotes cell death under GSH depletion induced by glutamate The GSH content in C6 cells decreased after treatment with 10 mM glutamate. The GSH level in the cells treated with glutamate was 1.0 nmol/mg protein, and corresponded to approximately 7% of the control levels at 9 h. The T1/2 (half life time) of GSH depletion was estimated to be approximately 3 h after exposure to glutamate (Fig. 1A). AA promoted slightly the GSH-reduction that was induced by glutamate. In this study, we used α-Toc and NDGA as an antioxidant and a lipoxygenase inhibitor. α-Toc is a hydrophobic antioxidant with the ability to scavenge lipophilic radicals and peroxyl radicals within membrane (Halliwell and Gutteridge, 1989). Nordihydroguaiaretic acid (NDGA), a natural antioxidant, inhibits lipoxygenase, an enzyme involved in the metabolism of AA (Bokoch and Reed, 1981). Both α-Toc and NDGA did not suppress the GSH-reduction induced by glutamate in the presence of AA. NAC suppressed completely the glutamateinduced GSH-reduction, even if the presence of AA. NAC is converted to cysteine, which is a precursor for GSH by deacylation and thereby this maintaines intracellular GSH level (Issels et al., 1988). The cytotoxic effect of glutamate on C6 cells was examined by a 51Cr release assay (Fig. 1B). A significant 51 Cr release from the 51 Cr-labeled cells was observed 9 h after exposure to 10 mM glutamate, and was approximately 58% and 80% at 12 h and 24 h, respectively. AA enhanced 51Cr release caused by glutamate by shortening the lag-times of 51Cr release. All of α-Toc, NDGA and NAC suppressed the 51Cr release. AA alone (50 μM) did not affect the intracellular GSH basal levels and not induce any cytotoxic effect in C6 cells. Effect of AA on the GSH levels and the cell death (51Cr release) of C6 cells treated with or without glutamate was determined. AA promoted the glutamate-induced GSH-reduction and thereby 51Cr release in C6 cells in dose-

1862

Y. Higuchi et al. / Life Sciences 80 (2007) 1856–1864

response manner. AA alone at the concentration of 10 μM or 50 μM except at 250 μM did not affect GSH levels at 6 h (Fig. 1C) and cell viability at 9 h (Fig. 1D) after the incubation. AA suppresses internucleosomal DNA fragmentation induced by glutamate DNA fragmentation and aggregation in nuclei of apoptotic cells under GSH depletion induced by glutamate were examined by AGE analysis. Internucleosomal (ladder-forming) DNA fragments were observed in C6 cells treated with glutamate. AA at concentrations of 10 μM and 50 μM suppressed the internucleosomal DNA fragmentation in C6 cells treated with glutamate in a dose-response manner (Fig. 2). NDGA inhibited the AA-induced suppression of ladder-forming DNA fragmentation. Alpha-Toc suppressed the internucleosomal DNA fragmentation induced by glutamate, even if AA was present. AA enhances accumulation of ROS or hydroperoxides and promotes lipid peroxidation under GSH depletion induced by glutamate Intracellular oxidative products such as ROS and lipid hydroperoxides under GSH depletion induced by glutamate were determined using DCFH reagent and the TBARS assay. The amount of DCFH-reactive substances (oxidized DCF) in C6 cells treated with glutamate for 12 h were approximately 3.2-fold of the control (Fig. 3A). AA approximately doubled the amount of oxidized DCF in C6 cells treated with glutamate. In contrast, the DCF enhancement by glutamate was inhibited by α-Toc (Fig. 3A). The amount of TBARS in C6 cells was increased and was approximately 2.2-fold of the control by exposure to glutamate for 12 h (Fig. 3B). AA enhanced TBARS formation in glutamate-treated C6 cells. NDGA and α-Toc strongly suppressed the TBARS formation. AA alone did not affect TBARS formation in the control C6 cells. Enhancement of 12-lipoxygenase during C6 cell death caused by glutamate The enzyme activity of 12-lipoxygenase was determined in C6 cells. The activity of 12-lipoxygenase in C6 cells treated with glutamate alone was increased but was not significantly (Fig. 4A). AA affected the activity of 12lipoxygenase by approximately 2 fold in glutamate-treated C6 cells. The protein amount of 12-ipoxygenase in C6 cells was increased by treatment with glutamate (approximately 50%) and was increased more by addition of AA to the glutamate treatment system (Fig. 4A). Both NDGA and αToc suppressed the induction and the enzyme activity of 12lipoxygenase caused by AA. Several lipoxygenase inhibitors such as NDGA and baicalein inhibited the glutamate-caused 51 Cr release, even in the presence of AA. ETYA, a competitive inhibitor, also inhibited the cytolysis but weakly in the presence of at higher concentration than that of ETYA. In contrast, indomethacin, a cyclooxygenase inhibitor, did not inhibit the glutamate-caused 51 Cr release (Fig. 4B).

AA reduces the substances involving in apoptosis caused by glutamate Cytochrome c release into cytosol from mitochondria of C6 cells treated with glutamate was determined (Fig. 5A). The cytochrome c in cytosol of C6 cells treated with glutamate was remarkably reduced in the presence of AA but was not changed together with α-Toc, even if the presence of AA. NDGA suppressed the reductions of both cytochrome c release into cytosol from mitochondria and caspase-3 induction caused by glutamate in the presense of AA (Fig. 5A). To determine whether the cell death caused by glutamate is related to apoptosis, we examined caspase-3 enzyme activity (DEVDase) in C6 cells and the effect of a specific inhibitor for caspase-3 on cell death induced by glutamate. Caspase-3 activity did not significantly increase in C6 cells treated with glutamate, compared with the control at both 12 and 24 h after the incubations (Fig. 5B). Caspase-3 activity was significantly reduced by addition of AA. In cells treated with glutamate, the amount of pro-caspase-3 protein (32 kDa) was slightly increased but was markedly decreased by addition of AA (Fig. 4A). Whereas, activated caspase-3 protein (17 kDa) was not detectable in western analysis of C6 cells that were treated with glutamate in both the presence and the absence of AA. Ac-DEVD-CHO, a cell permeable tetra peptide inhibitor specific for caspase-3, also did not significantly inhibit the 51Cr-release from C6 cells induced by glutamate both in the absence and the presence of AA (Fig. 5C). TPCK and DCI, which are inhibitors for serine proteases, prevented the glutamate-induced cell death. DCI also prevented the C6 cell death even in the presence of AA (Fig. 5C). AA reduces mitochondrial membrane potentials, intracellular NAD and ATP levels during cell death caused by glutamate We wanted to determine whether the C6 cell death induced by glutamate was apoptosis or necrosis. Therefore, we examined mitochondrial membrane potentials using the TPP+ releasing assay (Shimizu et al., 1996) that shows mitochondrial membrane potentials associated with mitochondrial membrane dysfunction not only in apoptosis but also in necrosis (Lemaster et al., 1998). In addition, we also measured NAD and ATP levels in C6 cells treated with glutamate. A large decrease of NAD and ATP levels and lipophilic TPP+ activity in C6 cells treated with glutamate for 12 h was observed in the addition of AA (Fig. 6). Although NAD level in C6 cells was not changed by treatment with glutamate, it was reduced by addition of AA (Fig. 6A). ATP level in C6 cells treated with glutamate for 12 h was slightly lower than the control and the level was very low in the presence of AA (Fig. 6B). TPP+ release was observed in C6 cells treated with glutamate for 12 h and its release was increased in the presence of AA (Fig. 6C). α-Toc and NDGA suppressed the reduction of both the NAD (Fig. 6A) and the ATP levels (Fig. 6B) and recovered the TPP+ reduction to the control levels with time-dependence (Fig. 6C) in C6 cells treated with glutamate in the presence of AA. AA alone did not affect the NAD and ATP levels and the TPP+ release in the control cells.

Y. Higuchi et al. / Life Sciences 80 (2007) 1856–1864

AA enhances plasma membrane integrity loss induced by glutamate PI is a widely used indicator of membrane integrity (Bevensee et al., 1995). In attempt to quantify cytoplasmic membrane integrity damaged by lipid peroxidation, we monitored the uptake of PI by incubating the cells with PI after being treated with glutamate (Fig. 7). The C6 cells treated with glutamate had a PI uptake of approximately twice that of controls. AA enhanced the PI uptake by C6 cells treated with glutamate. Both α-Toc and NDGA suppressed the AA-activated PI uptake by C6 cells treated with glutamate. Discussion AA promoted cell death under glutamate-induced GSH depletion in C6 cells by enhancing intracellular hydrogen peroxide or lipid hydroperoxides (Figs. 1 and 3). The results suggest an involvement of lipid peroxidation initiated by enzyme reactions of lipoxygenases or by hydroxyl radicals in cell death induced by glutamate-induced GSH depletion. In this study, both the enzyme activity and the amount of 12lipoxygenase in the glutamate-treated C6 cells were increased by addition of AA (Fig. 4), suggesting that the hydroperoxides metabolized from AA by 12-lipoxygenase were involved as an initiator in lipid peroxidation. Twelve-hydroperoxyeicosatetraenoic acid (12-HPETE), a lipid hydroperoxide and metabolic product of 12-lipoxygenase, may amplify lipid peroxidation through a chemical chain reaction under GSH depletion, even if the metabolite is at very low levels (Halliwell and Gutteridge, 1989). Li et al. (Li et al., 1997) have reported that 12lipoxygenase was enhanced and played a role in nerve cell death in primary neuronal cells and in HT-22 hippocampus-derived cells under GSH depletion. AA may play a role for modulation of cell survival in glial cell death caused by oxidative stress under cellular GSH depletion. The activation of caspase-3 is an early biochemical marker of general apoptosis in certain types of cells induced by various triggers for apoptosis. Glutamate induced weak but no significant caspase-3 activation in C6 cells (Fig. 5). AcDEVD-CHO did not significantly inhibit the glutamate-induced cell death both in the presence and the absence of AA, wherease, serine protease inhibitors such as TPCK and DCI inhibited glutamate-induced cell death (Fig. 5). Zhivotovsky et al. (Zhivotovsky et al., 1996) have reported on the inhibitory activity of TPCK for cell degradation accompanying chromatin fragmentation. These results imply that some serine proteases might contribute to the GSH depletion-induced C6 cell death. AA reduced caspase-3 activity in glutamate-treated C6 cells, in which caspase-3 might leak out under the low cell membrane integrity. AA attenuated internucleosomal DNA fragmentation under the glutamate-induced GSH depletion (Fig. 2). Armstrong et al. (Armstrong et al., 1997) have reported on the activation of caspase-3 in cerebellar granule cells undergoing apoptosis but not necrosis. AA may promote lipid peroxidation leading to reduction of mitochondrial membrane potentials and to plasma membrane

1863

integrity loss and thereby convert apoptosis to necrosis, which is associated with rapid and intensive cell death through NAD and ATP depletion and could be suppressed by antioxidants and lipoxygenase inhibitors such as α-Toc and NDGA. Internucleosomal DNA fragmentation is associated with not only caspases and endonucleases but also intracellular ATP levels or the ATP synthesis system (Kass et al., 1996; Leoeur, 2002). Furthermore, intracellular ATP can convert necrosis to apoptosis in oxidantinduced endothelial cells (Eguchi et al., 1997). Taken together, it is likely that the glutamate-induced GSH depletion causes apoptosis, activating proteinases such as serine proteinase and the related endonucleases associated with internucleosomal DNA fragmentation. The accumulation of AA or its metabolites may lead apoptotic cell death to necrosis, and diminish caspase3, which appears to be inactivated or leaked out from the plasma membrane together with ATP and NAD. Some phospholipases are activated during necrosis, particularly cytosolic Ca2+-dependent PLA2 (Cummings et al., 2000). PLA2 is specific to substrate with AA at the sn-2 position. Lysophosphatidic acid, which is produced from phospholipids by removal of AA, also induces both apoptosis and necrosis in hippocampal neurons by unknown mechanisms (Holtsberg et al., 1998). Mild and severe insults with nitric oxide/superoxide induce two distinct events by apoptosis and necrosis in cortical cell cultures (Bonfoco et al., 1995). Alpha-Toc, a chain-breaking antioxidant, suppressed production of DCF reactive substances and lipid peroxidation in glutamate-treated C6 cells, even in the presence of AA (Fig. 4). The change of apoptosis to necrosis by diminishing internucleosomal DNA fragmentation might be dependent on the intensity of oxidative stress or lipid peroxidation. Our study in this model experimental system suggests that lipid peroxidation that is associated with GSH and AA levels may control the necrotic cell death of neuronal cells. The results from exogenously added AA indicates the possible mechanism for conversion of apoptosis to necrosis under GSH depletion through induction of lipid peroxidation. Therefore antioxidants, including 12-lipoxygenase inhibitors capable of inhibiting lipid peroxidation, could be useful for suppression of glial cell death caused by oxidative stress. Acknowledgments This work was supported in part by a Grant-in-Aid for Scientific Research (C10680580 and C15590268) from the Ministry of Education, Culture, Sports, Science and Technology of Japan and the Hokkoku Foundation for Cancer Research in Japan. References Armstrong, R.C., Aja, T.J., Hoang, K.D., Gaur, S., Bai, X., Alnemri, E.S., Litwack, G., Karanewsky, D.S., Fritz, C., Tomaselli, K.J., 1997. Activation of the CED3/ICE-related protease CPP32 in cerebellar granule neurons undergoing apoptosis but not necrosis. Journal of Neuroscience 17, 553–562. Bevensee, M.O., Schwiening, C.J., Boron, W.F., 1995. Use of BCECF and propidium iodide to assess membrane integrity of acutely isolated CA1 neurons from rat hippocampus. Journal of Neuroscience Methods 58, 61–75.

1864

Y. Higuchi et al. / Life Sciences 80 (2007) 1856–1864

Bokoch, G.M., Reed, P.W., 1981. Evidence for inhibition of leukotriene A4 synthesisby 5,8,11,14-eicosatetraynoic acid in guinea pig polymorphonuclear leukocytes. Journal of Biological Chemistry 256, 4156–4159. Bonfoco, E., Krainc, D., Ankarcrona, M., Nicotera, P., Lipton, S.A., 1995. Apoptosis and Necrosis: two distinct events induced, respectively, by mild and intense insults with NMDA or nitric oxide/superoxide in cortical cell cultures. Proceedings of the National Academy of Sciences of the United States of America 92, 7162–7166. Buege, J.A., Aust, S.D., 1978. Microsomal lipid peroxidation. Methods in Enzymology 52, 302–310. Campagne, M.L., Lucassen, P.J., Vermeulen, J.P., Balazs, R., 1995. NMDA and kainate induce internucleosomal DNA cleavage associated with both apoptotic and necrotic cell death in the neonatal rat brain. European Journal of Neuroscience 7, 1627–1640. Carson, D.A., Ribeiro, J.M., 1993. Apoptosis and disease. Lancet 341, 1251–1254. Cummings, B.S., Mchowat, J., Schnellmann, R.G., 2000. Phospholipase A2 in cell injury and death. Journal of Pharmacology and Experimental Therapy 294, 793–799. Cyle, J.T., Puttfarcken, P., 1993. Oxidative stress, glutamate and neurodegenerative disorders. Science 262, 689–695. Eguchi, Y., Shimizu, S., Tsujimoto, Y., 1997. Intracellular ATP levels determine cell death fate by apoptosis or necrosis. Cancer Research 57, 1835–1840. Halliwell, B., Gutteridge, J.M.C., 1989. Lipid peroxidation: a radical chain reaction. In: Halliwell, B., Gutteridge, J.M.C. (Eds.), Free Radicals in Biology and Medicine. Clarendon Press, Oxford, pp. 188–276. Higuchi, Y., 1996. Antitumor effect of Streptococcus pyogenes by inducing hydrogen peroxide production. Japanese Journal of Cancer Research 87, 1271–1279. Higuchi, Y., 2002. Measurement of DNA double-strand breaks with giant DNA and high molecular weight DNA fragments by pulsed-field gel electrophoresis. Methods in Molecular Biology 186, 161–170. Holtsberg, F.W., Steiner, M.R., Keller, J.N., Mark, R.J., Mattson, M.P., Steiner, S.M., 1998. Lysophosphatidic acid induces necrosis and apoptosis in hippocampal neurons. Journal of Neurochemistry 70, 66–76. Ishyama, M., Watanabe, T., Ueda, N., Tsukamoto, H., Watanabe, K., 1993. Arachidonate 12-lipoxygenase is localized in neurons, glial cells and endothelial cells of canine brain. Journal of Histochemistry and Cytochemistry 41, 111–117. Issels, R.D., Nagele, A., Eckert, K.G., Wilmanns, W., 1988. Promotion of cystine uptake and its utilization for glutathione biosynthesis induced by cysteamine and N-acetyl cysteine. Biochemical Pharmacology 5, 881–888. Kass, G.E.N., Eriksson, J.E., Weis, M., Orrenius, S., Chow, S.C., 1996. Chromatin condensation during apoptosis requires ATP. Biochemical Journal 318, 749–752. Kawajiri, H., Qiao, N., Zhuang, D., Yoshimoto, T., Hagiya, H., Yamamoto, S., Sei, H., Morita, Y., 1997. Diurnal change of arachidonate 12-lipoxygenase in

rat pineal gland. Biochemical Biophysical Research Communication 238, 229–233. Leoeur, H., 2002. Nuclear apoptosis detection by flow cytometry: influence of endogenous endonucleases. Experimental Cell Research 277, 1–14. Lemaster, J.J., Nieminen, A.L., Qian, T., Trost, L.C., Elmore, S.P., Nishimura, Y., Crowe, R.A., Cascio, W., Bradham, C.A., Brenner, D.A., Herman, B., 1998. The mitocondrial permeability transition in cell death: a common mechanism in necrosis, apoptosis and autophagy. Biochimica Biophysica Acta 1336, 177–196. Li, Y., Maher, P., Schubert, D., 1997. A role for 12-lipoxygenase in nerve cell death caused by glutathione depletion. Neuron 19, 453–463. Murakami, M., Kambe, T., Shinbara, S., Kudo, I., 1999. Functional coupling between various phospholipase A2s and cyclooxyg nases in immediate and delayed prostanoid biosynthetic pathways. Journal of Biological Chemistry 274, 3103–3115. Murphy, T.H., Miyamoto, M., Sastre, A., Schnaar, R.L., Coyle, J.T., 1989. Glutamate toxicity in a neuronal cell line involves inhibition of cystine transport leading to oxidative stress. Neuron 2, 1547–1558. Pereicra, C.M., Oliveira, C.R., 1997. Glutamate toxicity on a PC12 cell live involves GSH depletion and oxidative stress. Free Radical Biology and Medicine 23, 637–647. Royall, J.A., Ischiropoulos, H., 1993. Evaluation of 2′, 7′-dichlorofluorescein and dihydro-rhodamine 123 as fluorescent probes for intracellular H2O2 in cultured endothelial cells. Archives of Biochemistry and Biophysics 302, 348–355. Shimizu, S., Eguchi, Y., Kamiike, W., Waguri, S., Uchiyama, Y., Matsuda, H., Tsujimoto, Y., 1996. Bcl-2 blocks loss of mitochondrial membrane potential while ICE inhibitors act at a different step during inhibition of death induced by respiratory chain inhibitors. Oncogene 13, 21–29. Simonian, N.A., Coyle, J.T., 1996. Oxidative stress in neuro-degenerative diseases. Annual Review of Pharmacology and Toxicology 36, 83–106. Spiteller, G., 1996. Enzymic lipid peroxidation-A consequence of cell injury. Free Radical Biology and Medicine 21, 1003–1009. Swartz, L.M., Smith, S.W., Jone, M.E., Osbone, B.A., 1993. Do all programmed cell deaths occur via apoptosis? Proceedings of the National Academy of Sciences of the United States of America 90, 980–984. Tang, D.G., Chen, Y.Q., Honn, K.V., 1996. Arachidonate lipoygenases as essential regulations of cell survival and apoptosis. Proceedings of the National Academy of Sciences of the United States of America 93, 5241–5246. Toyook, T., Uchiyama, S., Saito, Y., Imai, K., 1988. Simultaneous determination of thiols and disulfides by high performance liquid chromatography with fluorescence detection. Analytical Chimica Acta 205, 29–41. Zhivotovsky, B., Burgess, D.H., Orrenius, S., 1996. Proteases in apoptosis. Experientia 52, 968–978.