Www-user.tu-chemnitz.de

ALS and other motor neuron disorders 2003 4(Suppl 1), 142–160# 2003 ALS and other motor neuron disorders. All rights reserved. ISSN 1466-0822 expressing G93A-SOD1 (G93A-glia) by standard immuno- chemical methods; NF-kB activity was determined by EMSA.
Ferri A1,2, Cozzolino M1, Casciati A1, Ferraro E3, We have demonstrated that high-level, transient expression ofseveral mutant FALS-SOD1s (A4V, G37R, G85R, G93A, 1Centro di Neurobiologia Sperimentale ‘‘Mondino-Tor Vergata-S.Lucia’’, Rome, I113T) induces programmed cell death in ETNA cells. This Italy; 2Ist. di Neuroscienze del CNR, Sez. Psicobiologia e Psicofarmacologia, PCD is mitochondria- and Apaf-1-dependent, but AIF- Rome, Italy; 3Dipartimento di Biologia, Universita` di Roma ‘‘Tor Vergata’’, Rome, independent. Induction of mitochondria damage follows NO-dependent modulation of signalling pathways. Furtherexperiments are in progress to further elucidate the details of E-mail address for correspondence: carri@Bio.uniroma2.it Activation of various apoptosis-related molecules has been Our data confirm that mitochondria play a pivotal role in reported to be involved in selective motor neuron death that FALS-SOD1-induced cell death, acting as a target of the occurs in human ALS and in animal and cellular models of noxious function of mutant SOD1s and as an effector of the SOD1-associated ALS.1 We have previously demonstrated that a deleterious interplay between neuronal and glial cells isnecessary for the pathogenesis of the disease. In co-cultureconditions, activation of apoptotic pathways in human neuroblastoma cells expressing G93A-SOD1 is dependenton activation of inflammatory processes and NO production 1. Gue´gan C, Przedborski S. Programmed cell death in amyotrophic in G93A-glioblastoma cells stimulated by G93A-neuroblas- lateral sclerosis. J Clin Invest 2003; 111: 153–161.
toma.2 We have also observed3 that the link between 2. Ferri A, Nencini M, Casciati A et al. Cell death in amyotrophic pathogenic properties of FALS-SOD1s and the activation of lateral sclerosis: interplay between neuronal and glial cells the apoptotic cascade is represented by Apaf-1 (apoptotic protease-activating factor-1), a molecular adaptor coupling 3. Cozzolino M, Ferraro E, Ferri A et al. Apoptosome deficiency mitochondria-released cytochrome c to downstream activa- prevents neuronal loss induced by mutant human SOD1s in vivo tion of executioner caspases. However, a complete picture of the mechanisms underlying programmed cell death (PCD) inALS is still lacking.
The aim of this study was to analyse in detail the role of mitochondria damage and of several mediators of signal transduction in the Apaf-1-dependent apoptotic cascade PATIENTS AND IN SOD1-MUTATEDNEUROBLASTOMA CELL LINES Bazzini E1, Alimonti D1, Cereda C1, Corato M1,2,Ceroni M1,2,3 To study the role of mitochondria and Apaf-1 in mutantFALS-SOD1s-induced apoptosis, we have used embryonic telencefalic cell lines derived from Apaf1x/x and z/z mice Neurological Institute IRCCS ‘‘C. Mondino’’, Pavia, Italy; 2Department of Neurological Science, Pavia University, Pavia, Italy; 3Policlinico of Monza, (ETNAx/x and ETNAz/z, respectively) where we have assessed apoptosis-related signalling events, such as activationof caspases, cytochrome c release, mitochondrial metabolismusing standard biochemical procedures and specific fluores- cence probes. NO signalling was studied in co-cultures ofhuman SH-SY5Y neuroblastoma cells expressing G93A-SOD1 Abnormal SOD1 mRNAs were identified in brain tissue of (G93A-neuro) with human U373 MG glioblastoma cells sporadic amyotrophic lateral sclerosis patients (SALS) as well as in neuronal and non-neuronal tissues from a subject with no neurological disease (Akihiro Kawata et al. 2000).
Other different transcripts of SOD1 gene were foundin various tissues of SOD1 mutated patients (M. Hirano et al. 2000). In summary, five alternative splicing transcriptsof the SOD1 gene were described. We have studied the Bunte H1, Sakharov D1, van Muiswinkel FL2, Ba¨r PR2, SOD1 mRNAs in the ALS cellular model SH-SY5Y neuro- blastoma line carrying the G93A, L84F, H46R mutations andin the wild type line. We have also considered lymphocytes from fresh blood of ten SALS patients’ lymphocytes non- Department of Biochemistry of Lipids, Institute of Biomembranes, Utrecht University, Utrecht, The Netherlands; 2Department of Experimental Neurology, stimulated and stimulated with H2O2 and from healthy R. Magnus Institute of Neuroscience, University Medical Centre, Utrecht, The E-mail address for correspondence: h.bunte@chem.uu.nl In this cell line the extraction of total RNA was followed bya retrotranscription reaction using oligo dT primers; then by a PCR, using two specific external primers (the forward in Neuronal cells cannot proliferate and therefore it is very exon 1 and the reverse primer in exon 5), we have amplified important that these cells are well protected against any the entire codifying sequence (cds). With another pair of damage caused. Protection of neuronal cells against the primers we have also divided the cds into three parts and the overproduction of reactive oxygen species (ROS) occurs by an amplifications by PCR. The abnormal transcript was auto-matically sequenced by dye terminator technology. The same effective anti-oxidant defence system. However, the genera- analysis was carried out in lymphocytes from fresh blood of tion of oxidative stress is suggested to play an important role ten SALS patients and controls. An agent causing oxidative in the selective degeneration of motor neurons in amyo- trophic lateral sclerosis (ALS). Evidence supporting the role of 2O2) was added in primary culture of lymphocytes at different concentrations (25, 50, 100 mM). Cells were oxidants in ALS has been provided by the finding that about collected at different time intervals (2, 4, 8, 16, 24, 36 20% of patients suffering from familiar ALS (10% of all cases) have mutations in the gene encoding for the anti-oxidantenzyme Cu,Zn-superoxide dismutase (SOD1). Investigationshave led to the model in which mutant SOD1 is somehowcytotoxic to motor neurons, the so-called toxic gain-of- function model, instead of the expected loss-of-function In all cell line samples, a unique band corresponding to the model. There is also evidence for a toxic role of ROS in entire cds, (about 500 bp) was detected, while in the case of patients with sporadic ALS. Neuronal damage caused by the G93A line another band that differs from the coding oxidants affects DNA, lipids and proteins and eventually leads sequence (between 200 and 300 bp) was present. This abnormal band is a fragment of the cds lacking the initial The aim of this project is 1) to establish a model which part. Sequence data showed that the band is a mRNA lacking allows monitoring of ongoing protein oxidation in cultured a part of exon 1. Aberrant mRNA was not found in non- motor neurons, and 2) to investigate the relationship between stimulated lymphocytes of SALS patients. We found different mutant SOD1 expression, protein oxidation, and the vulner- aberrant transcripts produced by alternative splicing in ability of G93A-expressing motor neurons to ROS inducing primary culture of lymphocytes from ALS patients, following the stimulation. All of the abnormal transcripts are smallerthan the entire cds.
Primary motor neuron cultures derived from transgenic mice overexpressing human mutant SOD1 (G93A mutation) Probably the aberrant transcript, found in the G93A and non-transgenics are used as a model. Cells are cultured at 6% neuroblastoma line, depends on the type of mutation. In O2 and treated with several stressors, including H2O2 and NOC- fact the G93A amino acid substitution, unlike the other two 18 (a NO-donor). To detect protein oxidation, cells are incu- mutations, causes loss of enzyme activity because it is located bated with acetylTyrFluo. This is a fluorescein-labelled tyramine in the active enzymatic site. All of these abnormal transcripts, (tyrosine analogue) that, upon oxidation, is converted into a found in stimulated patient’ lymphocytes, depends on tyrosyl radical, which is able to cross-link with oxidised tyrosine oxidative stimulus. Oxidative stress or ‘severe’ mutation of residues. As a result, oxidised proteins become fluorescently the SOD1 gene seems to increase transcription of aberrant labelled and can be visualised and identified by fluorescence SOD1 mRNAs smaller than the entire cds.
microscopy and 2D-electrophoresis, respectively.
Western blot analysis revealed a basal signal in non-treated, The pharmacological effect of the two drugs were assessed non-transgenic motor neurons. In addition, a concentration- in term of cytoprotective effect and prevention of reactive dependent increase in protein oxidation was observed upon oxygen species (ROS) generation. Human neuroblastoma cultures were exposed to MTT (0.5 mg/mL) for 40 minutes 2O2 (0–300 mM) and NOC-18 (0–50 mM) treatment. Com- parison of the immuno-stained blot with the overall protein at 37‡C; subsequently cells were lysed in DMSO and pattern shows that specific proteins are affected.
formazan was spectrophotometrically quantified (570 nm).
The dye 2’,7’-dichlorofluorescein diacetate (DCF-DA) wasused to quantify levels of whole-cells ROS. Cells wereincubated for 45 minutes in Locke’s buffer containing 10 mM DCF-DA, washed two times, collected and fluorescencewas fluorimetrically quantified (excitation 488 nm, emission Our experiments show that acetylTyrFluo is a useful probe 525 nm), after resuspension of the cell pellet in Locke’s buffer to monitor ongoing protein oxidation in primary cultured mouse motor neurons, either under basal conditions orexperimentally induced oxidative stress. While the subcellularlocalisation of oxidised proteins will be visualised by Confocal Laser Scanning Microscopy, 2D-electrophoresis Our findings indicate that human SH-SY5Y cells do not will be used to identify the oxidised proteins. Using the display any susceptibility to excitotoxic cell death. However, established model, we are currently investigating the protein this cell line undergoes apoptosis when exposed to oxidative oxidation pattern in motor neurons derived from mice stress generated by addition of hydrogen peroxide. The carrying the G93A mutant SOD1 gene and wild type controls.
cytoprotective effect shown by riluzole against hydrogenperoxide toxicity cannot be mediated by its anti-glutamateproperties, but it is expression of a direct antioxidant effect,as further confirmed by the preventive effect of riluzole on whole-cell ROS production. Interestingly, riluzole exerted a significant effect on oxidative stress at the lowest concentra- tions tested (0.5–1 mM), which are the same as those presentin the CSF of ALS patients chronically treated with the drug.
On the other hand, alpha tocopherol showed only a modestantioxidative effect at 0.1 mM, which corresponds to the CSF levels reached by this drug during a high-dose supplementa-tion. At the highest concentrations, alpha tocopherol (10 mM)resulted in an almost complete protective effect, which had 1Department of Neuroscience and Biomedical Technologies, University of not occurred with riluzole (10–30 mM).
Milano-Bicocca, San Gerardo Hospital, Monza The therapeutic use of most antioxidants is limited since The relative efficacy of riluzole in the treatment of they do not cross the blood-brain barrier after systemic amyotrophic lateral sclerosis (ALS), compared to other anti- administration. Our findings suggest that riluzole is likely glutamatergic drugs, points to other mechanisms of action for to exert an antioxidant effect in the central nervous system this agent, in addition to its anti-excitotoxic properties. In the of ALS patients, when chronically administered following present study, riluzole was compared to the biological the usual clinical schedule. Also, despite being a weaker antioxidant alpha tocopherol in human SH-SY5Y neuroblas- antioxidant compared to alpha tocopherol in absolute toma cells undergoing hydrogen peroxide-induced oxidative terms, riluzole reaches CSF concentrations that are more cell death. The concentrations tested included the actual efficacious in counteracting oxidative stress than those cerebrospinal fluid levels reached by each drug in human subjects, treated according to the usual clinical schedule.
NEUROTOXICITY THROUGH AREDUCTION OF FREE RADICALS Koh S-H1,2, Kim SH1, Yu H-J3, Kim M4, Lee K-W4,Jung HK2 Kim HJ1, Cho WM1, Park JH1, Hong YH2, Kim JM2,Sung JJ2, Kim MH2, Lee KW2 1Department of Neurology, College of Medicine, Hanyang University, Seoul,Korea; 2Department of General Toxicology, National Institute of Toxicological 1Department of Neurology, and Clinical Research Institute, Seoul National Research, KFDA, Seoul, Korea; 3Department of Neurology, Pundang Jaesaeng University Hospital, Seoul, Korea; 2Department of Neurology, Seoul National Hospital, Seoul, Korea; 4Department of Neurology, College of Medicine, Seoul Amyotrophic lateral sclerosis (ALS) is a neurodegenerative The effects of epigallocatechin gallate (EGCG) on the disorder characterized by the progressive loss of motor phosphoinositide 3-kinase (PI3K)/Akt and glycogen synthase neurons leading to weakness and death. Mutations in kinase-3 (GSK-3) pathway during oxidative stress induced the Cu,Zn-superoxide dismutase (SOD1) gene are respon- injury were studied using H2O2 treated VSC4.1 cells (motor sible for familial ALS (FALS). Although mutant SOD1- induced motor neuronal death is still unknown, alteredzinc and copper binding capacity within the SOD1 proteinmay be related to the toxic gain of function. The purpose of our study is to investigate the toxic effect of copper/zincon the SOD1 mutation and evaluate the protective Following 100 mM H2O2 exposure, the viability of VSC4.1 effect of co-factors of mitochondrial enzymes against this cells (EGCG or z-VAD-fmk pretreated vs. not pretreated) was evaluated by using the MTT assay. Additionally expression ofcytochrome c, caspase-3, poly (ADP-ribose) polymerase(PARP), PI3K/Akt, and GSK-3 were examined using Western Motor neuron-neuroblastoma hybrid (VSC4.1) cells consti-tutively expressing human SOD1 with mutations (G93A, A4V), or wild-type (W.T) were treated with copper or zinc.
Viability was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5- EGCG or z-VAD-fmk pretreated VSC4.1 cells showed an diphenyltetrazolium bromide (MTT) assay. To verify the increased viability compared to untreated VSC4.1 cells, and characteristics of neuronal cell death, cells treated with pretreatment of VSC4.1 cells with either agent induced dose- copper or zinc were fixed and stained with Hoechst 33342.
dependent inhibition of caspase-3 activation and PARP Caspase inhibitor (Z-VAD-FMK) was co-treated with copper cleavage. However, inhibition of cytochrome c release was or zinc in each cell group. To examine the exposure of only detected in EGCG pretreated cells. Upon examination of copper/zinc induced free radical formation, intracellular the PI3K/Akt and GSK-3 upstream pathway, Western blots of concentration of peroxides was measured by flow cytometry EGCG pretreated cells showed decreased immunoreactivity of the fluorescence emitted by DCFH oxidation. Free radical (IR) of Akt and GSK-3 and increased IR of p85a PI3K, scavenger (trolox) was preincubated before treatment with phosphorylated Akt, and phosphorylated GSK-3. In contrast, copper or zinc. Other co-factors of mitochondrial enzymes no changes were seen in z-VAD-fmk pretreated cells.
(pyruvate, thiamine or lipoic acid) were co-incubated withcopper or zinc to investigate protective effects. After treatmentof co-factors, the expressions of Bax and PARP cleavage were These results show that EGCG affects the PI3K/Akt, GSK-3pathway as well as downstream signaling, including the cytochrome c and caspase-3 pathways. Therefore, it issuggested that EGCG mediated activation of PI3K/Akt and Copper/zinc decreased viability and increased the endo- inhibition of GSK-3 could be a new potential therapeutic genous levels of peroxides in the cells expressing mutant strategy for motor neuron diseases associated with oxidative SOD1. In copper/zinc treated cells, staining with Hoechst 33342 showed nuclear condensation and fragmentation, however this toxicity was attenuated by caspase inhibitor (Z- little investigation, in spite of their much shared pathogenic VAD-FMK) or free radical scavenger (Trolox). Among mechanism. We previously showed that HC induces selective the various co-factors of mitochondrial enzymes, thiamine cytotoxicity to SOD1 mutants (A4V and marginally to G93A).
and lipoic acid were ineffective, whereas pyruvate signifi- Thus we investigated the effect of HC on the intracellular cantly increased the viability of cells expressing mutant SOD1.
calcium concentration and oxidative protein injury in motor In addition, pyruvate reduced bax expression and PARP neuronal cell-line (VSC4.1) transfected with human SOD1 of wild-type or mutants (WT, G93A, and A4V respectively).
These results indicate that pyruvate, which acts as hydroxyl After establishing WT, G93A and A4V cell-line by transfection radical scavenger or energy supplier, may protect motor of SOD1s, expression of SOD1s was confirmed by the neuron degeneration in FALS with SOD1 mutation. Low Western blotting assay. We measured the intracellular calcium toxicity and ability to cross the blood-brain barrier could be concentration of WT, G93A, and A4V before and after the of therapeutic value in some forms of FALS.
treatment of 10 mM HC for various time durations (1 to 24hours) by calculating the ratio of emissions at 360 and380 nm after loading with Fura2-acetoxymethyl (Fura2-AM)for 20 mins. We determined the production of intracellular generation of peroxides using 2’,7’-dichlorofluorescin diace- tate (DCF-DA) in WT, G93A, and A4V cells responding to the HC (100 mM, 10 mM)-treatment. We adopted Western blot-ting using the anti-nitrotyrosine antibody for assaying the protein nitrosylation and using the anti-2,4-dinitrophenylhy- drazine (DNPH) antibody after incubation of HC (100 mM,10 mM)-treated whole-cell extraction with DNPH for assayingthe protein carbonylation.
Sung JJ2, Kim HJ1, Hong YH1, Kim NH1, Cung YM1,Min JH1, Kim M1, Lee KW1 1Department of Neurology, and Clinical Research Institute, and NeuroscienceResearch Institute in SNU Medical Research Center, Seoul National University HC induced no change of intracellular calcium concentration College of Medicine, 28 Yongon-Dong, Jongno-Gu, Seoul 110-744, Korea; in each cell group until the 24-hrs HC-treatment, as well as 2Department of Neurology, Seoul Municipal Boramae Hospital and College of no difference of calcium content between WT and mutants. In Medicine, Seoul National University, 395 Sindaebang-Dong, Dongjak-Gu, Seoul the determination of intracellular generation of peroxides, the peroxide contents increased gradually along the increase ofHC-incubation time in all cell groups, but the significantincrease was observed in the HC-treated mutants compared with the HC-treated WT. Both protein nitrosylation andcarbonylation increased in the mutants compared with WT, Mutations in Cu,Zn-superoxide dismutase (SOD1) cause and HC increased the protein nitrosylation and carbonyla- ,20% of cases of familial amyotrophic lateral sclerosis tion. Interestingly, we observed the increase of protein (ALS). Mutant SOD1 triggers the disease through dominant carbonylation and nitrosylation in the transfected SOD1.
cytotoxic gain-of-functions including increased excitotoxic,and oxidative damage, increased nitrosylation and proteinaggregation. Late-onset motor neuron degeneration despite the innate presence of mutant SOD1 has no clear explana-tion. Parkinson’s disease and dementia were reported to have Here we show that the HC-induced cytotoxicity is not an association with hyperhomocysteinemia. A potential mediated by the excitotoxicity, but by oxidative cytotoxicity mechanism for cytotoxicity of homocysteine (HC) includes including the protein nitrosylation and carbonylation, at least autoxidation of the thiol group leading to the production of in this cell-line model. The mutated SOD1 damage and the reactive oxygen species mediated by copper, and excitotoxicity increased protein nitrosylation with the absence of excito- mediated by reaction of HC as an agonist of the glutamate toxicity implicate that the loss-of-function theory should be receptor. Relationships between HC and ALS have received included in familial ALS pathogenesis.
toxicity of copper/zinc. However, HSP 70 significantly increased the cell death induced by copper in G93A without affecting formation of peroxides. Increased toxicity was attenuated byantioxidant (trolox), caspase inhibitor (Z-VAD-FMK) or calpain inhibitor (calpeptin). A more protective tendency was observed in calpain inhibitor than in caspase inhibitor.
Kim HJ1, Cho WM1, Park JH1, Hong YH1, Sung JJ2,Kim MH, Lee KW2 Our data show that the protective effect of HSP 70 againstcopper was inhibited by mutation of SOD1 (G93A). Thisprocess seemed to induce activation of calpain, which results 1Department of Neurology, and Clinical Research Institute, Seoul National in motor neuronal cell death. These findings suggest that University Hospital, Seoul, Korea; 2Department of Neurology, Seoul National dysfunction of HSP 70 through mutant SOD1 may be a factor Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disorder characterized by the progressive loss of motor neurons leading to paralysis and death. Mutations of the human Cu,Zn superoxide dismutase (SOD1) are found insome cases of familial ALS. Recent evidence suggests thatmutant SOD1s bind to and sequester heat-shock proteins Park KS1, Choi WJ2, Kim HJ3, Kim NH2, Kim MH2, (HSPs). This sequestration by mutant SOD1 decreases the neuroprotective effects of HSP against a variety of insults. Inaddition, normal antioxidant function of SOD1 may be 1Department of Neurology, Seoul Paik Hospital, Inje University, , Seoul Korea; affected, thereby promoting oxidative stress. The purpose of 2Department of Neurology, Seoul National University Hospital, Seoul, Korea; our study is to investigate the effect of HSP70 on the toxic 3Department of Neurology and Clinical Research Institute, Seoul National gain of function of mutant SOD1 through altered copper and To test the effect of HSP70 on constitutively expressing mutant SOD1, motor neuron-neuroblastoma hybrid cells (VSC4.1) co- (SOD1) gene have been identified in some cases of familial expressing HSP70 and human mutant SOD1 (G93A, A4V) were amyotrophic lateral sclerosis (FALS). Many studies have compared. These cells were treated with copper or zinc, and the demonstrated that SOD1 mutation promotes neural death viability was determined by 3-(4,5-dimethylthiazol-2-yl)-2,5- at the intracellular level. The aim of this study was to diphenyltetrazolium bromide (MTT) assay. In order to determine investigate whether SOD1 mutant motor neurons have an the features of neuronal cell death, we co-incubated each cell extracellular effect that may affect the viability of adjacent group with caspase inhibitor (Z-VAD-FMK) or calpain inhibitor (calpeptin, ALLM, ALLN) during the treatment of these metalions. Cells treated with copper or zinc were fixed and stained withHoechst 33342. To confirm that exposure of copper/zinc inducedfree radical formation, production of peroxides was monitored by flow cytometry using DCFH-DA. Analog of vitamin E (trolox)was preincubated before addition of copper/zinc.
We developed motor neuron-neuroblastoma hybrid cells(VSC 4.1), which constitutively expressed a mutant (G93A)or wild type (WT) SOD1. For co-culture of the different kinds of cell lines, VSC 4.1 cells were grown and differ-entiated with G93A or WT cells. For co-culture of the same Extracellular addition of copper/zinc increased the endogenous kinds of cell lines, VSC 4.1, G93A or WT cells were grown and levels of peroxides and death of cells expressing mutant SOD1 differentiated under the same conditions. Cell viability was (G93A, A4V). Staining with Hoechst 33342 showed that in determined using MTT assay. To investigate the mechanism of copper/zinc-treated cells, the chromatin was dispersed into viability changes, z-VAD-fmk (caspase inhibitor), trolox multiple small nuclear fragments. HSP 70 reduced the produc- (antioxidant), kynurenic acid (glutamate receptor blocker), tion of peroxides and increased the viability in A4V against the 2-amino-5-phosphonovaleric acid (APV; NMDA receptor blocker), or 6-cyano-7-nitroquidoxaline-2, 3-dione (CNQX; enzyme. Although a number of hypotheses exist, including AMPA/Kainate receptor blocker) were added to the culture aberrant free radical handling, metal toxicity and protein aggregation, the exact mechanism(s) of selective motorneuron degeneration have not been precisely defined. Usingthe motor neuron-like NSC34 cell line, transfected with normal or mutant SOD1 protein, we used the AffymetrixGeneChip system to identify changes in gene expression. This The viability of G93A alone was lower than that of allowed the profiling of 6,000 well characterized mouse genes VSC4.1 or WT. In the co-culture of VSC 4.1 and G93A and 6,000 ESTs, such that specific pathways, rather than just cells, the viability was lower than predicted from each cell individual genes altered in the presence of mutant SOD1 line. The expression of bax, nNOS or PARP cleavage was not different among the conditions. Z-VAD-fmk, kynurenic acid,APV, or CNQX did not inhibit the reduced viability, but trolox, the antioxidant reversed the viability of G93A. Inaddition, co-culture with WT cells increased the viability of RNA was extracted from NSC34 cells previously transfected with vector, normal human SOD1 or G93A mutant humanSOD1. Following cDNA synthesis, labelled cRNA wastranscribed and applied to the murine GeneChip U74Av2, according to Affymetrix protocols. After stringency washesand staining, MASv5 was used to analyse the data.
Our data demonstrated that motor neuron cells with Transcription profiles from cells transfected with nomal mutant SOD1 (G93A) altered the viability of VSC4.1 cells human SOD1 or mutant human SOD1, were compared when they were cultured together. The decreased viability with those transfected with vector using dCHIP. Semi- was reversed by antioxidant trolox, supporting that oxida- quantitative RT-PCR was used to confirm a selection of tive stress may play a primary role. This result suggests those changes identified. Western blotting and functional that motor neuron cells with mutant SOD1 affect the viability assays were also used to provide further insight into the of neighboring motor neurons by generating oxidative On average, hybridisation signals were obtained from 37% of the genes arrayed on the murine GeneChip. Of 280 genes differentially expressed in the presence of G93A mutant SOD1, 211 were decreased and 69 increased. Severalcategories of genes whose expression was altered were identified, including those involved in the antioxidant response, ubiquitin-proteasome pathway, apoptosis andcell signalling, as well as genes previously implicated inneurodegeneration. In contrast, in the presence of normal Kirby J1, Heath PR1, Allen S1, Lowes D2, Halligan E2, SOD1, 6 genes were decreased, and 10 genes increased, compared to both vector and G93A mutant SOD1 transfectedcells.
1Academic Neurology Unit, University of Sheffield, Sheffield, UK; 2OxidativeStress Group, University of Leicester, Leicester, UK E-mail address for correspondence: jkirby@hgmp.mrc.ac.uk Identification of alterations in antioxidant defence and ubiquitin-proteasome pathways provides further supportfor a role for oxidative stress and protein aggregation in Mutations in the Cu,Zn superoxide dismutase gene (SOD1) SOD1-associated motor neuronal cell death. Gene alterations are causative in 20% of familial amyotrophic lateral sclerosis which may be amenable to therapeutic manipulation as a (FALS) cases, through a toxic gain of function of the mutant neuroprotective strategy will be discussed.
Under basal conditions, cells with mutant G93A SOD1, compared to cells transfected with normal SOD1 or empty vector, showed changes suggesting an increased propensity of apoptosis with the activation of caspase-9 (P,0.001 byANOVA). There was significant activation of caspases-3(P,0.01 by ANOVA) and x8 (P,0.01 by ANOVA) in the presence of mutant SOD1 in cells induced to undergo oxidative stress compared to control cell groups. Similarly inthe G93A transgenic mouse model, caspase-9 activationpreceded the activation of caspases-3 and -8.
1Academic Neurology Unit, University of Sheffield, UK; 2Laroratory of In the cellular model, there was significant inhibition of Neurogenetics, National Institute of Aging, NIH, Maryland, USA cell death by specific caspase-9 (P,0.05 by t-test) andcaspase-8 (P,0.05) inhibitors in cells expressing the E-mail address for correspondence: s_sathasivam@hotmail.com mutant G93A SOD1 upon serum withdrawal. However, noneuroprotection was observed with specific caspase-3 inhibi- tion. Minocycline alone produced a significant 17.8%reduction in cell death in the G93A mutant cells undergoing In amyotrophic lateral sclerosis (ALS), approximately 10% of oxidative stress (P,0.05 by t-test). Nifedipine alone did not cases are familial, with one fifth of these associated with offer any significant neuroprotection. The combination of mutations in the gene encoding the enzyme copper/zinc minocycline and nifedipine reduced cell death by 24.3% superoxide dismutase-1 (SOD1). There is increasing evidence (P,0.05). Upon serum withdrawal, there were significant that the final cell death pathway responsible for motor decreases in caspase-3 activity in the NSC34 mutant cell neuron death in ALS is apoptosis. The aim of this study was line treated with minocycline alone (P,0.05 by t-test), to investigate biochemically and pharmacologically the cell nifedipine alone (P,0.05 by t-test), or both drugs in death pathways involved in motor neuron death by utilising a combination (P,0.01). Both drugs in combination pro- duced a synergistic decrease in caspase-3 activity. Inaddition, increased cytosolic cytochrome c levels wereobserved upon oxidative stress which were ameliorated in the presence of minocycline in the NSC34 cells expressingmutant SOD1 (P,0.05 by t-test) or empty vector (P,0.05 by Using a mouse motor neuron cell line (NSC34) stably transfected with mutant or normal forms of human SOD1,or with vector alone, we examined the cell death pathwayunder basal conditions and serum withdrawal-induced oxida-tive stress. Oxidative stress is thought to be a key mechanism involved in the pathogenesis of ALS. We characterized thesequential activation of caspases by immunoblotting and Mutant SOD1 predisposes motor neuronal cells to apop- fluorogenic caspase activity detection. The caspase activation tosis and increases the likelihood of cell death upon oxidative cascade in the cell model was confirmed in the G93A stress. Activation of the mitochondrion-dependent apop- trangenic mouse model of familial ALS by studying the totic pathway appears crucial in the cell death process.
trangenic mice and their littermates at different stages of the In ALS, caspase-8 appears to play at least some role as disease process. In addition, we examined the pharmaco- a downstream caspase, although its role as an initiator caspase may still be important. This study suggests that inhibitors, the antimicrobial agent minocycline and the for useful neuroprotection to take place in ALS, interfer- voltage-gated calcium channel blocker nifedipine, upon ence of the apoptotic pathway at the initiator stage is serum withdrawal in the NSC34 cell model. We then went required. In particular, once the activation of caspase-3 on to examine the possible mechanisms of neuroprotection has occurred, the death process appears irreversible and afforded by minocycline and nifedipine.
Three proteins were found to be up regulated in cells expressing G93A mutant human SOD1 compared to those expressing normal SOD1 and two were downregulated. To date, 3 of these changed spots have been identified byMALDI-TOF-MS. Of those spots identified by MALDI-TOF MS, Western blotting has thus far confirmed the identity anddownregulation of one protein, peroxiredoxin 3. Work toconfirm the identities of the other changed protein spots by Academic Neurology Unit, Sheffield University Medical School, E FloorMedical School, Beech Hill RoadSheffield, S10 2RX E-mail address for correspondence: c.wood-allum@sheffield.ac.uk Peroxiredoxin 3 is a thioredoxin-dependent hydroperoxidase,which acts to convert hydrogen peroxide to water within themitochondrial matrix. Mitochondria are particularly vulner- able to oxidative damage from superoxide and its derivatives(generated as a byproduct of the activities of the electron Amyotrophic lateral sclerosis (ALS) is an incurable, adult- transport chain), yet lack catalase, the most efficient onset neurodegenerative disorder causing progressive muscle hydroperoxidase. Given the pre-existing evidence for oxidative weakness and death after 2–5 years from respiratory failure.
damage to mitochondria in ALS, it is interesting that Some 10% of ALS is familial, of which 20% is the result peroxiredoxin 3, an enzyme which might be expected to of mutations to Cu,Zn superoxide dismutase (SOD1), reduce oxidative damage due to hydrogen peroxide within a ubiquitously expressed free-radical scavenging enzyme.
mitochondria, is downregulated in cells expressing mutant Recently, increasing evidence of mitochondrial dysfunction SOD1. Investigation of the downstream effects of this in the tissues of patients with ALS and in models of the downregulation and a wider examination of the mitochon- disease has led to suggestions that mitochondrial dysfunction drial antioxidant system are underway in response to this may be important in disease pathogenesis.
In this study we aim to identify changes in mitochondrial protein expression attributable to the expression of G93A mutant SOD1 in a cell-culture model of SOD1 familial ALS.
We also plan to investigate the downstream, functional consequences of these changes and by doing so hope toclarify the molecular basis of the mitochondrial dysfunction NEURON-LIKE AND IN NON-NEURONALCELLULAR MODELS OF ALS Raimondi A1, Mangolini A1, Vanoni C1,Conforti L2, Francolini M2, Rizzardini M2, NSC34 cells are motor neuron-like cells, which although immortalised, retain many characteristics of adult motorneurons. Mitochondrially-enriched preparations of NSC34 cells stably transfected with empty vector, normal human Department of Medical Pharmacology, C.E.N.D., University of Milan, IN-CNR SOD1 or G93A mutant human SOD1 were made and 2-D gel Cellular and Molecular Pharmacology section, Via Vanvitelli 32, 29129 Milan, electrophoresis performed to identify differentially expressed Italy; 2Istituto di Ricerche Farmacologiche Mario Negri, Via Eritrea 62, 20157Milan, Italy protein spots of statistical significance (non-parametric,paired Wilcoxon t-test). Spots whose expression was shown E-mail address for correspondence: a.raimondi@in.cnr.it to change significantly were manually excised from Coomas-sie-stained 2-D gels and MALDI-TOF mass spectroscopyperformed. Database searching was then carried out to generate candidate protein identities. Confirmation of bothspot identity and of the change in expression level was then A still unsolved issue regards the selectivity of motor neuron damage caused by SOD1 mutants. Evidence obtained in various experimental systems suggests that mutated SOD1 may exert its toxic functions in mitochondria. In this study we have tested the hypothesis that selective damage of motorneurons may depend on a higher localisation of SOD1 in the mitochondria of these neurons than in the same organelles of TRANSPORT CHAIN IN A CELLULARMODEL OF FALS The human wt SOD1 and the ALS-associated G93A Rizzardini M, Mangolini, A Ubezio P, Lupi M, mutant have been stably expressed in a motor neuron-like cellular system (NSC34) and in a non-neuronal cell line(MDCK). In these cell lines we have measured the Istituto di Ricerche Farmacologiche Mario Negri, Milan, Italy total expression of wt and mutant SOD1 by Western blotanalysis and their mitochondria/cytosol distribution by E-mail address for correspondence: rizzardini@marionegri.it immunoelectron microscopy on cryosections stained forhuman SOD1. Alterations in the structure and functionof The significance of the altered functioning of mitochondrial electron transport chain in ALS is still elusive. This studyinvestigated in a cellular model of FALS whether the presence of the G93A mutant form of human SOD1 increases thesusceptibility of motor neurons to inhibition of the Western blot analysis showed a four- and two-fold higher expression respectively of wt and mutant SOD1 expression instably transfected MDCK cells than NSC34 cells. Morpho-metric analysis on ultrathin cryosections revealed a higher mitochondria/cytosol distribution of the mutated SOD1 inboth MDCK and NSC34 cell lines. In NSC34 even the Motor neuron-like cells NSC-34 were stably transfected with endogenous SOD1 showed a marked enrichment in mito- normal (wt) or G93A mutant human SOD1. The expression chondria. However, ultrastructural alterations of the mito- of human SOD1 was determined by Western blotting.
chondria were observed only in NSC34-G93A cells. In these Mitochondrial function and proliferation were measured by cells, morphometry on trasmission electron micrographs the MTT and the alamarBlue assay respectively. Cell death revealed the appearance of a population with a higher size and mitochondrial membrane potential were measured by due to increase of mitochondrial volume. This sub-popula- flow cytometry using propidium iodide (PI) and DiIC1(5) tion of swollen mitochondria showed pale and patchy matrix and sometimes abnormal cristae. We did not notice anyappreciable difference in the staining intensity for cytochromeoxidase activity. No mitochondrial or functional modifica- tions were ever seen in the MDCK-G93A cells, although theyexpressed a higher level of the G93A mutant protein than The level of human SOD1 proteins was about 25–50% of that of the endogenous murine SOD1. The doubling oftimes of the G93ASOD1 and untransfected cells duringexponential growth were comparable. An effect of mutant SOD1 on mitochondrial function was evident when serum One possible explanation of these results is that toxi- was removed for three days from culture medium. In these city of SOD1 mutants does not depend on the amount conditions, MTT conversion was lower in G93ASOD1 cells of SOD1 in mitochondria but on a higher sensitivity (57% of control) than in the untransfected or wtSOD1 cells.
of motor neuron mitochondria to the G93A-mediated Using Newman-Keuls post hoc test, the difference among toxicity. Experiments are in progress to identify the the cell lines was significant. Rotenone, malonate, antimycin molecular specificity underlying the selective vulnerability of A, KCN and oligomycin, specific inhibitors of different motor neuron mitochondria to ALS-associated SOD1 muta- steps of the mitochondrial respiratory chain, were then added to the cells in routine culture conditions and theMTT conversion of the three cell lines was measured after 6hrs of incubation. Antimycin A (2–10 mM) did not causesignificant mitochondrial impairment. After exposure to malonate (10–100 mM), a comparable dose-dependent loss This work is supported by Ministero della Salute, grant of mitochondrial function (10–60% of decrease) was IRCCS-2002, Fondazione Mondino, Roma, Italy.
observed in the three lines. Conversely rotenone (12.5 mM), KCN (1mM) and oligomycin (5 mM) showed a preferential neurons. The SMA-determining gene has been termed toxicity towards G93ASOD1 cells, all causing in this line Survival Motor Neuron (SMN) and is deleted or mutated in a significant decrease of MTT conversion (65%, 82% and over 98% of patients. We have recently demonstrated that in 70% of control respectively); their effect was negligible or rodents SMN is expressed as two main isoforms with Mr of lower in untransfected or wtSOD1. The difference among 32 kDa and 35 kDa, respectively. Both isoforms are localised the cell lines was significant by Newman-Keuls post hoc test.
in nuclear coiled (Cajal) bodies, but the 32 kDa form is also The effect of rotenone was further evaluated by flow cytoplasmic, whereas the 35 kDa form is also microsomal.
cytometry. Treatment caused a larger increase of PI permeable SMN is a novel protein involved in spliceosomal complex cells in G93ASOD1 than in wtSOD1 cells (39% versus formation and pre-mRNA splicing. However, its role in the 28.5%) and, among viable cells, in G93ASOD1 there was selective motor neuron degeneration in SMA is still unknown.
a larger fraction of cells with depolarised mitochondrialmembrane.
To study the molecular relationship between the two mainSMN isoforms and their potential post-translational mod- mitochondrial dysfunction caused by inhibition of mito-chondrial complex I, IV and V indicating an interactionbetween the mutant SOD1 and the functioning of the mitochondrial electron transport chain. The most consistenteffect was observed after treatment with rotenone, an Rat SMN cDNA was subcloned in oCMV-Tag1 vector to give a inhibitor of complex I, probably the major ROS source SMN fusion protein with FLAG epitope at the N-terminal and in mitochondria. This gives support to the hypothesis c-MYC epitope at the C-terminal. Transient transfections with that in motor neurons an altered free radicals handling the double-tagged rat SMN were performed in COS-7 cells contributes to the mitochondrial toxicity of G93ASOD1 and protein expression analysed by immunoblots and immunocytochemistry with confocal image analysis. SMNputative glycosylation and phosphorylation were studied in cell extracts using deglycosylation and alkaline phosphataseassays.
This work was supported by Ministero della Salute, grantIRCCS-2002/Fondazione Mondino and by MIUR, projectRBNEO1B5WW-004, Rome, Italy.
Immunoblot and immunostaining studies of the transfectedCOS-7 cells demonstrated that the 32 kDa SMN isoform derives from the full-length 35 kDa through a proteolytically cleavage at the C-terminal. Furthermore, while neither isoform undergoes glycosylation, the 35 kDa SMN isoformis physiologically phosphorylated in vivo.
SMN undergoes a complex post-translational rearrangement.
1INSERM U382 – IBDM, Campus de Luminy, Marseille, 13288 France; 2ALS In physiological conditions, the protein is proteolytically Clinical Research Centre, Institute of Neuropsychiatry, Palermo, 90129 Italy truncated at the C-terminal, a putative intermediate step inits degradation pathway. SMN full length (i.e. the 35 kDa E-mail address for correspondence: labella@unipa.it isoform) is also heavily phosphorylated in vivo, and this maydrive its ability to interact with its molecular partners, either proteins or nucleic acids. Our findings expand the under-standing of the biochemistry of SMN and may represent an Spinal muscular atrophy (SMA) is an autosomal recessive important step in our knowledge of the pathophysiology of disease characterized by a progressive loss of the spinal motor SMA and a basis for developing specific therapies.
DISMUTASE (GLY93AALA) MUTATIONTO KAINATE INDUCE TOXICITY Spalloni A1, Albo F1, Ferrari F1, Bernardi G1,2, Department of Neuroscience, University of Rome ‘‘Tor Vergata’’, Via Montpellier, 1, 00133 Roma, Italy; 2Fondazione S. Lucia, I.R.C.C.S.Via Ardeatina,306, 00179 Roma, Italy 1IRCCS Fondazione Santa Lucia Rome, Italy; 2Clinica Neurologica, University of E-mail address for correspondence: neurolab@virgilio.it Rome Tor Vergata, Rome, Italy; 3Department of Neuroscience, University ofRome Tor Vergata, Rome, Italy (2-amino-6-[trifluoromethoxy]-benzothiazole) Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenera- known as a neuroprotective agent with anticonvulsive, tive disorder, characterized by the progressive loss of motor anxiolytic and anaesthetic properties. Clinical trials have neurons (MN) in brainstem, spinal cord and motor cortex.
shown that riluzole prolongs the life of patients with ALS occurs in two forms: sporadic and familial ALS (FALS).
amyotrophic lateral sclerosis (ALS), a disease leading to FALS accounts for approximately 10% of all ALS cases, with degeneration of motor neurons. Several mechanisms of 20% of FALS cases associated with dominantly inherited action have been suggested for this drug and recently our mutations in the Cu2z,Zn2z superoxide dismutase (SOD1) studies have shown that interactions with glutamatergic gene. The development of a transgenic mice (G93A) over- neurotransmission, modulates the AMPA/kainate receptors expressing this mutant form of SOD1 has provided a valuable in rat cortical neuron.1 The aim of this work was to study if animal model of the disease. Among several theories of ALS, riluzole, as in cortical neurons, interacts with the AMPA/ a substantial body of literature suggests that MN degeneration kainate receptors in motor neurons in order to better may be a consequence of a disturbance in glutamatergic characterize its neuroprotective effect and its therapeutic neurotransmission involving excessive activation of glutamate receptors and the disruption of intracellular Ca2z homeo-stasis. A pathogenic role for AMPA receptor (AMPAR)mediated neurotoxicity has been identified in the selective Thus we are examining in mixed spinal cord cultures if MN Primary cultures were obtained from spinal cords of mice from G93A mice are more susceptible to toxicity following embryos on day 15 of gestation. Motor neurons were visually prolonged AMPAR stimulation compared to control and identified by their morphological appearance and by thres- SOD1 MN and exploring different pathways than the Ca2z hold diameter criteria (w28 microm). Whole-cell configura- permeable AMPAR. Firstly, we have tested if G93A MN are tion of the patch clamp technique was used to record the more vulnerable to AMPAR mediated excitotoxicity, by currents induced by the perfusion of kainate at different incubating MN cultures in the presence of Kainate. Cultures have been exposed to various concentrations of the AMPARagonist Kainate (10, 25, 50 and 100 mM) in the presenceof 10 mM MK-801 to block NMDA receptor activation.
From preliminary data the results indicate that at thelowest concentration used (10 mM), all three preparations In all tested motor neurons, when clamped at x60 mV, bath are equally resistant to Kainate-induced toxicity,whereas application of kainate produced inward currents without any higher concentrations of the agonist (from 25 to 100 mM) evident desensitization. These currents gated by kainate were are causing a significantly higher incidence of MN death in concentration dependent, with an EC50=35 mM. The applica- the G93A population with respect to the control and SOD1 tion of different concentrations of riluzole (0.5–100 mM) always decreased the currents induced by 50 mM kainate, ina dose dependent manner. The minimal effective effect wasobserved at 0.5 mM and an almost maximal inhibition wasobtained with 10 mM riluzole (IC50=1.54 mM). In order todeepen its mechanism of action on cultured motor neurons,different concentrations of kainate were applied in presenceof 1 mM riluzole. Riluzole reduced the kainate responseswithout significantly changing the EC50. These data indicateda noncompetitive inhibitory mechanism of riluzole on thekainate action site.
channels and heterooligomeric GluR1/2 channels wereexpressed in HEK293 cells.
These results indicate that riluzole decreases the kainate-induced currents in motor neurons in a dose-dependentmanner with a IC50 lower compared to cortical neurons. This IC50 value is comparable with the plasma concentrationmeasured in ALS patients during the pharmacological Co-application of the drugs with 10 mM glutamate resulted in a slight dose-dependent depression of the peak currentamplitude by 10–20% at a concentration of 1 mM. Thecurrent decay in presence of glutamate was faster in presence of the blockers. Resensitisation of GluR1 or GluR2 receptorchannels follows a monoexponential time course with a time This work was supported by grant from Telethon, Italy constant in the range between 10 ms and 150 ms. Resensitiza- tion had a biexponential course when glutamate and theblocker were co-applied. The slower component of recoverywas independent on blocker concentration and had values of around 2 s to 4 s, at least two orders of magnitude slower thanthe fast resensitisation. The proportion of the second, slowercomponent of recovery on the whole recovery process 1. Zona C, Cavalcanti S, De Sarro G, et al. Synapse 2002; 43: 244– increased with blocker concentrations and reached 90% at GluR 2 flip, and 50% at GluR1 flop and GluR2 flop channelsat 1 mM. The non-desensitising L506Y GluR2 channelsshowed a marked, dose-dependent current decay upon co- application of glutamate and RPR119990/RPR117824. A similar slow component of recovery was found as compared to wild-type GluR2 channels in the presence of the blockers.
At kainate-type GluR6 receptors no block was observed RPR117824 in single and double pulse experiments. AMPA- Krampfl K, Schlesinger F, Cordes A-L, Dengler R, type channels were competitively blocked by RPR119990 or RPR117824 with an IC50 around 10x8 M. At GluR6 channelsthe dose-response relation of the observed competitive blockwas shifted to higher concentrations.
1Neurologische Klinik, Medizinische Hochschule Hannover,Carl-Neuberg-Str. 1, 30625 Hannover, Germany E-mail address for correspondence: Krampfl.Klaus@MH-Hannover.de Our results show a specific block of recombinant AMPA-type GluR channels and synaptic AMPA receptors by these novel Antagonising glutamatergic neurotransmission by the block- pyrazine derivatives. There is evidence from the kinetic ade of AMPA type glutamate receptors is a promising analysis of the block effect that the antagonists act by pharmacological strategy for neuroprotection in neurodegen- competing with glutamate at the glutamate binding site.
erative diseases such as amyotrophic lateral sclerosis (ALS) or We performed a quantitative analysis of the experimentally Parkinson’s disease (PD). We investigated the interaction of observed block mechanism by computer simulation and two new pyrazine derivatives (RPR 119990 and RPR 117824) could predict the experimental results by the addition of a with native and recombinant AMPA type glutamate receptors.
blocked state connected with the unliganded closed stateof the receptor in a Markov model. Notably, the IC50 as lowas 10x8 M at AMPA-type glutamate receptors points to a potential role of the novel drugs in preventing glutamate-triggered excitotoxic cell damage and correlates with positive Recombinant human homooligomeric GluR1 flop, GluR2 results from pharmacological testing on the transgenic model flip, GluR2 flop, GluR6, non-desensitising L506Y GluR2 These results constitute initial evidence that, in accordance with clinical data, the firing properties of single cortical neurons in a genetic model of ALS are altered to induceneuronal hyperexcitability. Further studies are needed toidentify and characterize the relevance of such current Pieri M1, Albo F1,2, Longone P2, Zona C1,2 1Department of Neuroscience, University of Rome ‘‘Tor Vergata’’, Via Montpellier, 1–00133 Roma, Italy; 2Fondazione S. Lucia, I.R.C.C.S., ViaArdeatina, 306–00179 Roma, Italy This work was supported by grant from Telethon, Italy(n‡ 1185).
E-mail address for correspondence: lab.fisiologia@libero.it Amyotrophic lateral sclerosis (ALS) is a neurodegenerative 1. Caramia MD, Palmieri MG, Desiato MT, et al. Neurology 2000; disease characterized by the selective degeneration of motor neurons in the spinal cord, brain stem and cerebral cortex.
Mutations in the gene coding for Cu,Zn superoxide dismutase(SOD1) have been identified in 10% of patients affected bya familial form of ALS. Previous observations have shownin ALS patients by transcranial magnetic stimulation (TMS), a reduction in corticomotor threshold with a consequentcorticomotor hyperexcitability.1 To analyse if the firing properties of single cortical neurons in a genetic mouse model of ALS are altered to induce neuronal hyperexcitability, the electrical activity of cortical neurons has been tested in a transgenic mouse model of a familial form of ALS, associatedwith a mutation in Cu,Zn SOD (Gly93- Ala).
Vanoni C1, Massari S1, Losa M1, Perego C1,Conforti L2, Pietrini G1 Cortical neurons were isolated from control and transgenic 1Department of Medical Pharmacology, C.E.N.D., University of Milan, IN-CNR mice embryos expressing either the wild-type form of human Cellular and Molecular Pharmacology section, Via Vanvitelli 32, 29129 Milan SOD1 or the mutant Gly93- Ala (G93A) and neuronal cultures Italy; 2Mario Negri Institute of Pharmacology, Via Eritrea 62, 20157 Milan, Italy were prepared. Current-clamp recordings were obtainedusing whole-cell configuration of the patch-clamp technique.
E-mail address for correspondence: grazia.pietrini@unimi.it The following parameters have been measured: 1) restingmembrane potential, 2) membrane input resistance, 3) membrane time constant and 4) membrane capacitance.
The astroglial GLT-1 glutamate transporter is the most Action potential (AP) properties have been measured from important transporter in maintaining the extracellular con- the repetitive firing derived from the first two spikes of the AP centration of glutamate below neurotoxic levels. Loss of GLT- discharges elicited on the same injection of current.
1 and increased extracellular concentration of glutamate havebeen documented in sporadic and familial cases of ALS, and in FALS transgenic mice models expressing mutated SOD1.
The molecular mechanisms of this regulation remain to be Our results indicate that all membrane passive properties in elucidated, and aim of this project is to clarify how the G93A control, SOD1 and G93A cortical neurons are comparable. In mutant of human SOD1 induces loss of GLT-1.
contrast, with the same injection of the current, the firingfrequency in G93A cortical neurons (36.93±10.8 Hz, n=15)is higher with respect to control (25.73±8.34 Hz, n=17) and SOD1 (29.61±8.54 Hz, n=17) neurons. Moreover, thethreshold and the rate of repolarisation calculated from the We have developed and characterized a cellular model of ALS first AP were significantly different (P,0.05) in G93A cortical stably expressing high levels of the human G93A mutant of neurons (23.39±4.78 mV; 26.32±15.13 mV/ms) with respect SOD1 (MDCK-G93A cell lines). In these cell lines and in cell to control (28.24±3.46 mV; 15.85±4.77 mV/ms) and SOD1 lines expressing WT human SOD1 we have transiently (27.59±4.33 mV; 20.91±13.03 mV/ms) neurons.
transfected cDNAs encoding glutamate transporters, and the expression of the transporters have been analysed by biochemical (Western blot, surface biotinylation assays) and morphological assays (indirect double immunofluorescence Wisman LAB1, Coenen S1, Van Wingerden W1,Hol EM2, Verhaagen J2, Van Muiswinkel FL1, Ba¨r PR1 Similarly to ALS cases and animal models of ALS, Western 1Department of Experimental Neurology, Rudolf Magnus Institute of blot analysis of total cell extracts indicated a lower expression Neuroscience, University Medical Center Utrecht, Utrecht, The Netherlands;2 of GLT-1 in cell lines expressing the G93A mutant of SOD1 Graduate School for Neurosciences, Netherlands Institute for Brain Research, (20–30% of the expression measured in SOD1 expressing MDCK cells). The effect is specific for GLT-1, as the total E-mail address for correspondence: L.A.B.Wisman@neuro.azu.nl expression of the EAAC1 glutamate transporter isoform isnot altered. Moreover, morphological analysis showed anintracellular redistribution of the GLT-1 glial glutamate transporter in MDCK-G93A cell lines, whereas the surfacedistribution of the neuronal EAAC1 or the glial GLAST were Glutamate excitotoxicity plays an important role in the not affected. Together with a surface expression, GLT-1 also pathogenesis of amyotrophic lateral sclerosis (ALS). In ALS appeared in intracellular structures co-localising with inter- patients the glutamate concentration in the cerebrospinal nalised FITC-WGA labelled surface glycoproteins. No redis- fluid is higher than in controls, suggesting an abnormal tribution of GLT-1 was observed in cells over-expressing WT glutamate metabolism. Normally, glutamate is removed from SOD1. Surface biotinylation experiments indicated that the synaptic cleft by glutamate transporters. In ALS, the approximately 50% of the total GLT-1 was in intracellular glutamate transporter EAAT2 is decreased, which may result structures. The nature and functions of these structures in toxic glutamate levels. In our study we aim to explore the was demonstrated by double-immunofluorescence experi- use of gene therapy to increase EAAT2 expression, in order to ments using markers of endocytotic compartments. These lower the glutamate concentration in the spinal cord, in a experiments revealed a selective accumulation of GLT-1 in lysosomal degradative compartments. In line with GLT-1being targeted to degradation in MDCK-G93A cell lines, inhibition of degradative compartments with chloroquinecaused an increase of GLT-1 expression.
To explore the possibilities of gene therapy for EAAT2,genetically engineered cells over-expressing EAAT2 werecreated to test the neuroprotective efficacy of EAAT2 onprimary motor neurons. Also, lentiviral vectors (LVV) coding for EAAT2 were constructed and characterized using organo-typic spinal cord cultures. After characterization in vitro, GFP- Our results demonstrated that the G93A mutant of SOD1 and EAAT2-LVV were injected into the spinal cord of G93A- mediates a decreased cell surface expression of GLT-1 and hSOD1 mice with the use of a spinal adaptor, at the level of targets the transporter to degradation, thus suggesting the vertebra L1. In a first experiment, LVV coding for GFP were mechanism underlying GLT-1 loss in ALS.
injected, while GFP expression was examined two weeks later.
Subsequently, ALS mice were treated with LVV-EAAT2 to evaluate its effect on disease onset, progression and survival.
This work was supported by ALS Association, starter grant2001 to Pietrini G.
Genetically engineered cells showed an increase in functionalglutamate transporters compared to wild-type cells. Whenprimary motor neurons, cultured on an astroglial feeders,were exposed to glutamate in the presence of cells over-expressing EAAT2, an increased survival was observed. Similarto the results obtained in organotypic spinal cord cultures,injections of LVV-GFP in ALS mice lead to a profuseexpression of GFP in the ventral spinal cord without overttissue damage. While mainly localised in astrocytes, labellingof NeuN/Chat-immunoreactive motor neurons was onlyoccasionally observed.
While in vitro studies have demonstrated the neuroprotective efficacy of an EAAT2 based gene therapy approach, in vivo studies using GFP-LVV have indicated that spinal injection of LVV is an effective and safe means to transduce foreign genes in the ventral spinal cord of mice. Currently, we are inves- tigating the neuroprotective effect of EAAT2-LVV gene therapyin ALS mice. Results of these studies will be presented.
Dewil M, Lemmens G, Robberecht W,Van Den Bosch L P119 MITOMYCIN C DOWNREGULATESEXPRESSION OF CU,ZN SUPEROXIDE Laboratory for Neurobiology, Department of Experimental Neurology,University of Leuven, Belgium E-mail address for correspondence: wim.robberecht@uz.kuleuven.ac.be Broom WJ, Ay I, Pasinelli P, Brown RH Jr.
Day Neuromuscular Laboratory, Department of Neurology, Massachusetts In the mutant SOD1 mouse model for human ALS, mino- General Hospital, Harvard Medical SchoolBoston, MA 02115, USA cycline, a semi-synthetic tetracycline, slows disease progressionand significantly prolongs survival. The mechanism of action E-mail address for correspondence: wbroom@partners.org of minocycline remains unknown, but in vitro evidencesuggests that microglial activation and the p38 mitogen Familial ALS accounts for 10% of all ALS cases and activated protein kinase (MAPK) pathway may be involved. In approximately 25% of these cases are due to mutations in the present study we investigated the effect of minocycline on the Cu,Zn superoxide dismutase gene (SOD1). More than this protein kinase system in vivo and in vitro.
100 different mutations have been identified in the SOD1gene; these span all five exons. Several lines of evidence arguethat the mutant SOD1 protein is neurotoxic through an acquired, adverse function, as yet not explicitly defined.
Western blot analysis showed p38 MAPK and in particular its Particularly convincing is the observation that transgenic active, phosphorylated form to be upregulated in the spinal expression of high levels of mutant SOD1 protein produces a cord of mutant (G93A) SOD1 mice, as compared to wild type motor neuron disease phenotype in transgenic mice, with age SOD1 over-expressing mice. This upregulation was almost of onset and disease duration dependent on copy number.
exclusively present in the ventral part of the spinal cord as These considerations predict that measures which decrease compared to the dorsal part. Treatment of mice with levels of mutant SOD1 protein should ameliorate the minocycline significantly inhibited the p38 MAPK activation.
phenotype in transgenic mice and potentially in patients Immunohistochemical studies of mutant (G93A) SOD1 mice with SOD1-mediated disease. Mitomycin C (MMC) is an spinal cord suggest that the phosphorylated p38 MAPK is antibiotic used in chemotherapy for the treatment of present in both ventral horn neurons, glial and microglial primarily gastric, bladder and colorectal cancer. MMC is celIs. To elucidate whether minocycline acts on the p38MAPK activated in vivo to bind and cross-link DNA, thereby system in microglial and neuronal cells or both, we studied inhibiting DNA synthesis and transcription. A previous the effect of the drug on microglial activation and on mutant report suggested that MMC inhibits SOD1 expression.1 We (G93A) SOD1-dependent motor neuron death in vitro.
have completed experiments to confirm and extend this observation. Our Western immunoblotting data confirm that induced phosphorylated p38 MAPK activation in purified the drug reduces SOD1 protein levels in rat and human cells’ microglial cultures. To evaluate the protective action of the in vitro system in parallel with an increase in level of p53.
compound on neuronal cells, we studied its effect on an in Data from in vitro expression studies of transcription as well vitro model for selective mutant SOD1-induced cell death, as in vivo animal experiments will be presented.
namely cyclosporin A-induced mutant SOD1-dependentapoptotic neuronal death, described before. Minocyclinesignificantly inhibited this mutant SOD1-dependent cell death of N2A cells. As SB203580, an inhibitor of the activityof p38 MAPK pathway, yielded similar protection, the 1. Cho, et al. Biochem Mol Biol Int 1997; 42: 949–956.
involvement of the p38 MAPK pathway in the cyclosporinA-induced neuronal death is likely. To evaluate the signifi-cance of these findings, primary mutant SOD1 motor neuroncultures were treated with cyclosporin A. In this model both minocyclin and SB203580 were confirmed to protect against in terms of growth curves (Trypan blue exclusion method) cyclosporin A-induced apoptotic cell death.
and differentiation capacity (by immunostaining with specificneural antibodies, such as Tau1 for neurons and GFAP for These results demonstrate that the beneficial effect of minocycline in the mutant (G93A) SOD1 mouse is accom-panied by inhibition of the p38 MAPK system. This drug Embryonic and adult neural stem cells share the great appears to act on both microglial activation mechanisms as majority of expressed genes (96.8%) while specific embryonic well as on intrinsic neuronal death pathways.
transcripts are mainly connected to developmental stagerelated or migration genes (i.e. Engrailed2, Reelin and Zic1).
Proliferation curves show that NSC mitotic cycles are quite similar between adult and embryonic stem cells since no statistically relevant differences can be observed. Even thepotential is apparently independent from developmental age; production of neurons, astrocytes and oligodendrocytes,expressed as a percentage of total cell number, is similar.
Cova L1,2, Ratti A1,2, Fogh I1,2, Mantegna L1,2,Fantozzi R1,2, Silani V2 Expression profiles seem to be highly conserved between 1Department Neurological Sciences, IRCCS Ospedale Maggiore; 2Department embryonic and adult NSC; nevertheless their environmental of Neurology and Laboratory of Neuroscience, ‘‘Dino Ferrari’’, University of Milan niche of origin and their state are quite dissimilar. We Medical School, IRCCS Istituto Auxologico Italiano, Milan, Italy show that adult NSC potential is limited mainly by a non-permissive environment and this may be exploited for cell E-mail address for correspondence: lidicova@hotmail.com therapy of neurodegenerative disorders such as ALS, incombination with epigenetic stimulation in order to obtain Neural stem cells (NSC) can be derived from adult mammalian central nervous system, expanded in vitro asfloating clones (neurospheres) and induced to terminaldifferentiation in the three major neural phenotypes (neurons, astrocytes and oligodendrocytes). Recent papers have reported their wide potential in vivo both in thedevelopmental capacity (integration in chimeric animals) andneuronal repair in traumatic spinal cord injury or, more recently, in chronic models of multiple sclerosis. Never- theless, clinical use of adult stem cells is still limited by thescarce knowledge of their integration capacity and benefits in Department of Neurology, Osaka University, Osaka, Japan comparison with the foetal cells, since extensive molecularand cellular comparative data are still missing.
E-mail address for correspondence: sakoda@neurol.med.osaka-u.ac.jp We aim to investigate if adult neural stem cells possessthe same potential as their embryonic counterpart. A wide The aim of the present study is to establish an organotypic definition of adult stem cell features will have an immediate slice culture using mouse spinal cord as a reproducible in vitro practical benefit in the treatment of neurodegenerative model of ALS. This culture technique is useful for the drug diseases and, in particular, of ALS.
screening of ALS, because this is applicable to the spinal cordof transgenic ALS mice.
In order to address this issue we compared neural stem cellpopulations obtained from the murine embryonic day 12 After the spinal cords were carefully dissected, lumbar spinal (E12) representative of primary neurogenesis stage with adult- cords were sectioned transversely at 350-mm intervals with a derived NSC. An expression profile study has been performed tissue chopper. The sectioned slices were transferred on using the Affymetrix MG-U74Av2 Genechip arrays in order to Millicell-CM porous membrane (Millipore), and then incu- define the ‘overall’ NSC molecular characteristics. Briefly, total bated in a defined medium containing 25% horse serum. At RNA was extracted and hybridised to the commercial chips days in vitro 14 (DIV14), 100 mM of pyrrolidine decarboxylic and statistical analysis on obtained data was conducted using acid (PDC), an inhibitor of glutamate uptake, was added the MAS 5.0 software. Contemporaneously cellular behaviour, to the medium. To rescue the MN death induced by PDC, CNQX, non-NMDA antagonist, was applied at DIV 14. The cultured slices were fixed with paraformaldehyde, and spinal MNs were counted by immunohistochemical staining with The cytoarchitecture of the slices was maintained at least Department of CNS Research, Boehringer-Ingelheim Pharma GmbH & Co. KG, for one month. More MNs were observed in slices from post-natal day 2 (P2) than from P8 at DIV 14. PDC treat-ment induced MN death, which was rescued by CNQX E-mail address for correspondence: kussmaul@bc.boehringer-ingelheim.com The dopamine receptor agonist (-)Pramipexole ((-)PPX) hasbeen shown to be neuroprotective in vivo, and several in vitro studies showed that it might be a mitochondrio-protective We established organotypic slice culture using mouse spinal agent. In addition to its well-known dopaminergic agonism, cord. Using this culture technique, screening molecules for its physicochemical property of being a lipophilic cation with a low redox potential has led to the assumption that (-)PPXmight concentrate in mitochondria, the main source ofreactive oxygen species. We therefore examine, if [3H]-labelled(-)PPX is taken up by brain mitochondria derived from mice.
We found that (-)PPX uptake in mitochondria is 1) a fast and 2) non-saturable process, which depends on 3) mitochondrial volume and 4) mitochondrial membrane potential (DyM).
Furthermore we found (-)PPX to accumulate in mitochondriaby a factor of 3–4 which depends on mitochondrial Kussmaul L, Mierau J, Fleissner S, Gillardon F, membrane potential and the relative concentrations of (-)PPX2z, (-)PPXz and the electrogenic neutral (-)PPXmolecule. We conclude that (-)PPX is able to accumulate in Department of CNS Research, Boehringer-Ingelheim Pharma GmbH & Co. KG, mitochondria by a mechanism which is driven by the DyM (negative inside) and therefore might be targeted tomitochondria where it can act as an antioxidant or inhibitor E-mail address for correspondence: kussmaul@bc.boehringer-ingelheim.com The dopamine receptor agonist Pramipexole ((-)PPX) hasbeen shown to be neuroprotective in vitro and in vivo,although the mechanism has not yet been resolved. In addition to its well-known dopaminergic agonism, (-)PPX is a lipophilic cation with a low redox potential. Therefore it has been anticipated that (-)PPX might accumulate in mitochondria, the main source of reactive oxygenspecies, where it might act as a powerful antioxidant. Wetherefore examined if [3H]-labelled (-)PPX is taken up by Kussmaul L, Mierau J, Neumaier M, Boehringer M, neural cells. We found that (-)PPX uptake in neural cells is a Dorner-Ciossek C, Mendla K, Fleissner S, Maschke I, 1) fast and 2) non-saturable process, which depends on (3) ion homeostasis, 4) pH and 5) mitochondrial membranepotential, especially in neurons. We thus conclude that (-)PPX Department of CNS Research, Boehringer-Ingelheim Pharma GmbH & Co.
is able to enter the cells and might function there as a neuroprotective agent by a mechanism which has to beelucidated.
E-mail address for correspondence: kussmaul@bc.boehringer-ingelheim.com The dopamine receptor agonist Pramipexole ((-)PPX) hasbeen shown to be neuroprotective in vitro and in vivo. Inaddition to its dopaminergic agonism, (-)PPX is also able toenter neural cells and to accumulate in mitochondria.
Here we show the equipotent detoxification of in vitro We thus conclude that EUK-134 is a powerful antioxidant SND919CL2x. Furthermore (-)PPX is also able to detoxify which lowers oxidative stress in vivo, as it has been shown reactive oxygen species, generated in vitro by hypoxanthine/ previously.1 However, caution has to be taken if EUK134 xanthine oxidase. However, the capacity of (-)PPX to detoxify is administered chronically e.g. in models of neurode- ROS was 100-fold less compared to the SOD-Catalase- generative diseases associated with oxidative stress such as mimetic EUK-134. In order to investigate if EUK-134 is Alzheimer’s and Parkinson’s disease owing to an accumula- able to lower oxidative stress in mitochondria in vivo, we tion of the compound and/or its metabolites. Additionally, treated SOD-2 (-/-) mice with EUK-134 (30 mg/kg i.p daily).
the SOD-2 (-/-) model might be a useful tool to examine if EUK-134 prolonged lifespan from 7±3 days in vehicle treated the ROS and NO scavenging capacity of (-)PPX and its (z) enantiomer in vitro is able to translate into a therapeutical Furthermore, EUK-134 treated SOD-2 (-/-) animals showed an increase in body weight for the first 15 days, whichwas significantly higher than in untreated animals. However,after 15 days of treatment the mice lost body weightand developed a progressive movement disorder. The paramagnetic properties of EUK-134 allowed us to examineits biodistribution by magnetic resonance imaging (MRI), 1. Melov S, Doctrow SR, et al. Lifespan extension and rescue of where it causes shortening of T1 values. We treated adult spongiform encephalopathy in superoxide dismutase 2 nullizy- mice by injection of 30 mg/kg EUK-134 i.p. and found gous mice treated with superoxide dismutase-catalase mimetics. J after 24 h increased signal intensities in liver and kidney.

Source: http://www-user.tu-chemnitz.de/~ducom/ALSSymposiumMilan/Abstracts/Theme_4.pdf