MELAS mitochondrial DNA mutation A3243G reduces glutamate transport in cybrids cell lines

Experimental Neurology 212 (2008) 152–156

Contents lists available at ScienceDirect

Experimental Neurology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y e x n r

MELAS mitochondrial DNA mutation A3243G reduces glutamate transport in cybrids cell lines Jacopo C. DiFrancesco a,c,d,⁎, J. Mark Cooper d, Amanda Lam d, Paul E. Hart d, Lucio Tremolizzo a,b, Carlo Ferrarese a,b,c, Antony H. Schapira d,e a

Department of Neuroscience and Biomedical Technologies, University of Milano-Bicocca, Monza, Italy Department of Neurology, San Gerardo Hospital, Monza, Italy Scientific Institute “E. Medea”, Bosisio Parini (LC), Italy d University Department of Clinical Neurosciences, Royal Free and University College Medical School, London, UK e Institute of Neurology, University College London, London UK b c

a r t i c l e

i n f o

Article history: Received 20 November 2007 Revised 10 March 2008 Accepted 15 March 2008 Available online 26 March 2008 Keywords: MELAS A3243G Excitotoxicity Glutamate Mitochondria Cybrid

a b s t r a c t MELAS (mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes) is commonly associated with the A3243G mitochondrial DNA (mtDNA) mutation encoding the transfer RNA of leucine (UUR) (tRNA Leu(UUR)). The pathogenetic mechanisms of this mutation are not completely understood. Neuronal functions are particularly vulnerable to alterations in oxidative phosphorylation, which may affect the function of the neurotransmitter glutamate, leading to excitotoxicity. In order to investigate the possible effects of A3243G upon glutamate homeostasis, we assessed glutamate uptake in osteosarcoma-derived cytoplasmic hybrids (cybrids) expressing high levels of this mutation. High-affinity Na+-dependent glutamate uptake was assessed as radioactive [3H]-glutamate influx mediated by specific excitatory amino acid transporters (EAATs). The maximal rate (Vmax) of Na+-dependent glutamate uptake was significantly reduced in all the mutant clones. Although the defect did not relate to either the mutant load or magnitude of oxidative phosphorylation defect, we found an inverse relationship between A3243G mutation load and mitochondrial ATP synthesis, without any evidence of increased cellular or mitochondrial free radical production in these A3243G clones. These data suggest that a defect of glutamate transport in MELAS neurons may be due to decreased energy production and might be involved in mediating the pathogenic effects of the A3243G mtDNA mutation. © 2008 Elsevier Inc. All rights reserved.

Introduction Mitochondrial encephalomyopathies are a diverse range of disorders caused by a number of different mutations of either the mitochondrial or nuclear genomes (DiMauro and Schon, 2003; Schapira, 2006). The MELAS phenotype may be caused by several different mitochondrial DNA (mtDNA) mutations of which the A3243G is the most common (Goto et al., 1990). The clinical effects are diverse and encompass not only the full MELAS phenotype, but also monosymptomatic features such as deafness or diabetes mellitus. The strokelike episodes represent one of the most important clinical features of MELAS. Their pathogenesis is likely to be multifactorial and it is still Abbreviations: Cybrids, cytoplasmic hybrids; DHE, dihydroethidium; DHR, dihydrorhodamine 123; EAATs, excitatory amino acid transporters; Km, glutamate uptake affinity; LHON, Leber's hereditary optic neuropathy; MELAS, mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes; mtDNA, mitochondrial genome; ROS, reactive oxygen species; THA, l-(-)-threo-3-hydroxyaspartic acid; Vmax, glutamate uptake maximal rate. ⁎ Corresponding author. University of Milano-Bicocca, Department of Neuroscience – Section of Neurology, Via Cadore, 48 – 20052 Monza (MI) – Italy. Fax: +39 02 6448 8108. E-mail address: [email protected] (J.C. DiFrancesco). 0014-4886/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2008.03.015

debated: the ischemic vascular hypothesis suggests they are caused by “mitochondrial angiopathy” and the generalized cytopathic theory proposes that neuronal hyperexcitability may initiate, then maintain and develop the cascade of stroke-like events caused by “mitochondrial cytopathy”. Once neuronal hyperexcitability is developed in a localized brain region, this could depolarize the adjacent neurons, spreading at the surrounding cortex, in agreement with the non vascular distribution of the stroke-like events (Iizuka and Sakai, 2005). The molecular consequences of the A3243G mutation are not completely understood, but may include effects on both transcription and translation of mtDNA. In fact, this mutation has been linked to a marked decrease in both the rates of synthesis and the steady-state levels of the mitochondrial translation products; moreover a small but consistent increase in the levels of an unprocessed RNA containing the tRNALeu(UUR) sequence (RNA 19) has been reported and it could contribute to the observed inhibition of mitochondrial protein synthesis (King et al., 1992). Other authors have shown a diminished steady-state level of the tRNALeu(UUR) in mutant cellular models. This has been linked to post-transcriptional modifications such as diminution of methylation, which could be responsible for a slower processing rate of the precursor transcript and may accelerate the rate

J.C. DiFrancesco et al. / Experimental Neurology 212 (2008) 152–156

of degradation of the tRNA, lowering its steady-state level (Helm et al., 1999). Other results suggest that the primary biochemical defect in cells with high levels of A3243G mutated mtDNA is the inability to translate the UUR leucine codons (El Meziane et al., 1998). Nevertheless, all these different mechanisms seem to have in common a decreased ability to translate the tRNA of leucine. While the exact mechanism is still debated, the A3243G mutation results in dysfunction of the respiratory chain, which can lead to cellular energy impairment and cell death by failure of different key homeostatic processes and activation of the apoptotic cascade (Liu et al., 2004). Glutamate uptake is a cellular process strictly dependent upon energy supply and a mitochondrial respiratory chain defect may induce a reduction of glutamate transport (Danbolt, 2001), leading to excitotoxic damage, a feature reported in several CNS diseases (Ferrarese and Biel, 2004). Moreover, the potential contribution of abnormal glutamate handling in the pathogenesis of mtDNA diseases has recently been highlighted in Leber's hereditary optic neuropathy (LHON), a common cause of blindness, most frequently related to mutations of mtDNA genes encoding subunits of complex I (Beretta et al., 2004). Interestingly, LHON-MELAS overlap syndromes have been described in association with mtDNA mutations G13513A (Pulkes et al., 1999) and G3376A (Blakely et al., 2005). We hypothesized that the MELAS phenotype, including the strokelike episodes, may be related to the reduced ATP supply associated with the A3243G mutation, influencing the activity of specific excitatory amino acid transporters (EAATs) with disruption of glutamate transport, resulting in excitotoxicity and subsequent neuronal loss. Therefore, the aim of this paper was to analyze glutamate transport in cybrids cell lines carrying the A3243G mtDNA mutation, in order to identify a putative dysfunction of this homeostatic process. Materials and methods

153

restriction site and, after cleavage, generated 2 products of 214 bp and was used to assess enzyme efficiency. Samples from each cell line were spiked with the control product and digested with ApaI (1 U/µg DNA) (Stratagene, CA, USA) overnight at 37 °C. Products from enzyme digestion were separated on ethidium bromide-stained 3% agarose gels and radioactivity signal acquired and quantified using a phosphorimager (Storm™, Amersham). A3243G mutant load for each cybrid line was calculated by the ratio between 315 bp band (product of 630 bp band digestion) and the residual 630 bp band, uncut by ApaI. Restriction enzyme efficiency was assessed by the ratio between 214 bp and 428 bp bands and was always N98%. High-affinity sodium-dependent glutamate uptake Growing cybrid cells were placed in 24-well plates at a seeding density of 105 cells/ml. After 24 h, near confluent cells were washed with HEPES-buffered saline (10 mM HEPES, 135 mM choline chloride, 5 mM KCl, 0.6 mM MgSO4, 7 H2O, 2.5 mM CaCl2, 6 mM D-glucose, pH 7.4) and pre-incubated for 20 min with HEPES-buffered saline (475 µl/ well) containing either 135 mM NaCl (high sodium) or choline chloride (no sodium), shaking at 37 °C in order to study the sodium dependency of glutamate uptake (O'Neill et al., 1994). The uptake was started with the addition of glutamate (2 to 240 µM) containing L-[3H]glutamate (2 µCi each well, specific activity 51 Ci/mmol) and was stopped after 30 min by placing the 24-well plates in an ice-cold bath and rapidly washing cells three times with 0.5 ml/well of ice-cold 0.32 M sucrose. The time of radioactive incubation was chosen based on the linearity of the signal of glutamate uptake activity in this cellular model (see also Beretta et al., 2004). Cells were rapidly lysed with 0.25 M NaOH (0.5 ml/ well) and used for protein analysis (BCA) and radioactivity assessment using a β-counter (Beckman Coulter, CA, USA). The specific activity of tritiated glutamate was used to convert disintegrations per minute (dpm) into nmoles glutamate/mg of cell protein/30 min.

Cybrid cell lines generation and culture conditions Measurement of mitochondrial ATP synthesis Three control (H2-1, H2-2, D2-2) and five mutant (RS#3, RS#9, RS#17, E2#18, E2#23) human osteosarcoma-derived cytoplasmic hybrid (cybrid) clonal lines expressing high levels of A3243G mutation were generated by the fusion of platelets obtained from two separate encephalopathic MELAS patients and two age matched healthy controls into 206 ρ° cells (ρ° cells obtained from Prof G. Attardi), as previously described (Gu et al., 1998). Cybrid lines were grown in Dulbecco's Modified Eagle's Medium (DMEM) containing glucose (10 mM), sodium pyruvate (1 mM),10% foetal bovine serum, L-alanyl-glutamine 4.03 mM, penicillin (100 U/ml), streptomycin (100 µg/ml) at 37 °C in an incubator with a humidified atmosphere of 5% CO2. The medium was changed every 2–3 days and always on the day of every experiment. A3243G mtDNA mutation load analysis DNA was extracted from cultured cells using Nucleon BACC extraction and purification kit (Amersham, NJ, USA). Polymerase chain reactions (PCR) were performed using standard conditions (Turner et al., 1998). In order to generate a 630 bp band containing the site of A3243G mutation, primers M4 (5′-CCT AGG GAT AAC AGC GCA AT-3′, nucleotides 2928–2947) and M5 (5′-TAG AAG AGC GAT GGT GAG AG-3′, nucleotides 3558–3539) (R and D Systems, Oxford, UK) were used with an annealing temperature of 58 °C. Preliminary experiments indicated that after 20 cycles the product levels were not saturated. In order to perform the quantification of A3243G mutant load, nucleotide 32 P dTTP (5 µCi each sample, specific activity 3000 Ci/mmol) was added to the last cycle of every reaction. A second control product (428 bp) was generated using primers M106 (5′-TGT AAA ACG ACG GCC AGT ATA TGT CTC CAT ACC CAT-3′, nucleotides 4215–4251) and M107 (5′-CAG GAA ACA GCT ATG ACC CTT GAT GGC AGC TTC TGT GG-3′, nucleotides 4643–4605) using an annealing temperature of 58 °C. This had an ApaI

ATP synthesis was measured in digitonin permeabilized cells as previously described (Korlipara et al., 2004). Briefly cells at a concentration of 106/ml, were incubated for 15 min at 37 °C in a buffer solution containing 150 mM KCl, 2 mM K2-EDTA, 10 mM K2HPO4, 25 mM Tris pH 7.4, 0.1% (w/v) BSA, 40 µg digitonin, 10 mM glutamate, 10 mM malate and 1 mM ADP. The reaction was terminated by the addition of 0.5 ml of 60% HClO4. As preliminary experiments, control cybrids were treated with rotenone 1 µM before the measurement of ATP synthesis, confirming that the measured ATP was synthetized primarily by mitochondrial activity. The conditions used for the experiments were chosen following the linearity of the signal of ATP production in this cellular model. Following neutralisation, the ATP content was assayed using a luminescent method (Biothema, Haninge, Sweden) and a Jade Tube Luminometer (Labtech, Uckfield, UK). Each sample was assayed in triplicate, corrected for ATP levels in the cells at the start of the incubation and results were expressed as moles/µg protein/15 min. Measurement of cellular ATP Cells at 80% confluence, growing in 10-cm culture dish, were washed with PBS at 37 °C and 1 ml of 6% HClO4 was added. From this step, the levels of ATP were analysed as described above. Results were expressed as moles/µg protein. Reactive Oxygen Species (ROS) detection Dihydroethidium staining (DHE; Sigma-Aldrich, St. Louis, MO) was used to evaluate ROS total cellular level in growing cybrids. DHE is freely permeable to cells and, in the presence of free radicals, is

154

J.C. DiFrancesco et al. / Experimental Neurology 212 (2008) 152–156

oxidized to fluorescent ethidium bromide which is trapped intracellularly by intercalation into the genomic DNA. Growing cells were placed into 24-well plates at a seeding density of 105 cells/ml for 24 h and Phenol red-serum free medium with DHE 10 µM was added for 45 min. After incubation at 37 °C, DHE stimulus was removed and cells were lysed with 10% Triton-X-100 solution in PBS and transferred to 96-well plates. Fluorescence was determined using a Synergy HT MultiDetection Microplate Reader (excitation 530 nm, emission 590 nm) (BioTek, VT, USA). Preliminary experiments confirmed that DHE staining accumulated in a linear way up to 120 min in control cybrid lines. DAPI (1 µg/ml) was added to solubilised cells incubated at 37 °C for 30 min and the fluorescence read (emission 360 nm, excitation 460 nm). Each sample was assessed in triplicate and results were expressed as the ratio between DHE and DAPI fluorescence units (f.u.). Mitochondrial free radical levels Dihydrorhodamine 123 (DHR; Sigma-Aldrich) is a fluorescence dye used to measure the level of mitochondrial free radical production. DHR is able to cross mitochondrial membranes and it is oxidized and transformed to Rhodamine 123 in the presence of ROS. Cybrid lines cultured in 10 cm dishes were washed and incubated for 40 min in the dark with Locke's solution (154 mM NaCl, 5.6 mM KCl, 3.6 mM NaHCO3, 2.3 mM CaCl2, 5.6 mM glucose, 5 mM HEPES, 1.2 mM MgCl2, pH 7.4) containing 10 µM DHR, then washed twice with Locke's solution without dye. Cells were scraped, centrifuged and the pellet was resuspended with 3 ml of 0.25 M NaOH. Rhodamine 123 fluorescence was quantified using a Cary Eclipse fluorimeter (excitation 488 nm, emission 525 nm) (Varian, CA, USA). Each sample was assessed in triplicate and results were expressed as DHR f.u./µg protein.

Fig. 1. The kinetics of sodium-dependent glutamate uptake in A3243G cybrid lines. High-affinity sodium-dependent glutamate uptake was determined using increasing concentrations of extracellular glutamate (2 to 240 µM) in control (CT) and A3243G (RS, E2) cybrid clones. Each point represents the mean ± SD of 4 separate experiments.

Protein oxidation detection (OxyBlot)

Statistical analysis

OxyBlot (Chemicon, CA, USA) was used to measure protein carbonylation level, an index of cellular oxidation. Cybrid lines at 80% confluence, growing in 10-cm dishes, were scraped and spun down. Cellular pellet was resuspended and sonicated in ice-cold solution of PBS, protease cocktail inhibitors (Sigma-Aldrich) and 2% βmercaptoethanol. Total protein solutions (15 µg for each cybrid line, measured by BCA assay) were allowed to react with 2,4-dinitrophenylhydrazine (DNP) for 15 min at 25 °C. Samples not DNP-derivatized were used as negative controls while positive controls were obtained by pre-treating cells with 2 mM H2O2 for 2 h (data not shown). Samples were then separated by 12% SDS polyacrilamide electrophoresis gel and proteins transferred to nitrocellulose membrane. Preliminary experiments with different protein concentrations were performed in order to assess the linearity of the signal. Blots were blocked with 3% milk solution in PBS, incubated with a rabbit antiDNP primary antibody followed by a goat HRP-conjugated anti-rabbit secondary antibody. Chemioluminescence signals were visualized on

Results were expressed as mean ± SD and analyzed with Instat software (GraphPad, CA, USA). Two-tailed Mann–Whitney U-test was used for comparison between groups. The Spearman test was used to correlate two sets of data. p values lower than 0.05 were considered significant.

Table 1 The kinetics of sodium-dependent glutamate uptake in control and A3243G cybrid lines Cell line

A3243G mutant load

CT E2#18 E2#23 RS#17 RS#3 RS#9

0 82.3 85.9 92.4 97.5 100

Vmax nmols/mg prot/30 min 10.2 ± 3.9 4.5 ± 1.3 3.7 ± 1.7 3.8 ± 0.9 4.7 ± 1.8 4.7 ± 1.1

Km µM

Vmax % respect control

127.9 ± 74.3 85.5 ± 37.9 89.6 ± 41.8 122.1 ± 20.4 68.7 ± 39.3 91.9 ± 9.8

100 44* 36** 37** 46* 46*

Vmax and Km values for each cell line were calculated from Eadie–Hofstee plots. Values represent the mean ± SD for n = 4 analyses. The Vmax of sodium-dependent glutamate uptake was significantly decreased in the A3243G cybrid lines (RS, E2) with respect to the control cybrid lines (CT; *p b 0.05, **p b 0.01), while the Km values were not significantly changed.

X-ray film using ChemiLucent detection system (Chemicon) and protein expression was evaluated by imaging densitometer. BCA protein assay Total protein concentration for every sample was analyzed with the BCA protein assay using BSA as a standard (Pierce, IL, USA).

Results Level of A3243G mtDNA mutation load The level of the A3243G mtDNA mutation was assessed in five mutant (E2#18, E2#23, RS#17, RS#3, RS#9) and three control cybrid clones (H2-1, H2-3, D2-2). As expected the control lines did not express the mutation while the A3243G clones expressed between 82.3 – 100% mutant mtDNA content (Table 1). Sodium-dependent glutamate transport in A3243G and control cybrid clones Control cybrids exhibited high-affinity, sodium-dependent glutamate uptake that fitted a hyperbolic kinetic curve increasing until plateau levels, according to time of exposure and concentration of radioactive glutamate (Fig. 1). This confirmed the glutamatergic transport properties of osteosarcoma cybrids that have been previously described in this cellular model (Beretta et al., 2004). All A3243G clones showed saturation of glutamate uptake, but at a rate significantly lower relative to controls (Fig. 1). Calculation of the glutamate uptake maximal rate (Vmax) and affinity (Km) values using the Eadie–Hofstee plots demonstrated a broadly similar decrease (54– 64%) in the Vmax of the A3243G cybrids compared to the controls. The Km values were not significantly different between mutant and control cybrids, suggesting that the affinity for glutamate uptake was similar in the A3243G and control lines (Table 1).

J.C. DiFrancesco et al. / Experimental Neurology 212 (2008) 152–156

In order to suggest a putative link between energy production and glutamate transport, we measured glutamate uptake in control cybrids, at the final concentration of 240 µM, after incubation with the complex I inhibitor rotenone (100 nM/15 h), showing a decrease of 48.6% relative to vehicle treated cells (p b 0.05, n = 3). Cybrid bio-energetic features Since glutamate uptake is influenced by cellular energy supply, we decided to investigate the ATP steady-state and the mitochondrial synthesis of ATP in the A3243G clones. There was no significant difference in the steady-state cellular ATP levels of the A3243G cybrids compared to controls (Fig. 2A). However, all the mutant clones showed a significant decrease in glutamate/malate dependent ATP synthesis with a reduction ranging from 67 to 91% (Fig. 2B) that was correlated to the A3243G mutant load (r = −0.44, p b 0.05). Free radical generation and oxidative damage in A3243G osteosarcoma cybrids Measurement of the oxidation of DHE showed an increase in free radical generation in paraquat treated cells (0.3 mM for 24 h) used as a positive control of ROS production; however there was no evidence for the enhancement of free radicals in the A3243G cybrids relative to the controls (Table 2). Mitochondrial ROS production was also assessed by measuring DHR fluorescence. There was no difference between the values for the A3243G and control cybrids (Table 2). Moreover, the assessment of protein carbonyls using the OxyBlot technique was used to assess oxidative damage of cellular proteins. The pattern of cross reactive material and their relative intensity was

Fig. 2. Bio-energetic status of the A3243G cybrids. (A) Cellular ATP levels in the A3243G clones (RS, E2) were not significantly different to those in the control cybrids (CT). (B) Mitochondrial ATP synthesis was markedly decreased in all the mutant clones in comparison with the control cybrids (⁎p b 0.05, ⁎⁎p b 0.01). Mutant loads for each cell line are stated in Table 1.

155

Table 2 Free radical generation in A3243G and control cybrids Cell line CT E2#18 E2#23 RS#17 RS#3 RS#9

DHE f.u./DAPI f.u.

DHR f.u./μg of protein

100 ± 17 91 ± 3 89 ± 15 93 ± 11 98 ± 4 91 ± 17

5.0 ± 0.8 6.9 ± 1.8 6.6 ± 2.2 5.7 ± 0.8 6.2 ± 0.3 4.6 ± 0.8

DHE and DHR were used to measure the cellular and mitochondrial production of free radical, respectively. There was no significant change in DHE or DHR fluorescence in the A3243G cybrids with respect to control lines (CT).

similar in the control and A3243G cybrid lines again, suggesting there was no increase of oxidative damage in the A3243G cybrid lines (data not shown). Discussion In this study we generated a series of stable cybrid lines carrying high levels of the A3243G mtDNA mutation, ranging from 82.3 to 100%. According to previous observations, the A3243G mtDNA mutation caused a reduction in mitochondrial ADP phosphorylation (Ciafaloni et al., 1992; Ichiki et al., 1988; King et al., 1992; Kariya et al., 2005; Rusanen et al., 2000), which inversely correlated with the level of the mutant load (Shepherd et al., 2006). There was no influence upon ATP steady-state levels in these cells under quiescent conditions, in agreement with a high glycolytic supplementation of ATP in cell cultures, which have been previously reported (James et al., 1999). All these clones showed a reduced capacity for glutamate uptake, measured as [3H]-glutamate influx through the EAATs. The Vmax was significantly reduced in all mutant cybrids compared to controls, while the Km of glutamate uptake was unchanged. Sodium-dependent glutamate uptake mediated by EAATs requires cellular ion gradients which are maintained by ATP-dependent activities. Thus, even if the presence of a direct link between the decrease of ATP synthesis and the reduction of glutamate uptake is missing, we could hypothesize that a decrease of energy supply might lead to reduced cellular and mitochondrial membrane potential, with an impairment of glutamate transporters' activity, as already previously hypothesized (James et al., 1999). Furthermore, the putative relationship between impaired oxidative phosphorylation and decreased glutamate transport might be supported by the decrease of glutamate uptake in control cells following incubation with rotenone, even if MELAS pathophysiology clearly encompasses mechanisms broader than the isolated complex I dysfunction. According to our results, the inhibition of glutamate uptake has been recently been reported in synaptosomes from mice following rotenone treatment (Hirata et al., 2008). Our finding of abnormal glutamate transport in association with the A3243G mutation may have particular relevance for some of the most important neurological features of MELAS, such as the strokelike episodes and the encephalopathy. A “weak excitotoxic hypothesis” has already been proposed to explain neuronal cell death in mitochondrial encephalomyopathies (Albin and Greenamyre, 1992; Sparaco et al., 1993): reduced glutamate uptake and the consequent enhanced extracellular concentration of the excitatory aminoacid outside of the cells could activate NMDA receptors affecting cytosolic Ca2+ homeostasis and contribute to premature cell death (Moudy et al., 1995; Ferrarese and Beil, 2004). Our results strength the hypothesis of the possible role played by glutamate excitotoxicity in the pathological conditions characterized by mitochondrial respiratory chain deficiency, such as MELAS. Interestingly, we previously demonstrated a reduction of glutamate uptake in LHON cybrids, suggesting that it may relate to an excess of mitochondria-derived free radicals, from the complex I defect in this disorder (Beretta et al., 2004; Wong et al., 2002). Even if

156

J.C. DiFrancesco et al. / Experimental Neurology 212 (2008) 152–156

MELAS and LHON syndromes appear to have pathogenetic aspects in common (Pulkes et al., 1999; Blakely et al., 2005), the biochemical consequences of the different mtDNA mutations may diverge in some respects. The various mitochondrial mutations linked to LHON (11778/ ND4, 3460/ND1 and 14484/ND6) all selectively target complex I with a consequent loss of function in addition to an increased production of free radicals which may be considered as a toxic gain of function (Beretta et al., 2004; Beretta et al., 2006; Wong et al., 2002). However, while the A3243G mutation predominantly affects complex I activity, in contrast to the LHON mutations, there is also an impairment of complex III and IV activities at high mutant loads (Yoneda et al., 1989) with no evidence of increased free radical generation. In fact, even in different cybrid lines harbouring high mutant loads of the A3243G mtDNA mutation causing decreased mitochondrial oxidative phosphorylation, we could not find any evidence of increased free radical production or oxidative damage using DHE, DHR and OxyBlot techniques. Our results suggest that cybrid models harbouring the A3243G mutation do not acquire a toxic gain of function resulting in an enhancement of free radicals production. This is in general agreement with previous studies showing no direct evidence of increased free radical generation in glioblastoma cybrids carrying the A3243G mutation (Sandhu et al., 2005), although these clones were more susceptible to pro-oxidant stress stimuli (Wong and Cortopassi, 1997). However, in contrast to these data, a reduced GSH/GSSG ratio (Pang et al., 2001) and increased activity of antioxidant enzymes (Rusanen et al., 2000) have been shown in cellular models harbouring the A3243G mutation, but a direct demonstration of higher ROS production has not been reported. Our results allow suggesting a putative role for excitotoxicity in MELAS. We might in fact hypothesize that in conditions of elevated energetic demand, a reduced ATP synthesis might result in a dysfunction of those homeostatic systems, strictly dependent on the energetic supply for they correct functioning, such as glutamate uptake. To date, no drug intervention has been demonstrated to be conclusively neuroprotective in mitochondrial encephalomyopathies. However, the high frequency of stroke-like episodes in MELAS suggests that the use of an effective anti-glutamatergic therapy may be worth evaluating, at least in those patients exhibiting the more severe phenotypes. It may be that such therapy could be more effective than in ischemic stroke where there is no underlying bioenergetic defect influencing glutamate uptake. The observations that mitochondrial respiratory chain dysfunction leads to decreased glutamate uptake has a relevance to the pathogenesis of neurodegenerative disorders associated with mitochondrial dysfunction, including both diseases with primary mtDNA mutations and those where the defect may be secondary. However, more studies are required to understand the mechanisms involved and their relevance to the pathogenesis, in particular the role of ATP synthesis, free radical generation and glutamate transporter regulation. This will also help understand the link between the biochemical effects of the A3243G mtDNA mutation and the MELAS phenotype. Acknowledgments This study was supported by the Parkinson's Disease Society (UK) and the Uzi Kattan Trust. References Albin, R.L., Greenamyre, J.T., 1992. Alternative excitotoxic hypotheses. Neurology 42, 733–738. Beretta, S., Mattavelli, L., Sala, G., Tremolizzo, L., Schapira, A.H., Martinuzzi, A., Carelli, V., Ferrarese, C., 2004. Leber hereditary optic neuropathy mtDNA mutations disrupt glutamate transport in cybrid cell lines. Brain 127, 2183–2192. Beretta, S., Wood, J.P., Derham, B., Sala, G., Tremolizzo, L., Ferrarese, C., Osborne, N.N., 2006. Partial mitochondrial complex I inhibition induces oxidative damage and perturbs glutamate transport in primary retinal cultures. Relevance to Leber Hereditary Optic Neuropathy (LHON). Neurobiol. Dis. 24, 308–317.

Blakely, E.L., de Silva, R., King, A., Schwarzer, V., Harrower, T., Dawidek, G., Turnbull, D.M., Taylor, R.W., 2005. LHON/MELAS overlap syndrome associated with a mitochondrial MTND1 gene mutation. Eur. J. Hum. Genet. 13, 623–627. Ciafaloni, E., Ricci, E., Shanske, S., Moraes, C.T., Silvestri, G., Hirano, M., Simonetti, S., Angelini, C., Donati, M.A., Garcia, C., 1992. MELAS: clinical features, biochemistry, and molecular genetics. Ann. Neurol. 31, 391–398. Danbolt, N.C., 2001. Glutamate uptake. Prog. Neurobiol. 65, 1–105. DiMauro, S., Schon, E.A., 2003. Mitochondrial respiratory-chain diseases. N. Engl. J. Med. 348, 2656–2668. El Meziane, A., Lehtinen, S.K., Hance, N., Nijtmans, L.G., Dunbar, D., Holt, I.J., Jacobs, H.T., 1998. A tRNA suppressor mutation in human mitochondria. Nat. Genet. 18, 350–353. Ferrarese, C., Biel, M.F., 2004. Excitotoxicity in neurological diseases, new therapeutic challenge. Kluwer Academic Publisher. Goto, Y., Nonaka, I., Horai, S., 1990. A mutation in the tRNA(Leu)(UUR) gene associated with the MELAS subgroup of mitochondrial encephalomyopathies. Nature 348, 651–653. Gu, M., Cooper, J.M., Taanman, J.W., Schapira, A.H., 1998. Mitochondrial DNA transmission of the mitochondrial defect in Parkinson's disease. Ann. Neurol. 44, 177–186. Helm, M., Florentz, C., Chomyn, A., Attardi, G., 1999. Search for differences in posttranscriptional modification patterns of mitochondrial DNA-encoded wild-type and mutant human tRNALys and tRNALeu(UUR). Nucleic Acids Res. 27, 756–763. Hirata, K., Akita, Y., Povalko, N., Nishioka, J., Yatsuga, S., Matsuishi, T., Koga, Y., 2008. Effect of l-arginine on synaptosomal mitochondrial function. Brain Dev. 30, 238–245. Ichiki, T., Tanaka, M., Nishikimi, M., Suzuki, H., Ozawa, T., Kobayashi, M., Wada, Y., 1988. Deficiency of subunits of Complex I and mitochondrial encephalomyopathy. Ann. Neurol. 23, 287–294. Iizuka, T., Sakai, F., 2005. Pathogenesis of stroke-like episodes in MELAS: analysis of neurovascular cellular mechanisms. Curr. Neurovasc. Res. 2, 29–45. James, A.M., Sheard, P.W., Wei, Y.H., Murphy, M.P., 1999. Decreased ATP synthesis is phenotypically expressed during increased energy demand in fibroblasts containing mitochondrial tRNA mutations. Eur. J. Biochem. 259, 462–469. Kariya, S., Hirano, M., Furiya, Y., Ueno, S., 2005. Effect of humanin on decreased ATP levels of human lymphocytes harboring A3243G mutant mitochondrial DNA. Neuropeptides 39, 97–101. King, M.P., Koga, Y., Davidson, M., Schon, E.A., 1992. Defects in mitochondrial protein synthesis and respiratory chain activity segregate with the tRNA(Leu(UUR)) mutation associated with mitochondrial myopathy, encephalopathy, lactic acidosis, and strokelike episodes. Mol. Cell. Biol. 12, 480–490. Korlipara, L.V., Cooper, J.M., Schapira, A.H., 2004. Differences in toxicity of the catecholO-methyl transferase inhibitors, tolcapone and entacapone to cultured human neuroblastoma cells. Neuropharmacology 46, 562–569. Liu, C.Y., Lee, C.F., Hong, C.H., Wei, Y.H., 2004. Mitochondrial DNA mutation and depletion increase the susceptibility of human cells to apoptosis. Ann. N.Y. Acad. Sci. 1011, 133–145. Moudy, A.M., Handran, S.D., Goldberg, M.P., Ruffin, N., Karl, I., Kranz-Eble, P., DeVivo, D.C., Rothman, S.M., 1995. Abnormal calcium homeostasis and mitochondrial polarization in a human encephalomyopathy. Proc. Natl. Acad. Sci. U. S. A 92, 729–733. O'Neill, C.M., Ball, S.G., Vaughan, P.F., 1994. Effects of ischaemic conditions on uptake of glutamate, aspartate, and noradrenaline by cell lines derived from the human nervous system. J. Neurochem. 63, 603–611. Pang, C.Y., Lee, H.C., Wei, Y.H., 2001. Enhanced oxidative damage in human cells harboring A3243G mutation of mitochondrial DNA: implication of oxidative stress in the pathogenesis of mitochondrial diabetes. Diabetes Res. Clin. Pract. 54 (Suppl 2), S45–S56. Pulkes, T., Eunson, L., Patterson, V., Siddiqui, A., Wood, N.W., Nelson, I.P., MorganHughes, J.A., Hanna, M.G., 1999. The mitochondrial DNA G13513A transition in ND5 is associated with a LHON/MELAS overlap syndrome and may be a frequent cause of MELAS. Ann. Neurol. 46, 916–919. Rusanen, H., Majamaa, K., Hassinen, I.E., 2000. Increased activities of antioxidant enzymes and decreased ATP concentration in cultured myoblasts with the 3243A–N G mutation in mitochondrial DNA. Biochim. Biophys. Acta 1500, 10–16. Sandhu, J.K., Sodja, C., McRae, K., Li, Y., Rippstein, P., Wei, Y.H., Lach, B., Lee, F., Bucurescu, S., Harper, M.E., Sikorska, M., 2005. Effects of nitric oxide donors on cybrids harbouring the mitochondrial myopathy, encephalopathy, lactic acidosis and stroke-like episodes (MELAS) A3243G mitochondrial DNA mutation. Biochem. J. 391, 191–202. Schapira, A.H., 2006. Mitochondrial disease. Lancet 368, 70–82. Shepherd, R.K., Checcarelli, N., Naini, A., De Vivo, D.C., DiMauro, S., Sue, C.M., 2006. Measurement of ATP production in mitochondrial disorders. J. Inherit. Metab. Dis. 29, 86–91. Sparaco, M., Bonilla, E., DiMauro, S., Powers, J.M., 1993. Neuropathology of mitochondrial encephalomyopathies due to mitochondrial DNA defects. J. Neuropathol. Exp. Neurol. 52, 1–10. Turner, L.F., Kaddoura, S., Harrington, D., Cooper, J.M., Poole-Wilson, P.A., Schapira, A.H., 1998. Mitochondrial DNA in idiopathic cardiomyopathy. Eur. Heart J. 19, 1725–1729. Wong, A., Cortopassi, G., 1997. mtDNA mutations confer cellular sensitivity to oxidant stress that is partially rescued by calcium depletion and cyclosporin A. Biochem. Biophys. Res. Commun. 239, 139–145. Wong, A., Cavelier, L., Collins-Schramm, H.E., Seldin, M.F., McGrogan, M., Savontaus, M. L., Cortopassi, G.A., 2002. Differentiation-specific effects of LHON mutations introduced into neuronal NT2 cells. Hum. Mol. Genet. 11, 431–438. Yoneda, M., Tanaka, M., Nishikimi, M., Suzuki, H., Tanaka, K., Nishizawa, M., Atsumi, T., Ohama, E., Horai, S., Ikuta, F., 1989. Pleiotropic molecular defects in energytransducing complexes in mitochondrial encephalomyopathy (MELAS). J. Neurol. Sci. 92, 143–158.