Cell death in rat cerebellar granule neurons induced by hydrogen peroxide in vitro: Mechanisms and protection by adenosine receptor ligands

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –2 02

a v a i l a b l e a t w w w. s c i e n c e d i r e c t . c o m

w w w. e l s e v i e r. c o m / l o c a t e / b r a i n r e s

Research Report

Cell death in rat cerebellar granule neurons induced by hydrogen peroxide in vitro: Mechanisms and protection by adenosine receptor ligands Amos A. Fatokun, Trevor W. Stone, Robert A. Smith⁎ Division of Neuroscience and Biomedical Systems, Institute of Biomedical and Life Sciences, West Medical Building, University of Glasgow, Glasgow G12 8QQ, Scotland, UK

A R T I C LE I N FO

AB S T R A C T

Article history:

Oxidative stress, resulting from excessive production of reactive oxygen species (ROS), is a

Accepted 3 November 2006

pathological state that causes profound cellular damage and eventual death resulting from

Available online 26 December 2006

the overactivation of glutamate receptors, and the generation of nitric oxide, superoxide and hydrogen peroxide (H2O2). As such, H2O2 represents an important model for studying the

Keywords:

neuropathology of oxidative stress in a variety of CNS disorders. The effects of H2O2 on the

Cerebellar granule neuron

viability of post-natal cerebellar granule neurons (CGNs), the nature of the cell death

In vitro

involved and the potential protection by adenosine receptors against the damage were

Hydrogen peroxide

examined in the current study. Hydrogen peroxide (10–400 μM) reduced CGN viability in a

Oxidative stress

concentration- and time-dependent manner. The addition of catalase (100 U/ml) prevented

Purines

this effect, and the non-specific COX inhibitor aspirin (1 mM) also alleviated the damage. A

Neuroprotection

combination of H2O2 (5 μM) and Cu2+ (0.5 mM) resulted in a significant damage that was not prevented by the hydroxyl radical scavenger mannitol (50 mM). The permeability transition pore blocker cyclosporin A, the caspase-3 inhibitor Z-DEVD-fmk (40 μM) and the PARP-1 inhibitor DPQ (10 μM) each significantly protected against peroxide damage. While the A1 adenosine receptor agonist CPA and the A2A receptor antagonist ZM241385 (each at 100 nM) elicited protection, the A1 adenosine receptor blocker DPCPX and the A2A receptor agonist CGS21680 (each at 100 nM) showed no effect. The data demonstrate that H2O2 induced oxidative stress in CGNs, involving both apoptotic and necrotic death, and this can be ameliorated by A1 receptor activation or A2A receptor blockade. © 2006 Elsevier B.V. All rights reserved.

⁎ Corresponding author. Fax: +44 0 141 330 2923. E-mail address: [email protected] (R.A. Smith). Abbreviations: AIF, apoptosis-inducing factor; ASA, aspirin; ATP, adenosine triphosphate; CAT, catalase; CGN, cerebellar granule neuron; CGS21680, 2-p-(2-carboxyethyl)phenethylamino-5′-N-ethylcarboxyamidoadenosine hydrochloride; 2-ClA, 2-chloroadenosine; COX-2, cyclooxygenase-2; CPA, N6-cyclopentyladenosine; CsA, Cyclosporin A; DPCPX, 8-cyclopentyl-1,3-dipropylxanthine; DPQ, 3,4dihydro-5-[4-(1-piperidinyl)butoxyl]-1(2H)-isoquinolinone; H2O2, hydrogen peroxide; MPT, membrane permeability transition; NAD+, nicotinamide adenine dinucleotide oxidized; NMDA, N-methyl-D-aspartate; PARP, poly (ADP-ribose) polymerase; PTP, permeability transition pore; ROS, reactive oxygen species; SBTI, soybean trypsin inhibitor; Z-DEVD-fmk, benzyloxycarbonyl-Asp(OMe)-Glu(OMe)-ValAsp(OMe)-fluoromethylketone; ZM241385, 4-(2-[7-amino-2-[2-furyl][1,2,4]triazolo[2,3-a][1,3,5]triazo-5-yl-amino]ethyl)phenol 0006-8993/$ – see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2006.11.008

194

1.

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –20 2

Introduction

Hydrogen peroxide (H2O2) is a member of the reactive oxygen species (ROS), which are known to induce toxic oxidative stress and damage by oxidizing cellular biomolecules, including lipids (Callaway et al., 1998, Garcia et al., 2000), proteins and nucleic acids (Stadtman and Levine, 2000; Marnett, 2000; Gilgun-Sherki et al., 2001) and by cross-linking membrane constituents (Coyle and Puttfarcken, 1993). This frequently results in cell death although, at low concentrations, H2O2 may be mitogenic in a number of cell types (Burdon, 1994, 1995). Antioxidant defense systems within the body, including catalase and gluthathione peroxidase, are capable of protecting against damage by H2O2 by reducing it to water and oxygen. Produced by almost all tissue types, the pathological relevance of H2O2 is underpinned by its very stable nature and by its ability to freely traverse cellular membranes (unlike the superoxide anion, of which H2O2 is a dismutation product), thereby targeting many intracellular and extracellular sites (Reiter et al., 2002). In addition, it can be converted, in the presence of transition metals such as iron or copper, to the highly toxic hydroxyl radical (Mazzio and Soliman, 2003). The brain is particularly susceptible to oxidative damage (Gaeta and Hider, 2005), and increased production of H2O2 has been implicated in the pathogenesis of several neurodegenerative diseases, including Parkinson's and Alzheimer's diseases, as well as in the damage produced by ischemia and reperfusion (Mazzio and Soliman, 2003). One of the major sources of ROS in tissues results from the overactivation of glutamate receptors, especially the NMDA receptor (Coyle and Puttfarcken, 1993; Behl et al., 1995), which is believed to mediate excitotoxic neuronal death in conditions such as stroke, induced by hypoxia and ischemia (Obrenovitch and Urenjak, 1997), especially when followed by reperfusion (Hou and MacManus, 2002), and also in neurodegeneration. We aimed therefore to examine the effects of hydrogen peroxideinduced oxidative stress on neurons, the death pathways involved, and the potential attenuation afforded by adenosine receptors, known to play significant roles in neuroprotection (Jones et al., 1998a,b; Stone, 2002). Cerebellar granule neurons were studied since they have been employed previously in elucidating the neurotoxic effects of agents such as glutamate (Ankarcrona et al., 1995; Slagsvold et al., 2000; Hou et al., 2006) and kainic acid (Smith et al., 2003) and in investigating apoptosis (Kalda and Zharkovsky, 1999).

2.

Results

2.1. Hydrogen peroxide reduces the viability of cerebellar granule neurons Cerebellar granule neurons (CGNs) were treated with H2O2 concentrations ranging from 10 to 400 μM for 0.25, 1, 3 and 6 h, at the end of which the agent was removed. Two recovery intervals (6 and 24 h) were examined to test whether the duration of recovery influenced treatment outcomes. With a 6 h recovery period, H2O2 reduced CGN viability significantly in both a concentration- and time-dependent manner, compared to the controls (P < 0.01) (Fig. 1). Even after only a 0.25 h exposure to

Fig. 1 – Effects of hydrogen peroxide (10 μM–400 μM) on CGN cultures for 0.25, 1, 3 and 6 h exposures at 8 days in vitro with viability determined 6 h after the end of treatment period. The 6 h recovery period was the duration immediately following treatment for which alamar blue was incubated with the cultures before determination of viability. Each column represents the mean ± SEM for n = 4–7 cultures. *P < 0.05, **P < 0.01 compared to untreated control.

10 μM H2O2, viability was significantly reduced to 69.45%± 3.09 of the control value. When cultures were allowed a 24 h recovery interval, viability remained above 50% of control levels with 10 and 30 μM exposures for 0.25 h. Survival was negligible with longer treatments (data not shown), with higher concentrations yielding progressively lower viability. Microscopical examination revealed that, compared to the untreated control (Fig. 2A), H2O2 caused progressive loss of neurons and the elimination of neurite processes (Fig. 2B). Some of the cell bodies also became shrunken in their appearance.

2.2.

Prevention of H2O2 effects by catalase

In order to determine whether the observed effect of H2O2 was due to its direct action on the CGNs, the ability of catalase in preventing the damage was examined. Catalase at an activity of 100 U/ml completely prevented the deleterious effects of H2O2 (30 μM) when tested for 1 or 6 h (n = 4) (Fig. 3). At 1 h, it improved viability from 13.86% ± 3.04 (for peroxide alone) to 109.66% ± 7.61 (P < 0.001) when combined with H2O2 treatment, and at 6 h from 11.98% ± 2.34 to 108.52% ± 5.17 (P < 0.001). Morphological observations confirmed that H2O2-treated neurons in the presence of catalase retained a control rather than an H2O2-challenged phenotype (Fig. 2C). Catalase alone in these experiments actually enhanced significantly the basal viability levels of CGNs to up to 130.73% ± 3.02 (P < 0.001) of the control, respectively (Fig. 3), but this in itself is clearly not sufficient to account for the vast protection to neurons exposed to the doses of H2O2.

2.3. Effect of aspirin (acetylsalicylic acid, ASA) on the cytotoxicity of H2O2 In view of the ability of the H2O2-degrading enzyme catalase to induce COX-2, we have examined the effects of acetylsalicylic acid/aspirin (ASA) on the cytotoxicity induced by H2O2. Catalase has been reported to induce cyclooxygenase-2 (COX-2) (Chen et al., 1998; Litvinov and Turpaev, 2004) and

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –2 02

195

Fig. 2 – Photomicrographs showing the neurotoxic effects of hydrogen peroxide (H2O2) on the morphology of CGNs and the protection afforded by catalase. (A) Control untreated neurons (arrows) with extensive neurite processes. (B) Neurons treated with 30 μM H2O2 for 3 h were reduced in numbers and displayed damaged and often shrunken perikarya (arrows) with a loss of neurites. (C) In the presence of CAT 100 U/ml, neurons retained a phenotype similar to untreated control cells, even after a 6 h treatment with 30 μM H2O2. Scale bar = 50 μm.

ASA blocks cyclooxygenases. Aspirin at 1 mM had no effect on basal viability (98.93% ± 4.47 of the untreated controls). When 1 mM ASA was applied along with H2O2 (30 μM), there was a slight (11%) improvement compared to H2O2 alone; however, this difference did not reach statistical significance (data not shown).

2.4.

Hydrogen peroxide, copper (II) ion and mannitol

Through the Fenton reaction, hydrogen combines with transition metals like iron and copper to yield the very toxic hydroxyl radical (Mazzio and Soliman, 2003), for which mannitol is known to be a scavenger. Interactions between H2O2, copper (II) ion and mannitol were therefore investigated for 6 h. Five micromolar H2O2 induced a reduction in viability, although this was not significant (Fig. 4) (n = 8). Copper (II) ion, generated from CuSO4 (0.5 mM), caused a time-dependent reduction of CGN viability, lowering it significantly to 48.42% ± 9.23 of the control value (P < 0.001). Co-administration of H2O2 (5 μM) and CuSO4 (0.5 mM) also significantly reduced viability.

Fig. 3 – Effects of catalase 100 U/ml on the viability of CGN cultures and hydrogen peroxide-induced cytotoxicity in these cultures. Each column shows the mean ± SEM for n = 4 cultures. **P < 0.01, ***P < 0.001 compared to control; ### P < 0.001 compared to corresponding H2O2 30 μM.

In the presence of copper and H2O2, viability was reduced significantly compared to cells treated with H2O2 alone (P < 0.001), although not significantly lower than the effect of copper alone. Microscopical examination revealed that the neurons had become shrunken, with extensive loss of neurites, following treatment. The hydroxyl radical scavenger, mannitol, failed to prevent the damage induced by this combination of hydrogen peroxide and copper (data not shown).

2.5.

Alleviation of H2O2 damage by cyclosporin A

Cyclosporin A (CsA) blocks the permeability transition pore (PTP) thereby preventing the membrane permeability transition (MPT) that precedes the activation of caspases and the subsequent cellular death (Ruiz et al., 2000). Cyclosporin A (0.5 and 10 μM) had no significant effect on viability when on its own following 1 h treatments (92.29% ± 5.03 and 93.97% ± 3.05 of the control, respectively; n = 3). With 1 h treatments, the 10 μM H 2 O 2 -mediated reduction of CGN viability was

Fig. 4 – Effects on the viability of CGN cultures of 6 h exposure to a combination of hydrogen peroxide (5 μM) and copper (II) ion (0.5 mM). Each column shows the mean ± SEM for n = 8 cultures. ***P < 0.001, a = non-significant compared to control; ###P < 0.001, ns = non-significant.

196

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –20 2

potentiated in the presence of CsA 0.5 μM to 55.62% ± 2.37 (P < 0.05), but was attenuated by 10 μM CsA to 88.05% ± 5.32 (P < 0.01) (Fig. 5). Exposure to 100 μM H2O2 for 1 h lowered viability to 12.04% ± 0.63 of the control (P < 0.001, n = 3). The addition of 10 μM CsA ameliorated this to 28.51% ± 5.87 of the control (P < 0.01) (Fig. 5).

2.6.

Induction of caspase-3 and PARP-1

Z-DEVD-fmk is an inhibitor specific for caspase-3 (Bravarenko et al., 2006), while DPQ is a potent PARP-1 inhibitor (Takahashi et al., 1997). Basal CGN viability was not affected by 0.25 h or 1 h exposures to 40 μM Z-DEVD-fmk, and similarly, DPQ (10 μM) had no effect when applied on its own. With 10 μM H2O2 for 0.25 and 1 h exposures, viability was reduced significantly to 77.89% ± 3.45 and 73.09% ± 7.72 of the control, respectively (P < 0.01, n = 4) (Fig. 6A). While Z-DEVD-fmk failed to alleviate these effects, DPQ elicited significant improvements in viability. At a higher H2O2 concentration (100 μM), viability decreased to 32.57% ± 3.33 at 0.25 h and to 6.07%± 1.59 at 1 h (Fig. 6B). Z-DEVD alleviated the damage resulting from a 0.25 h exposure and DPQ raised viability levels to 58.85% ± 3.16 of the untreated controls (P < 0.001). Neither of these agents was able to alleviate the damage caused by 1 h exposure to H2O2 100 μM.

2.7. Roles of adenosine receptors in protection against oxidative damage by H2O2 The neuroprotective roles of adenosine receptors against oxidative damage were investigated by testing a non-selective ligand, 2-chloroadenosine, and adenosine A1 and A2A receptor ligands (each at 100 nM) against a low (10 μM) or a high (30 μM) concentration of peroxide. The ligands were applied 15 min prior to the subsequent 1 h treatment with H2O2 in the continuing presence of the ligands. 2-Chloroadenosine (1 μM), while having no effect on viability on its own (103.43% ± 2.87 of the control), failed to prevent the damage caused by peroxide,

Fig. 5 – Effects of the permeability transition pore blocker cyclosporin A (0.5, 10 μM) on the neurotoxicity mediated by 1 h application of H2O2 (10 μM and 100 μM) in CGN cultures. Each column represents the mean ± SEM for n = 3 cultures. ***P < 0.001 compared to untreated control. #P < 0.05, ##P < 0.01 and ns = non-significant compared to corresponding H2O2 concentration.

Fig. 6 – Effects of the caspase-3 inhibitor Z-DEVD-fmk (40 μM) and the PARP inhibitor DPQ (10 μM) on the neurotoxicity mediated by (A) H2O2 10 μM and (B) H2O2 100 μM in CGN cultures. Each column represents the mean ± SEM for n = 4 cultures. **P < 0.01, ***P < 0.001 compared to control; #P < 0.05, ##P < 0.01, ###P < 0.001, ns = non-significant compared to corresponding H2O2.

with a value of 34.59% ± 6.24 compared to 31.22% ± 9.72 for peroxide (30 μM) alone (P < 0.001, n = 3) (data not shown). In the presence of the selective A1 receptor agonist CPA, the effect of 10 μM H2O2 was significantly attenuated (P < 0. 01) (Fig. 7A). When tested against 30 μM H2O2, CPA improved damage (P < 0.001) from 36.02% ± 1.68 for peroxide alone to 73.34 ± 3.53 (n = 3). On the other hand, the significant reduction in viability with both 10 μM and H2O2 30 μM was unaltered in the presence of the selective A1 receptor antagonist DPCPX (Fig. 7B). With the selective A2A receptor agonist CGS21680, the reduction in viability caused by exposure to 10 μM H2O2 for 1 h remained unaffected (Fig. 8A), as was the reduction by 30 μM H2O2 (20.59% ± 2.97 vs. 22.14% ± 2.36, for peroxide alone and H2O2 in the presence of CGS21680, respectively; n = 4). The selective A 2A receptor antagonist ZM241385, however, improved viability significantly (P < 0.01) when tested against 10 μM H2O2 (n = 4) (Fig. 8B) and also against 30 μM H2O2, where it improved damage (P < 0.05) to 57.69% ± 13.42 of the control when present with 30 μM H2O2 (n = 3), compared to a value of 34.10% ± 0.53 (P < 0.001) in the presence of H2O2 alone.

3.

Discussion

Concentration- and time-dependent damaging effects of hydrogen peroxide on CGN viability and morphology were

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –2 02

197

observed in this current study. A greater sensitivity to H2O2 by CGNs compared to cortical neurons has been previously reported (Klein et al., 2002). Neuronal survival was less when the recovery duration was increased from 6 to 24 h, possibly reflecting the initiation of secondary damage to cells that might have escaped the initial oxidative insult. Catalase prevented the toxic effects of H2O2, suggesting that H2O2 elicited the damage by a direct action. This study has shown that oxidative stress induced by H2O2, at least in the CGNs, involved activation of both apoptotic and necrotic death pathways since partial protection against cell death was achieved with the permeability transition pore blocker cyclosporin A, the caspase-3 inhibitor Z-DEVD-fmk, and with the PARP-1 inhibitor DPQ. While adenosine A1 receptor activation by CPA and A2A receptor blockade by ZM241385 were significantly protective against peroxide damage, A1 receptor blockade by DPCPX and A2A receptor activation by CGS21680 failed to modify it.

3.1.

Hydrogen peroxide and neuronal damage

Hydrogen peroxide is an ROS that induces significant toxic oxidative stress in a variety of tissues and causes damage by

Fig. 8 – (A) The lack of effect of the selective A2A adenosine receptor agonist CGS21680 (100 nM) and (B) the protective effect of the selective A2A adenosine receptor antagonist ZM241385 (100 nM) against the neurotoxicity mediated by either 10 μM or 30 μM H2O2 in CGN cultures. Each column represents the mean ± SEM for n = 3 or 4 cultures. ***P < 0.001, a = non-significant compared to control; #P < 0.05, ##P < 0.01, ns = non-significant compared to H2O2 alone.

Fig. 7 – (A) The protective effect of the selective A1 adenosine receptor agonist CPA (100 nM) against the neurotoxicity mediated by either 10 μM or 30 μM H2O2 in CGN cultures and (B) the lack of effect by the selective A1 adenosine receptor antagonist DPCPX (100 nM) against the neurotoxic effect of 10 μM or 30 μM H2O2 in CGN cultures. Each column represents the mean ± SEM for n = 3 or 4 cultures. **P < 0.01, ***P < 0.001, a = non-significant compared to control; ##P < 0.01, ### P < 0.001, ns = non-significant compared to H2O2 alone.

oxidizing cellular biomolecules, eventually leading to cell death (Gilgun-Sherki et al., 2001). Although a number of reports have argued that the pathological consequences of H2O2 are often mediated by the hydroxyl radical, and not by H2O2 per se (Halliwell, 1992; Avshalumov et al., 2000), concentrations can build up significantly in living tissues as it can pass through cellular membranes. The brain is exceptionally susceptible to oxidative damage due to its high oxygen consumption, the critically high levels of both iron and ascorbate and the relatively low levels of antioxidant protective agents (Gaeta and Hider, 2005). Neurodegeneration increases with ageing, particularly in the CNS, and this is related to attack by free radicals and a decrease in defense mechanisms (Reiter, 1995). It has been proposed that oxidative stress is involved in the ageing process, partly by inducing damage to mitochondrial DNA (Cadenas and Davies, 2000; Finkel and Holbrook, 2000). The lipid peroxidation induced by H2O2, e.g., in rat brain homogenates (Garcia et al., 2000), is a major consequence of oxidative stress, which occurs due to its interactions with polyunsaturated lipids in cell membranes (Callaway et al., 1998; Neely et al., 1999). In Parkinson's disease, H2O2 generated from presynaptic Lewy body α-synuclein may be associated with neurodegeneration in the substantia nigra and destruction to the

198

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –20 2

nigrostriatal tract (Turnbull et al., 2001), while in Alzheimer's disease, β-amyloid plaque accumulation causes increased intracellular levels of hydrogen peroxide (Behl et al., 1995). Overactivation of glutamate receptors, e.g., the N-methyl-Daspartate (NMDA), leads to the production of ROS including nitric oxide, superoxide and H2O2 (Behl et al., 1995).

3.2.

Effect of aspirin on H2O2-induced damage

The tendency shown by the non-specific COX inhibitor, aspirin, to protect against H2O2-induced oxidative damage suggests that the induction of COX-2 may be involved. Whereas COX-1 is constitutive in most cells, COX-2 is inducible by cytokines, endotoxins, growth factors or tumor promoters, in addition to being constitutively expressed in some brain regions, reproductive tissues, kidney and thymus (Baigent and Patrono, 2003; Rocca et al., 1999). COX-2 inhibition protects cultured CGNs from glutamate-mediated cell death (Strauss and Marini, 2002), and COX-2 is known to contribute to NMDA-mediated death in primary cultured cortical neurons (Hewett et al., 2000). We also considered the possibility that COX might modulate damage by hydrogen peroxide since the enzyme catalase, which destroys hydrogen peroxide, has been shown to induce COX-2 (Chen et al., 1998; Litvinov and Turpaev, 2004). However, aspirin had no effect on the damage by this ROS, indicating that the COX-2-inducing capacity of catalase may be dissociated from its degradative action on peroxide.

3.3. Damage by H2O2 and copper (II) ion: lack of effect of mannitol Mannitol, which is known to scavenge the hydroxyl radical (Khan et al., 2005), did not reduce the damage resulting from treatment with a combination of H2O2 and Cu2+ and also did not elicit any effect on its own. One explanation of this finding is that the damaging effect of simultaneous exposures to the two neurotoxic agents was not caused by the generation of hydroxyl radicals.

3.4. Involvement of apoptosis and necrosis in H2O2-induced neuronal damage and death Our current investigation of apoptotic mechanisms focused on membrane permeability transition (MPT), which precedes the release of cytochrome c from the mitochondria, leading to the induction of caspase-3, the executive caspase in apoptosis. Both the intrinsic and the extrinsic death pathways converge on the induction of caspase-3, a central player in precipitating apoptosis (Stepanichev et al., 2005). Prior to caspase-3 induction, MPT opens the permeability transition pore (PTP), with the collapse of the mitochondrial membrane potential, ψm. PTP formation is enhanced by increased production of ROS (Connern and Halestrap, 1994) with the release of cytochrome c preceded by its dissociation from binding to cardiolipin in the inner mitochondrial membrane (Fariss et al., 2005). Cytochrome c leaks through the multiprotein complex (Beutner et al., 1996) and combines with apoptotic protease-activating factor-1 (Apaf-1) to activate the

apoptotic cascade. Cyclosporin A (CsA) blocks the PTP and also blocks the calcium-calmodulin regulated phosphatase calcineurin (Wang et al., 1996, 1999). The protection by CsA against glutamate excitotoxicity has been reported to occur through both calcineurin-dependent and -independent mechanisms (Dawson et al., 1993; Ruiz et al., 2000). The opening of the PTP is also involved in cellular apoptosis mediated by the release of the apoptosis-inducing factor (AIF) from the mitochondria (Susin et al., 1996). The significant protection exhibited by 10 μM CsA against peroxide damage in this study provides evidence of some involvement of apoptosis in peroxide-mediated oxidative death of neurons, although it may not fully explain cell death mediated by the oxidant. The paradoxical potentiation of H2O2 (10 μM) damage by the low CsA dose (0.5 μM) adds to the continuing controversy surrounding the anti-apoptotic effects of CsA in neuronal preparations (see Fall and Bennett, 1998; McDonald et al., 1996; Chang and Johnson, 2002; Canudas et al., 2004). The agent Z-DEVD-fmk is an irreversible, cell-permeable caspase-3-selective inhibitor. The partial protection by Z-DEVD-fmk (40 μM) against damage by short exposure to 100 μM H2O2, but not against damage by a lower concentration of H2O2 (10 μM), suggests involvement of caspase-dependent apoptosis only in specific phases or combinations of circumstances in cell damage and death mediated by brief exposures of CGN cultures to high concentrations of the oxidant. The role of necrosis was examined by inhibition of poly (ADP-ribose) polymerase 1 (PARP-1), which is activated by necrosis-inducing stimuli but also may be implicated in the non-caspase type of apoptotic cell death involving the AIF (Dawson and Dawson, 2004, Hong et al., 2004). PARP is a family of nuclear proteins responsible for the repair of DNA strand nicks and breaks (Cosi et al., 1994). The most important is PARP-1, which generates more than 95% of the total ADPribose polymers in a cell and whose activity rapidly increases up to 500-fold following DNA damage (Dawson and Dawson, 2004; Hong et al., 2004). Poly (ADP)-ribosylation is a unique biochemical pathway, with poly ADP-ribose (PAR) synthesis and degradation known to occur in all mitotic and postmitotic mammalian cells (de Murcia and Shall, 2000). However, overactivation of PARP-1 leads to cell death by metabolic derangement resulting from the depletion of NAD+ and adenosine triphosphate (ATP) (Ha and Snyder, 1999). Moroni et al. (2001) reported that PARP inhibitors attenuated necrotic, but not apoptotic, neuronal death in experimental models of cerebral ischemia. A number of environmental and chemical stimuli and free radical/oxidant attacks (e.g., induced by H2O2) have been reported which can trigger the overactivation of PARP-1 in response to DNA damage (Yu et al., 2002; Dawson and Dawson, 2004; Hong et al., 2004). PARP-1 might also modulate neuronal cell death by regulating transcriptional activity via nuclear factor kappa B (NF-κB) (Koh et al., 2005), p53 or other effector proteins (Chiarugi, 2002). Pharmacological inhibition of PARP-1 or the use of PARP-1 knockouts has demonstrated a therapeutic efficacy in experimental models of disorders characterized by DNA damage, including ischemia, and excitotoxic neuronal death (Virag and Szabo, 2002; Skaper, 2003; Dawson and Dawson, 2004). The ability of the potent PARP-1 inhibitor DPQ to ameliorate significantly

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –2 02

damage by H2O2 in the current study demonstrates an involvement of necrosis, although the possibility that apoptosis mediated by AIF could also have been additionally involved in the phenomenon cannot be ruled out.

3.5.

199

the prospect of finding suitable therapeutic interventions for the treatment of neurological and neurodegenerative conditions characterized by increased levels of reactive oxygen species.

Adenosine receptors and peroxide damage

The broad-spectrum adenosine receptor agonist 2-chloroadenosine failed to modulate peroxide damage in the current study. Adenosine has been previously reported to promote neuronal recovery from ROS-induced lesion in hippocampal slices (Almeida et al., 2003), which suggests that protection by adenosine or its analogues against oxidative damage may not be uniform across brain regions. Protection against H2O2 damage was afforded by both CPA and ZM241385, indicating that either A1 receptor activation or A2A blockade is partially effective, while no exacerbation by DPCPX or CGS21680 suggests that in CGNs, at least, blockade of A1 receptors or the activation of A2A receptors may not worsen oxidative damage in the absence of other potentiating factors or cellular events. Adenosine concentration in the extracellular fluid can increase dramatically in hypoxia, hypoglycemia or ischemia due to increased ROS production, during which endogenous adenosine becomes sufficient to elicit a degree of neuroprotection (Stone and Addae, 2002). Antagonists at the A2A receptors, such as SCH58261 and ZM241385, may protect neurons against damage caused by a range of toxins, including glutamate and NMDA receptor agonists (Jones et al., 1998a,b), and in situations involving oxidative stress such as cerebral hypoxia and ischemia (Stone, 2002). They are more promising neuroprotectants than A1 receptor agonists as A1 receptors can be susceptible to inhibition by NMDA or ROS. A2A receptor antagonists are neuroprotective in animal models of ischemia (Monopoli et al., 1998; Ongini et al., 1997) and excitotoxicity (Jones et al., 1998a,b; Stone et al., 2001) and have been developed for use in clinical trials (Phillis and Goshgarian, 2001). The use of knockout models has confirmed that the effects of A2A receptor antagonists are due to the blockade of receptors, rather than to an unrecognized or non-specific action (Stone, 2005). Activation of A2A receptors – and even A3 receptors (von Lubitz, 1999) – could inhibit A1 receptor activation. The protection by A1 receptor activation or A2A receptor blockade in this study further suggests that the mechanisms of adenosine protection against glutamateinduced excitotoxicity could involve antioxidant action and that oxidative stress might be a critical downstream effector. In contrast, no neuroprotection was found with A2A receptor activation in this study, although in vivo data have shown that both adenosine A2A receptor agonists and antagonists are neuroprotective, with the protection elicited by A2A receptor activation involving multiple mechanisms (Cronstein et al., 1985; Lafon-Cazal et al., 1993; Reynolds and Hastings, 1995; Patel et al., 1996; Cunha and Ribeiro, 2000; Arslan et al., 1997; Arslan and Fredholm, 2000; Heese et al., 1997; Lee and Chao, 2001). Overall, this study demonstrates the profound susceptibility of neurons to H2O2-induced oxidative damage, which involves activation of both apoptotic and necrotic pathways and also provides evidence of the promising potential of adenosine receptors in mediating protection, thus advancing

4.

Experimental procedures

4.1.

Preparation of CGNs

Cerebellar granule neurons were prepared from 8-day-old Sprague–Dawley rat pups, anesthetized by intraperitoneal injection of 0.1 ml Euthatal (Fort Dodge Animal Health, Southampton, UK). Brains were removed and bathed in filter-sterilized buffer containing 250 mg D(+)-glucose, 300 mg bovine serum albumin and 1 ml of 3.82% w/v MgSO4, made up to 100 ml with phosphate-buffered saline (pH adjusted to 7.4 with 1 M NaOH) in order to excise the cerebella and clean them of meninges and blood vessels. The cerebella were then chopped with a scalpel and transferred to a 0.25 mg/ml trypsin solution at 37 °C for 20 min, with gentle swirling every 5 min. At the end of this step, buffer containing soybean trypsin inhibitor (SBTI) (at a final concentration of 50 μg/ml) and DNase (50 U/ml) was added prior to a number of gentle centrifugation and trituration steps using flame-polished glass pipettes of decreasing diameters. The supernatant was aspirated and the final pellet resuspended in 2 ml of sterilized Minimum Essential Medium (MEM) supplemented with 10% fetal calf serum (FCS), 2 mM glutamine, 25 mM KCl and 50 μg/ ml gentamicin, pre-warmed to 37 °C, before more medium was added to a volume of 15 ml. Cell viability was assessed by trypan blue exclusion to be greater than 95%. The density of the suspension was determined and adjusted to 1 × 106 cells/ ml and 100 μl aliquots seeded into 96-well plates pre-coated with 50 μl of 15 μg/ml poly-D-lysine. Cultures were maintained at 37 °C in a humidified atmosphere of 5% CO2/95% O2. Cytosine arabinoside (final concentration of 10 μM) was added after 20 h to eliminate proliferating non-neuronal cells (Smith and Orr, 1987).

4.2. Determination of viability and morphological assessment Cell viability was assessed by the alamar blue (AB) assay (dye purchased from BioSource International, Nivelles, Belgium) as recently described (Fatokun et al., 2006). In brief, medium was aspirated at the end of each treatment period, and 100 μl of fresh medium containing 10% v/v AB added to control and treated wells. Reagent blanks were included. Plates were incubated at 37 °C for a further 6 h prior to measuring the absorbance at 540 nm and at 595 nm wavelengths using a spectrophotometric plate reader (DYNEX Technologies, USA). The absorbance of medium blank at each of the wavelengths was subtracted from the absorbance of control and test wells with AB to give the absorbance of the oxidized form. Alamar blue reduction was then calculated according to the manufacturer's formula. Experimental data were normalized to control values. Control and treated cell morphology was monitored on an Olympus DP50 inverted phase-contrast microscope with a

200

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –20 2

digital camera system to capture images by DPSoft software (Olympus UK Ltd., Southall, UK).

4.3.

Cell treatments and data analysis

Experiments were conducted at 8 days in vitro to ensure functional maturity of the CGNs (Kato et al., 1991). Cultures were aspirated and fresh medium containing test agents at the desired concentrations was added for the experimental period. At the end of treatment, cultures were restored to culture medium for either 6 or 24 h before determination of viability. Cell viability was expressed as a percentage of the control, and the results presented as mean ± standard error of the mean (SEM) for multiple wells tested in at least three separate experiments. Two sets of means were compared using a paired or unpaired t-test as appropriate. For comparison of more than two means, one-way analysis of variance (ANOVA) was used, followed by either the Dunnett's (comparison to control) or the Student–Newman–Keuls' post hoc test, with a P-value of <0.05 considered statistically significant.

4.4.

Drugs and reagents

Receptor ligands and all other drugs were administered at concentrations typical of those used in the published literature. Chemicals and enzymes were obtained from the following sources: acetylsalicylic acid (ASA, aspirin), bovine serum albumin, catalase (CAT), 2-chloroadenosine (2-ClA), N6-cyclopentyladenosine (CPA), copper (II) sulfate (CuSO4) pentahydrate, cytosine arabinoside, 8-cyclopentyl-1,3dipropylxanthine (DPCPX), 3,4-dihydro-5-[4-(1-piperidinyl) butoxyl]-1(2H)-isoquinolinone (DPQ), DNase 1, fetal calf serum (FCS), hydrogen peroxide solution, poly-D-lysine, soybean trypsin inhibitor (SBTI), superoxide dismutase (SOD) and trypsin were all from Sigma Ltd., Poole, Dorset, UK. 2-p-(2-Carboxyethyl)phenethylamino-5′-N-ethylcarboxamidoadenosine hydrochloride (CGS21680), benzyloxycarbonyl-Asp(OMe)-Glu(OMe)-Val-Asp(OMe)-fluoromethylketone (Z-DEVD-fmk), cyclosporin A (CsA) and L-glutamine were all from Tocris, Bristol, UK. Diethylpyrocarbonate (DEPC)-treated water, Minimum Essential Medium (MEM) and gentamicin were from GIBCO Invitrogen, Paisley, UK. 4-(2-[7-Amino-2-[2furyl][1,2,4]triazolo[2,3-a][1,3,5]triazo-5-yl-amino]ethyl)phenol (ZM241385) was a gift from Dr. Poucher (AstraZeneca, Alderley Edge, UK). Stock solutions of drugs were dissolved in distilled water except the following: CAT and SOD were dissolved in normal saline, DPQ and Z-DEVD-fmk in dimethyl sulfoxide (DMSO), and CsA and DPCPX in ethanol. The final concentration of DMSO or ethanol never exceeded 0.1%, except for Z-DEVD-fmk, which had a final DMSO concentration of 0.2% as recommended by the supplier, and for which control solutions therefore included the same concentration of DMSO. None of the vehicles, when tested alone, had any effect on the cultures. Cultures were pretreated with Z-DEVDfmk or DPQ for 1 h and maintained in medium containing the inhibitor for the entire period of recovery. When adenosine receptor antagonists were used, cultures were pretreated with them for 15 min before H2O2 was added together with them.

REFERENCES

Almeida, C.G., de Mendonca, A., Cunha, R.A., Ribeiro, J.A., 2003. Adenosine promotes neuronal recovery from reactive oxygen species induced lesion in rat hippocampal slices. Neurosci. Lett. 339, 127–130. Ankarcrona, M., Dypbukt, J.M., Bonfoco, E., Zhivotovsky, B., Orrenius, S., Lipton, S.A., Nicotera, P., 1995. Glutamate-induced neuronal death: a succession of necrosis or apoptosis depending on mitochondrial function. Neuron 15, 961–973. Arslan, G., Fredholm, B.B., 2000. Stimulatory and inhibitory effects of adenosine A2A receptors on nerve growth factor-induced phosphorylation of extracellular regulated kinases 1/2 in PC12 cells. Neurosci. Lett. 292, 183–186. Arslan, G., Kontny, E., Fredholm, B.B., 1997. Down-regulation of adenosine A2A receptors upon NGF-induced differentiation of PC12 cells. Neuropharmacology 36, 1319–1326. Avshalumov, M.V., Chen, B.T., Rice, M.E., 2000. Mechanisms underlying H2O2-mediated inhibition of synaptic transmission in rat hippocampal slices. Brain Res. 882, 86–94. Baigent, C., Patrono, C., 2003. Selective cyclooxygenase 2 inhibitors, aspirin, and cardiovascular disease: a reappraisal. Arthritis Rheum. 48, 12–20. Behl, C., Widmann, M., Trapp, T., Holsboer, F., 1995. 17β-oestradiol protects neurons from oxidative stress induced cell death in vitro. Biochem. Biophys. Res. Commun. 216, 473–482. Beutner, G., Ruck, A., Riede, B., Welte, W., Brdiczka, D., 1996. Complexes between kinases, mitochondrial porin and adenylate translocator in rat brain resemble the permeability transition pore. FEBS Lett. 396, 189–195. Bravarenko, N.I., Onufriev, M.V., Stepanichev, M.Y., Ierusalimsky, V.N., Balaban, P.M., Gulyaeva, N.V., 2006. Caspase-like activity is essential for long-term synaptic plasticity in the terrestrial snail Helix. Eur. J. Neurosci. 23, 129–140. Burdon, R.H., 1994. Free radicals and cell proliferation. In: Rice-Evans, C.A., Burdon, R.H. (Eds.), Free Radical Damage and Its Control. Elsevier, Amsterdam, pp. 155–185. Burdon, R.H., 1995. Superoxide and hydrogen peroxide in relation to mammalian cell proliferation. Free Radical Biol. Med. 18, 775–794. Cadenas, E., Davies, K.J., 2000. Mitochondrial free radical generation, oxidative stress, and aging. Free Radical Biol. Med. 29, 222–230. Callaway, J.K., Beart, P.M., Jarrott, B., 1998. A reliable procedure for comparison of antioxidants in rat brain homogenates. J. Pharmacol. Toxicol. Methods 39, 155–162. Canudas, A.M., Jorda, E.G., Verdaguer, E., Jimenez, A., Sureda, F.X., Rimbau, V., Camins, A., Pallas, M., 2004. Cyclosporin A enhances colchicine-induced apoptosis in rat cerebellar granule neurons. Br. J. Pharmacol. 141, 661–669. Chang, L.K., Johnson, E.M., 2002. Cyclosporin A inhibits caspase-independent death of NGF-deprived sympathetic neurons: a potential role for mitochondrial permeability transition. J. Cell Biol. 157, 771–781. Chen, G., Kamal, M., Hannon, R., Warner, T.D., 1998. Regulation of cyclo-oxygenase gene expression in rat smooth muscle cells by catalase. Biochem. Pharmacol. 55, 1621–1631. Chiarugi, A., 2002. Poly(ADP-ribose) polymerase: killer or conspirator? The “suicide hypothesis” revisited. Trends Pharmacol. Sci. 23, 122–129. Connern, C.P., Halestrap, A.P., 1994. Recruitment of mitochondrial cyclophilin to the mitochondrial inner membrane under conditions of oxidative stress that enhance the opening of a calcium-sensitive non-specific channel. Biochem. J. 302, 321–324. Cosi, C., Suzuki, H., Milani, D., Facci, L., Menegazzi, M., Vantini, G., Kanai, Y., Skaper, S.D., 1994. Poly(ADP-ribose)

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –2 02

polymerase: early involvement in glutamate-induced neurotoxicity in cultured cerebellar granule cells. J. Neurosci. Res. 39, 38–46. Coyle, J.T., Puttfarcken, P., 1993. Oxidative stress, glutamate, and neurodegenerative disorders. Science 262, 689–695. Cronstein, B.N., Rosenstein, E.D., Kramer, S.B., Weissmann, G., Hirschhorn, N.R., 1985. Adenosine: a physiologic modulator of superoxide anion generation by human neutrophils. Adenosine acts via an A2 receptor on human neutrophils. J. Immunol. 135, 1366–1371. Cunha, R.A., Ribeiro, J.A., 2000. Purinergic modulation of [H-3] GABA release from rat hippocampal nerve terminals. Neuropharmacology 39, 1156–1167. Dawson, V.L., Dawson, T.M., 2004. Deadly conversations: nuclear-mitochondrial cross-talk. J. Bioenerg. Biomembr. 36, 287–294. Dawson, T.M., Steiner, J.P., Dawson, V.L., Dinerman, J.L., Uhl, G.R., Snyder, S.H., 1993. Immunosuppressant FK506 enhances phosphorylation of nitric oxide synthase and protects against glutamate neurotoxicity. Proc. Natl. Acad. Sci. U. S. A 90, 9808–9812. de Murcia, G., Shall, S. (Eds.), 2000. From DNA Damage and Stress Signalling to Cell Death: Poly ADP-ribosylation Reactions. Oxford Univ. Press, New York. Fall, C.P., Bennett, J.P., 1998. MPP+ induced SH-SY5Y apoptosis is potentiated by cyclosporin A and inhibited by aristolochic acid. Brain Res. 811, 143–146. Fariss, M.W., Chan, C.B., Patel, M., Houten, B.V., Orrenius, S., 2005. Role of mitochondria in toxic oxidative stress. Mol. Interv. 5, 94–111. Fatokun, A.A., Stone, T.W., Smith, R.A., 2006. Hydrogen peroxide-induced oxidative stress in MC3T3-E1 cells: the effects of glutamate and protection by purines. Bone 39, 542–551. Finkel, T., Holbrook, N.J., 2000. Oxidants, oxidative stress and the biology of ageing. Nature 408, 239–247. Gaeta, A., Hider, R., 2005. The crucial role of metal ions in neurodegeneration: the basis for a promising therapeutic strategy. Br. J. Pharmacol. 146, 1041–1059. Garcia, J.J., Martinez-Ballarin, E., Robinson, M., Allue, J.L., Reiter, R.J., Osuna, C., Acuna-Castroviejo, D., 2000. Protective effects of β-carbolines and other antioxidants on lipid peroxidation due to high peroxide in rat brain homogenates. Neurosci. Lett. 294, 1–4. Gilgun-Sherki, Y., Melamed, E., Offen, D., 2001. Oxidative stress induced neurodegenerative diseases: the need for antioxidants that penetrate the blood brain barrier. Neuropharmacology 40, 959–975. Ha, H.C., Snyder, S.H., 1999. Poly (ADP-ribose) polymerase is a mediator of necrotic cell death by ATP depletion. Proc. Natl. Acad. Sci. U. S. A. 96, 13978–13982. Halliwell, B., 1992. Reactive oxygen species and the central nervous system. J. Neurochem. 59, 1609–1623. Heese, K., Fiebich, B.L., Bauer, J., Otten, U., 1997. Nerve growth factor (NGF) expression in rat microglia is induced by adenosine A2A receptors. Neurosci. Lett. 231, 83–86. Hewett, S.J., Uliasz, T.F., Vidwans, A.S., Hewett, J.A., 2000. Cyclooxygenase-2 contributes to N-methyl-D-aspartate-mediated neuronal cell death in primary cortical cell culture. J. Pharmacol. Exp. Ther. 293, 417–425. Hong, S.J., Dawson, T.M., Dawson, V.L., 2004. Nuclear and mitochondrial conversations in cell death: PARP-1 and AIF signaling. Trends Pharmacol. Sci. 25, 259–264. Hou, S.T., MacManus, J.P., 2002. Molecular mechanisms of cerebral ischemia-induced neuronal death. Int. Rev. Cytol. 221, 93–148. Hou, S.T., Jiang, S.X., Desbois, A., Huang, D., Kelly, J., Tessier, L., Karchewski, L., Kappler, J., 2006. Calpain-cleaved collapsin

201

response mediator protein-3 induces neuronal death after glutamate toxicity and cerebral ischemia. J. Neurosci. 26, 2241–2249. Jones, P.A., Smith, R.A., Stone, T.W., 1998a. Protection against hippocampal kainate excitotoxicity by intracerebral administration of an adenosine A2A receptor antagonist. Brain Res. 800, 328–335. Jones, P.A., Smith, R.A., Stone, T.W., 1998b. Protection against kainate-induced excitotoxicity by adenosine A2A receptor agonists and antagonists. Neuroscience 85, 229–237. Kalda, A., Zharkovsky, A., 1999. Metabotropic glutamate receptor agonists protect from oxygen-glucose deprivation- and colchicine-induced apoptosis in primary cultures of cerebellar granule cells. Neuroscience 92, 7–14. Kato, K., Puttfarcken, P.S., Lyons, W.E., Coyle, J.T., 1991. Developmental time course and ionic dependence of kainate-mediated toxicity in rat cerebellar granule cell cultures. J. Pharmacol. Exp. Ther. 256, 402–411. Khan, F.H., Sen, T., Maiti, A.K., Jana, S., Chatterjee, U., Chakrabarti, S., 2005. Inhibition of rat brain mitochondrial electron transport chain activity by dopamine oxidation products during extended in vitro incubation: implications for Parkinson's disease. Biochim. Biophys. Acta 30, 65–74. Klein, J.A., Longo-Guess, C.M., Rossmann, M.P., Seburn, K.L., Hurd, R.E., Frankel, W.N., Bronson, R.T., Ackerman, S.L., 2002. The harlequin mouse mutation downregulates apoptosis-inducing factor. Nature 419, 367–374. Koh, D.W., Dawson, T.M., Dawson, V.L., 2005. Poly(ADP-ribosyl) ation regulation of life and death in the nervous system. Cell. Mol. Life Sci. 62, 760–768. Lafon-Cazal, M., Pietri, S., Culcasi, M., Bockaert, J., 1993. NMDA-dependent superoxide production and neurotoxicity. Nature 364, 535–537. Lee, F.S., Chao, M.V., 2001. Activation of Trk neurotrophin receptors in the absence of neurotrophins. Proc. Natl. Acad. Sci. U. S. A 98, 3555–3560. Litvinov, D., Turpaev, K., 2004. Extracellular catalase induces cyclooxygenase 2, interleukin 8, and stromelysin genes in primary human chondrocytes. Biochimie 86, 945–950. Marnett, L.J., 2000. Oxyradicals and DNA damage. Carcinogenesis 21, 361–370. Mazzio, E.A., Soliman, K.F., 2003. Cytoprotection of pyruvic acid and reduced beta-nicotinamide adenine dinucleotide against hydrogen peroxide toxicity in neuroblastoma cells. Neurochem. Res. 28, 733–741. McDonald, J.W., Goldberg, M.P., Gwag, B.J., Chi, S.I., Choi, D.W., 1996. Cyclosporine induces neuronal apoptosis and selective oligodendrocyte death in cortical cultures. Ann. Neurol. 40, 750–758. Monopoli, A., Lozza, G., Forlani, A., Mattavelli, A., Ongini, E., 1998. Blockade of adenosine A2A receptors by SCH 58261 results in neuroprotective effects in cerebral ischemia in rats. NeuroReport 9, 3955–3959. Moroni, F., Meli, E., Peruginelli, F., Chiarugi, A., Cozzi, A., Picca, R., Romagnoli, P., Pellicciari, R., Pellegrini-Giampietro, D.E., 2001. Poly(ADP-ribose) polymerase inhibitors attenuate necrotic but not apoptotic neuronal death in experimental models of cerebral ischemia. Cell Death Differ. 8, 921–932. Neely, M.D., Sidell, K.R., Graham, D.G., Montine, T.J., 1999. The lipid peroxidation product, 4-hydroxynoneneal inhibits neurite outgrowth, disrupts neuronal microtubules and modifies cellular tubulin. J. Neurochem. 72, 2323–2333. Obrenovitch, T.P., Urenjak, J., 1997. Altered glutamatergic transmission in neurological disorders: from high extracellular glutamate to excessive synaptic efficacy. Prog. Neurobiol. 51, 39–87. Ongini, E., Adami, M., Ferri, C., Bertorelli, R., 1997. Adenosine A2A receptors and neuroprotection. Ann. N. Y. Acad. Sci. 825, 30–48. Patel, M., Day, B.J., Crapo, J.D., Fridovich, I., Mcnamara, J.O., 1996.

202

BR A I N R ES E A RC H 1 1 3 2 ( 2 00 7 ) 1 9 3 –20 2

Requirement for superoxide in excitotoxic cell death. Neuron 16, 345–355. Phillis, J.W., Goshgarian, H.G., 2001. Adenosine and neurotrauma: therapeutic perspectives. Neurol. Res. 23, 183–189. Reiter, R.J., 1995. Oxidative processes and antioxidative defence mechanisms in the aging brain. FASEB J. 9, 526–533. Reiter, R.J., Tan, D.X., Burkhardt, S., 2002. Reactive oxygen and nitrogen species and cellular and organismal decline: amelioration with melatonin. Mech. Ageing Dev. 123, 1007–1019. Reynolds, I.J., Hastings, T.G., 1995. Glutamate induces the production of reactive oxygen species in cultured forebrain neurons following NMDA receptor activation. J. Neurosci. 15, 3318–3327. Rocca, B., Spain, L.M., Pure, E., Langenbach, R., Patrono, C., FitzGerald, G.A., 1999. Distinct roles of prostaglandin H synthases 1 and 2 in T-cell development. J. Clin. Invest. 103, 1469–1477. Ruiz, F., Alvarez, G., Ramos, M., Hernandez, M., Bogonez, E., Satrustegui, J., 2000. Cyclosporin A targets involved in protection against glutamate excitotoxicity. Eur. J. Pharmacol. 404, 29–39. Skaper, S.D., 2003. Poly (ADP-ribose) polymerase-1 in acute neuronal death and inflammation: a strategy for neuroprotection. Ann. N. Y. Acad. Sci. 993, 217–228. Slagsvold, H.H., Marvik, O.J., Eidem, G., Kristoffersen, N., Paulsen, R., 2000. Detection of high molecular weight DNA fragments characteristic of early stage apoptosis in cerebellar granule cells exposed to glutamate. Exp. Brain Res. 135, 173–178. Smith, R.A., Orr, D.J., 1987. The survival of adult mouse sensory neurons in vitro is enhanced by natural and synthetic substrata, particularly fibronectin. J. Neurosci. Res. 17, 265–270. Smith, R.A., Walker, T., Xie, X., Hou, S.T., 2003. Involvement of the transcription factor E2F1/Rb in kainic acid-induced death of murine cerebellar granule cells. Mol. Brain Res. 116, 70–79. Stadtman, E.R., Levine, R.L., 2000. Protein oxidation. Ann. N. Y. Acad. Sci. 899, 191–208. Stepanichev, M.Y., Kudryashova, I.V., Yakovlev, A.A., Onufriev, M.V., Khaspekov, L.G., Lyzhin, A.A., Lazareva, N.A., Gulyaeva, N.V., 2005. Central administration of a caspase inhibitor

impairs shuttle-box performance in rats. Neuroscience 136, 579–591. Stone, T.W., 2002. Purines and neuroprotection. Adv. Exp. Med. Biol. 513, 249–280. Stone, T.W., 2005. Adenosine, neurodegeneration and neuroprotection. Neurol. Res. 27, 161–168. Stone, T.W., Addae, J.I., 2002. The pharmacological manipulation of glutamate receptors. Eur. J. Pharmacol. 447, 285–296. Stone, T.W., Jones, P.A., Smith, R.A., 2001. Neuroprotection by A2A receptor antagonists. Drug Dev. Res. 52, 323–330. Strauss, K.I., Marini, A.M., 2002. Cyclooxygenase-2 inhibition protects cultured cerebellar granule neurons from glutamate-mediated cell death. J. Neurotrauma 19, 627–638. Susin, S.A., Zamzami, N., Castedo, M., Hirsch, T., Marchetti, P., Macho, A., Daugas, E., Geuskens, M., Kroemer, G., 1996. Bcl-2 inhibits the mitochondrial release of an apoptogenic protease. J. Exp. Med. 184, 1331–1341. Takahashi, K., Greenberg, J.H., Jackson, P., Maclin, K., Zhang, J., 1997. Neuroprotective effects of inhibiting poly (ADP-ribose) synthetase on focal cerebral ischemia in rats. J. Cereb. Blood Flow Metab. 17, 1137–1142. Turnbull, S., Tabner, B.J., El-Agnaf, O.M., Moore, S., Davies, Y., Allsop, D., 2001. Alpha-synuclein implicated in Parkinson's disease catalyses the formation of hydrogen peroxide in vitro. Free Radical Biol. Med. 30, 1163–1170. Virag, L., Szabo, C., 2002. The therapeutic potential of poly (ADP-ribose) polymerase inhibitors. Pharmacol. Rev. 54, 375–429. von Lubitz, D.K., 1999. Adenosine and cerebral ischaemia: therapeutic future or death of a brave concept? Eur. J. Pharmacol. 365, 9–25. Wang, H.G., Rapp, U.R., Reed, J.C., 1996. Bcl-2 targets the protein kinase Raf-1 to mitochondria. Cell 87, 629–638. Wang, H.G., Pathan, N., Ethell, I.M., Krajewski, S., Yamaguchi, Y., Shibasaki, F., Mckeon, F., Bobo, T., Franke, T.F., Reed, J.C., 1999. Ca2+-induced apoptosis through calcineurin dephosphorylation of BAD. Science 284, 339–343. Yu, S.W., Wang, H., Poitras, M.F., Coombs, C., Bowers, W.J., Federoff, H.J., Poirier, G.G., Dawson, T.M., Dawson, V.L., 2002. Mediation of poly (ADP-ribose) polymerase-1-dependent cell death by apoptosis-inducing factor. Science 297, 259–263.