Neuroscience Vol. 48, No. 4, pp. 903-914, 1992 Printed in Great Britain

0306-4522/92$5.00+ 0.00

Pergamon PressLtd 0 1992IBRO

A MECHANISM FOR GLUTAMATE TOXICITY IN THE C6 GLIOMA CELLS INVOLVING INHIBITION OF CYSTINE UPTAKE LEADING TO GLUTATHIONE DEPLETION S. KATO,*~ K. NEGISHI,* K. MAWATARI~ and C.-H. Ku@

Molecular Neurobiology Group, *Department of Neurophysiology, Neuroinformation Research Institute (NIRI), School of Medicine, IDepartment of Medical Technology, School of Allied Medical Professions and $Department of Pharmacology, Cancer Research Institute, University of Kanazawa, 13-1 Takaramachi, Kanazawa 920, Japan Abstract-We

have demonstrated that addition of L-glutamate in millimolar amounts to a culture of C6 glioma cells induced cell death within 24 h. The glutamate-induced toxicity in the C6 glioma cells was completely suppressed by adding L-cystine (0.4-l .OmM), while the C6 cells degenerated in L-cystine-deprived culture medium. Kinetic studies of [%]cystine and [3H]glutamate uptake showed that cystine competitively inhibited glutamate uptake, and conversely glutamate inhibited cystine uptake competitively, suggesting that C6 cells have a cystine/glutamate antiporter (system CG or XI?) similar to that already described in the periphery. Exogenous cystine (1 mM) stimulated a release of endogenous glutamate from C6 cells in a Na+ -independent Cl--dependent fashion. Thus, the antiporter normally transports glutamate out of and cystine into the cells. With the glutamate analogues tested, there was a good correlation between cytotoxicity and inhibition of cystine uptake. The de nouo synthesis of glutathione was largely dependent upon the uptake of extracellular cystine. Intracellular levels of glutathione were dramatically decreased within S-10 h by culture in glutamate-added or cystine-free medium. Vitamin E (100 PM), an antioxidant, rescued the death of C6 cells induced by glutamate exposure or by culture in cystine-deprived medium, but did not restore the apparent decrease of intracellular glutathione. Taken together, the present data strongly indicate that glutamate-induced cell death is initially due to inhibition of cystine uptake through the antiporter XS: system; such inhibition leads to glutathione depletion exposing the cells to oxidative stress. Excess of extracellular glutamate introduced from endogenous or exogenous roots might disorder this mechanism, resulting in cell death.

L-Glutamate is thought to serve as the major excitatory neurotransmitter in the CNS* and produce its neurophysiological effects by acting at three subtypes of glutamate receptors, each commonly referred to by preferred pharmacological agonists: quisqualate, Nmethyl-D-aspartate (NMDA) and kainate.37 In fact, glutamate excites virtually all central neurons and is present in nerve terminals at millimolar levels.’ Normally, the extracellular levels of glutamate rise to high

but was a feature common to the actions of all excitatory amino acids on central neurons. He postulated therefore that glutamate or related compounds, might be the cause of the neuronal cell loss found in certain degenerative neurological disorders such as

levels only in the brief and spatially localized fashion appropriate to synaptic transmission. Lucas and Newhouse” in 1957 first showed that sustained exposure to glutamate can destroy retinal neurons. In a subsequent set of experiments, 01neyz4s26established that this toxicity, which he later called excitotoxicity, was not unique to glutamate or to retinal neurons,

sorts: acute swelling of neuronal dendrites and cell bodies and a more slowly evolving neuronal degeneration.2s Axons and glia are relatively spared, although high levels of excitatory amino acids can produce some swelling of glia.3*15 Recent studies indicate that quisqualate interacts with Na + -independent Cl - -sensitive glutamate transporter enriched in brain sonicates’* and synaptosomes,4o which was originally misunderstood to be a quisqualate subtype of glutamate receptor;9s38 and L-cystine inhibits glutamate sequestration at this quisqualate-sensitive glutamate transporter.u@ This carrier site may be an expression for the cystine/ glutamate antiporter, designated as system XE, in amino acid transporters as already shown in the peripheral fibroblasts. ‘v2Recently, a possible suggestion of such an antiporter system has been presented

tTo whom correspondence should be addressed. Abbreviations: ABD-F, 4-(amino sulphonyl)-7-fluoro-2,1,3benzoxadiazole; AMPA, (RS)-a-amino-3-hydroxy-5methyl-isoxazole-4-propionic acid; BSO, buthionine sulphoximine; CNQX, 6-cyano-7-nitroquinoxaline-2,3dione; DMEM, Dulbecco’s modified Eaale’s medium; EDTA, ethylenediaminetetra-acetic acid; GSSG, gluta: thione disulohide: HPLC. hiah nerfonnance liauid chromatography; LDH, lactate dehydrogenase; _NMDA, N-methyl-o-aspartate; PBS, phosphate-buffered saline; PCA, perchloric acid; SBD-F, ammonium 7-fluoro2,1,3-benzoxadiazole-4-sulphonate.

Huntington’s disease or ischemic brain attack, as occurs in a stroke with abrupt deprivation of blood ~upply.~~~’The toxic changes produced by glutamate or related excitatory amino acids in vivo are of two

in a neural and glial cell line.4,22 903

904

S. KATO (‘I i/l.

In a previous paper,” we have demonstrated that glutamate (in the mM range) induced a delayed cell death of rat clonal C6 glioma cells in culture. Some anionic amino acids of longer chain length, such as homocysteate, tl -aminoadipate and a -aminopimelate were also toxic to the cells while aspartate, the shorter anionic amino acid, failed to destroy the cells. Only quisqualate, but not NMDA and kainate, among the three classes of glutamate receptor agonist was toxic to the C6 cells.‘2 Moreover, neither competitive nor non-competitive glutamate antagonists blocked the glutamate toxicity. These pharmacological characteristics of glutamate toxicity on C6 glioma cells were quite similar to those of cystine transport inhibition by glutamate or analogues in the peripheral fibroblasts and in the neuronal cell line as mentioned above.‘s22 These data strongly suggest that the glutamate toxicity in the C6 cells is not mediated through glutamate receptors. Therefore, using the C6 glioma cells, we now test this hypothesis of Coyle’s laboratory to our glutamate damage of C6 glioma cells and report a glial expression of the cystine/glutamate antiporter (XC) and a vital role of cystine uptake through this antiporter system for maintaining a cellular level of glutathione (y-glutamylcysteinylglycine). Glutathione is an important substrate against oxidative stress in the living cells. The inhibition of cystine uptake by excess extracellular glutamate ultimately reduces the levels of cellular glutathione, resulting in a delayed cell death. Depletion of cellular glutathione by some drugs also induces cell death as occurs with glutamate exposure, while vitamin E, an antioxidant, powerfully blocks the cell death by glutamate. All of the data obtained can be linked to the glutamate-induced cell death by oxidative stress. A preliminary form of this work has been published elsewhere.” EXPERIMENTAL PROCEDURES Chemicals

Monosodium L-glutamate, L-cystine, L-cysteine, maleic acid diethyl ester and vitamin E were purchased from Wako Pure Chemicals (Osaka). Monosodium L-aspartate and DLG(-aminopimelic acid were purchased from Nakarai Chemicals (Kyoto). DL-Homocysteic acid was purchased from Calbiochem-Behring Corp. (La Jolla). Quisqualic acid, kainic acid, DL-a-aminoadipic acid, L-buthionine (S, R) sulphoximine, glutathione and glutathione disulphide were obtained from Sigma (St Louis). (RS)-a-Amino-3-hydroxyhydrobromide 5-methyl- isoxazole-4-propionic acid (AMPA) was purchased from Research Biochemicals Inc. y-Glutamylcysteine was generously supplied by Dr N. Sakura (School of Pharmacy, Hokuriku University). The structural determination of- j-glutamylcysteine was performed bv FAB MS soectrometrv. L-135S1Cvstine (> 3000 Ci/mmol), L-[35S]methionine (> lb00 Ciin&oi) and L-[3,4-‘Hlglutamic acid (50 Ci/mmol) were purchased from Dupont/NEN Research Products (Boston). Cell cullure C6 glioma cells (1.5 x 106) were plated in 60-mm plastic dishes (Falcon) and cultured at 37°C in a humidified atmosphere of 10% CO, in a standard Dulbecco’s modified

Eagle’s medium (DMEM) containmg 4 mM ol’~,-glu~~~n~inc 5% fetal calf serum. penicillin (100 units/ml) and \trcplomycin (100 pg/ml) as previously described.” Exposure 7-fluoro-2,l,3-benzoxadiazole (ABD-F) for thiols and ammonium 7-fluoro-2,1,3-benzoxadiazole-4- sulphonate (SBD-F) for disulphides as described by Toyooka et al. 33 C6 glioma cells cultured in control or treated with various drugs were collected and suspended in 0.6 N perchloric acid (PCA) solution containing 1 mM of EDTA. The cells were then homogenized with a sonicator (HeatSystem Inc.) for 1 min and centrifuged at 12,000 g for 20 min at 4°C. The resulting supematant was used for measurement of cellular thiols or disulphides. The mixture of thiols and disulphides in the supematant was treated with ABD-F at 60°C for Smin in borate buffer containing 1 mM EDTA (pH 9.3). After the removal of excess non-binding ABD-F by extraction with ethyl acetate, the remaining disulphides in the aqueous phase were treated with SBD-F at 60°C for 20 min in the presence of t&n -butylphosphine, a reducing agent. The ABD-thiols and SBD-thiols thus produced were separated by reversed phase chromatography (Inertsil ODS column, Gasukuro Kogyo) and detected by fluorimetry (380 nm excitation, 510 nm emission). SBD-cysteine (L-cystcine), ABD-cystine (L-cysteine), SBD-glutathione (GSSG), ABD-glutathione (GSH) and ABD-glutamylcysteine (y-glutamylcysteine) were well separated by linear gradient elution for over 30 min from 0.15 M H,PG,/CH,CN (96:4) to 0.15 M H,PG&H,CN (85: 15) followed by an isocratic elution with 0.15 M H,PG,/CH,CN (85 : 15) for 10 min. The PCA insoluble fraction was dissolved in 1 N NaOH and protein concentration was determined by the method of Lowry et al.” Incorporation of [35S]cystine into glutathionefraction Incorporation of [?I]cystine into various thiols such as cysteine, glutathione, y -glutamylcysteine was measured by counting the radioactivity of each peak fraction of thiols separated. Glioma C6 cells (6 x 106) were incubated with 5 PM of [35S]cystine (2 pCi/tube) for various time periods (15 mm-6 h) and the reaction was terminated with ice-cold PBS. The cells were washed twice with PBS and homogenized in 0.6 N PCA containing 1 mM EDTA. After centrifugation thiols in the supematant were then separated using HPLC and the eluate was collected by a fraction collector (FRAC-100, Pharmacia), one per minute. The radioactivity of each fraction was counted using a scintillation spectrometer. Agarose gel electrophoresisfor DNA Genomic DNA was extracted according to a method of Watanabe et a1.36C6 glioma cells (1 x 10’) cultured with or without 10mM of glutamate were collected at various cultivation times and centrifuged at 2,000 r.p.m. for 5 min at 4°C. The precipitate was suspended with lOm1 of 20mM Tris-HCl (DH 8.0)/140 mM NaCl and centrifueed as above. The result&g peiiet was resuspended in 3 ml of 50mM Tris-HCl (pH 8.0)/25 mM EDTA and incubated for 30 min at 4°C after the addition of 120~1 lysozyme (30mg/ml, Sigma). The precipitate was then dissolved by 120 ~1 10% sarcosyl and incubated with 35~1 RNase (lOmg/ml, P-L Biochemicals) for 30 min followed by overnight treatment at 37°C with 40~1 proteinase K (20 mg/ml, Boehringer Mannheim). Genomic DNA was extracted twice by addition of phenol saturated with 20mM Tris-HCl (pH 8.0) containing 140mM NaCl plus chloroform/isoamyl alcohol, and d: alvsed against 10 mM Tris-HCl (DH 7.5)/O.1 mM EDTA (TE) andthen precipitated with ethinol. The precipitate was dried and dissolved in 100-200 ~1 TE. DNAs (0.5 pg) from control and glutamate-treated C6 cells were then loaded onto 0.8% (w/v) agarose horizontal gels. Electrophoresis was performed in x0.5 Tris-acetate

905

(40mM Tris-HCl, 20mM Na-acetate, 2mM EDTA, pH 7.5) buffer at 50 V. Genomic DNAs on the agarose gels were

stained with ethidium bromide and photographed under UV light with Polaroid type 667 film. Hind III digested bacteriophage 1 DNA fragments were used as marker DNAs. RESULTS Suppressive effect of cystine to glutamate-induced cell death in C6 glioma cells

In control culture, 1.5 x lo6 C6 glioma cells were plated in 60-mm dishes and incubated with the standard DMEM. The cells covered the whole surface of the dish as a confluent monolayer which formed two days after culture (Fig. 1A). When 10 mM glutamate was added to the medium at the beginning of the incubation, morphological degeneration of the cells progressed 14-16 h after culture and almost all cells died 24 h after culture, resulting in detachment from the surface of the dish and flotation of cells in the medium (Fig. 1B). The ED,, of glutamate on the cell death was about 4mM.12 The surviving fraction of the glutamate-treated cells on day 2 was less than 0.1% of the untreated cell population. The glutamate toxicity was not attenuated by any glutamate receptor antagonists or uptake inhibitors.‘* As such, we examined the role that the quisqualate-sensitive Cl - -dependent glutamate transport system played in the resulting glutamate damage to C6 cells. Addition of cystine (0.5 mM) with glutamate at the beginning of the culture completely suppressed the glutamate-induced cell death, and the morphology on day 2 was similar to that of a control culture (Fig. 1C). A range of 0.4l.OmM cystine effectively suppressed the glutamate (7SmM)-induced cytotoxicity of C6 cells (Fig. 2). When cystine was removed from the standard DMEM (containing 0.25 mM cystine), C6 cells rapidly degenerated within 12 h even in the absence of glutamate (Fig. 1D). The cell death in the cystinefree medium was more severe and prominent than that induced by glutamate. Cystinejglutamate uptake in the C6 cells The above data clearly showed that addition of cystine rescued C6 cells from the glutamate-induced

death, and deprivation of cystine from the culture medium resulted in a cell death similar to that seen by glutamate exposure. Therefore, we examined cystine uptake by C6 cells. The cystine uptake by C6 cells cultured in standard DMEM was performed at 37°C for 10min in Krebs-Ringer solution with [3SS]cystine (10 PM to 1 mM). The uptake of [3sS]cystine by C6 cells was linearly increased until 15 min and then saturated. It was also dependent upon the number of C6 cells and the temperature of incubation. Lineweaver-Burk analysis of cystine uptake into C6 cells cultured for four days shows that the apparent K,,, and V,,,, for cystine transport were 192 PM and 9.1 nmo1/2.2 x lo6 cells/l0 min, respectively (Fig. 3). Glutamate (Xl-100 FM) clearly inhib-

906

s. KATOet

(21.

Fig. 1. Glioma C6 cells cultured under various conditions. 1.5 x IO’ cells were plated and incubated for 48 h (A, C and E) and 24 h (B, D and F), respectively. (A) Control culture in standard DMEM, (B) with lOmM glutamate; (C) with 1OmM glutamate plus I mM cystine; (D) in cystine-free medium: (E) with 1OmM glutamate plus IOOpM vitamin E; (F) with 0.5 mM buthionine sulphoximine

ited the [3sS]cystine uptake in a competitive fashion (Fig. 3; X; = 80pM). The IC, of glutamate for cystine uptake was less than 100 PM (Table 1). Glutamate receptor agonists or related compounds were then tested for their effect on the cystine uptake system. Of these compounds, DL-hom~ysteic acid, t%a-aminoadipic acid, DL-a -aminopimelic acid and quisqualate competitively inhibited the [35S]cystine uptake. Quisqualate was the most effective and the I& for cystine uptake was 50pM (Table I). In contrast, aspartic acid, L-glutamine, GABA, NMDA,

kainate and AMPA (a powerful agonist for quisqua~ate receptor) produced less than 10% inhibition of cystine uptake. There was a good correlation between inhibition of cystine uptake and cytotoxicity by glutamate analogues tested (Table I; see also Kato et ul. I’). We further examined glutamate uptake by C6 glioma cells in the same way for cystine uptake as shown in Fig. 3. Glutamate uptake was performed with rH]glutamate (iOpM-l mMf at 37°C for I5 min. A Lineweaver-Burk plot of glutamate uptake

907

Glutamate toxicity and glutathione in the C6 glioma cells 100

+-+h with

s8 =

L-Glu

7.5 mM

50

6 S

C L-Cystine

(PM)

Fig. 2. Suppressive effect of additional cystine on glutamate (L-Glu) damage in two-day culture of C6 glioma cells. An addition of cystine (0. l-l .OmM) to the culture medium at beginning completely blocked the glutamate-induced cell death in C6 cells. The C6 cells were proliferated as seen as control culture (cf. Fig. 1C). Each value is the mean of three different experiments with duplicate samples. Vertical bars show SD.

into C6 glioma cells cultured for four days shows that the apparent K,,, and V,,,,, for glutamate transport were 200 PM and 1.0 nmol/2.2 x lo6 cells/l5 min, respectively (Fig. 4). The amount of glutamate uptake was about one-tenth that of cystine (cf. Fig. 3). Cystine competitively inhibited the [3H]glutamate uptake, and just conversely glutamate inhibited the cystine uptake (Fig. 4, K = 176 PM), indicating that one carrier protein is coupled with both amino acids for transport in the C6 cells. Releasable glutamate by cystine in the C6 cells A spontaneous release of glutamate from C6 cells (5-7.5 x 106) incubated in Krebs-Ringer solution without Ca*+ ions for 5 min at 37°C was analysed in a chromatogram (Fig. 5A). The retention time of glutamate eluted was 17.1 + 1.2 min in our. HPLC method. Addition of 1 mM cystine to the C6 cells suspended in normal Krebs-Ringer solution for 5 min stimulated a release of glutamate about twice that of the control (Fig. 5B). Release of other amino

/

I.

0 0.001

0.01 I/S

0.02 IL-Cystlne

0.03 plvl)

Fig. 3. Lineweaver-Burk plots of the rate of [3sS]cystine uptake by C6 glioma cells as a function of exogenous cystine concentrations. The C6 cells were cultured for four days and then collected. Uptake of cystine was measured after IO-mm incubation at 37°C in Krebs-Ringer solution supplemented with 0.01-l mM [“S]cystine (-a-). Glutamate (L-Glu; 50 PM, -O-and 100 PM, -W--) clearly inhibited this cystine uptake in a competitive fashion. Each value is the mean of three different experiments with duplicate samples. Vertical bars show SD.

acids was not significantly evoked by cystine. This releasable glutamate stimulated by cystine was reduced by addition of cystine in a Cl--free but not in a Ca*+-free or Na+-free solution (Fig. 5C). The induction of releasable glutamate by cystine was blocked by the co-existence of 1 mM quisqualate (Fig. 5D). The glutamate release was also temperature-dependent: incubation of C6 cells at 4°C shows only residual glutamate release (data not shown). Incorporation of [‘*S]cystine into glutathione Glutathione plays a major role in the oxidation-reduction state of living cells.*’ The synthesis of glutathione is largely dependent on the cellular level of cysteine. In cell culture system, cysteine released from the cells is easily auto-oxidized to

Table 1. Cytotoxicity and inhibition of [“Slcystine uptake by glutamate or related substances on the glioma C6 cells

Substances L-Glutamic acid L-Glutamine DL-Homocysteic acid GABA DL-a-Aminoadipic acid DL-a-Aminopimelic acid N-methyl-D-aspartic acid Kainic acid Quisqualic acid AMPA

Toxicity ED, 4mM 2mM 4mM 4mM ImM -

Inhibition of [‘%]cystine uptake fC50 0.1 mM 0.08 mM 0.1 mM 0.2 mM 0.05 mM -

C6 glioma cells were cultured for two days with various substances and cytotoxicity was assayed by counting the number of viable cells (Kato et al.“). Cystine uptake experiment was done with [35S]cystine for C6 glioma cells (2.2 x 106). The inhibition of cystine uptake was estimated by counting the radioactivity in the presence or absence of various substances.

S. KATO er ul

908

A

B

> 1

A’

0

0.01

I/S

0.025 (~-Glutamate

005

PM)

Fig. 4. Lineweaver-Burk plots of the rate of [3H]glutamate uptake by C6 glioma cells as a function of exogenous glutamate concentrations. The C6 cells cultured for four days were collected and used in the uptake experiment. Uptake of glutamate was measured for 15-min incubation at 37°C in Krebs-Ringer solution supplemented with 0.01-l mM [‘HIglutamate (-•-). Cystine (L-Cys; 200 PM, -_O-) competitively inhibited the glutamate uptake. Each value is the mean of three different experiments with duplicate samples. Vertical bars show S.D.

cystine in the medium.’ Therefore, cystine uptake is thought to be a rate-limiting process in glutathione synthesis. A simultaneous separation of cystine, cysteine, y-glutamylcysteine, glutathione or GSSG was performed by an HPLC method with SH group specific fluorigenic reagents, ABD-F and SBD-F. The retention times in our HPLC system were as follows: cystine, 4.0 + 0.2 min; cysteine, 5.5 f 0.2 min; GSSG, 6.7 &-0.1 min; glutathione, 11.9 + 0.1 min; y -Glutamylcysteine, 13.3 + 0.1 min. C6 ghoma cells (6 x 106) were incubated with 1OpM [3SS]cystine (2 pCi/tube) for 15 min-6 h, and the cells were pelleted and suspended in PCA. After deproteination, incorporation of [35S]cystine into cysteine, y -glutamylcysteine and glutathione in the PCA-soluble fractions were measured by counting the radioactivity in each peak fraction separated with HPLC method. The ratio of r5S]cystine into cysteine and y-glutamylcysteine was very low and was maximal after 60 min incubation followed by the low constant level (Fig. 6A). In contrast, the ratio of [35S]cystine into glutathione was very high and increased linearly until 2-h-incubation (Fig. 6A). The elution profile of cystine, cysteine, glutathione and y -glutamylcysteine labeled with [35S]cystine for 4 h is shown in Fig. 6B. The largest peak (retention time = 4 min) was that of cystine taken up into the C6 cells. The incorporation of [3sS]cystine measured in PCA-insoluble fraction is also shown in Fig. 6A (a dashed line). The ratio (radioactivity) into the PCA insoluble fraction was 3-4 times higher than that into glutathione. A similar

0

10

20

0

10

20min

Fig. 5. Cystine-induced glutamate release from C6 cells. C6 cells (5 7.5 x 106) cultured for four days were collected and incubated with various agents for 5min at 37°C. After centrifugation, the resulting supematant was loaded directly to HPLC and amino acids were separated by ion-exchange chromatography. (A) Incubation with normal Krebs-Ringer solution. Tau. taurine; Asp, aspartic acid; Thr, threonine; Ser. serine; Glu, ghrtamic acid. (B) With 1 mM cystine in normal Ringer solution. (C) With I mM cystine in a Cl -free Krebs-Ringer solution. (D) With 1 mM cystine plus I mM quisqualic acid (@is). Retention time of glutamate eluted was 17.I f I .2 min. Cystine clearly induced an endogenous glutamate release from C6 cells in a Cl--dependent manner (B, C). *P < 0.01 compared with control. This cystine-induced glutamate release was blocked by co-existence of quisqualate (D). Experiments were repeated quadricated and a representative chromatogram was shown. **P < 0.01 compared with B. Scale shows 250 pmol for glutamate.

experiment of [3sS)methionine incorporation into glutathione was also performed in C6 cells. The ratio of conversion into glutathione from [‘5S]methionine was less than one tenth of that with [35S]cystine (data not shown). Reduction qfter

of cellular

treatment

glutathione with glutamate

level.~ in the C6 cell.~

We further examined effects of glutamate exposure on the cellular levels of glutathione and GSSG in the C6 cells. A marked decrease in cellular levels of glutathione and GSSG by treatment with 1OmM glutamate is illustrated in Fig. 7. After 4 h incubation, the cellular glutathione gradually decreased and a significant decrease was obtained at 8-10 h after exposure. It reached a level of about 1.5 pmol; pg protein (one tenth of control) at 24 h after exposure. The GSSG content transiently increased at 2- 3 h culture and then gradually decreased to an almost undetectable level at 24 h (Fig. 7). The total protein

909

Glutamate toxicity and glutathione in the C6 glioma cells

B

0

05

20

4.0

IO

60

Culture (h )

20

30

Fraction number

Fig. 6. Incorporation of [35S]cystineinto glutathione (GSH) in the C6 cells. C6 cells (6 x 106)cultured for four days were collected and incubated with 10 PM of [s5S]cystine at 37°C for 15 min-6 h. The cells were washed twice with cold PBS, suspended with 0.6 N PCA and sonicated for 1 min. After centrifugation, the PCA-soluble fraction was loaded to HPLC with a reversed phase column. Thiols in the supernatant from C6 cells incubated for various time-periods were separated. Incorporation of [35S]cystineinto each thiol was estimated by measuring radioactivities of peak fractions separated by HPLC method. (A) A time-course of [35S]cystine incorporation into thiols (GSH, -@--; cysteine, -_O--; y -glutamylcysteine, --A--). Incorporations into PCA insoluble fraction of C6 cells were also shown in a dashed line (--n---). (B) An elution profile of thiols in the PCA soluble fraction from C6 cells incubated with [35S]cystine for 4 h. The peak of cystine was confirmed by using authentic [35S]cystine. content was not significantly affected during this period (S-10 h), although it was slightly reduced at 24 h after glutamate exposure (Fig. 7; dashed line). A

200

dose-response curve for glutamate’s effect on cellular glutathione levels by 24-h treatment is shown in Fig. 8. The I&, of glutamate for reduction of glutathione levels was about 4 mM. The value is the same as seen in the case of glutamate toxicity (see Table 1). Effects of other compounds on GSH content in the C6 glioma cells are also summarized in Table 2.

150

100

50

0

Duration of exposure (IO mM

L-

Glu;h)

Fig. 7. Reduction of cellular levels of glutathione (GSH) after treatment by glutamate (L-Glu). C6 cells were cultured with or without 1OmM glutamate and collected at the time-periods indicated. Cellular levels of glutathione and GSSG were measured by the HPLC method. The levels of glutathione (-•--) were dramatically decreased by this 8-10-h glutamate treatment. GSSG (-_O-) was also much decreased by 8-10-h incubation. In contrast, protein concentrations were not so decreased by the 8-10-h glutamate exposure (- -w- -), although a slight decrease was seen at 24 h after exposure. Each value is the mean of five different experiments. Vertical bars show SD.

I 0

5

IO

15

L- Glutamate (mM)

Fig. 8. A dose-response curve of glutamate in reduction of cellular level of glutathione (GSH). Cellular glutathione in C6 cells cultured with various concentrations (1-15 mM) of glutamate for 24 h was measured by the HPLC method. Glutamate clearly induced a decrease in cellular glutathione level in a dose-dependent fashion. The IC, was about 4mM. Bach value is the mean of three to five different experiments. SD. value is less than 10% of the corresponding mean.

910 Table 2. Cellular levels of glutathione in the C6 and T24 cells after varrous treatments Cellular glutathione levels 24 h after treatment (IO mM) +cystine (I mM) tvitamin E (IOOpM) Quisqualate (4 mM) NMDA (IOmM) Kainate (IO mM) Cystine-free medium L-Buthionine sulphoximine (1 mM)

Control Glutamate (10 mM)

15.1 k 3.6 13.4 * I.8

Glutamate

C6 cells

T24 cells

--...-

15.4 * 2. I pmol/ I .5 * 0.x 14.2 & 1.x I.4 &0.X I.1 i_o.7 15.0 * 2.0 14.8 * I.7 0 2.4 + 1.o

Control

j~p protem

C6 glioma cells or T24 cells were cultured for 24 h with various substances and cellular levels of GSH were measured by an HPLC method with fluorigenic reagents (see Experimental Procedures). Each value is the mean + S.D. of three to five determinations.

Glutathione

modifVing agents to the C6 cells

oxidant or reducing agents on the C6 cells treated

Exposure of buthionine sulfoximine (BSO, 0.5 mM), an inhibitor of glutathione synthesis,” in-

duced a delayed cell death of C6 cells as seen after glutamate treatment (Fig. 1F). The relationship between C6 cell death and BSO concentration is shown in Fig. 9. More than 0.5 mM BSO is lethal to the cell. By the co-existence of BSO and glutamate, the cell death was further accelerated and the C6 cells degenerated within 10-11 h after exposure of them (data not shown). Any concentrations of cystine tested did not suppress the BSO-induced cell death. This BSO treatment of C6 cells induced a marked reduction of glutathione content at 24 h after incubation (Table 2). Addition of maleic acid diethyl ester (OSmM), a depleting agent of glutathione2’ also caused the death of C6 cells within 12-16 h (data not shown). Vitamin E protects duced toxicity

the C6 cells against glutamate-in

with glutamate (10 mM). Of these compounds, vitamin E (100 p M) protected the C6 cells against glutamate-induced cell death (Fig. 1E), but did not prevent the reduction of glutathione content by glutamate exposure (Table 2). The protective effect of vitamin E was ascertained by the measurement of releasable LDH into the culture medium (Fig. 10). A significant release of LDH was evoked 13-14 h after glutamate exposure. The release was gradually increased thereafter. The addition of vitamin E clearly blocked the release of LDH even after two-day culture, as in the control culture. The addition of vitamin E was effec-

tive not only at the beginning of the culture but also at 10-12 h after glutamate exposure (data not shown). Two reducing agents, glutathione and cysteinc (more than 1 mM) did not protect against glutamate toxicity.

-

In order to see whether glutamate toxicity in the C6 cells was primarily caused by depletion of glutathione itself or, secondarily, caused by exposing the cells to oxidative stress. we examined effects of an anti100

-o-Control + Glu --•-- Glu + vitamin E

.

t

r;i

0

6

12

18

24

x)

36

Time of exposure (h)

L-Buthionine-&RI-sulphoximine

(+M)

Fig. 9. Buthionine sulphoximine (BSO)-induced cell death in the C6 glioma cells. Addition of BSG (0.1-W mM) to the culture medium induced a cell death just as seen after glutamate exposure (cf. Fig. IB and 1F). Each value is the mean of three different experiments with duplicate samples. S.D. value is less than 10% of the corresponding mean.

Fig. IO. LDH release by glutamate and protection by vitamin E. C6 cells were cultured in standard DMEM (-_O-), with IOmM glutamate (-•--) or with IOmM glutamate plus 100 PM vitamin E (- -O- -) and the supernatant of each culture condition was collected at the tiwPeriods indicated (O-36 h). LDH activity was expressed by measuring the reduction of NADH (AA”/3 rain Per 20 rll). Glutamate evoked a signifiamt release of LDH at 13-14 h after exposure. In contrast. addition of vitamin E ctearly protects against the release of LDH into the culture medium (cytotoxicity) induced by glutamate. F&h value is the mean of three different experiments with duplicate samples. Vertical bars show SD.

Glutamate toxicity and glutathione in the C6 glioma cells

DNA fragmentations by glutamate treatment In the previous paper,‘* we reported decompositions of cellular DNA in glutamate-treated C6 cells by flow cytometric analysis. In the present study, DNA damage of C6 glioma cells caused by glutamate (10 mM) was examined by analysis of genomic DNA prepared from the cells using 0.8% agarose gel electrophoresis. Genomic DNAs from glutamatetreated C6 glioma cells for 24 h show DNA fragmentation as smear bands of 4-20 kbp on a 0.8% agarose gel (Fig. 11; Glu), whereas genomic DNAs from control C6 cells show a single band of more than 20 kbp (Fig. 11; C6). The DNA fragmentation by glutamate was also protected with vitamin E treatment (data not shown). Lack of effect of glutamate on T24 cells In order to test whether or not the cytotoxicity of glutamate is universal to any tumor cells, T24 cells (originally derived from human urinary bladder carcinoma cells; Japanese Cancer Research Resources Bank) were cultured with glutamate. The T24 cells did not degenerate in the presence of more

911

than 1OmM glutamate (data not shown). The IC, of glutamate on the inhibition of [3sSlcystine uptake was more than 1 mM for T24 cells, and the cellular glutathione level in T24 cells cultured with 1OmM glutamate for 24 h did not change (Table 2).

DISCUSSION

Cystine/glutamate antiport transporter system in the C6 cells The glutamate-toxicity in our C6 glioma cells is characterized as follows: (i) the effective concentration of glutamate was very high, at mM levels; (ii) only quisqualate was toxic to the cells, but not kainate, NMDA or even AMPA; (iii) the toxicity was not attenuated by glutamate receptor antagonists (including CNQX) or glutamate uptake inhibitor. These data suggest that the glutamate toxicity is not mediated through glutamate receptors. And quisqualate does not interact with glutamate receptor

C6 Glu

Fig. 11. DNA fragmentations of C6 cells by glutamate treatment. Genomic DNAs (0.5 tig) from C6 cells treated with or without 10 mM glutamate for 24 h were extracted and separated bn b.%$&agarose gel electrophoresis. Bacteriophage 1 DNA with Hind III digestion was used as marker DNAs. Bv elutamate treatment (Glu), DNA fr&&ntations became evident ia smear of 4-20 kb size as compared &iih control (C6). An arrow indicates the starting origin of electrophoresis.

912

s. KATO

sites or with ordinary Na+-dependent glutamate carrier sites. According to the recent paper by Cho and Bannai,4 glutamate transport in the C6 glioma cells was divided into two components; Na + dependent and Na +-independent. The presence of a Na + -independent glutamate transport was reported in brain synaptosomes,@ neuroblastoma-retina hybrid Nl8-RE-105 cells2* and in glioma LRM-55 cells.35 The glutamate transport in these cells may reflect the expression of antiporter for cystine/ glutamate in the nervous tissue. The present kinetic studies of cystine and glutamate uptake for C6 cells clearly showed that both transport systems have a common carrier site: glutamate is a good substrate for a cystine transporter, and conversely cystine is a good substrate for a glutamate transporter (Figs 3,4). Furthermore, we clearly demonstrated that exogenous cystine (1 mM) certainly induced an endogenous glutamate release from C6 cells in a Na +-independent Cl--dependent fashion and that quisqualate inhibited this cystine-induced glutamate release. Therefore, the present paper is elucidating and confirming the presence of cystine/glutamate antiporter system in C6 glioma cells. The protective effect of cystine (0.4-l .OmM) on glutamate toxicity in our C6 cells can be reasonably explained by this antiporter system. It is well known that glutamate is the most abundant free amino acid in the central nervous system. The intracellular level of glutamate in our C6 cells cultured for four days indeed is high, 50.6 + 10.4 nmol/mg protein (the intracellular concentration is around 10 mM). Therefore, the antiporter for cystine/glutamate usually functions to transport glutamate outside and cystine inside. If excess of glutamate is added in the culture medium, it may compete with cystine for the transporter, resulting in cell death. Addition of cystine (0.4-1.0 mM) restores the driving force for cystine uptake against glutamate exposure, while the experiment in the cystine-free medium further supports this idea. The C6 cells cultured in the cystine-free medium degenerated more rapidly within 12 h than after glutamate exposure (24 h). Such a hypothesis was also supported by a good correlation between cytotoxicity and inhibition of cystine uptake (Table 1). In T24 cells which showed a poor expression of the cystine/glutamate antiporter, over 10 mM of glutamate was not cytotoxic to the cells. This suggests that the mechanism for glutamate toxicity in the present study is not generally applied to any tumour cells, but is limited to cells or cell lines having the cystine/glutamate antiporter system. GIutathione depletion and glutamate toxicity in the C6 cells The incorporation of [35S]cystine into glutathione in the C6 glioma cells was very rapid, and the incorporation ratio was about ten-times higher than that of [3sS]methionine into glutathione, indicating that the de nova synthesis of this tripeptide in our C6

Cl a/.

cells was largely dependent on the uptake of cxtrscellular cystine but not methionine through thrr cystine/glutamate antiporter. The deprivation o! cystine from the culture medium rapidly reduced the glutathione level in C6 cells and was lethal to the cells. while addition of cystine maintained a normal level of cellular glutathione even with glutamate exposure (Table 2). Furthermore, both glutathione inhibiting and depleting agents induced a delayed death of Ch cells as did glutamate exposure. The BSO and glutamate were additively toxic to the cells. These data together suggest that the inhibition of cystine uptake by glutamate leads to glutathione depletion and finally results in delayed death of C6 cells (16 24 h) The glutamate-induced DNA damage has been also confirmed by DNA fragmentations in agarosc gel electrophoretic analysis. Glutathione is a major non-protein SH compound in the living cells and plays a key role for protection against oxidative stress and the conjugation for various toxins.“.?’ Therefore, the glutathione depletion by glutamate appears to induce a cell death. After 8-10 h glutamate exposure, the glutathione content of C6 cells significantly declined, although the significant release of LDH into the culture medium was evoked only 13-14 h after glutamate exposure. The data strongly suggest that the decrease in intracellular level of glutathione by glutamate treatment precedes the release of LDH (cytotoxicity). GSSG levels were also decreased by this treatment, although its concentration increased transiently 2-3 h after glutamate exposure. Such an imbalance between glutathione and GSSG contents at the initial period of glutamate exposure has never been reported and could form the basis of future studies. The present vitamin E experiment suggests that the death of C6 cells by glutamate is not simply due to glutathione depletion but due to the generation of some oxidative processes, such as membrane lipid peroxidation as reported by others. *‘Jo However, questions as to which types of oxidative stress (including active oxygens or free radicals) involved in the glutamateinduced cell death and DNA damage remain unsolved.

CONCLUSION

Glutathione undoubtedly is important in brain metabolism. The intracerebral level is high, ranging from 1 to 3 mM in human and monkey.)2 All the enzymes for the y -glutamyl cycle are present in the brain and choroid plexus. *’Alterations of glutathione metabolism may accompany derangements of brain function. In particular, hypoxia is associated with a reduction of brain glutathione levels.5,MFurthermore. Raps et aLB recently observed that glutathione levels in astrocytes are high compared with those in neurons. Yudokoff et al.” also showed that the glial glutathione pool is quite large and its turnover is high

Glutamate toxicity and glutarthione in the C6 glioma cells enough to be of potential importance to brain glutamate metabolism. The cellular pools of glutathione and glutamate must be considered more important with respect to cystine/glutamate antiporter in glial cells, since the antiporter is necessary to support a maintenance of cellular level of glutathione.’ Furthermore, as the existence of such an antiporter system has been recently reported in primary culture of newborn rat brain astrocytes,4 the gliotoxicity induced by glutamate with high concentrations on

913

CNS glia may be partly mediated through this mechanism. Acknowledgements-We thank Dr N. Sakura and N. Asano (School of Pharmacy, Hokuriku University) for preparing y-glutamylcysteine. We also thank Dr Y. F. Mahe for reading our manuscript and Mrs Tami Urano for her secretarial assistance. This study was supported in part by research grants (Nos 61480105 and 01650506 to S.K.) from the Ministry of Education, Science and Culture, Japan and from the Fund for Medical Treatment of the Elderly, School

of Medicine Kanazawa University, 1991.

REFERENCES

1. Bannai S. (1986) Exchange of cystine and glutamate across plasma membrane of human fibroblasts. J. biof. Chem. 261, 2256-2263. 2. Bannai S. and Kitamura E. (1980) Transport interaction of L-cystine and L-glutamate in human diploid fibroblasts in culture. J. biol. Chem. 255, 2372-2376. 3. Casper D. S., Trelstad R. L. and Reif-Lehrer L. (1982) Glutamate-induced cellular injury in isolated chick embryo retina: Milller cell localization of initial effects. J. camp. Neurof. 289, 79-90. 4. Cho Y. and Bannai S. (1990) Uptake of glutamate and cystine in C6 glioma cells and in cultured astrocytes. J. Neurochem. 55, 2091-2097. 5. Cooper A. J. L., Pulsinelli W. A. and Duffy T. E. (1980) Glutathione and ascorbate during &hernia and postischemic reperfusion in rat brain. J. Neurochem. 35, 1242-1245. 6. Coyle J. T. and Schwartz R. (1976) Lesion of striatal neurones with kainic acid provides a model for Huntington’s chorea. Nature 263, 244-246. 7. Curtis D. R. and Johnston G. A. R. (1974) Amino acid transmitters in the mammalian central nervous system. Ergebn Physiol. 69, 98-188. 8. Fonnum F. (1984) Glutamate: a neurotransmitter in mammalian brain. J. Neurochem. 42, l-11. 9. Foster A. C. and Fagg G. E. (1984) Acidic amino acid binding sites in mammalian neuronal membranes: their characteristics and relationship to synaptic receptors. Brain Res. Rev. 7, 103-164. 10. Griffith 0. W. and Meister A. (1979) Potent and specific inhibition of glutathione synthesis by buthionine sulfoximine (S-n-butylhomocysteine sulfoximine). J. biol. Chem. 254, 7558-7560. 11. Kato S. (1990) Suppressive effect of L-cystine on glutamate damage to C6 glioma cells. Trans. Am. Sot. Neurochem. 21, 199. 12. Kato S., Higashida H., Higuchi Y., Hatakenaka S. and Negishi K. (1984) Sensitive and insensitive states of cultured glioma cells to glutamate damage. Brain Res. 383, 365-373. 13. Kato S., Negishi K., Hayashi Y. and Miki N. (1983) Enhancement of neurite outgrowth and aspartate-glutamate uptake systems in retinal explants cultured with chick gizzard extract. J. Neurochem. 40, 929-938. 14. Kato S., Negishi K., Honma K., Sakai K. and Shimada Y. (1989) A high performance liquid chromatography assay for glutamine synthetase. Neurochem. Int. 14, 491496. 15. Kato S., Sugawara K., Matsukawa T. and Negishi K. (1990) Gliotoxic effects of a-aminoadipic acid isomers on the carp retina: a long-term observation. Neuroscience 36, i45-i53. 16. Kessler M.. Baudrv M. and Lvnch G. (1987) Use of cvstine to distinauish _ -alutamate bindine from elutamate ’ sequestration. Neurbsci. L&t. 81; 221-226.‘ 17. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall R. J. (1951) Protein measurement with the Folin phenol reagent. J. biol. Gem. 193, 265275. 18. Lucas D. R. and Newhouse J. P. (1957) The toxic effect of sodium tglutamate on the inner layers of the retina. Arch. Ophthalmol. 58, 193-201. 19. Meister A. (1988) Glutathione metabolism and its selective modification. J. biol. Chem. 263, 1720517208. 20. Meister A. and Anderson M. (1983) Glutathione. A. Rev. Biochem. 52, 711-760. 21. Miyamoto M., Murphy T. H., Schnaar R. L. and Coyle J. T. (1989) Antioxidants protect against glutamate-induced cytotoxicity in a neuronal cell line. J. Pharmac. exp. Ther. 258, 1132-I 140. 22. Murphy T. H., Miyamoto M., Sastre A., Schnaar R. L. and Coyle J. T. (1989) Glutamate toxicity in a neuronal cell line involves inhibition of cystine transport leading to oxidative stress. Neuron 2, 1547-1558. 23. Okonkwo P. O., Orlowski M. and Green J. P. (1974) Enzymes of the y-glutamyl cycle in the choroid plexus and brain. J. Neurochem. 22, 1053-1058. 24. Olney J. W. (1969) Brain lesions, obesity, and other disturbances in mice treated with monosodium glutamate. Science 164, 719-721. 25. Olney J. W. (1986) Inciting excitotoxic cytocide among central neurons. Adc. exp. med. Bioi. 283, 631-645. 26. Olney J. W., Ho 0. L. and Rhee V. (1971) Cytotoxic effects of acidic and sulfur containing amino acids on the infant mouse central nervous system. Expl Brain Res. 14, 61-70. 27. Ormstad K., Jones D. P. and Orrenius S. (1980) Characteristics of glutathione biosynthesis by freshly isolated rat kidney cells. J. biol. Chem. 255, 175-181. 28. Pin J. P., Bockaert J. and Recasesn M. (1984) The Ca2+/Cl- dependent [‘HI-glutamate binding: a new receptor or a particular transport process? Fedn Eur. biochem. Sots L&t. 175, 31-36. 29. Raps S. P., Lai J. C. K., Hertz L. and Cooper A. J. L. (1989) Glutathione is present in high concentrations in cultured astrocytes but not in cultured neurons. Brain Res. 493, 398401. 30. Rehncrona S., Folbergrova J., Smith D. S. and Siesjii B. K. (1980) Influence of complete and pronounced incomplete cerebral ischemia and subsequent recirculation on cortical concentrations of oxidized and reduced glutathione in the rat. J. Neurochem. 34, 477486.

914

s.

KATO

et

a/.

31. Rothman S. M. and Olney J. W. (1986) Glutamate and the pathophysiology of hypoxic-ischemic brain damagL,_4nn, Neurol. 19, 105-l 11. 32. Slivka A., Mytilineou C. and Cohen G. (1987) Histochemical evaluation of glutathione in brain. Brain KI , 409, 275-284. 33. Toyooka T., Uchiyama S., Saito Y. and Imai K. (1988) Simultaneous determination of thiols and disulfides hy high-performance liquid chromatography with fluorescence detection. Analyt. Chim. Acta 205, 2941. 34. Yudkoff M., Pleasure D., Cregar L., Lin Z. P., Nissim I., Stern J. and Nissim I. (1990) Glutathione turnover in cultured astrocytes: studies with [“Njglutamate. J. Nenrochem. 55, 137-145. 35. Waniewski R. A. and Martin D. L. (1984) Characterization of r_-glutamic acid transport by glioma cells in culture: evidence for sodium-independent, chloride-dependent high affinity-influx. J. Neurosci 4, 223712246. 36. Watanabe Y.. Kawasaki K.. Miki N. and Kuo C. H. (1990) Isolation and analvsis of the human MEKA gene encodinr a retina-specific protein. Bikhem. biophys. Res. C&nun.’ 170, 951-956. _ 37. Watkins J. C. and Olverman H. J. (1987) Agonists and antagonists for excitatory amino acid receptors. Trends Neurn.ccr. 10, 265-272. 38. Werling L. L. and Nadler J. V. (1983) L-Glutamate binding to hippocampal synaptic membranes: two binding sites discriminating by their differing affinities for quisqualate. J. Neurochem. 41, 586-593. 39. Wroblewski P. and LaDue J. ST(1955) Lactic dehyhrogenase activity in blood. Proc. Sot. exp. biol. Med. 90, 210.-213. 40. Zaczek R., Balm M., Arlis S., Drucker H. and Coyle J. T. (1987) Quisqualate-sensitive, chloride-dependent transport of glutamate into rat brain synaptosomes. J. Neurosci. Res. 18, 425-431. (Accepted 5

November

1991)

A mechanism for glutamate toxicity in the C6 glioma cells involving inhibition of cystine uptake leading to glutathione depletion.

We have demonstrated that addition of L-glutamate in millimolar amounts to a culture of C6 glioma cells induced cell death within 24 h. The glutamate-...
2MB Sizes 0 Downloads 0 Views