Brain Research 862 (2000) 83–89 www.elsevier.com / locate / bres
Research report
Intracellular glutathione levels determine cerebellar granule neuron sensitivity to excitotoxic injury by kainic acid a a, b c a Maddalena Ceccon , Pietro Giusti *, Laura Facci , Gianfranco Borin , Marta Imbesi , a b Maura Floreani , Stephen D. Skaper b
a Department of Pharmacology, University of Padova, 35131 Padova, Italy Department of Neuroscience, SmithKline Beecham Pharmaceuticals, New Frontiers Science Park, Third Avenue, Harlow, CM19 5 AW, Essex, UK c Biopolymer Research Center, CNR and Department of Organic Chemistry, University of Padova, 35131 Padova, Italy
Accepted 25 January 2000
Abstract Glutathione (GSH) is a key component of the cellular defence cascade against injury caused by reactive oxygen species. Kainic acid (KA) is a potent central nervous system excitotoxin. KA-elicited neuronal death may result from the generation of ROS. The present study was undertaken to characterize the role of GSH in KA-induced neurotoxicity. Cultures of cerebellar granule neurons were prepared from 8-day-old rats, and used at 8, 14 and 20 days in vitro (DIV). Granule neurons displayed a developmental increase in their sensitivity to KA injury, as quantified by an ELISA-based assay with the tetrazolium salt MTT. At DIV 14 and 20, a 30-min challenge with KA (500 mM) reduced cell viability by 45% after 24 h, significantly greater (P,0.01) than the 22% cell loss with DIV 8 cultures. Moreover acute (30 min) KA exposure concentration-dependently reduced intracellular GSH and enhanced reactive oxygen species generation (evaluated by 29,79-dichlorofluorescein diacetate). In comparison to control, KA (500 mM) lowered GSH levels in DIV 8 granule neurons by 16% (P50.0388), and by 36% (P50.0001) in both DIV 14 and DIV 20 neurons, after 30 min. Preincubation of granule neurons with the membrane permeant GSH delivery agent, GSH ethyl ester (5 mM), for 30 min significantly increased intracellular GSH content. Importantly, GSH ethyl ester reduced the toxic effects of KA, becoming significant at 1 mM (P50.007 vs. KA-treated group), and was maximal at $2.5 mM (P,0.0001). GSH ethyl ester displayed a similar dose-dependence in its ability to counteract KA-induced depletion of cellular GSH. The data strengthen the notion that cellular GSH levels have a fundamental role in KA-induced neurotoxicity. 2000 Elsevier Science B.V. All rights reserved. Themes: Neurotransmitters, modulators, transporters and receptors Topics: Excitatory amino acids: excitotoxicity Keywords: Glutathione; Excitatory amino acids; Cerebellum; Reactive oxygen species; Neurodegeneration; Development
1. Introduction Age-related neurodegenerative disorders (Alzheimer’s disease, Parkinson’s disease, Huntington’s disease and amyotrophic lateral sclerosis) are characterized by a selective neuronal vulnerability. Recent evidence suggests a role for the overstimulation of glutamate receptor-gated ion channels in these chronic neurodegenerative disorders, as well as in stroke, epilepsy, and cerebral trauma [2,5,19]. Reactive oxygen species (ROS) may also be involved in *Corresponding author. Tel.: 139-049-8275103; fax: 139-0498275093. E-mail address:
[email protected] (P. Giusti)
neuronal death caused by excitatory amino acid (EAA) neurotransmitters. The study of intracellular processes that lead to ROS formation is important both to our understanding of the mechanisms underlying neuronal excitotoxicity, and for the discovery of pharmacological modalities to prevent or slow such neuropathologies. ROS formation is a normal product of cellular metabolism [12,13]. Oxygen radicals can react with and damage any biological molecule, including DNA, proteins and membrane lipids [14,34]. When ROS production exceeds the antioxidant defence capacity of the cell, a condition of ‘oxidative stress’ can develop [28]. Brain cells are particularly vulnerable to oxidative stress. The brain consumes about 20% of total body oxygen, which results in a
0006-8993 / 00 / $ – see front matter 2000 Elsevier Science B.V. All rights reserved. PII: S0006-8993( 00 )02074-6
84
M. Ceccon et al. / Brain Research 862 (2000) 83 – 89
relatively high load of ROS in relation to its tissue mass [29]. Polyunsaturated fatty acids, which comprise one of the main constituents of neuronal membranes, are a preferential target of ROS. Moreover, the brain is relatively deficient in protective mechanisms compared to other tissues, such as liver. Reduced glutathione (GSH) is a major player in cellular defence against ROS. Given that ROS have been implicated in excitotoxin-associated pathologies, it is conceivable that an imbalance in GSH levels could participate, directly or indirectly, in the oxidative stress which may accompany such situations. Kainic acid (KA) is a potent central nervous system excitotoxin producing acute and subacute epileptiform activity, ultimately resulting in widespread neuropathological changes involving not only nerve cells, but also glia, myelin sheaths and blood vessels [31]. Lipid peroxidation and protein oxidation have been detected in brain concurrent with KA-induced neurotoxicity. Furthermore, the neurotoxic effects of KA were blocked by the centrally acting antioxidant idebenone [20]. Antioxidants also attenuated KA-induced lipid peroxidation and neurodegeneration in vitro [24]. Activation of KA receptors has been reported to cause decreases in the GSH pool in a number of brain areas and in cultured neurons but not astroglia, suggesting that disruption of intracellular GSH homeostasis is responsible for this injury [8]. That inborn errors of the GSH biosynthetic machinery produce neurological abnormalities [4] attests to the importance of GSH in normal brain function. The mechanism of KAinduced destabilization of GSH homeostasis is not known, but could involve ROS attack of enzymes like gluathione reductase and glutathione peroxidase, upon whom the GSH redox cycle is dependent. The present study was undertaken to further characterize the interactions between the excitotoxic effects of KA and the role of GSH in this process. Utilizing cultures of cerebellar granule neurons, we have investigated the relationship between KA-induced oxidative stress and the intracellular concentration of GSH. In addition, neuroprotection afforded by the membrane permeant GSH delivery agent GSH ethyl ester, and its capability to counteract KA-induced depletion of intracellular GSH, was examined.
10% fetal calf serum, 2 mM L-glutamine, 50 mg / ml gentamicin, and 25 mM KCl. For viability assays, dissociated cells were seeded in 24-well plates, at a density of 6.5310 5 cells per well in 700 ml medium. For GSH measurements, cells were plated in 35 mm dishes, at a density of 3310 6 cells per dish in 2.5 ml medium. All culture surfaces were coated with 10 mg / ml of poly-Llysine. Twenty-four hours after cell plating cytosine arabinofuranoside (10 mM) was added to inhibit growth of non-neuronal cells. Cultures were maintained at 378C in a 95% air–5% CO 2 incubator, and were used at 8, 14 and 20 days in vitro (DIV).
2.2. Neurotoxicity assays Glutamate and NMDA were dissolved in Mg 21 -free Locke’s solution (154 mM NaCl, 5.6 mM KCl, 3.6 mM NaHCO 3, 2.3 mM CaCl 2, 5.6 mM glucose, 5 mM Hepes, pH 7.4), while KA was prepared in complete Locke’s solution containing 1.2 mM MgCl 2 . GSH ethyl ester was dissolved in Locke’s solution. Conditioned culture medium from the granule neurons was collected and set aside. Monolayers were washed once with 0.5 ml Mg 21 -free Locke’s solution, and the desired EAA receptor agonist added. All incubations were performed at 22–248C for 30 min. To allow for full activation of NMDA-sensitive ionotropic receptors, treatment with glutamate and NMDA was carried out in the absence of Mg 21 . GSH ethyl ester was added 30 min before KA treatment. Upon conclusion of EAA agonist challenge, the cultures were washed twice with 0.5 ml of Locke’s solution and returned to their culture-conditioned medium for a further 24 h. Neuron survival was quantified by a colorimetric method utilizing the metabolic dye 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyltetrazolium bromide (MTT) [30]. Culture medium was removed and replaced with 500 ml per well of Locke’s solution containing 0.18 mg / ml MTT. After incubation for 30 min at 378C, the MTT solution was removed and the reaction product dissolved in 350 ml of dimethylsulfoxide. MTT is reduced to a blue formazan by mitochondrial dehydrogenases in living cells, but not by dying cells or their lytic debris [21]. Absorbance was read at 570 and 630 nm, and the results expressed as percentage viability relative to the control culture (5100%).
2. Materials and methods
2.3. GSH assay
2.1. Cell culture
Cellular GSH levels were measured enzymatically by using a modification of the procedure of Tietze [33], as described by Floreani et al. [7]. The method is based on the determination of a chromophoric product, 2-nitro-5thiobenzoic acid, resulting from the reaction of 5-59dithiobis-(2-nitrobenzoic acid) with GSH. In this reaction, GSH is oxidized to GSSG, which is then reconverted to GSH in the presence of GSH reductase (GRx) and NADPH. The rate of 2-nitro-5-thiobenzoic acid formation
Cultures of cerebellar granule neurons were prepared from 8-day-old rat pups (Sprague–Dawley, Harlan, S. Pietro in Natisone, UD, Italy), as described previously [30]. Briefly, cerebella were minced with a razor blade, trypsinized, and triturated in the presence of DNase and trypsin inhibitor. Cells were then collected by centrifugation and resuspended in Eagle’s basal medium containing
M. Ceccon et al. / Brain Research 862 (2000) 83 – 89
is followed at 412 nm. Immediately following the 30 min of EAA agonist exposure (see Neurotoxicity Assays), cell monolayers were washed once with Locke’s solution, and 2 ml of 6% metaphosphoric acid was added per dish. After 10 min on ice, the acid extract was collected, centrifuged, and processed for GSH. To 0.1 ml of supernatant was added 0.78 ml of 0.1 M potassium phosphate–5 mM EDTA buffer (pH 7.4), 0.025 ml of 20 mM 5-59dithiobis(2-nitrobenzoic acid) (prepared in 0.1 mM phosphate buffer, pH 7.4) and 0.08 ml of 5 mM NADPH. After a 3 min equilibration period at 258C, the reaction was started by adding 2U GRx (type III Sigma, from bakers yeast, diluted in 0.1 M phosphate–EDTA buffer, pH 7.4). The formation of 2-nitro-5-thiobenzoic acid was continuously recorded at 412 nm (258C) with a Shimadzu UV-160 recording spectrophotometer. The total amount of GSH in the samples was determined from a standard curve obtained by plotting known amounts of GSH (0.05–0.4 mg / ml) incubated in the same experimental conditions vs. the rate of change of absorbance at 412 nm. GSH standards were prepared daily in 6% metaphosphoric acid and diluted in phosphate–EDTA buffer (pH 7.4). The cellular content of GSSG was typically less than 2% of the GSH level and was not considered.
2.4. Assay of ROS formation ROS formation was determined by fluorescence using 29,79-dichlorofluorescein diacetate (DCF-DA) [17]. Upon entering cells DCF-DA is de-esterified to the ionized free acid (DCFH), which then reacts with ROS to form the fluorescent 29,79-dichlorofluorescein (DCF). In a typical experiment, cells were first washed with 1.5 ml of Locke’s solution, and then preincubated for 15 min (378C) with DCF-DA (50 nmol / mg cell protein) in 2 ml of Locke’s solution. DCF-DA was added from a stock solution in methanol; the quantity of methanol never exceeded 10 ml,
85
and was always added to the blank. After this time, cells were washed with 2 ml of Locke’s solution to remove extracellular DCF-DA. The cells were then treated with 0.1–1 mM KA for 30 min at 378C. After this time the test solution was removed and 1.2 ml of 0.1 M KH 2 PO 4 –0.5% Triton X-100 (pH 7.0) was added for 10 min. Cells were then scraped from the dish and the extract centrifuged (5 min, 12000 rpm, Eppendorf microfuge model 3223). The supernatants were collected and the fluorescence was immediately read using a Perkin-Elmer spectrophotometer (LS-3) (excitation 488 nm, emission 525 nm). ROS formation was expressed as the amount of DCF formed, utilizing a DCF standard curve (0.01–100 mM). To correct for autofluorescence, controls were run in which cells were sham-treated but otherwise not exposed to KA. The basal DCF value was 4.3260.43 pmol / mg protein / min.
2.5. Statistics All data were analyzed by ANOVA and the Bonferroni test. Significance was taken as P,0.05.
3. Results
3.1. Effect of EAA receptor agonists on DIV 14 cerebellar granule neuron viability In the first series of experiments, mature (DIV 14) cerebellar granule neurons were used. All EAA receptor agonists tested produced concentration-dependent decreases in vitality of granule neurons at this age (Fig. 1). No loss in neuron survival was detected within the first 4 h after EAA challenge (data not shown). At 24 h glutamate was the most potent among the agonists, reducing cellular viability to 2661% at 200 mM (LC 50 57062.4 mM). KA and NMDA were less active: at 1000 mM, these EAA
Fig. 1. Excitatory amino acid receptor agonists display differential toxicity for DIV 14 cerebellar granule neurons. Cells at DIV 14 were exposed to the indicated concentrations of glutamate (s), NMDA (d), or KA (j) for 30 min. Neuron survival was quantified 24 h later. Results are expressed relative to sham (buffer only) treated cultures (5100). Values are means6S.D. (six experiments).
86
M. Ceccon et al. / Brain Research 862 (2000) 83 – 89
agonists decreased neuron viability to 4663.9% and 6464.8%,, respectively, of control. This cell death did not differ significantly from that produced by 500 mM KA or NMDA. Therefore, 500 mM was used in all subsequent experiments.
Table 1 GSH concentration of cerebellar granule neurons at different DIV a
3.2. Effect of culture age on granule neuron responsiveness to EAA receptor agonists
a Data are means6S.D. from six different cell preparations. *P,0.01 vs. DIV 8.
Glutamate (200 mM), KA (500 mM) and NMDA (500 mM) were next evaluated for their neurotoxic potential on cerebellar granule neurons at different ages in vitro. At DIV 8 both glutamate and KA already caused significant (P,0.001) decreases in cell survival by 37 and 26%, respectively (Fig. 2). Granule neurons at DIV 8 were resistant to the neurotoxic action of NMDA; however, by DIV 14 a 30-min pulse with NMDA resulted in the death of 2968.6% of the neurons 24 h later (Fig. 2). In these same DIV 14 cultures glutamate and KA now reduced cell numbers by 68 and 45%, respectively, a significantly (P,0.01) greater death than at DIV 8. The cultures thereafter appeared to achieve a stable degree of sensitivity to excitotoxin challenge, as no EAA agonist at DIV 20 caused cell death that differed significantly from DIV 14 (Fig. 2).
The intracellular GSH content of cerebellar granule neurons varied as a function of culture age. Granule
neurons at DIV 8 contained 2.260.20 nmol GSH / 10 6 cells (Table 1). Cellular GSH underwent a significant decline with time, reaching an ¯35% lower value at DIV 14 and DIV 20 (Table 1). Because the protein content of the granule neurons remained relatively invariant between DIV 8–20 (6263.1 mg / 10 6 cells), GSH values were routinely expressed on a cell number basis. When cellular GSH was measured immediately after the 30-min challenge with an EAA agonist, neither NMDA (500 mM) nor glutamate (200 mM) altered granule neuron GSH content at any of the culture ages (DIV 8–14–20) examined (data not shown). In contrast, granule neurons treated with KA (500 mM) at DIV 8 displayed a significant reduction (216%, P50.0388) in their GSH levels after 30 min, in comparison to control (Fig. 3). A further and significant decrease (236%, P50.0001) in cellular GSH was evident in both DIV 14 and DIV 20 cerebellar granule neurons following a 30-min exposure to KA (Fig. 3). The decreases in neuronal survival caused by KA, measured after 24 h, paralleled the alterations in GSH (Fig. 3). KA reduced cellular GSH in a concentration-dependent manner, being maximally effective at 250 mM of the EAA agonist (Table 3, left column).
Fig. 2. In vitro age specifies responsiveness of cerebellar granule neurons to excitatory amino acid receptor agonist-induced toxicity. Granule neurons at DIV 8, 14 and 20 were exposed for 30 min to 200 mM glutamate (h), 500 mM KA (9), or 500 mM NMDA (j). Neuron survival was quantified 24 h later. Results are expressed relative to sham (buffer only) treated cultures (5100). Values are means6S.D. (six experiments). Glutamate and KA vs. NMDA at DIV 8 (P,0.01); glutamate vs. KA or NMDA at Div 14 and DIV 20 (P,0.01).
Fig. 3. KA reduces survival and GSH levels concomitantly in cerebellar granule neurons: influence of age in vitro. Granule neurons at DIV 8, 14 and 20 were treated with 500 mM KA, and intracellular GSH content (h) or survival (j) assessed after 30 min and 24 h, respectively. Results are expressed relative to sham (buffer only) treated cultures (5100). Values are means6S.D. (six experiments). GSH levels and survival at DIV 14 and 20 differed significantly (P,0.05) from corresponding values at DIV 8.
3.3. Relationship between cytoplasmic GSH levels and viability of glutamatergic agonist-treated granule neurons
DIV
nmol GSH / 10 6 cells
8 14 20
2.260.20 1.560.35* 1.560.18*
M. Ceccon et al. / Brain Research 862 (2000) 83 – 89
87
3.4. DCF production as an index of ROS generation
Table 2 GSH ethyl ester increases cerebellar granule neuron GSH content
Addition of KA to DIV 14 cerebellar granule neurons led to a significant increase in ROS production, as assessed by DCF generation. The effect was time-dependent, becoming maximal after 30 min incubation with KA (data not shown). Thereafter, the fluorescent signal remained constant or decreased slightly. The KA-induced increase in DCF production was concentration-dependent between 100 and 1000 mM. Net generation of DCF with 1000 mM KA was 3.360.97 pmol DCF / mg protein / min (Fig. 4).
GSH ethyl ester (mM)
nmol GSH / 10 6 cells
0 1.0 2.5 5.0 7.5 10
1.560.25 a 1.660.13 a 1.760.14 a,b 1.960.11 b 1.960.09 b 1.960.07 b
3.5. GSH ethyl ester reduces KA toxicity to cerebellar granule neurons Incubation of DIV 14 cerebellar granule neurons with the membrane permeant GSH delivery agent, GSH ethyl ester, significantly increased cellular GSH content after 30 min (Table 2). The minimum effective concentration of GSH ethyl ester was 5 mM; higher concentrations (7.5–10 mM) did not produce further elevations in GSH. Preincubation of granule neurons with GSH ethyl ester for 30 min significantly reduced the toxic effects of 500 mM KA. The neuroprotective action of GSH ethyl ester first attained significance at 1 mM (P50.007 vs. KA-treated group), and was maximally evident with GSH ethyl ester concentrations $2.5 mM (P,0.0001) (Fig. 5). Moreover, GSH ethyl ester counteracted the KA-induced depletion of cellular GSH stores (Table 3). Importantly, a 30-min pretreatment of DIV 14 granule neurons with 5 mM GSH ethyl ester strongly quenched the ROS generated by a subsequent challenge with KA, at all concentrations of the
Cerebellar granule neurons at DIV 14 were incubated for 30 min with different concentrations of GSH ethyl ester, and GSH content then measured. Data are means6S.D. from four cell preparations. Values having different superscript letters indicate significant differences (P,0.01) between those groups.
Fig. 5. GSH ethyl ester protects cerebellar granule neurons from KA toxicity. Neurons at DIV 14 were preincubated with different concentrations of GSH ethyl ester for 30 min. Cells were then challenged with 500 mM KA for 30 min in the presence of GSH ethyl ester. Cell survival was quantified 24 h later. Results are expressed relative to sham (buffer only) treated cultures (5100). Values are means6S.D. (six experiments). Different superscript letters indicate significant differences (P,0.01) between those groups.
EAA agonist (data not shown). For example, pretreatment with GSH ethyl ester reduced KA (500 mM) DCF production by 85% (from 1.8360.26 to 0.2960.09 pmol / mg protein / min). Table 3 GSH ethyl ester counteracts KA-induced GSH depletion in cerebellar granule neurons
Fig. 4. KA dose-dependently increases ROS production in DIV 14 cerebellar granule neurons. Cells were challenged with KA for 30 min. Cultures were processed after this time for measurement of ROS. Results are expressed relative to sham (buffer only) treated cultures (5100). Values are means6S.D. (six experiments). Values having different superscript letters indicate significant differences (P,0.01) between those groups.
KA (mM)
nmol GSH / 10 6 cells GSH ethyl ester 2
5 mM
0 100 250 500
1.560.35 a,b 1.260.11 b,c 1.060.07 c 0.9860.08 c
1.960.23 a 1.760.14 a 1.760.21 a 1.660.09 a
Data are means6S.D. from four cell preparations. Values having different superscript letters indicate significant differences (P,0.01) between those groups.
88
M. Ceccon et al. / Brain Research 862 (2000) 83 – 89
4. Discussion Kainic acid is a known potent neuroexcitotoxin, although the biochemical mechanism producing its underlying neurotoxic effect is not clear. KA binds to and activates a subclass of non-NMDA excitatory amino acid receptors [23]. Histopathological examination of gerbil brains 24 h after systemic injection of KA revealed severe neuronal lesions in different brain regions, especially the cerebellum and hippocampus [32]. Besides inducing brain lesions directly, KA can provoke the release of potentially neurotoxic amounts of glutamate [5]. KA-elicited neuronal death may also result from the generation of ROS and subsequent membrane destruction [32]. A key finding of the present study is that the cellular level of GSH is a critical factor in determining the sensitivity of cerebellar granule neurons to KA toxicity. The generation of free radicals appears central to the mechanism of KA-elicited neuronal death. Free radical formation in the brain has been detected 1 h after KA administration, together with increased lipid peroxidation [32]. In cultured cerebellar granule neurons, the toxicity of KA can be prevented by inhibiting xanthine oxidase, a cellular source of cytotoxic superoxide radicals [6]; attenuation in the ability of KA to induce neuronal damage was also obtained with various lipophilic antioxidants [24]. Ionotropic, but not metabotropic glutamate receptor agonists have been reported to increase the rate of ROS formation in an isolated synaptoneurosomal fraction derived from rat cerebral cortex [1,11]. In cortical neuron cultures, free radical scavengers protected from excitotoxicity triggered by agonists for non-NMDA glutamate receptors but not from the toxicity of NMDA [3]. Cerebellar granule neuron sensitivity to the toxic actions of EAA receptor agonists varied as a function of age in culture. Before DIV 8, granule neurons responded to neither glutamate, NMDA nor KA. Between DIV 8 and DIV 20, granule neurons displayed an increasing sensitivity to all three EAA agonists, with a rank of potency: glutamate.KA.NMDA. From DIV 14 onward the cultures appeared to achieve a stable degree of sensitivity to excitotoxin challenge, as no EAA agonist at DIV 20 caused a level of cell death that differed significantly from DIV 14. Interestingly, the granule neurons did not become sensitive to NMDA toxicity until DIV 14. A similar time course for the development of KA toxicity in cultured cerebellar granule neurons has been observed by others [16]. One of the most important functions of reduced glutathione lies in its ability to limit oxidative damage caused by ROS, many of which are generated as a consequence of normal metabolic activity. The present findings support the notion that cytoplasmic GSH content is a determining factor in neuronal sensitivity to injury by KA. Granule neuron GSH content declined by about one-third between DIV 8 and DIV 14. While neither NMDA nor glutamate
affected cellular GSH levels at any culture age, KA provoked reductions in granule neuron GSH which increased as a function of both rising agonist concentration and in vitro age of the cells. Moreover, KA-induced decreases in neuron survival measured 24 h post-agonist paralleled the alterations in GSH. The timing of the observed KA-triggered decrease in GSH indicates that the latter may represent an index of cellular damage prior to death; no changes in granule neuron integrity could be detected in the first 2 h after KA washout. This view is strengthened by the fact that although GSH content was already significantly reduced in the forebrain, hippocampus, and amygdala (KA receptor-rich areas) by 2.5 h after KA administration, no tissue injury was evident at this time [8]. Not unexpectedly, exposure of cerebellar granule neurons to KA led to significant increases in intracellular levels of ROS. As with its effect on GSH content, the capability of KA to stimulate ROS production was dependent on the EAA agonist concentration. Furthermore, KAtreated DIV 14 granule neurons generated greater amounts of ROS than did DIV 8 neurons (unreported results). Under these conditions, neither NMDA nor glutamate produced any changes in cellular ROS. Together, the data suggest that the capacity of granule neurons to cope effectively with the neurotoxin challenge by KA is tightly linked to a developmentally-dependent elaboration of GSH in vitro. To provide further support for this idea, cerebellar granule neurons at DIV 14 were incubated with the membrane permeant GSH delivery agent, GSH ethyl ester. This was found to significantly increase cellular GSH even after 30 min. Additionally, preincubation of granule neurons with GSH ethyl ester counteracted not only the depletion of cellular GSH stores and ROS formation following challenge with KA, but the neurotoxic effect of exposure to the EAA agonist, as well. The brain has a large oxidative capacity, but its ability to combat oxidative stress is limited. A major endogenous protective system is the glutathione redox cycle: GSH acts as both a nucleophilic scavenger of toxic compounds and as a substrate in the glutathione peroxidase-mediated destruction of hydroperoxides [18]. That inborn errors of the GSH biosynthetic machinery produce neurological abnormalities [4] attests to the importance of GSH in normal brain function. Inhibition of GSH synthesis leads to a pronounced decrease of cortical GSH and to a striking enlargement and degeneration of mitochondria in brains of newborn rats [15]. Treatment with GSH reportedly protects against KA-induced neuropathological changes in rat brain [27], suggesting the value of pharmacological strategies directed toward the regulation of endogenous GSH levels. For example, it has recently been shown that the neurotoxic actions of KA can be counteracted by the pineal hormone melatonin [9,10], via maintenance of cellular GSH homeostasis [8]. Melatonin neuroprotection probably derives from its antioxidant properties [25,26]. Molecules
M. Ceccon et al. / Brain Research 862 (2000) 83 – 89
that maintain glutathione balance in the brain may thus be considered putative neuroprotective agents useful for the treatment of central nervous system pathologies that involve excitotoxicity or where oxidative damage may contribute to neuropathogenesis, e.g. Parkinson’s disease [22], Alzheimer’s disease [12], or amyotrophic lateral sclerosis [5].
Acknowledgements This work was supported in part by MURST Anno 1998-prot. 9805089988]007 and Ricerca Sanitaria Finalizzata-Anno 1997, Regione Veneto.
References [1] S.C. Bondy, D.K. Lee, Oxidative stress induced by glutamate receptor agonists, Brain Res. 610 (1993) 229–233. [2] D.W. Choi, Glutamate neurotoxicity and diseases of the nervous system, Neuron 8 (1988) 623–634. [3] H.S. Chow, J.J. Lynch III, K. Rose, D.W. Choi, Trolox attenuates cortical neuron injury induced by iron, ultraviolet light, glucose deprivation, or AMPA, Brain Res. 639 (1994) 101–108. [4] A.J.L. Cooper, A. Meister, Glutathione in the brain: disorders of glutathione metabolism, in: R. Rosenberg, S. Prusiner, S. DiMauro, R. Barchi, L. Kunkel (Eds.), The Molecular and Genetic Basis of Neurological Disease, Butterworths, Stoneham, 1993, pp. 209–238. [5] J.T. Coyle, P. Puttfarcken, Oxidative stress, glutamate and neurodegenerative disorders, Science 262 (1993) 689–695. [6] J.A. Dykens, A. Stern, E. Trenker, Mechanism of kainate toxicity to cerebellar neurons in vitro is analogous to repurfusion tissue injury, J. Neurochem. 49 (1987) 1222–1228. [7] M. Floreani, M. Petrone, P. Debetto, P. Palatini, A comparison between different methods for determination of reduced and oxidized glutathione in mammalian tissues, Free Radic. Res. 26 (1997) 449–455. [8] M. Floreani, S.D. Skaper, L. Facci, M. Lipartiti, P. Giusti, Melatonin maintains glutathione homeostasis in kainic acid-exposed rat brain tissues, FASEB J. 11 (1997) 1309–1315. [9] P. Giusti, M. Gusella, M. Lipartiti, D. Milani, W. Zhu, S. Vicini, H. Manev, Melatonin protects primary cultures of cerebellar granule neurons from kainate but not from N-methyl-D-aspartate excitotoxicity, Exp. Neurol. 131 (1995) 39–46. [10] P. Giusti, M. Lipartiti, D. Franceschini, N. Schiavo, M. Floreani, H. Manev, Neuroprotection by melatonin from kainate-induced excitotoxicity in rats, FASEB J. 10 (1996) 891–896. [11] P. Giusti, D. Franceschini, D. Petrone, H. Manev, M. Floreani, In vitro and in vivo protection against kainate-induced excitotoxicity by melatonin, J. Pineal Res. 20 (1996) 226–231. [12] B. Halliwell, Reactive oxygen species and the central nervous system, J. Neurochem. 59 (1992) 1609–1623. [13] B. Halliwell, J.M.C. Gutteridge, Free Radicals in Biology and Medicine, 3rd Edition, Oxford University Press, Oxford, 1999. [14] J.A. Imlay, S. Linn, DNA damage and oxygen radical toxicity, Science 240 (1988) 1302–1309.
89
˜ [15] A. Jain, J. Martensson, E. Stole, P.A.M. Auld, A. Meister, Glutathione deficiency leads to mitochondrial damage in brain, Proc. Natl. Acad. Sci. USA 88 (1991) 1913–1917. [16] K. Kato, P.S. Puttfarcken, W.E. Lyons, J.T. Coyle, Developmental time course and ionic dependence of kainate-mediated toxicity in rat cerebellar granule cell cultures, J. Pharmacol. Exp. Ther. 256 (1991) 402–411. [17] C.P. Lebel, S.C. Bondy, Sensitive and rapid quantitation of oxygen reactive species formation in rat synaptosomes, Neurochem. Inter. 17 (1990) 435–440. [18] A. Meister, M.E. Anderson, Glutathione, Annu. Rev. Biochem. 52 (1983) 711–760. [19] B. Meldrum, Amino acids as dietary excitotoxins: a contribution to understanding neurodegenerative disorders, Brain Res. Rev. 18 (1993) 293–314. [20] M. Miyamoto, J.T. Coyle, Idebenone attenuates neuronal degeneration induced by intrastriatal injection of excitotoxins, Exp. Neurol. 108 (1990) 38–45. [21] T. Mosmann, Rapid colorimetric assay for cellular growth and survival: application to proliferation and cytotoxicity assays, J. Immunol. Methods 65 (1983) 55–63. [22] C.W. Olanow, An introduction to the free radical hypothesis in Parkinson’s disease, Ann. Neurol. 32 (1992) 52–59. [23] J. Olney, V. Rhee, O.L. Ho, Kainic acid: a powerful toxic analogue of glutamate, Brain Res. 77 (1974) 507–512. [24] P.S. Puttfarcken, R.L. Getz, J.T. Coyle, Kainic acid-induced lipid peroxidation: protection with butylated hydroxytoluene and U78517F in primary cultures of cerebellar granule cells, Brain Res. 624 (1993) 223–232. [25] R.J. Reiter, Oxidative processes and antioxidative defence mechanisms in the aging brain, FASEB J. 9 (1995) 526–533. ˜ [26] R. Reiter, L. Tang, J.J. Garcia, A. Munoz-Hoyos, Pharmacological actions of melatonin in oxygen radical pathophysiology, Life Sci. 60 (1997) 2255–2271. [27] A. Saija, P. Princi, A. Pisani, M. Lanza, M. Scalese, E. Aramnejad, R. Cesarini, G. Costa, Protective effect of glutathione on kainic acid-induced neuropathological changes in rat brain, Gen. Pharmacol. 25 (1997) 97–102. [28] H. Sies, Oxidative stress: introductory remarks, in: H. Sies (Ed.), Oxidative Stress, Academic Press, London, 1985. [29] B.K. Siesjo, Free radicals and brain damage, Cerebrovasc. Brain Metab. Rev. 1 (1989) 165–211. [30] S.D. Skaper, L. Facci, D. Milani, A. Leon, G. Toffano, Culture and use of primary and clonal neural cells, in: P.M. Conn (Ed.), Methods in Neuroscience, Vol. 2, Academic Press, San Diego, 1990, pp. 17–33. [31] G. Sperck, Kainic acid seizures in the rat, Prog. Neurobiol. 42 (1994) 1–32. [32] A.Y. Sun, Y. Cheng, Q. Bu, F. Oldfield, The biochemical mechanisms of the excitotoxicity of kainic acid. Free radical formation, Mol. Chem. Neuropathol. 17 (1992) 51–63. [33] F. Tietze, Enzymatic method for quantitative determination of nanogram amounts of total and oxidized glutathione: applications to mammalian blood and other tissues, Anal. Biochem. 27 (1969) 502–522. [34] S.P. Wolff, A. Garner, R.T. Dean, Free radicals, lipids and protein degradation, Trends Biol. Sci. 11 (1986) 27–31.
