www.elsevier.com/locate/ynbdi Neurobiology of Disease 24 (2006) 308 – 317
Partial mitochondrial complex I inhibition induces oxidative damage and perturbs glutamate transport in primary retinal cultures. Relevance to Leber Hereditary Optic Neuropathy (LHON) Simone Beretta, a,⁎ John P.M. Wood, b Barry Derham, c Gessica Sala, a Lucio Tremolizzo, a Carlo Ferrarese, a and Neville N. Osborne b a
Department of Neuroscience and Biomedical Technologies, University of Milano-Bicocca, Via Cadore 48, 20052 Monza (MI), Italy Nuffield Laboratory of Ophthalmology, University of Oxford, Walton Street, Oxford OX2 6AW, UK c Department of Physiology, University of Oxford, Parks Road, Oxford OX1 3PT, UK b
Received 26 July 2005; revised 12 July 2006; accepted 16 July 2006 Available online 7 September 2006
Leber Hereditary Optic Neuropathy (LHON) is a maternally inherited form of visual loss, due to selective degeneration of retinal ganglion cells. Despite the established aetiological association between LHON and mitochondrial DNA mutations affecting complex I of the electron transport chain, the pathophysiology of this disorder remains obscure. Primary rat retinal cultures were exposed to increasing concentrations of rotenone to titrate complex I inhibition. Neural cells were more sensitive than Müller glial cells to rotenone toxicity. Rotenone induced an increase in mitochondrial-derived free radicals and lipid peroxidation. Sodium-dependent glutamate uptake, which is mostly mediated by the glutamate transporter GLAST expressed by Müller glial cells, was reduced dose-dependently by rotenone with no changes in GLAST expression. Our findings suggest that complex I-derived free radicals and disruption of glutamate transport might represent key elements for explaining the selective retinal ganglion cell death in LHON. © 2006 Elsevier Inc. All rights reserved. Keywords: Complex I; Leber hereditary optic neuropathy (LHON); Reactive oxygen species; Excitotoxicity; GLAST; Retina; Rotenone
Introduction The proton-translocating NADH-ubiquinone oxidoreductase (complex I) is the first enzyme complex of the electron transport chain (ETC). Complex I is one of the largest and most complicated enzyme systems known, consisting of more than 40 protein subunits, 7 of which are encoded by the mitochondrial genome (mtDNA) while the remainder originate from nuclear DNA (Walker, 1992). This enzyme is embedded in the mitochondrial
⁎ Corresponding author. Fax: +39 02 6448 8108. E-mail address:
[email protected] (S. Beretta). Available online on ScienceDirect (www.sciencedirect.com). 0969-9961/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.nbd.2006.07.016
inner membrane and catalyzes electron transport from NADH to ubiquinone, which is coupled to vectorial proton movements. Structural and functional defects in complex I are a characteristic of some mitochondrial diseases, which are associated with a wide spectrum of clinical phenotypes, preferentially affecting muscle and the central nervous system (McFarland et al., 2002; DiMauro and Hirano, 2005). Leber Hereditary Optic Neuropathy (LHON) is a maternally inherited neurodegenerative disease, clinically characterized by subacute, bilateral loss of central vision due to degeneration of retinal ganglion cells (RGCs) and their axons (Newman, 1998). The aetiology of LHON has been definitely linked to mtDNA mutations in genes encoding for subunits of complex I (Wallace et al., 1988; Chalmers and Schapira, 1999). Nevertheless, the pathophysiology of selective RGCs degeneration remains unknown (Man et al., 2002; Howell, 2003), mostly because methodological problems in manipulating the mitochondrial genome limit the development of proper experimental models for this disease (Wallace, 2002). In addition to the suggested alteration in electron transfer through complex I (Baracca et al., 2005; Carelli et al., 2004a,b), a body of evidence also indicates that an increased production of reactive oxygen species (ROS) might play a relevant role (Wong et al., 2002; Battisti et al., 2004; Floreani et al., 2005). Considering the exquisite sensitivity of RGCs to the toxicity mediated by NMDA glutamate receptors (Luo et al., 2001; Osborne et al., 2004), the involvement of excitotoxicity has also been advocated. A recent study from our group (Beretta et al., 2004) demonstrated that the three primary LHON mutations decreased the activity of the excitatory amino acid transporter-1 (EAAT1) in transmitochondrial cell lines. EAAT1 (GLAST in mice) is the major glutamate transporter in the mammalian retina and it is predominantly expressed by glial Müller cells (Harada et al., 1998; Rauen et al., 1998). An impaired function of EAAT1/GLAST in Müller cells would lead to abnormally elevated levels of extracellular glutamate, exposing RGCs to excitotoxicity. This mechanism might contribute to the selective degeneration of RGCs, consistently with
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317
the evidence that pharmacological inhibition of glutamate transport or administration of antisense oligodeoxynucleotides against EAAT1/GLAST cause glutamate-mediated toxicity to retinal ganglion cells in vivo (Vorwerk et al., 2000). Rotenone is the most potent member of the rotenoids, a family of isoflavonoids extracted from Leguminosae plants and since the earliest reports it has become the classical inhibitor of mitochondrial complex I, with a reported Ki of 4 nM in the standard assay of NADH-Q reductase (Degli Esposti, 1998). Rotenone inhibition is selective for complex I, non-competitive with endogenous ubiquinone, markedly time-dependent and is likely to involve two binding sites at the ND1 and ND4 subunits (Degli Esposti, 1998). Rotenone acts by antagonizing the semiquinone intermediate stabilized within the complex, blocking the reduction of Q by the electron transferred from NADH, through oxidoreduction of cluster N2 (Schuler et al., 1999). In the present study, we explored the oxidative stress–excitotoxicity interplay in primary cultures of retinal cells exposed to various concentrations of rotenone, in order to obtain a pharmacological modulation of complex I activity. This approach was aimed at assessing the effect of mild-versus-severe complex I inhibition in retinal neural and glial cells. Materials and methods Materials Modified Eagle’s medium (MEM) + Earl’s salts—glutamine, fetal calf serum, trypsin, gentamicin were supplied by Gibco (Paisley, UK). 3 L-[ H]glutamate (42.9 Ci/mmol) was obtained by Amersham (Little Chalfont, UK). L-(−)-threo-3-hydroxyaspartic acid, L-aspartate-βhydroxamate and dihydrokainic acid were obtained by Tocris (Bristol, UK). Guinea pig anti-EAAT1 polyclonal antibody were supplied by Chemicon (Temecula, CA, USA). Rabbit anti-HNE Michael adducts polyclonal antibody, which specifically recognizes chemically reduced amino acid-(4)-HNE adducts, and MnTBAP were obtained by Calbiochem (Nottingham, UK). All the other reagents and immunochemicals were supplied by Sigma (St. Louis, MO, USA). Primary rat retinal cell cultures (RCC)
309
with the primary antibodies against protein gene product (PGP) 9.5 (1:300), vimentin (1:50), or reduced HNE Michael adducts (1:300) in PBS-T. After incubation with the primary antiserum, coverslips were washed three times (5 min each) in PBS-T and immunolabelling was revealed with fluoroscein isothiocyanate (FITC)-linked secondary antibodies (1:100). In some experiments, coverslips were stained with standard hematoxylin–eosin staining. Visualisation was obtained using a fluorescence microscope (200×). Oxygen consumption studies Oxygen consumption in retinal cell suspension (0.5 million cells/mL) was measured polarographically with a Clark type oxygen electrode in a thermostatically controlled micro chamber (Instech, Plymouth Meeting, PA, USA). The solubility of oxygen in a air-satured medium at 37°C was taken to be 390 ng-atoms/mL. The rate of oxygen consumption was measured over 30 min and expressed as nmoles oxygen/million cells/hour. Rate of 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolimium (MTT) reduction Cell cultures were exposed to MTT (0.5 mg/mL final concentration) in standard medium for 20 min at 37°C in an atmosphere of 5% CO2 in air. After cell lysis with dimethylsulfoxide, MTT formazan was spectrophotometrically quantified (570 nm) in a microplate reader (Biorad, Hercules, CA, USA). The mean value of optical density (OD) of untreated cells was assumed to be 100% of cell viability and OD values of treated cells and standard deviations were proportionally converted. Superoxide levels Superoxide generation was imaged by use of the redox-sensitive, cell-permeable fluorophore dihydroethidium (HEt) (Bindokas et al., 1996). HEt is oxidized by superoxide to a novel product which binds to DNA enhancing nuclear fluorescence (Kalivendi et al., 2003). Cells were cultured on coverslips and treated with different concentrations of rotenone. Following treatment, culture medium was aspirated, cells were washed in Locke’s buffer and loaded in the dark with HEt (5 μM) for 30 min. After 3 washes in Locke’s buffer, cells were observed using a fluorescence microscope (Zeit, Hercules, USA).
Retinal cells were isolated from retinas of 1- to 5-day-old Wistar rats. All procedures were performed in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Briefly, rat pups were decapitated, the eyes were enucleated and the retinas were gently peeled off with fine forceps and placed in sterile Hanks’ balanced salt solution (HBSS). Retinas were collected and incubated at 37°C in 0.1 mg/ml trypsin for 10 min. Soybean trypsin inhibitor (type I-S, 0.6 mg/ml) was added to the tube to halt the enzyme reaction. After centrifugation (180 g, 5 min, 4°C), retinal cells were obtained by trituration in growth medium (MEM + Earle’s salts—L-glutamine, 10% foetal calf serum, 1.33 M D-glucose, 25 mM glutamine, 1% gentamicin). Cultures were maintained at 37°C in a humidified atmosphere of 5% CO2–95% air, and half the growth medium was replaced every 3 days. Cultures were used for the experiments at 6–9 days in vitro.
The dye dihydrorhodamine-123 (DRH) (Lievre et al., 2001) was used to quantify levels of mitochondrial ROS. Cells were exposed to 10 μM DHR in Locke’s buffer (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) for 30 min. Fluorescence was quantified using a Cary Eclipse fluorimeter (Varian, Palo Alto, USA) (excitation 488 nm, emission 525 nm) and related to total protein content. Protein content was assessed using the method of Bradford. For some experiments, cells were cultured on coverslips and observed using Radiance 2100 confocal microscope (Biorad, Hercules, USA) (excitation 488 nm, emission 515–540 nm).
Immunocytochemistry
Western blots
RCC were grown on glass coverslips and fixed in 4% paraformaldehyde for 30 min. Cells were incubated overnight
Cells were collected by centrifugation (80 × g/8 min/4°C) and were sonicated in 250 μl of homogenisation buffer (20 mM Tris–
Mitochondrial reactive oxygen species (ROS) levels
310
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317
HCl, pH 7.4, containing 2 mM EDTA, 0.5 mM EGTA, 1 mM dithiothreitol, 50 μg/ml leupeptin, 50 μg/ml pepstatin A, 50 μg/ml aprotinin and 0.1 mM phenylmethylsulphonyl fluoride). An equal volume of sample buffer (62.5 mM Tris–HCl, pH 7.4, containing 4% SDS, 10% glycerol, 10% β-mercaptoethanol and 0.002% bromophenol blue) was added and samples were boiled for 3 min. Electrophoresis of samples (~ 40 μg of proteins) was performed as reported previously (Wood and Osborne, 1997) using 10% polyacrylamide gels containing 0.1% SDS. Proteins were transferred to nitrocellulose and blots were stained as previously described (Wood and Osborne, 1997), for the presence of GLAST (1:500; Chemicon, Temecula, CA, USA), HNE Michael adducts (1:500; Calbiochem, Nottingham, UK) or β-actin (1:2000, Chemicon, Temecula, CA, USA). Electrophoretic transfer of proteins from gel to nitrocellulose was performed overnight, at 30 mA (electrodes 8 cm apart). The nitrocellulose was washed for 5 min in Tris buffered saline (TBS; 10 mM Tris–HCl, pH 7.4, containing 140 mM NaCl) and then for 5 min in TBS containing 0.1% Tween-20 (TBS-T). Antibodies for immunoprobing were diluted accordingly in TBS-T and the appropriate nitrocellulose blots were then incubated with antibody for 3 h at room temperature. After washing for 5 min in TBS-T, followed by 5 min in high saline Tris buffer (HST; 10 mM Tris–HCl, pH 7.4, containing 1 M NaCl and 0.5% Tween-20) and a subsequent 5 min in TBS-T, to remove unbound immunoglobulin and serum proteins, nitrocellulose blots were incubated for 90 min with goat secondary antibodies conjugated to horseradish peroxidase (1:5000 dilution in HST). Following washing in TBS-T (4 × 5 min), HST (10 min) and TBS-T (3 × 5 min), nitrocellulose blots were developed with a 0.016% (w/v) solution of 3-amino-9-ethylcarbazole (AEC) in 50 mM sodium acetate (pH 5.0) containing 0.05% (v/v) Tween-20 and 0.03% (v/v) H2O2. Colour reaction on the blots was stopped with 0.05% sodium azide solution. Protein expression was evaluated by imaging densitometer (Multi-Analyst software, Bio-Rad) and relative immunoreactivity calculated as ratio between target proteins and beta-actin.
by placing the culture plates in an ice bath and rapidly washing cells three times with ice-cold 0.32 M sucrose. Cells were immediately extracted with 0.25 M NaOH and the protein content determined by the method of Bradford. Cell extracts were placed in scintillation vials with 5 ml of scintillation liquid (Packard, Monza, Italy). The level of radioactivity was measured with a β-counter with 60% efficiency (Beckman Coulter, High Wycombe, UK). The specific activity of tritiated glutamate was used to convert disintegrations per minute (dpm) values into the corresponding number of nanomoles. Glutamate uptake rates are expressed as nanomoles/mg of cell proteins/10 min. Studies of inhibition of glutamate uptake High-affinity sodium-dependent glutamate uptake was studied, as previously described, after treatment with unselective and selective glutamate transport inhibitors: L-(−)-threo-3-hydroxyaspartic acid (THA), an unselective glutamate transporter inhibitor; L-aspartate-β-hydroxamate (LABH), a EAAC1/EAAT3 selective inhibitor at low dose and EAAC1/EAAT3 + GLAST/EAAT1 inhibitor at high dose; dihydrokainic acid (DHK), an EAAT2selective inhibitor (see Results for Ki of single inhibitors). Details of the pharmacological properties of the glutamate uptake inhibitors used in this study are discussed by Bridges et al. (1999). Statistical analysis All results are expressed as mean ± standard deviation. One-way ANOVA analysis of variance, followed by Tukey’s multiple comparison test, was used to assess the significance of differences among values of different treatments. In a subset of experiments, Two-way ANOVA analysis of variance was performed to examine the main effects of different treatments and the interactions between them. Results
High-affinity sodium-dependent glutamate uptake
Characterization of retinal cell cultures
RCC were washed with HEPES-buffered saline (10 mM HEPES, 135 mM choline chloride, 5 mM KCl, 0.6 mM MgSO4·7H2O, 2.5 mM CaCl2, 6 mM D-glucose, pH 7.4) and preincubated for 15 min in HEPES-buffered saline with 135 mM NaCl or choline chloride, to study the sodium dependency of glutamate uptake, in a 37°C shaker (O’Neill et al., 1994). Uptake was started with the addition of glutamate-L-[3H]glutamate mixture (40:1), at concentrations ranging from 1 to 200 μM, and stopped after 10 min
Primary dissociated cultures of mixed rat retinal cells (retinal cell cultures, RCC) were used for the experiments at 6–9 days in vitro. The morphological and immunocytochemical analyses revealed that different cell types grew in RCC under our experimental conditions, as previously described (Beale et al., 1982). The general aspect of RCC was of clusters of neural cells growing on a uniform layer of large glial cells (Fig. 1A). Among the variety of retinal neurons growing in RCC, amacrine cells (PGP9.5 positive, Fig. 1B)
Fig. 1. Characterization of RCC. (A) Hematoxylin–eosin staining of RCC reveals a homogenous Müller glial cell layer over which retinal neural cells grow in clusters. (B) PGP9.5-immunocytochemical labelling shows that amacrine cells represent a major subpopulation of retinal neurons in RCC. (C) Vimentin-positive Müller glial cells are the almost exclusive glial subtype in RCC.
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317
311
represented the predominant subpopulation. On the contrary, RGCs were barely present in RCC, due to their specific requirements for being cultured in vitro (Luo et al., 2001). Müller glial cells (vimentin-positive, Fig. 1C) appeared to be the almost exclusive glial cell type, even though astrocytes and microglial cells were occasionally detected (data not shown). Titrating mitochondrial complex I inhibition in RCC A pharmacological modulation of mitochondrial complex I activity was obtained by exposure to selected concentrations of rotenone, within the known range of its inhibitory action (Degli Esposti, 1998). In consideration of the time-dependency of rotenone-induced inhibition, oxygen consumption was measured 1 h after the incubation with this agent. As shown in Fig. 2, rotenone-induced inhibition of cell respiration followed a hyperbolic shape, with about 12% of measurable cell respiration left at rotenone 100 nM. Effect of rotenone on the viability of retinal neural and glial cells RCC cultures were exposed to rotenone (1 to 100 nM) and cell viability was assessed after 48 h by the MTT assay. As shown in Fig. 3A, rotenone is dose-dependently toxic to RCC (25% cell loss at 10 nM, 70% at 100 nM). Under the same experimental conditions, this effect was immunocytochemically assessed on the neural and glial cell populations, using anti-PGP9.5 and antivimentin antibodies, respectively (Fig. 3B). Rotenone produced no toxicity at the concentration of 1 nM, which corresponds roughly to a 25% inhibition of cell respiration. A selective vulnerability of neurons, compared to Müller glial cells, was observed at rotenone 10 nM (60% respiratory inhibition). Conversely, 100 nM rotenone (90% respiratory inhibition) detrimentally affected the viability of both neural and glial cells. As shown in Fig. 3C and Table 1, rotenone (100 nM) toxicity was partially counteracted by coadministration of the water-soluble vitamin E analogue trolox (50 μM) or the AMPA/kainate glutamate receptor antagonist CNQX (50 μM). On the contrary, no protective effect was observed with the NMDA glutamate receptor antagonist MK-801 (5 μM), consistently with our preliminary experiments showing a high sensitivity of amacrine cells to glutamate toxicity mediated by
Fig. 2. Oxygen consumption in RCC exposed to complex I inhibition by rotenone. RCC were treated with various concentrations of rotenone (1–50 nM) for 1 h before measurement of cell respiration. Cell respiration was expressed as the rate of oxygen consumption per million of cells. Data were obtained from three separate experiments.
Fig. 3. Effect of rotenone on the viability of neural and glial retinal cells in culture. (A) RCC exposed to rotenone (1–100 nM) for 48 h showed a dosedependent decrease in cell viability compared to untreated cultures, assessed with the MTT assay. Data were obtained from twelve separate experiments. F(3, 47) = 97.99, **p < 0.01 compared to untreated. ***p < 0.001 compared to untreated. (B) The selective effect of rotenone (1–100 nM, 48 h) on the viability of neural and glial cells was tested by counting the number of PGP9.5 and vimentin-immunoreactive cells, respectively (see text). Five microscopic fields per coverslips were manually counted in six independent experiments. F(3, 23) = 157.2 for PGP9.5-IR cells. F(3, 23) = 213.5 for vimentin-IR cells. **p < 0.01 compared to untreated cells. ***p < 0.001 compared to untreated cells. (C) The decrease in cell viability caused by rotenone (100 nM, 48 h) in RCC was partially counteracted by simultaneous addition of the vitamin E analogue trolox or the AMPA/kainate glutamate receptor antagonist CNQX, but not by the NMDA glutamate receptor antagonist MK-801. Data were obtained by eight independent experiments. See Table 1 for 2-way ANOVA analysis. ***p < 0.001 compared to rotenone 100 nM.
312
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317
Table 1 Two-way ANOVA related to data represented in Fig. 3C Source
df
Sum of squares
Mean square
F
p
Corrected model 7 47,106.438 6729.491 932.801 < 0.001 Intercept 1 273,461.021 273,461.021 37,905.488 < 0.001 Rotenone 1 27,122.500 27,122.500 3759.557 < 0.001 MK 801 1 16.500 16.500 2.291 = 0.136 CNQX 1 1176.125 1176.125 163.027 < 0.001 Trolox 1 1164.031 1164.031 161.351 < 0.001 Rotenone * MK 1 0.031 0.031 0.004 = 0.948 Rotenone * CNQX 1 1225.125 1225.125 169.819 < 0.001 Rotenone * Trolox 1 1164.031 1164.031 161.351 < 0.001 Error 56 404.000 7.214 Total 64 391,786.000 R squared = 0.991
transporters GLAST/EAAT1 and GLT/EAAT2 was demonstrated in RCC extracts, whereas no signal was detected for the neuronal transporter EAAC/EAAT3 (data not shown). High-affinity, sodium-dependent glutamate transport was functionally assessed using radioactive glutamate. This allowed the construction of a hyperbolic kinetic curve describing transporter-mediated glutamate uptake (Km = 37.24 μM; Vmax = 70.66 nmoles/mg prot/10 min) (Fig. 6A). Exposure of RCC to the unselective glutamate transporter inhibitor THA (200 μM, Ki = 15–60 μM for EAAT1–3) completely abolished the sodium-dependent component of the glutamate uptake, while the partially selective inhibitor LABH (130–500 μM; Ki = 369 μM for GLAST; Ki = 133 μM for EAAC)
AMPA/kainate receptors, but not by NMDA receptors (unpublished data). Effect of rotenone on mitochondria-derived oxidative stress and lipid peroxidation Mitochondrial ROS production was quantitatively assessed with the fluorochromes DHR and HEt (Fig. 4). Incubation of RCC with rotenone (1–100 nM) for 6 h led to a dose-dependent increase in the production of ROS from mitochondrial sources (Fig. 4A), evident even at the lowest concentration of 1 nM. The fluorescence of DHR and HEt was also qualitatively imaged in RCC exposed to rotenone. The mitochondrially sequestered, oxidation-sensitive dye DHR (Dugan et al., 1995) showed increased fluorescence in the perinuclear mitochondria-rich regions of cells in rotenone-treated RCC compared to untreated cells (Figs. 4B–C). In particular, Müller glial cells showed a clear enhancement of the punctuate perinuclear labelling of DHR (Figs. 4B–C insert). Rotenone-treated RCC also showed a marked increase in the fluorescence of nuclei after incubation with HEt (Figs. 4D–E). This dye is oxidized to ethidium in the presence of superoxide anions and fluoresces on intercalation into DNA (Bindokas et al., 1996). 4-hydroxynonenal (HNE) is a reactive aldehyde that results from ROS-induced damage to lipids, a process termed lipid peroxidation. HNE was chosen as an index of rotenone-induced lipid peroxidation for its known ability to form adducts with a variety of cellular proteins (Nakashima et al., 2003), included glutamate transporters (Lauderback et al., 2001; Springer et al., 1997). An antibody which specifically recognizes the reduced Michael adducts formed by HNE was used for these experiments (see Methods). RCC exposed to rotenone (1–100 nM) for 6 h showed a dose-dependent increase in the intensity of multiple HNE-immunoreactive bands, scattered through the whole gel but especially evident in the lower molecular weight range (Fig. 5A). Immunocytochemical localization of HNE adducts revealed an increased fluorescence, apparently located in the cytosol and on the plasma membrane, in both neurons and Müller glial cells (Figs. 5B–C) treated with rotenone. Effect of rotenone on glutamate transport The expression of glutamate transporters GLAST/EAAT1, GLT/EAAT2 and EAAC/EAAT3 were investigated in RCC by Western blotting experiments. Expression of the glial glutamate
Fig. 4. Rotenone induces mitochondrial production of reactive oxygen species (ROS). (A) Mitochondrial ROS production, assessed with the fluorochrome DHR (expressed as rhodamine-123 fluorescence units/μg cell protein), in RCC treated with rotenone (1–100 nM) for 6 h. Data were obtained from seven separate experiments. F(3, 27) = 49.82, *p < 0.05 compared to untreated. ***p < 0.001 compared to untreated. DHR-related fluorescence was also imaged in RCC using a fluorescence microscope (200×) to further confirm the mitochondrial source of ROS production, as it is shown for untreated cultures (B) and cells exposed to rotenone 10 nM (C). Inserts in panels B and C show retinal glial cells at higher magnification (400×). Intracellular levels of superoxide, assessed with the fluorochrome HEt, are imaged as fluorescent nuclei from untreated cells (D) and cells treated with rotenone 10 nM (E) using a fluorescence microscope (200×).
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317
Fig. 5. Rotenone promotes the generation of 4-hydroxynonenal (HNE) adducts. RCC extracts of untreated and rotenone treated (1–100 nM, 6 h) cultures were analyzed by Western blot (A), using a primary antibody against reduced HNE-Michael adducts. A typical experiment is shown out of four independent experiments. The HNE-Michael adducts immunoreactive signal (upper panel) was normalized to the beta-actin signal (lower panel). The histogram represents the relative immunoreactivity for untreated and rotenone-treated cells, expressed as ratio between HNE-Michael adducts and beta-actin optical density (OD). F(3, 15) = 42.83. **p < 0.01 compared to untreated. ***p < 0.001 compared to untreated. Immunocytochemical detection (200×) of anti-HNE adducts for untreated cultures (B) and cells exposed to rotenone 10 nM (C) is shown. Inserts in panels B and C show retinal glial cells at higher magnification (400×). Abbreviations: M = molecular weight marker; C = untreated control cells.
reduced this process dose-dependently, with no effect at 130 μM and a 50% inhibition at 500 μM. The selective EAAT2 inhibitor DHK (250 μM; Ki = 20–80 μM for GLT) reduced the glutamate uptake rate by 21%. The results from these inhibition studies are consistent with a major functional role of GLAST/EAAT1 in retinal cells in culture. Glutamate uptake [10 μM] was also measured after exposure to rotenone (1–100 nM) for 6 h. Rotenone
313
Fig. 6. Glutamate uptake studies in rotenone-treated RCC. (A) RCC display a kinetics of sodium-dependent glutamate uptake (tested at 1–200 μM), consistent with a transporter-mediated type of glutamate transport. For each point of the cumulative curve, the data derived from five separate experiments were averaged. (B) Preincubation with MnTBAP and trolox prevented the rotenone-induced (100 nM, 6 h) decrease in glutamate uptake (10 μM). Data were obtained from six separate experiments. See Table 2 for 2-way ANOVA analysis. ***p < 0.001 compared to untreated cultures. °°°p < 0.001 compared to cultures treated with rotenone 100 nM. (C) A typical Western blot experiment, out of 4 independent experiments, showing expression of the glutamate transporter GLAST in RCC exposed to rotenone (1–100 nM, 6 h). The GLAST immunoreactive signal was normalized to the beta-actin signal. Histogram represents the relative immunoreactivity of untreated and rotenone-treated cells, expressed as ratio between GLAST and beta-actin optical density (OD).
314
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317
Table 2 Two-way ANOVA related to data represented in Fig. 6B Source
df
Corrected model 5 Intercept 1 Rotenone 1 MnTBAP 1 Trolox 1 Rotenone * MnTBAP 1 Rotenone * Trolox 1 Error 30 Total 36 R squared = 0.991
Sum of squares
Mean square
859.314 171,863 6773.501 6,773,501 421.268 421,268 41.213 41,213 131.602 131,602 55.358 55,358 127.882 127,882 7.502 0,250 7690.953
F
p
687,260 27,086,479 1684,602 164,805 526,260 221,372 511,385
< 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001 < 0.001
caused a significant dose-dependent decrease in the glutamate uptake rate compared to that measured in untreated cells (28% decrease at 1 nM, 46% at 10 nM and 75% at 100 nM; p < 0.001 using One-way ANOVA). As shown in Fig. 6B and Table 2, the detrimental effect of rotenone on glutamate transport was significantly counteracted when the mitochondrial superoxide dismutase (mtSOD) analogue MnTBAP (40 μM) or trolox (50 μM) were added to culture medium 1 h before exposure to rotenone. The rotenone-induced impairment of GLAST activity was not paralleled by a decrease in the total GLAST protein content, which did not show significant changes when assessed by Western blot experiments (Fig. 6C). Discussion Systemic inheritance of mtDNA point mutations affecting subunits of complex I is the established cause of LHON. Nevertheless, the pathogenetic pathways that link mtDNA mutations to the strikingly selective LHON neurodegeneration remain unclear. The issue of the selective vulnerability of RGCs to disturbances of complex I function might be addressed looking at the effect of reduced complex I activity in the retina. Most studies in the field of LHON pathogenesis have been performed using peripheral tissues from patients or transmitochondrial cell lines (reviewed by Carelli et al., 2004b). Albeit useful, the main disadvantage of these studies is that they do not explore the actual consequences of LHON mutations on retinal tissues. The present study investigated the effects of titrated complex I inhibition in dissociated mixed retinal cell cultures. A gradual modulation of complex I activity was obtained by exposure to increasing concentrations of rotenone under controlled conditions, in order to compare the effects of a mild-versus-severe inhibition. Our findings indicate that even a partial inhibition of complex I activity is sufficient to promote significant overproduction of mitochondrial-derived ROS in retinal cells and perturbation of glutamate transport mediated by Müller glial cells. A stronger inhibition of complex I activity induced a selective toxicity in retinal neurons, whereas Müller glial cells remained viable, but dose-dependently impaired. The neuron-glia crosstalk in the retina might be crucial in the pathogenetic mechanisms of LHON, as occurs in other neurodegenerative disorders (Bezzi et al., 2001; Rao and Weiss, 2004). Under this view, assessing a differential response to complex I inhibition characteristic of specific retinal cell types might be of
considerable importance for understanding the basis of selective degeneration in LHON. Unfortunately, a major limitation of retinal cultures is that RGCs need special requirements for being maintained in vitro and are only scarcely present in the mixed retinal cultures (RCC). Previously reported cultures of purified RGCs have been obtained using a complex panning procedure, resulting in the loss of all the other retinal cell types (Kawasaki et al., 2000). We chose RCC for this study because they represent a simple, well-characterized culture model, obtained directly from the whole retinal tissue of neonatal animals (Beale et al., 1982). The mixed rat retinal cultures used in the present study have been previously characterised to contain small spheroid cells (defined as small neurons including photoreceptors), larger process-bearing cells (larger retinal interneurons and ganglion cells) and flattened cells (Müller glial cells) after 7 days in culture (Beale et al., 1982). The major neuron type is the amacrine cell. It is noteworthy that differences in energetic metabolism have been demonstrated between neural and glial cells in the retina (Wood et al., 2005; Winkler et al., 2000), similarly to what occurs in the brain (Deitmer, 2001). While retinal neurons are strictly dependent on oxidative phosphorylation for energy production and survival, aerobic glycolysis appear to be the major pathway of glucose metabolism in Müller glial cells (Winkler et al., 2000). According to this metabolic profile, a more detrimental effect of mitochondrial dysfunction has to be expected in neurons compared to glial cells. On the other hand, if key functions physiologically exerted by glial cells were even mildly impaired by disturbed mitochondrial function, this could results in further harmful consequences for the nearby neurons. The reuptake of neurotransmitters is a well characterized function of glial cells, efficiently performed by Müller glial cells in the neuroretina. The impairment of glutamate reuptake observed in RCC after complex I inhibition may generate an excitatory overstimulation for the bioenergetically challenged RGCs, potentially acting as an excitotoxic insult. Our findings indicate that RCC are particularly sensitive to complex I inhibition. A selective toxicity to neural cells was observed with the intermediate concentration of rotenone, while both neural and glial cells were affected by complete complex I inhibition. This indicates that retinal neural cells are actually more prone to cell death induced by perturbed mitochondrial homeostasis than retinal glial cells. Interestingly, rotenone toxicity was partially counteracted by the administration of trolox or CNQX. The latter finding suggests that oxidative stress and excitotoxicity might be involved in the process of rotenoneinduced cell death. Rotenone promotes a local increase of ROS production originating from perturbed mitochondria, with a number of potentially harmful down-stream consequences for both neural and glial cells. Indeed, the pathogenicity of complex I defects is thought to be caused not only by the impairment of oxidative phosphorylation, but also by mitochondrial-derived ROS generated by the dysfunctional ETC. In particular, the superoxide anion is physiologically produced as a result of electrons leaking from the ETC and reacting inappropriately with oxygen (Brand et al., 2004). A recent study from Kudin et al. (2004) estimated that approximately 1% of respiratory chain electron transport is redirected to superoxide formation in isolated mitochondria under resting state conditions. Growing evidence points to complex I as the most important source of superoxide production by intact mammalian mitochondria, particularly in the brain (Brand et al.,
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317
2004; Kudin et al., 2004). In fact, the relative contributions of the different complexes of the ETC to total superoxide production may depend on the tissue examined and on the presence of specific inhibitors (Barja, 1999). Superoxide production has been reported to increase in the presence of the complex I inhibitor rotenone when mitochondria oxidize NAD-linked substrates (Barrientos and Moraes, 1999; Li et al., 2003), which are the primary source of reducing equivalents to ETC. Despite not being toxic, the lowest concentration of rotenone tested (1 nM) was sufficient to promote the generation of mitochondrial-derived oxidative stress as shown by the increased production of superoxide, mitochondrial ROS and HNE adducts. The intracellular generation of oxidative stress was induced dosedependently by rotenone in RCC, and apparently no gross differences were observed between neural and glial cells. Our data point to a principal role of superoxide in rotenone toxicity in RCC, which is generated by partially inhibited complex I. Our group recently reported a detrimental effect of primary LHON mutations on EAAT1/GLAST-mediated glutamate transport, using transmitochondrial cell lines (Beretta et al., 2004). The experiments performed in RCC provide further evidence that disturbances of mitochondrial complex I function affect glutamate transport in the retina. Rotenone induced a dose-dependent impairment of glutamate uptake in RCC, not associated with a decrease in total GLAST protein content and efficiently prevented when retinal cells where pre-incubated with antioxidants. These findings might be hypothetically ascribed to an oxidative modification of the transporter, according to the well-characterized sensitivity of glutamate transport activity to the intracellular oxidative state and the formation of peroxides (Miralles et al., 2001; Sitar et al., 1999; Trotti et al., 1998). Thus, EAAT1/GLAST may reasonably be within the number of the cellular proteins which bear the consequences of the perturbed redox homeostasis associated with complex I inhibition. The superoxide generation in complex I defects is of great interest for human diseases, in particular for LHON. Interestingly, recent reports showed that mtSOD-deficient animals develop optic neuropathy (Sandbach et al., 2001; Qi et al., 2003). Moreover, mice with partial deficiency of the mtSOD develop chronic mitochondrial oxidative stress and show an age-related impairment of glial glutamate transporters (Liang and Patel, 2004). The interplay among complex I dysfunction, mitochondrial-derived ROS and glutamate transport may occur with peculiar characteristics in the retina, due to its unique cytoarchitecture and physiology. The experimental model used in the present study allowed to demonstrate that even a mild inhibition of complex I activity is sufficient to promote ROS-associated damage and derangement of retinal excitatory neurotransmission. A number of physiological and biochemical features of RCGs might help to explain their unique vulnerability in LHON. Axons from RGCs were described to form varicosities rich in mitochondria in the unmyelinated part of the optic nerve, anterior to the lamina cribrosa (Andrews et al., 1999). The increased number of mitochondria at this site co-localizes with a high density of voltage-gated sodium channels (Barron et al., 2004). This is likely to reflect the functional needs of the optic nerve axons, which require greater energy to restore their action potential, because of their unmyelinated axons within the globe (Carelli et al., 2004b). This relatively high energy demand represents a bioenergetic disadvantage compared to other retinal cells. This could explain why optic neuropathies occur in primary mitochondrial diseases
315
such as LHON, Leigh’s syndrome, and myoclonic epilepsy with ragged red fibers (MERRF) and in the secondary mitochondrial dysfunctions described in dominant optic atrophy (DOA), Friedreich’s ataxia, tobacco alcohol amblyopia and other disorders (Carelli et al., 2002). Moreover, the asymmetrical distribution of mitochondria along the optic nerve axons is dependent on an efficient axonal transport (Hollenbeck, 1996), which is itself an ATP-dependent process (Sabri and Ochs, 1972). Interestingly, mutations in the OPA1 gene, encoding for a dynamin-related GTPase involved in mitochondrial structural organization, has been linked to DOA (Delettre et al., 2000). Nonetheless, the cascade of events leading to RGCs death in LHON is likely to be complex and the experimental data accumulated so far indicate that a pure energy-depletion theory is unsatisfactory. An element is needed which explains why other high energy-dependent retinal cells, such as photoreceptors or retinal pigment epithelium, are usually spared. Notably, the sensitivity to excitotoxicity is remarkably higher in RGCs compared to other retinal cell types (Luo et al., 2001; Vorwerk et al., 2000) and experimental evidence exists that neurons exposed to a defective energy status become vulnerable even to normally sublethal concentrations of glutamate (Nicholls, 2004). Further experimental evidence is needed to support a role for excitotoxicity in the pathogenesis of LHON, particularly in consideration of the paucity of extra-retinal manifestations in this disease. Nonetheless, it is tempting to speculate that glutamate toxicity could represent a key element to explain the selectivity of the neurodegeneration in LHON.
References Andrews, R.M., Griffiths, P.G., Johnson, M.A., Turnbull, D.M., 1999. Histochemical localisation of mitochondrial enzyme activity in human optic nerve and retina. Br. J. Ophthalmol. 83 (2), 231–235. Baracca, A., Solaini, G., Sgarbi, G., Lenaz, G., Baruzzi, A., Schapira, A.H., Martinuzzi, A., Carelli, V., 2005. Severe impairment of complex I-driven adenosine triphosphate synthesis in leber hereditary optic neuropathy cybrids. Arch. Neurol. 62, 730–736. Barja, G., 1999. Mitochondrial oxygen radical generation and leak: sites of production in states 4 and 3, organ specificity, and relation to aging and longevity. J. Bioenerg. Biomembr. 31, 347–366. Barrientos, A., Moraes, C., 1999. Titrating the effect of mitochondrial complex I impairment in the cell physiology. J. Biol. Chem. 274, 16188–16197. Barron, M.J., Griffiths, P., Turnbull, D.M., Bates, D., Nichols, P., 2004. The distributions of mitochondria and sodium channels reflect the specific energy requirements and conduction properties of the human optic nerve head. Br. J. Ophthalmol. 88, 286–290. Battisti, C., Formichi, P., Cardaioli, E., Bianchi, S., Mangiavacchi, P., Tripodi, S.A., Tosi, P., Federico, A., 2004. Cell response to oxidative stress induced apoptosis in patients with Leber’s hereditary optic neuropathy. J. Neurol., Neurosurg. Psychiatry 75, 1731–1736. Beale, R., Neuhoff, N.R., Osborne, N.N., 1982. The characterisation of cell cultures of neonatal retina. In: Clayton, R.M. (Ed.), Problems of Normal and Genetically Abnormal Retinas. Academic Press, Oxford, pp. 9–18. 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. Bezzi, P., Domercq, M., Vesce, S., Volterra, A., 2001. Neuron-astrocyte cross-talk during synaptic transmission: physiological and neuropathological implications. Prog. Brain Res. 132, 255–265.
316
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317
Bindokas, V.P., Jordan, J., Lee, C.C., Miller, R.J., 1996. Superoxide production in rat hippocampal neurons: selective imaging with hydroethidine. J. Neurosci. 16, 1324–1336. Brand, M.D., Affouritt, C., Esteves, T.C., Green, K., Lambert, A., Miwa, S., Pakay, J.L., Parker, N., 2004. Mitochondrial superoxide: production, biological effects and activation of uncoupling proteins. Free Radical Biol. Med. 37, 755–767. Bridges, R.J., Kavanaugh, M.P., Chamberlin, A.R., 1999. A pharmacological review of competitive inhibitors and substrates of high-affinity, sodium-dependent glutamate transport in the central nervous system. Curr. Pharm. Des. 5, 363–379. Carelli, V., Ross-Cisneros, F.N., Sadun, A.A., 2002. Optic nerve degeneration and mitochondrial dysfunction: genetic and acquired optic neuropathies. Neurochem. Int. 40, 573–584. Carelli, V., Rugolo, M., Sgarbi, G., Ghelli, A., Zanna, C., Baracca, A., Lenaz, G., Napoli, E., Martinuzzi, A., Solaini, G., 2004a. Bioenergetics shapes cellular death pathways in Leber’s hereditary optic neuropathy: a model of mitochondrial neurodegeneration. Biochim. Biophys. Acta 1658, 172–179. Carelli, V., Ross-Cisneros, F.N., Sadun, A.A., 2004b. Mitochondrial dysfunction as a cause of optic neuropathies. Prog. Retinal Eye Res. 23, 53–89. Chalmers, R.M., Schapira, A.H., 1999. Clinical, biochemical and molecular genetic features of Leber’s hereditary optic neuropathy. Biochim. Biophys. Acta 1410, 147–158. Degli Esposti, M., 1998. Inhibitors of NADH-ubiquinone reductase: an overview. Biochim. Biophys. Acta 1364, 222–235. Deitmer, J.W., 2001. Strategies for metabolic exchange between glial cells and neurons. Respir. Physiol. 129, 71–81. Delettre, C., Lenaers, G., Griffoin, J.M., Gigarel, N., Lorenzo, C., Belenguer, P., Pelloquin, L., Grosgeorge, J., Turc-Carel, C., Perret, E., Astarie-Dequeker, C., Lasquellec, L., Arnaud, B., Ducommun, B., Kaplan, J., Hamel, C.P., 2000. Nuclear gene OPA1, encoding a mitochondrial dynamin-related protein, is mutated in dominant optic atrophy. Nat. Genet. 26, 207–210. DiMauro, S., Hirano, M., 2005. Mitochondrial encephalomyopathies: an update. Neuromuscul. Disord. 15, 276–286. Dugan, L.L., Sensi, S.L., Canzoniero, L.M., Handran, S.D., Rothman, S.M., Lin, T.S., Goldberg, M.P., Choi, D.W., 1995. Mitochondrial production of reactive oxygen species in cortical neurons following exposure to N-methyl-D-aspartate. J. Neurosci. 15, 6377–6388. Floreani, M., Napoli, E., Martinuzzi, A., Pantano, G., De Riva, V., Trevisan, R., Bisetto, E., Valente, L., Carelli, V., Dabbeni-Sala, F., 2005. Antioxidant defences in cybrids harboring mtDNA mutations associated with Leber’s hereditary optic neuropathy. FEBS J. 272, 1124–1135. Harada, T., Harada, C., Watanabe, M., Inoue, Y., Sakagawa, T., Nakayama, N., Sasaki, S., Okuyama, S., Watase, K., Wada, K., Tanaka, K., 1998. Functions of the two glutamate transporters GLAST and GLT-1 in the retina. Proc. Natl. Acad. Sci. U. S. A. 95, 4663–4666. Hollenbeck, P.J., 1996. The pattern and mechanism of mitochondrial transport in axons. Front. Biosci. 1, d91–d102. Howell, N., 2003. LHON and other optic nerve atrophies: the mitochondrial connection. Dev. Ophthalmol. 37, 94–108. Kalivendi, S.V., Kotamraju, S., Cunningham, S., Shang, T., Hillard, C.J., Kalyanaraman, B., 2003. 1-Methyl-4-phenylpyridinium (MPP+)induced apoptosis and mitochondrial oxidant generation: role of transferrin-receptor-dependent iron and hydrogen peroxide. Biochem. J. 371, 151–164. Kawasaki, A., Otori, Y., Barnstable, C.J., 2000. Muller cell protection of rat retinal ganglion cells from glutamate and nitric oxide neurotoxicity. Invest. Ophthalmol. Visual Sci. 41, 3444–3450. Kudin, A.P., Bimpong-Buta, N.Y., Vielhaber, S., Elger, C.E., Kunz, W.S., 2004. Characterization of superoxide-producing sites in isolated brain mitochondria. J. Biol. Chem. 279, 4127–4135. Lauderback, C.M., Hackett, J.M., Huang, F.F., Keller, J.N., Szweda, L.I., Markesbery, W.R., Butterfield, D.A., 2001. The glial glutamate transporter, GLT-1, is oxidatively modified by 4-hydroxy-2-nonenal in
the Alzheimer’s disease brain: the role of Abeta1-42. J. Neurochem. 78, 413–416. Li, N., Raghed, K., Lawler, G., Sturgis, J., Rajwa, B., Melendez, J.A., Robinson, J.P., 2003. Mitochondrial complex I inhibitor rotenone induces apoptosis through enhancing mitochondrial reactive oxygen species production. J. Biol. Chem. 278, 8516–8525. Liang, L.P., Patel, M., 2004. Mitochondrial oxidative stress and increased seizure susceptibility in Sod2(−/+) mice. Free. Radical Biol. Med. 36 (5), 542–554. Lievre, V., Becuwe, P., Bianchi, A., Bossenmeyer-Pourie, C., Koziel, V., Franck, P., Nicolas, M.B., Dauca, M., Vert, P., Daval, J.L., 2001. Intracellular generation of free radicals and modifications of detoxifying enzymes in cultured neurons from the developing rat forebrain in response to transient hypoxia. Neuroscience 105, 287–297. Luo, X., Heidinger, V., Picaud, S., Lambrou, G., Dreyfus, H., Sahel, J., Hicks, D., 2001. Selective excitotoxic degeneration of adult pig retinal ganglion cells in vitro. Invest. Ophthalmol. Visual Sci. 42, 1096–1106. Man, P.Y., Turnbull, D.M., Chinnery, P.F., 2002. Leber hereditary optic neuropathy. J. Med. Genet. 39, 162–169. McFarland, R., Taylor, R.W., Turnbull, D.M., 2002. The neurology of mitochondrial DNA disease. Lancet Neurol. 1, 343–351. Miralles, V.J., Martinez-Lopez, I., Zaragoza, R., Borras, E., Garcia, C., Pallardo, F.V., Vina, J.R., 2001. Na+ dependent glutamate transporters (EAAT1, EAAT2, and EAAT3) in primary astrocyte cultures: effect of oxidative stress. Brain. Res. 922, 21–29. Nakashima, I., Liu, W., Akhand, A.A., Takeda, K., Kawamoto, Y., Kato, M., Suzuki, H., 2003. 4-hydroxynonenal triggers multistep signal transduction cascades for suppression of cellular functions. Mol. Aspects Med. 24, 231–238. Newman, N.J., 1998. Leber’s optic neuropathy. In: Miller, N.R., Newman, N.J. (Eds.), Walsh and Hoyt’s Clinical Neuro-Ophthalmology. Williams and Wilkins, Baltimore, pp. 742–753. Nicholls, D.G., 2004. Mitochondrial dysfunction and glutamate excitotoxicity studied in primary neuronal cultures. Curr. Mol. Med. 4, 149–177. O’Neill, C.M., Ball, S.G., Vaughan, P.F.T., 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. Osborne, N.N., Casson, R.J., Wood, J.P., Chidlow, G., Graham, M., Melena, J., 2004. Retinal ischemia: mechanisms of damage and potential therapeutic strategies. Prog. Retinal Eye. Res. 23, 91–147. Qi, X., Lewin, A.S., Hauswirth, W.W., Guy, J., 2003. Optic neuropathy induced by reductions in mitochondrial superoxide dismutase. Invest. Ophthalmol. Visual Sci. 44, 1088–1096. Rao, S.D., Weiss, J.H., 2004. Excitotoxic and oxidative cross-talk between motor neurons and glia in ALS pathogenesis. Trends Neurosci. 27, 17–23. Rauen, T., Taylor, W.R., Kuhlbrodt, K., Wiessner, M., 1998. High-affinity glutamate transporters in the rat retina: a major role of the glial glutamate transporter GLAST-1 in transmitter clearance. Cell Tissue Res. 291, 19–31. Sabri, M.I., Ochs, S., 1972. Relation of ATP and creatine phosphate to fast axoplasmic transport in mammalian nerve. J. Neurochem. 19, 2821–2828. Sandbach, J.M., Coscun, P.E., Grossniklaus, H.E., Kokoszka, J.E., Newman, N.J., Wallace, D.C., 2001. Ocular pathology in mitochondrial superoxide dismutase (Sod2)-deficient mice. Invest. Ophthalmol. Visual Sci. 42, 2172–2178. Schuler, F., Yano, T., Di Bernardo, S., Yagi, T., Yankovskaya, V., Singer, T.P., Casina, J.E., 1999. NADH-quinone oxidoreductase: PSST subunit couples electron transfer from iron–sulfur cluster N2 to quinone. Proc. Natl. Acad. Sci. U. S. A. 96, 4149–4153. Sitar, S.M., Hanifi-Moghaddam, P., Gelb, A., Cechetto, D.F., Siushansian, R., Wilson, J.X., 1999. Propofol prevents peroxide-induced inhibition of glutamate transport in cultured astrocytes. Anesthesiology 90, 1446–1453. Springer, J.E., Azbill, R.D., Mark, R.J., Begley, J.G., Waeg, G., Mattson, M.P., 1997. 4-hydroxynonenal, a lipid peroxidation product, rapidly
S. Beretta et al. / Neurobiology of Disease 24 (2006) 308–317 accumulates following traumatic spinal cord injury and inhibits glutamate uptake. J. Neurochem. 68, 2469–2476. Trotti, D., Danbolt, N.C., Volterra, A., 1998. Glutamate transporters are oxidant-vulnerable: a molecular link between oxidative and excitotoxic neurodegeneration? Trends Pharmacol. Sci. 19, 328–334. Vorwerk, C.K., Naskar, R., Schuettauf, F., Quinto, K., Zurakowski, D., Gochenauer, G., Robinson, M.B., Mackler, S.A., Dreyer, E.B., 2000. Depression of retinal glutamate transporter function leads to elevated intravitreal glutamate levels and ganglion cell death. Invest. Ophthalmol. Visual Sci. 41, 3615–3621. Walker, J.E., 1992. The NADH:ubiquinone oxidoreductase (complex I) of respiratory chains. Q. Rev. Biophys. 25, 253–324. Wallace, D.C., 2002. Animal models for mitochondrial disease. Methods Mol. Biol. 197, 3–54. Wallace, D.C., Singh, G., Lott, M.T., Hodge, J.A., Schurr, T.G., Lezza, A.M., Elsas II, L.J., Nikoskelainen, E.K., 1988. Mitochondrial DNA mutation
317
associated with Leber’s hereditary optic neuropathy. Science 242, 1427–1430. Winkler, B.S., Arnold, M.J., Brassell, M.A., Puro, D.G., 2000. Energy metabolism in human retinal Muller cells. Invest. Ophthalmol. Visual Sci. 41, 3183–3190. 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. Wood, J., Osborne, N.N., 1997. Induction of apoptosis in cultured human retinal pigmented epithelial cells: the effect of protein kinase C activation and inhibition. Neurochem. Int. 31, 261–274. Wood, J.P., Chidlow, G., Graham, M., Osborne, N.N., 2005. Energy substrate requirements for survival of rat retinal cells in culture: the importance of glucose and monocarboxylates. J. Neurochem. 93, 686–697.
