THEME 4 EPIDEMIOLOGY AND GENETICS

P59 ANALYSIS OF PHYSICAL ACTIVITY AS A RISK FACTOR FOR MOTOR NEURONE DISEASE

MOORE C¹, BESSON H², FINUCANE F², EKELUND U², MCDERMOTT C¹, WAREHAM N², SHAW P¹

¹Academic Unit of Neurology, University of Sheffield, Sheffield, South Yorkshire, United Kingdom, ²MRC Epidemiology Unit, Addenbrooke's Hospital, Cambridge, Cambridgeshire, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: Physical activity, risk factor, questionnaire validation

Background: Reports suggesting a higher incidence of motor neurone disease (MND) in professional sports people have lead to speculation of an aetiological relationship between physical activity (PA) and the development of MND ([1], [2]) This is supported by hypotheses regarding the underlying cellular mechanisms of MND, in particular oxidative stress and glutamate excitotoxicity. However, definitive evidence is lacking, with prior studies plagued by methodological problems and conflicting results.

Aims and Objectives: We hypothesise that a high level of previous PA is associated with an increased risk of MND in individuals with an underlying genetic susceptibility. Our objective is to create and validate a new adulthood PA questionnaire which can be used to evaluate PA as a risk factor for MND.

Methods: In collaboration with a unit who have expertise in PA epidemiology, we have designed a novel questionnaire to collect data regarding PA performed from aged 20 years onwards. Using closed questions, we ask

about PA performed at home, work and leisure, structured into time periods of 5 or 10 years. To validate the questionnaire we conducted 100 interviews using participants from a large population-based cohort created by the epidemiology unit to prospectively investigate factors linked to diabetes. Self-reported durations of PA were quantified into energy expenditure scores using an established coding scheme which classifies activities according energy expenditure.[3] These were compared with objective PA energy expenditure measurements taken from cohort members 10 and 15 years ago, calculated using the 4-day heart-rate monitoring technique with individual calibration.[4] The degree of correlation between objective and subjective measurements of PA was determined using the Spearman's rank correlation coefficient.

Results: A high degree of correlation between the questionnaire and objective PA variables was achieved. Correlation between total self-reported PA scores and objective PA energy expenditure was 0.34 for PA 10 years ago (p < 0.001) and 0.47 for PA 15 years ago (p < 0.001).

Discussion, Conclusions and Future Work: This novel adulthood PA questionnaire displays a level of validity beyond that reported in previous similar questionnaires. It will be invaluable in the analysis of PA as a risk factor for MND, providing a practical instrument for accurate quantification of historical PA and therefore addressing many of the challenges encountered in previous studies. As such, we will conduct a retrospective case-control study using incident MND cases and matched community controls without neurological diagnoses. Differences in PA participation between cases and controls will be analysed and any association between MND and PA determined.

P60 TOXICITY OF β-N-METHYLAMINO-L-ALANINE TO AQUATIC ORGANISMS

PURDIE E, SAMSUDIN S, METCALF J, REILLY M, YOUNG F, EDDY B, CODD G

Dundee University, Dundee, Scotland, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: Neurotoxicity, Animal models, BMAA

Background: The toxicity of β-N-methylamino-L-alanine (BMAA) a novel cyanobacterial amino acid, has been established over the past 40 years in a range of vertebrate models. The mechanism of BMAA toxicity has been determined in vitro in dissociated mixed spinal cell cultures, showing selective motor neurone loss [1] and with other insults, a potentiation of neuronal injury at 10µM BMAA [2].

The widespread production of BMAA by symbiotic and free-living cyanobacteria [3] raises questions of BMAA toxicity to aquatic organisms. Aquatic models are needed due to the risks of exposure of aquatic biota to BMAA, and their roles as both indicators of effects on community health and as primary cyanobacterial consumers in the food chain.

The Brine Shrimp, Artemia salina and Zebra Fish Danio rerio are valuable models for toxicity assessments and have, for example been used to establish toxicity of cyanotoxins, as well as to assess toxic effects on communities at large. In addition, much work has been done outlining neuronal developmental pathways and D. rerio is proving to be a valuable model in neuro-degeneration research [4].

Early stages of neuronal development are highly conserved among vertebrates; D. rerio is an ideal model organism to better understand the mechanisms associated with BMAA-induced motor neurone degeneration, as the homologous nature of early stage neuronal development across species, means that D. rerio analysis identifies common developmental principles, particularly neurone patterning, signalling pathways and the locomotor regions

Objective: To determine the toxicity of BMAA to aquatic organisms A. salina and D. rerio.

Methods: Multiple methodologies were used to assess the effects of BMAA exposure, including a 72-hour survival bioassay and 5-day behavioural assay for A. salina, and a 6-day bioassay for D. rerio, monitoring developmental, morphological-, and behavioural endpoints.

Results: BMAA caused time-and concentration-dependent mortality and loss of photo-tactic response in A. salina. Early stages of D. rerio development were more susceptible to BMAA, embryo hatching was delayed, and larval mortalities occurred due to BMAA exposure. BMAA reduced larval heart rate and induced oedema over a 6-day developmental period, during which neuro-muscular deformities occurred, including: occular deficiency, head deformation and trunk deformities.

Conclusions: The lowest observable adverse effect level (LOAEL) in both A. salina and D. rerio larvae (5µg/L; 42nM), occurred at environmentally-relevant BMAA concentrations. The increased sensitivity of D. rerio to BMAA indicates that it may indeed serve as a useful model to further investigate BMAA- induced neuro-degeneration; and that the bioavailability and neuro-toxicological significance of BMAA to aquatic biota require further investigation.

P61 A CLUSTER OF AMYOTROPHIC LATERAL SCLEROSIS IN CLOSE PROXIMITY TO TWO LAKES IN ENFIELD, NH, USA WITH KNOWN BLOOMS OF POTENTIALLY TOXIC CYANOBACTERIA

STOMMEL E¹, HARRIS B², COHEN J¹, WEST K³, BALL A³, COGBILL C², GRABER D², CALLER T¹, BOYER G⁴, HANEY J⁵

¹Department of Medicine/Section of Neurology, ²Department of Pathology, Dartmouth Medical School, Lebanon, NH, ³Dartmouth College, Hanover, NH, ⁴State University of New York College of Environemental Science and Forestry, Syracuse, New York, ⁵Department of Zoology, University of New Hampshire, Durham, NH, United states

E-mail address for correspondence: [email protected]

Keywords: epidemiology, cyanobacteria, environmental risk factors

Background: Cynaobacteria are associated with many neurotoxins and cytotoxins including β-methylamino L-alanine (BMAA), anatoxin (ATX), saxitoxin(STX), anatoxin a(S), curacin, microcystins and cylindrospermopsin (CYL). All of the neurotoxins have targets on the central or peripheral nervous system. BMAA has been linked to neurodegenerative disease as with amyotrophic lateral sclerosis/parkinsonism–dementia complex (ALS/PDC) that has been remarkably prevalent amongst the Chamorro people of Guam.

Methods: Meticulous review of patient records at our institution was undertaken to substantiate the diagnosis of sporadic ALS and document the exact dwelling address, family history and other relevant social and medical histories. Methods for examining water samples included phycological, chlorophyll and turbidity analysis. Liquid chromatography-mass spectrometry (LCMS) was used to measure free BMAA, ATX, STX, microcystins and CYL from filtered water samples.

Results: Water samples collected from Lake Mascoma in November 2007 were confirmed by microscopic examination to contain high numbers of potential-toxin producing cyanobacteria. However, we were unable to detect significant levels of any BMAA, microcystin, CYL or ATX in filtered samples of lake water using LCMS. Epidemiological evaluation of the geographical area of Lake Mascoma and a small adjoining Crystal Lake in Enfield, NH (population 4,854 in 2006; 120.5 persons per square mile of land area.), together encompassing an area of approximately 7.75 square miles (2.75 square miles of inland water excluded) shows an incidence of sporadic ALS that is approximately 25 times the expected incidence of 2/100,000/year. The 8 ALS patients who lived on or near the lakes (on average less than 0.15 miles from the shore) for a minimum of 9 years, were diagnosed between the years 1991–2007. Documented blooms of cyanobacteria have been reported in Lake Mascoma previously as part of a state monitoring program.

Discussion: We postulate that the high incidence of ALS in this patient cluster could be directly related to the chronic exposure of the cyanobacteria producing neurotoxins. Our inability to detect toxin levels in our samples may be a technical issue (we only analyzed from free BMAA and free microcystins, not the bound components) or may reflect the episodic occurrence of toxic algal blooms. We conjecture that the exposure might be related to aerosolization of toxin as with some other toxic algae; however the exposure could be potentially through other means such as swimming, eating fish from the lake or direct ingestion of water.

P62 THE NEUROTOXICITY OF BETA-N-METHYLAMINO-L-ALANINE (BMAA) IN DROSOPHILA

ZHOU X, ESCALA W, PAPAPETROPOULOS S, ZHAI G

University of Miami, Miami, United States

E-mail address for correspondence: [email protected]

Keywords: BMAA, Drosophila, neurodegenerative

The neurotoxic amino acid BMAA is produced by known groups of cyanobacteria in terrestrial, freshwater and marine environments, suggesting a potential for wide-spread human exposure. Evidence of such exposure has been demonstrated in brain tissues from patients from Guam with Amyotrophic Lateral Sclerosis (ALS)/Parkinsonism dementia complex and North American patients with ALS and Alzheimers disease [1]. Neuronal culture (in vitro) and primates/rodent model (in vivo) experiments have provided some insights to the neurodegenerative properties of BMAA.

We used the fruit fly Drosophila melanogaster as a model to investigate the effects of L-BMAA in vivo. We first conducted survival assays (1–2 day-old wild type flies were fed with a medium containing 0/2/4/6/8/10 mM BMAA). Another two amino acids, glutamate and alanine, were also used in similar concentrations to evaluate the toxicity of BMAA. We found that BMAA dramatically reduced the life span of flies in a dose dependent manner. Interestingly, glutamate also reduces lifespan but the effect is much less severe compared with BMAA at the same concentration. Alanine has no effects on the life span of flies. To compare the toxicity of BMAA with other environmental toxins, we used an herbicide paraquat in our assay. Paraquat is a reactive oxygen species inducing compound that causes neurodegeneration. We found that the toxicity of 4 mM BMAA is comparable with that of 2 mM paraquat. Second, we used negative geotaxis assays to evaluate the motoneural coordination and the integrity of the nervous system. The geotactic behavior was significantly affected in flies exposed to low (2mM) concentrations of BMAA and the effects are more severely affected in higher concentrations. Third, we evaluated the neuropathological changes using immunofluorescent confocal microscopy. Interestingly, we observed a possible up-regulation of pHistone in the brain of adult flies fed with BMAA. We demonstrated the toxic effects of BMAA on life span, motor function and neuronal function/morphology. The dramatic effect of BMAA on the nervous system that we observed in Drosophila not only provides strong evidence for an environmental cause for neurodegenerative disease but also validates and exemplifies the power of Drosophila as a model system.

P63 CYANOBACTERIA AND BMAA: POSSIBLE LINKAGES WITH AVIAN VACUOLAR MYELINOPATHY (AVM) IN THE SE UNITED STATES?

BIDIGARE R, CHRISTENSEN S

University of Hawaii, Honolulu, HI, United States

E-mail address for correspondence: [email protected]

Keywords: Cyanobacteria, BMAA AVM

Background: Avian vacuolar myelinopathy (AVM) is a neurological disease that produces uncoordinated behavior in affected waterfowl. AVM is characterized by bilaterally symmetrical vacuolation of the white matter of the brains and spinal cords of AVM-affected birds. Since 1994, AVM has caused the deaths of >100 bald eagles (Haliaeetus leucocephalus) and thousands of American coots (Fulica americana) at numerous sites in the southeastern (SE) United States. AVM is now believed to be the most significant unknown cause of eagle mortality in the history of the United States. Despite extensive screening for a wide rage of pathogens and toxins (prions, organic metals, pharmaceuticals and plant toxins), the causative agent for AVM has not been identified. More recently, feeding and sentinel trials, field surveys, and genetic studies have implicated exotic Hydrilla verticillata and an associated epiphytic cyanobacterial species (Order Stigonematales) as a causal link to AVM. This undescribed Stigonematales species covers up to 95% of the surface area of Hydrilla leaves in reservoirs where AVM-related bird deaths have occurred.

In order to explore possible linkages between BMAA and AVM, a variety of tissue samples were screened for BMAA. Samples included cultures of Stigonematales grown on BG-11 medium (with and without nitrate), H. verticillata minus the Stigonematales epiphyte, and H. verticillata plus the Stigonematales epiphyte. Samples were hydrolyzed and analyzed by HPLC/FL. BMAA was detected in the Stigonematales (plus NO₃^-) culture, the Stigonematales (minus NO₃^-) culture (30 µ g g^-1 dry weight), and the Hydrilla/Stigonematales field-collected sample (59 µ g g^-1 dry weight); BMAA was below the limit of detection (5 pmoles per injection) in the Hydrilla field-collected sample. The identification of BMAA in the H. verticillata plus the Stigonematales epiphyte sample was confirmed by LC/MS. We hypothesize that the BMAA concentration in the field-collected Stigonematales was greater than 59 µ g g^-1 since it “diluted” with non-BMAA containing H. verticillata biomass.

If BMAA is biomagnified in these wetland ecosystems, as has been observed in Guam, then the consumption of fish (e.g., shad and herring) and waterfowl (e.g. Canada geese and mallards) from local reservoirs could represent a significant health risk for humans. This risk also applies to other regions in the SE United States as AVM-positive sites have been identified in Arkansas, Texas, Georgia and North Carolina. Future studies will assess the degree of biomagnification of the neurotoxic amino acid, BMAA, in wetland ecosystems of the SE United States.

P64 CYANOBACTERIAL TOXIN BMAA IN NEURODEGENERATION

MASH D¹, PABLO J¹, BANACK S², COX PA³, JOHNSON TE⁴, PAPAPETROPOULOS S¹, BRADLEY W¹

¹University of Miami Miller School of Medicine, Miami, FL,, ²California State University Fullerton, Fullerton CA, ³Institute for Ethnomedicine, Provo, UT, ⁴San Jose, CA, United States

E-mail address for correspondence: [email protected]

Keywords: BMAA, environmental toxin, pathogenesis

Background: Cyanobacteria are found in almost every conceivable habitat, from oceans to fresh water to bare rock to soil. Most are found in fresh water, while others are marine, occur in damp soil, or even temporarily moistened rocks in deserts. The neurotoxic amino acid beta-N-methylamino-L-alanine (BMAA) is produced by laboratory strains isolated from all five known morphological groups of cyanobacteria, including cyanobacterial symbionts and diverse taxa of free-living cyanobacteria. The ubiquity of cyanobacteria in terrestrial, as well as freshwater, brackish, and marine environments, suggests a potential for wide-spread human exposure. BMAA is a naturally occurring, non-protein, non-essential amino acid that is found in high concentrations in brain tissues of patients with Amyotrophic Lateral Sclerosis-Parkinsonism-Dementia Complex (ALS/PDC) in the South Pacific island of Guam. More recently, BMAA has been measured in cohorts of Caucasian, North American patients with sporadic Alzheimer's disease and ALS. Although the role of BMAA in human degenerative disease is highly debated, there is evidence to suggest that BMAA may mimic glutamate toxicity.

Objective: To demonstrate human exposures to the neurotoxic amino acid beta-N-methylamino-L-alanine (BMAA) are associated with neurodegenerative disease. The neurotransmitter glutamate is an excitotoxin that has been implicated in a number of different neurodegenerative diseases. We report here that BMAA is detected in an independent cohort of ALS and AD cases that came to brain autopsy.

Methods: BMAA was measured in cryopreserved brain specimens taken from the cerebral cortex of ALS (N = 12), AD (N = 13) and age-matched control subjects (N = 12). BMAA was quantified in brain in protein-bound fractions using a validated fluorescence HPLC method confirmed by tandem quadrupole and Quadrupole time of flight (Q-ToF) mass spectrometry. We conducted biodistribution and ex vivo autoradiographic studies of BMAA in rodents, using custom synthesized (³H)BMAA.

Results: In human postmortem brain, BMAA was measured in varying concentrations as a bound component in archived brain tissues sampled from the frontal and temporal cortices. BMAA concentrations ranged from 30 to 250 µg/g in ALS (134 µ g/g +/- 12.8 (mean +/- SEM)) and 10 to 230 µg/g in AD patients(111 µ g/g +/- 14.6). In contrast, BMAA was only detected in one age-matched healthy control subject. In rodent studies, we observed specific uptake of radiolabeled BMAA that was localized to specific cortical and brainstem regions. Time course studies demonstrated that BMAA reached a plateau at 8 hours and remained elevated in brain at 48 hours post injection. In contrast to the brain, BMAA levels declined rapidly over time from other organs. Discussion/Conclusions: This observation suggests that BMAA is transported into the brain and that the amino acid may be incorporated into certain target areas due to selective incorporation in proteins or by glutamate uptake mechanisms. This is the second confirmation of BMAA exposures in sporadic ALS and AD patients from North America. These observations support a role for BMAA as a possible gene/environment interaction, serving as a potential trigger for certain types of neurodegenerative processes that share a common molecular pathogenesis but involve different anatomical regions of the brain.

Supported by grants from the Institute for Ethnomedicine, the Urbanek Family Trust and the National Parkinson Foundation, Inc.

P65 MULTIPLE NEUROTOXIC DIETARY ITEMS IN THE CHAMORRO DIET: IS A LINK TO ALS/PDC PLAUSIBLE?

BANACK S, BAKHSHI F, COX P

Institute for Ethnomedicine, Jackson, Wyoming, United States

E-mail address for correspondence: [email protected]

Keywords: Diet, Neurotoxicity, BMAA

Background: Initial reports [1] that cycad flour in the Chamorro diet may be linked to ALS/PDC resulted in five international cycad conferences and significant debate within the literature. The claim that washed cycad flour only has low amounts of BMAA [2] has been updated by the finding of 50–100 times the amount of protein-bound compared with free BMAA in cycad flour. Recently flying foxes were added as a possible source of BMAA in the Chamorro diet [3].

Objectives: We sought to determine if there are other possible inputs of BMAA in the traditional Chamorro diet.

Methods: Using AQC precolumn derivatization and HPLC-FD with LC/MS and triple quadrupole LC/MS/MS we tested for BMAA in a variety of dietary items prepared by the Chamorro people including washed cycad flour, cycad dough, cycad tortillas, cycad dumplings, cooked flying fox, flying fox stew, and flying fox broth, as well as 21 different flying fox specimens and a sample of feral deer.

Results: Although we did not detect significant concentrations of free-BMAA in traditional foods prepared by Chamorro villagers from washed cycad flour, washing does not remove protein-associated BMAA. We did detect significant quantities of protein-bound BMAA in cycad dough (4.7 µ g/g), cycad tortillas (334 µ g/g), and cycad dumplings (84 µ g/g). Feral deer had a low amount of BMAA (3 µ g/g in hair). Flying fox broth was particularly rich in protein-bound BMAA with a 1 litre portion resulting in ingestion of 12.4 mg of BMAA. BMAA may also occur in dietary items from other ALS/PDC foci.

Conclusions: Multiple inputs of BMAA in the Chamorro diet confound epidemiological analyses based on a single dietary item [4]. Thus determining whether consumption of cycad flour alone, or flying foxes alone, or even feral pigs, deer, and land crabs (which all may occasionally feed on cycad seeds) alone ignores total input of BMAA into the Chamorro diet from multiple sources. Furthermore, since BMAA is produced by symbiotic cyanobacteria in cycad roots, as well as in free-living marine cyanobacteria [5], inputs into the Chamorro diet from marine sources may also occur. (We are currently investigating BMAA in marine fish ingested by villagers in the coastal Kii ALS/PDC focus.) Recent laboratory findings of BMAA in double-blinded brain tissues of North American AD patients, with independent replication of the results by another laboratory indicate that further analysis of BMAA and its putative link to ALS/PDC among the Chamorro people is warranted.

P66 FRUIT AND VEGETABLE INTAKE AND RISK OF AMYOTROPHIC LATERAL SCLEROSIS.

OKAMOTO K¹, KIHIRA T², KONDO T³, NAGAI M⁴

¹Department of Public Health, Aichi Prefectural College of Nursing and Health, Nagoya, Japan, ^2Kansai College of Oriental Medicine, Osaka, Japan, ³Department of Neurology, Wakayama Medical University, Wakayama, Japan, ⁴Department of Public Health, Saitama Medical School, Saitama, Japan

E-mail address for correspondence: [email protected]

Keywords: amyotrophic lateral sclerosis, epidemiology, antioxidants

Objective: Few human studies have reported the relationship between a food rich in antioxidants and the risk of amyotrophic lateral sclerosis (ALS). We therefore analysed the relationship between fruits and vegetable intake and the risk of ALS using a case-control study in Japan.

Methods: The study comprised of 153 ALS patients diagnosed by El Escorial World Federation of Neurology criteria, and 306 population-based randomly selected controls matched to the cases by age (+/-5 years), gender, and county of residence. A self-administered food frequency questionnaire was used to estimate preillness intakes of food groups and nutrients. The strength of association between ALS and a potential risk factor was assessed by calculating odds ratios (ORs) and 95% confidence intervals (CIs).

Results: A high intake of fruit and vegetables (OR = 0.61; CI = 0.39–0.95; P = 0.04 for trend) and fruit (OR = 0.57; CI = 0.33–0.98; P = 0.04 for trend)was significantly associated with a reduced risk of ALS risk. In multivariate model, all vegetables, green and yellow vegetables and other vegetable intakes were observed an inverse association, although the association was not statistically significant.

Conclusion: Our findings suggest that dietary intake of fruit and vegetables intake may play a protective role in the development of ALS.

P67 HYPOCHOLESTEROLEMIANTS OR HYPERCHOLESTEROLEMIA IS ASSOCIATED WITH A LATER ONSET OF ALS BUT NOT WITH AN INCREASED RATE OF DECLINE

CAMU W, ALPHANDERY S, DIOU J, MOREAU L, PAGEOT N, MORALES R

CHU de Montpellier ALS center, Montpellier, France

E-mail address for correspondence: [email protected]

Key words: hypocholesterolemiants, treatment, cardiovascular risk factors

Background: It has recently been suggested that hypercholesterolemia may be protective regarding the neurodegenerative process. The authors tended to suggest that physicians should not prescribe or should stop statins of fibrates to an ALS patient. However, as ALS patients frequently have vascular risk factors, such a treatment withdrawal could have deleterious effect even in a rapidly evolving disorder.

Objectives: To determine whether ALS parameters are modified in a group of consecutive ALS patients recruited in our expert ALS center.

Material and Methods: We prospectively questioned ALS patients followed in our center for antecedent hypercholesterolemia and other vascular risk factors such as myocardic infarction or chronic ischemic cardiopathy. Antecedent or current statin or fibrate treatment was noted. ALS parameters were collected from clinical files (age of onset, site of onset, current disease duration and ALSFRS at the time of questioning, presence of familial ALS). At the time of analysis (April 2008), we also collected age of death and subsequent final ALS duration.

ALS parameters were compared between the groups with or without hypocholesterolemiants. We also calculated the rate of decline (ALSFRS decline/months of evolution), in order to evaluate a potential worsening of a given ALS group (with or without treatment).

Results: There were 174 sporadic ALS (SALS) patients and 15 familial ALS cases (FALS). In all treated SALS groups, mean age of onset of ALS was significant delayed: 60 years without treatment vs 64 yrs w/statins (p < 0.03), 65.2yrs w/fibrates (p < 0.03), 64.7 w/statins or fibrates (p < 0.003). In all treated SALS groups, the rate of decline was lower than without treatments but this was never significant: statins (0.49 vs 0.59), fibrates (0.48 vs 0.59), statins and fibrates (0.48 vs 0.62). There was no difference between the groups, for site of onset. A large majority of the patients are alive, precluding a reliable analysis regarding prognosis, but a Cox model will be processed every 6 months until mid-2008 to evaluate this point.

In FALS, the small number of patients restricted statistical analysis. Eleven patients belonged to the untreated group with mean age of onset of 48.5yrs and a rate of decline of 0.32. In the treated group of 4 patients, mean age of onset was 61.5yrs and the rate of decline was 0.93. Both mean age of onset and rate of decline were statistically significant (p < 0.05 and p <0.003, respectively).

Discussion: Treatment with hypocholemiants in associated both in SALS and FALS with a significant later onset. This has also been suggested recently in a large cohort of patients with Parkinson's disease. The same pattern was present in the Canadian cohort presented in the 18^th International Symposium on ALS/MND. In SALS, the largest group of our cohort, there is a lower but not significant rate of decline of ALSFRS in patients with statins and/or fibrates. Our data is supportive of a protective role of either hypocholesterolemiants or hypercholesterolemia regarding ALS. However, these data cannot support a worsening effect of the treatment with either statins or fibrates in SALS.

P68 DYSLIPIDEMIA IN ALS: NO POSITIVE EFFECTS ON SURVIVAL IN A TWO LARGE ITALIAN CLINICAL SERIES

CALVO A¹, ILARDI A¹, GIACONE S¹, MORA G², GHIGLIONE P¹, PESSIA A¹, MARINOU K², CAMMAROSANO S¹, CHIÒ A¹

¹Department of Neuroscience, Torino, ²Department of Neurological Rehabilitation, Pavia, Italy

E-mail address for correspondence: [email protected]

Keywords: hyperlipidemia, survival, cholesterol

Background: Recently hyperlipidemia has been found to positively influence survival in a series of ALS patients. This observation has been related to possible effects of metabolic changes during the course of the disease.

Aims: To evaluate in a series of ALS patients the serum lipid concentrations and their influence on patients survival.

Methods: Patients from two ALS referral series in Torino and Pavia, Italy, have been considered. The patients have been seen after January 1, 2000 and prospectively followed-up. Serum lipid concentrations are evaluated as a part of the systematic blood sampling at the time of diagnosis. No patient was consuming lipid-lowering drugs at the time of the assay. Survival times have been calculated as the duration between the onset of first symptoms and death or tracheostomy or the last day of follow-up (February 29, 2008).

Results: A total of 573 patients (mean age at onset, 61.8 (SD 11.8)) with definite or probable ALS were included in the study. The mean total level of cholesterol was 2.2 (±0.4) g/L, the mean level of triglycerides was 1.2 (±0.6) g/L, the mean level of HDL was 0.6 (±0.1) g/L, the mean level of LDL was 1.3 (±0.4) g/L, the mean LDL/HDL ratio was 2.4 (±0.8). The median survival from onset was 2.9 years. No difference could be found in survival for any of the lipids variables. In particular, using a LDL/HDL ratio cut-off of 2.99 the group with higher ratio showed only a slight, not significant, increase of survival (median survival, LDL/HDL ratio <2.99, 38.5±4.5 months; ≥2.99, 42.7±5.3 months). The LDL/HDL ratio was not significant also in Cox's multivariable analysis.

Conclusions: We could not find any significant effect of hyperlipidemia in a large series of ALS patients consecutively seen in Italy. Moreover, serum lipid concentrations were generally lower in our series than in that in the cases reported and was more similar to their controls. This difference could be due to different genetic backgrounds or to different dietary habits of Italian and French populations.[15]

P69 STATIN USE PRIOR TO AMYOTROPHIC LATERAL SCLEROSIS (ALS) AND PRIMARY LATERAL SCLEROSIS (PLS)

BROOKS BR

¹Carolinas Neuromuscular/ALS-MDA Center, Charlotte, NC, ²University of Wisconsin School of Medicine and Public Health, Madison, WI, United States

E-mail address for correspondence: [email protected]

Keywords: neuroepidemiology, statin-induced myopathy, toxicology

Objective: Define antecdent statin use in patient populations attending an ALS Clinic.

.Background: Statin use, including red yeast rice, has been rarely associated with myopathy and other neuromuscular symptoms. Some patients referred to an ALS Clinic have described muscle pain and weakness following initiation of statins leading to the diagnosis of ALS.

Methods: Review of prevalence of statin use in 240 (M-153; F-87) sporadic (s) or familial (f) patients with ALS or other motor neuron disease syndromes (MNDs) seen in a Midwest university-associated ALS Clinic.

Results: Among 31/164 statin-exposed sALS patients, 11 single statin treated (SST) and 3 multiple statin treated (MST) patients described treatment-associated pain and increased weakness before diagnosis. Only those 3 MST patients have been reported to MedWatch. In 13 SALS patient requiring thyroid (T) replacement, 1/3 SST and 2/2 MST patients had symptoms prior to ALS diagnosis. In 7 SALS patients with diabetes mellitus (DM) treated with oral agents, none required T but 1/3 SST and 1/1 MST patients had symptoms. In 4 sALS patients with DM requiring insulin, 2/4 required T but 0/2 SST and 1/2 MST patients had symptoms prior to diagnosis. Among 18 FALS patients none had DM or T and 0/2 SST patients had symptoms. Among 20 PLS patients none had DM or T and 1/2 SST and 1/1 MST patients had symptoms. There was no statin or T use, or DM among 14 dysimmune motor neuropathy/neuron disease or atypical MNDs prior to diagnosis.

Conclusions: Statin-associated pain and weakness is more common in MST SALS, hypothyroid SALS and DM SALS patients than FALS or other MNDs patients. Further studies are required to determine if statin use, particulary MST in the context of changing statins in response to statin-related symptoms is a trigger for sALS in some susceptible patients.

P70 CLINICAL AND BIOCHEMICAL INDICATORS OF VASCULAR DISEASE AND ALS RISK

SUTEDJA N, SIZOO E, FISCHER K, VELDINK JH, HUISMAN M, WOKKE J, VAN DEN BERG LH

University Medical Center, Utrecht, Netherlands

E-mail address for correspondence: [email protected]

Keywords: vascular, risk, hypoperfusion

Background: A role for hypoxia and reduced vascularisation in the pathogenesis of ALS has been suggested due to observed associations of ALS and mutations in two genes involved in angiogenesis, vascular endothelial growth factor (VEGF) and angiogenin (ANG).

Objective: To assess the association between vascular risk factors and ALS.

Methods: Traditional cardiovascular risk factors (smoking, hypertension, hypercholesterolemia, diabetes, and body mass index (BMI)) and cardiovascular diseases (myocardial infarction, angina pectoris, cerebrovascular accidents, and peripheral arterial disease) prior to onset of disease were established by a questionnaire (prevalent disease, smoking, hypertension, hypercholesterolemia, diabetes and body mass index (BMI)) in 303 patients with ALS. Biochemical assessments (total cholesterol, LDL-cholesterol, HDL-cholesterol, glucose, hs-CRP, homocysteine and fibrinogen) at diagnosis were measured in blood samples of 233 patients with ALS. Comparison was made with prospectively collected data from 2100 population-based control subjects. Association with survival in ALS was also analysed.

Results: Multivariate regression analysis adjusting for age, education and smoking showed a lower body mass index (BMI) in patients (in women: OR 0.9; 95% CI 0.9–0.99; P = 0.02; in men: OR 0.9; 95% CI (0.8–0.9); P < 0.001). Age-adjusted odds ratio showed lower levels of total (in women OR 0.8; 95% CI 0.7–0.97; P < 0.001; in men OR 0.7; 95% CI 0.6–0.8; P < 0.001) and LDL- (in women: OR 0.5; 95% CI 0.4–0.7; P < 0.001; in men OR 0.7; 95% CI 0.6–0.8; P < 0.001) but higher HDL-cholesterol (in women: OR 2.6; 95% CI 1.7–4.0; P < 0.001; in men: OR 2.0; 95% CI 1.2–3.3; P < 0.001) in patients. Fibrinogen levels, measured in women only, were higher in patients with ALS (OR 5.2; 95% 3.9–6.8; p < 0.001). Within the patient group, no association was found between survival and biochemical indicators of vascular risk.

Conclusions: Vascular risk factors, measured clinically and biochemically, did not increase ALS risk. Lower weight and atherogenic cholesterol levels in ALS patients suggest increased ALS risk may be related to a higher metabolic rate. The implication of increased fibrinogen in female ALS patients need to be determined.

P71 ELEVATED SERUM LEVELS OF SEX HORMONES IN AMYOTROPHIC LATERAL SCLEROSIS

HUISMAN MHB¹, VAN ES MA¹, LENTJES E², VELDINK JH¹, VAN DEN BERG LH¹

¹Department of Neurology, Rudolf Magnus Institute of Neuroscience, ²Laboratory of Clinical Chemistry and Hematology, University Medical Center Utrecht, Utrecht, Netherlands

E-mail address for correspondence: [email protected]

Keywords: Hormones, estrogen, testosterone

Background: The incidence of amyotrophic lateral sclerosis (ALS) is higher amongst men than women, with a male to female ratio varying from 1.1 to 1.8. It has been suggested that this difference points to an involvement of sex hormones in ALS.[1] This is supported by both in vitro and in vivo studies, which have demonstrated that androgens and estrogens have trophic effects on motor neurons.[17–19], [5]

Objective: The objective of our study is to determine the relationship between sex hormone levels and ALS.

Methods: Using a case-control design, the levels of estrone (E1) and 17β-estradiol (E2) were determined in 257 men and 161 postmenopausal women with ALS and in 195 male and 105 female healthy controls. Serum levels of testosterone (T), free testosterone (FT) and sex hormone binding globulin (SHBG) were determined in 121 male and 82 postmenopausal female ALS patients and in 93 male and 55 female healthy controls. Results were adjusted for body mass index (BMI) and age using logistic regression.

Results: The mean levels of E1, adjusted for BMI and age, in both men and post-menopausal women with ALS were significantly higher than in controls (men: 195.8 nmol/l vs. 150.6 nmol/l, p value 6.04*10⁻¹², women: 211.1 nmol/l vs. 118.4 nmol/l, p value 5.38*10⁻¹²). The level of T and FT in post-menopausal women with ALS was also significantly higher than in controls (T: 1.317 nmol/l vs. 0.982 nmol/l, p value 0.002, FT: 20.39 pmol/l vs. 14.63 pmol/l, p value 0.0001), but no significant difference in the levels of T and FT in male and SHBG in both sexes could be detected between patients with ALS and their controls. More results are in progress.

Discussion: Levels of E1 are elevated in male and female patients with ALS and levels of T and FT are elevated in female patients only. Elevated E1 levels amongst patients with ALS are probably the result of a secondary phenomenon, caused by adrenal dysregulation and increased peripheral conversion from adipose and muscle tissue due to muscular atrophy. Elevated levels of T and FT in female ALS patients may suggest a causal relationship between androgens and ALS, supporting the male preponderance in the incidence of ALS or it could also be the result of a secondary phenomenon. There is no clear evidence for a neuroprotective role of sex hormones in ALS.

P72 CLINICAL AND GENETIC ANALYSIS OF A LARGE CHINESE FAMILY WITH AMYOTROPHIC LATERAL SCLEROSIS

FANG D, ZHENG H, GUO X, ZENG Y, ZHOU D, ZHANG S, SHANG H

Department of Neurology, West China Hospital of Sichuan University, Chengdu, Sichuan, China

E-mail address for correspondence: [email protected]

Keywords: SOD1 gene, clinical feature, prognosis

Background: Amyotrophic lateral sclerosis (ALS) is the most common and fatal degenerative motor neuron disease in adults. Approximately 20% familial ALS (FALS) have been found Cu/Zn superoxide dismutase (SOD1) gene as the causative gene, though the aetiology of the ALS is unclear.

Objective: To investigate the clinical features, progression, prognosis genetics of a big Chinese family with ALS.

Patients and Methods: All living subjects who consented for this study were examined by at least two independent neurologists for the diagnosis of ALS (according to the El Escorial and revised El Escorial criteria). They also underwent electromyographic (EMG) examination to establish lower motor neuron (LMN) dysfunction in non-symptomatic body regions. SOD1 gene mutation screening was performed by directly sequencing after informed consent.

Results: This non consanguineous Chinese family consisted of 5 generations and had an autosomal dominate inheritance pattern. Twelve patients including 9 males and 3 females were found. Ten were dead. Mean onset age was 40 year old. The initial symptom of one limb weakness was followed soon by atrophy and fasciculation in the limb and progressed to other limb, swallowing difficulty. The duration of the disease is about 1.5 years. The cause of death is breath difficulty. All the patients presented with upper motor neurone pattern such as hyperreflexia and increased muscle tone besides the lower motor neuron signs. No mutation of SOD1 gene was detected.

Conclusions: The clinical phenotype of this family is quite similar to that described in other ethnic groups. However, no mutation detected in the SOD1 gene suggests ALS is a genetic heterogeneous disease and mutation of SOD1 gene is uncommon in familial Chinese ALS.

P73 MOTOR NEURON DISORDERS IN CHINA

LU M¹, WANG X², FAN D¹

¹Department of Neurology, Peking University Third Hospital, Beijing, China, ²Department of Medical Genetics, Peking University, Health Science Center, Beijing, China

E-mail address for correspondence: [email protected]

Keywords: case report, Kennedy's disease, variant

Objectives: To study the clinical features of Chinese patients with motor neuron disorders.

Methods: We reviewed retrospectively 552 cases of Chinese patients with motor neuron disorders in the database of Peking University Third Hospital from 2004 to 2008. Patients with amyotrophic lateral sclerosis (ALS) whose onset age was ≥ 70 years old as well as Kennedy's disease (KD) were included. The clinical features including symptoms and sings, courses of the disease, and biochemical indicators were analyzed.

Results: A total of 26 cases of elder ALS patients (4.7%) were found. The mean age at onset was 72.3±3.3 years, and gender ratio (male to female) was 2.7 vs 1. It was the most frequent that the first symptom was the single upper extremity weakness and atrophy (38.5%), especially the right side (26.9%), and 11.5% showed the first symptom was bulbar involvement. Among these 26 cases, 5 patients died (19.2%), 4 performed tracheastomy (15.4%), 5 manifested the clinical characteristics of “flail arm syndrome” (19.2%) which was found to have a longer median survival. The mean duration from onset to death or tracheastomy was 30.3±15.3 months. In addition, it was interesting that about 34.6% of these 26 cases of ALS patients had family history of tumors, which seemed too high to explain. On the other hand, 27 KD patients were found. There was mild motor functional lesion which correlated with the course of the disease (r = 0.77, P=0.000). The degrees of creatine kinase (920.10±495.54 U/L) and triglyceride (3.27±2.78 mmol/L) increased remarkably. Among them, a 27-year-old man, developed weakness of his lower limbs at age 25 and two of his three brothers have the same symptoms, was noticed that his physical examination showed all of his deep tendon reflexes were active, and Hoffman sign on the right side was positive but Babinski signs negative bilaterally. The serum level of creatine kinase is mildly increased (289 U/L). Direct sequencing showed that the exact number of CAG repeats in the patient was 52.

Conclusion: The clinical features of elder ALS patients in this study were a little dissimilar to those of typical ALS patients. Meanwhile, in KD, there were some distinctive characteristics such as elevated triglyceride level. In a young KD patient, we found his manifestation was relatively special because his age of onset was earlier and his creatine kinase level was lower than typical patients. His active tendon reflexes showed that there may have some lesions of upper motor neurons. Because of that, the patient should be differentiated with ALS carefully. We wonder whether this was a clinical variant of KD.

P74 STATUS OF CARE FOR ALS PATIENTS AND HEALTHCARE SYSTEMS IN JAPAN RESULTS OF 2006 NATIONAL SURVEY

OGURA A¹, KAWAMURA S², NAKAYAMA Y³, KAKUNO F¹, MURATA M⁴, KONISHI K⁵, HIRABAYASHI K⁶, OGAWA S⁷, UENO K⁸, SATO M⁹, AMAMOTO H¹⁰, MATSUDA C¹, ITAGAKI Y¹, NAGASAWA T¹

¹Tokyo Metropolitan Institute for Neuroscience, Fuchu-city, Tokyo, ²Aomori University of Health and Welfare, Aomori-city, Aomori, ³Showa University, Yokohama-city, Kanagawa, ⁴Higashi-oumi public health center, Oumi-city/Shiga, ⁵Department of Health and Welfare, Ibaraki Prefecture, Ibaraki-city/Ibaraki, ⁶Kokugakuin University,, ⁷Japanese Nursing Association, ⁸The National Association for Home-Visit Nursing Care, ⁹Japan Visiting Nursing Foundation, Shibuya-ku/Tokyo, ¹⁰The Japan Medical Association, Bunkyo-ku, Tokyo, Japan

E-mail address for correspondence: [email protected]

Keywords: the number, public support system, medical services at home

Background: In Japan, ALS patients are entitled to receive the public support system, but there are many problems to provide long-term medical care and support of patients’ lives.

Objectives: ALS patients were surveyed regarding where they were cared for (home or hospital/care facility) and whether they were using a ventilator. Home-care ALS patients who “needed sputum suctioning” and medical system providing their care were also studied to evaluate the system.

Methods: ALS patients registered at health centers in all prefectures were surveyed using a questionnaire by mail.

1: The number of ALS patients using and not using a ventilator, and numbers of patients using a ventilator who were cared for at home and a hospital/care facility were investigated. 2: Concerning home-care ALS patients “needing sputum suctioning”, the gender, age, history of illness, medical care provided,and living-support services used were investigated. All data were coded and analysed using the statistical package SPSS.

Results and Discussion: The number of ALS patients registered at 400 organisations (58.4% of the 685) was 5,335, of which 1,591 (29.8%) were ventilator users. Of the 1,591, 803 (54.1%) being cared for at home.

Concerning profiles of home-cared ALS patients answers were obtained from 326 organizations (47.6% of 685), the number of home-cared ALS patients requiring “sputum suctioning” was 837, and 749(89.5% of 837, Group (A) used a ventilator, but 88 (10.5% of 837, Group (B) did not.

Gender A: 472 males (63.0%), B: 47 males (53.4%); age (mean±SD) A: 57.96±20.01, B: 63.21±10.73; the duration of illness A: 3–8 years in 349 patients (46.6%), 9–18 years in 266 35.5%), B:3–8 years in 44 patients(50% of 88) and less than 3 years in 24 (27.3% of 88).

Total support for ADLwas necessary A: 706 (94.3%), B:59 (67.0%), having a specialized medical care A: 562 (75.0%) B: 76 (86.4%), having a primary care physician A: 637 (85.0%) B: 68 (77.3%), emergency visit by a physician A: 584 (78.0%), B: 53 (60.2%), emergency visit by a visiting nurse A: 636 (84.9%), B:74 (84.1%).

It is suggested that both Groups received a medical care, but not all were covered. Also, some ALS patients received care for 20 years or longer using a ventilator, so that it is necessary to sustain this care system for long time.

Conclusion: A survey of ALS patients showed that 29.8% of them were using a ventilator, and 54.1% of these ventilator users were cared for at home, and suggested problems with the securing of beds for their hospitalization, having specialized medical care, and home-visiting nursing.

P75 SPORADIC AMYOTROPHIC LATERAL SCLEROSIS IN JAPAN: AGE AT ONSET INFLUENCES ON WIDE-RANGED CLINICAL PROFILES

ATSUTA N¹, WATANABE H¹, ITO M¹, TANAKA F¹, NAKANO I², AOKI M³, TSUJI S⁴, YUASA T⁵, TAKANO H⁶, HAYASHI H⁷, KUZUHARA S⁸, SOBUE G¹

¹Department of Neurology, Nagoya University Graduate School of Medicine, Nagoya, ²Division of Neurology, Department of Medicine, Jichi Medical University School of Medicine, Tochigi,, ³Department of Neurology, Tohoku University School of Medicine, Sendai, ⁴Department of Neurology, Graduate School of Medicine, University of Tokyo, Tokyo, ⁵Kohnodai Hospital, National Center of Neurology and Psychiatry, Tokyo, ⁶Department of Neurology, Brain Research Institute, Niigata University, Niigata, ⁷Department of Neurology, Tokyo Metropolitan Neurological Hospital, Tokyo, ⁸Musashi Hospital, National Center of Neurology and Psychiatry, Tokyo, Japan

E-mail address for correspondence: [email protected]

Keywords: age at onset, initial symptom, Japan

Background: Symptomatic manifestations of ALS such as the frequency of various initial symptoms, their influence on clinical features, the frequency of rare symptoms such as oculomotor disturbance, the influence of age at onset on clinical features and progression, and many other symptomatic details have not been well characterised, particularly based on a large scale sample.

Objectives: To profile the detailed clinical features of sporadic ALS on large scale samples in Japan.

Methods: A nationwide registration of ALS patients has been conducted in Japan since 1974. In 2003, the registration system was revised and made available for research use. The registration of each patient is annually renewed. We analyzed 3428 newly registered cases after 2003, 4202 cases registered in a single year (2005) including those initially registered before and after 2003, and a total of 2128 cases with tracheostomy positive pressure ventilation (TPPV) from all of the registered cases. All the analysed patients were classified as definite, probable or possible ALS according to the revised El Escorial Criteria for the diagnosis of ALS.

Results: In the newly registered cases after 2003, the female to male ratio was 1.43:1 and the mean age at onset was 65.3±10.7 years. In the cross-sectional view of a single year (2005), the percentage of patients using either a gastrostomy tube or a nasogastric tube was 28.7% and 7.8%, respectively. The percentage of patients with TPPV was 29.3%.

The patients with an older age at onset progressed more rapidly to the TPPV stage than those with a younger age at onset. The subpopulation of patients with long-standing TPPV showed ophthalmoplegia, while its appearance rate was less in the patients with an older age at onset than in those with a younger age at onset. Furthermore, age at onset strongly influenced the frequency of initial symptoms: dysarthria, dysphagia, neck weakness and respiratory disturbance were more frequent in patients with an older age at onset, while upper or lower limb weakness was observed more frequently in patients with a younger age at onset. In addition, the initial symptoms were still the most prominent symptoms at the follow-up stage, suggesting that the initial symptoms determine the major clinical features even in the advanced illness.

Discussion and Conclusions: Our present study demonstrated that symptomatic features are strongly influenced by the age at onset, and thus clinical features may be altered in a super-aging society.

P76 SURVIVAL AND FACTORS PREDICATING SURVIVAL IN ALS PATIENTS 15-YEARS-SURVEY PERIOD

KOSTIC S¹, STEVIC Z², PEKMEZOVIC T³, DEDIC V⁴, RAKOCEVIC V¹, LAVRNIC D¹

¹Institute of Neurology at Clinical Centre of Serbia, Belgrade, ²Department of Neurology at Clicical Centre Zvezdara, Belgrade, ³Department of Mathematics, Statistics and Computer Science at BK University, Belgrade, ⁴Belgrade University School of Medicine, Belgrade, Serbia

E-mail address for correspondence: [email protected]

Keywords: ALS, survival, prognostic factor

Background: Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disorder. Median survival from symptom onset is about 3.5 years, but some patients survive more than 5 years.

Objectives: To determine the mean duration of the disease and statistical probability of survival in ALS population in Belgrade during a 15-years-survey period (1992–2007)

Methods: A total of 292 (178 males and 114 females) ALS patients were discovered in the Belgrade district and were hospitalized between 1992 and 2007. All patients fulfilled El Escorial criteria for probable or definite ALS. Eighty patients had a bulbar onset and 212 had spinal onset of the disease. The outcome in 183 cases was determined by directly contacting patients or their families. 3 patients (men) could not be reached for follow-up and dropped out of the study after diagnostic hospitalization. Survival was assessed with the Kaplan-Meier method in the whole sample, by selected prognostic indicators (age, sex, bulbar or spinal onset, and disease duration). Multivariate analysis was done with the Cox proportional hazard function.

Results: As of December 2007, 220 ALS patients (75%) have died. Nineteen (7%) patients survived five years after diagnosis. The mean age at diagnosis was 59.4±11.2 years. Mean survival time from diagnosis for all patients was 42±4.4 months (median survival time: 24±2 months). Mean age at symptoms onset was 56.8±12.3 years for men, and 59.2±11.1 years for women but this difference was not statistically significant. Mean age at symptoms onset was 60.3±11.3 years for patients with bulbar onset, and 56.7±12.0 years for patients with spinal onset of the disease and this difference was also not statistically significant. Mean survival time from symptoms onset for all patients was 70.8±8.5 months (median survival time: 36±2.4 months). Mean survival time from diagnosis for patients with spinal onset was 43.2±4 months and 38.4±5.5 months for patients with bulbar onset (Kaplan-Meier analysis, log-rank test, p = 0.37, no significance). Mean survival time from diagnosis for men was 46.8±5 months and 37.0±4.0 months for women (Kaplan-Meier analysis, log-rank test, p = 0.43, no significance). Mean survival time from diagnosis for patients younger than 49 years was 55.6±8.4 months and 36.0±3.0 months for older than 49 years (Kaplan-Meier analysis, log-rank test, p = 0.03, significant at 95% CI). Cox regression model was applied to adjust survival times from diagnosis for sex and age group, but hazard ratios were not found to be significant.

Conclusions: Five years’ survival was 7% after diagnosis of ALS. Positive effect on survival was demonstrated in the ALS patients with the onset of the disease before the age of 49.

P76A A 10-YEAR EPIDEMIOLOGICAL PROSPECTIVE STUDY OF ALS IN PIEMONTE AND VALLE D'AOSTA, ITALY

CHIÒ A¹, CALVO A¹, GHIGLIONE P¹, LEONE M², MAZZINI L², BOTTACCHI EDO³, MUTANI R¹

¹Department of Neurology, Torino, Italy, ²Ospedale Maggiore, Novara, Italy, ³Ospedale Civile, Aosta, Italy

E-mail address for correspondence: [email protected]

Keywords: Incidence, prevalence, survival

Background: Although the implementation of population-based registers has accurately measured the epidemiology of ALS in European and North American populations, it is not clear whether ALS incidence, clinical characteristics, and survival changed over time.

Aim:.To assess the trend of epidemiological and clinical characteristics of ALS in two Italian regions, using a prospective population-based registry, in a 10-year period (1995-2004).

Methods:. The PARALS is a prospective register collecting all cases of ALS incident in two regions of north-western Italy (total population, 4,332,842) established in 1995, based on the collaboration of all the neurological departments of the area. Several concurrent sources of cases are used. Patients are prospectively followed-up from diagnosis. Survival is calculated with life tables.

Results:. In the 10-year period of observation, a total of 1259 new cases of ALS were detected, 687 men and 572 women. The mean age at onset was 64.9 years (SD 11.2) (mean 64.6 (SD 10.1); women, 65.4 (SD 10.7); p= n.s.). The mean time delay from onset to diagnosis was 10.4 months (SD 10.1), with a significant decrease in the last 5-year period (1995-1999, 11.0 months; 2000-2004, 9.7 months; p = 0.001). The mean annual crude incidence rate in the period 1995-2004 was 2.90 (95% c.i., 2.72 to 3.09). The lifetime risk of developing ALS is 1/278 among men and 1/431 among women. Incidence rates were significantly higher in men than in women. The incidence rates did not change when comparing the first 5 years (1995-1999, 2.87/100,000) and the second five years (2000-2004, 3.00/100,000) of the study. A total of 319 patients were alive at the prevalence day (December 31^st, 2004), corresponding to a crude prevalence rate of 7.4 (95% c.i., 6.6-8.3)/100,000 population. There was a slight but not significant increase of median survival time (1995-1999, 915 days; 2000-2004, 1005 days). There was a significant increase of the number of patients who underwent endoscopic nutrition (PEG) (18.1% vs. 29.2%; p = 0.0001), and non-invasive ventilation (NIV) (8.9% vs. 16.7%; p = 0.0001), whereas the number of patients who performed tracheotomy only slightly modified (8.3% vs. 10.6%).

Discussion: ALS incidence did not change during the 10-year period of observation in this population-based on the prospective register. The significant reduction of the mean time delay is likely to indicate an increased awareness of ALS among the neurologists and other physicians in the examined area. Despite the increased use of PEG and NIV, there was only a modest increase of median survival.

P77 THE AUSTRALIAN MND DNA BANK: A CONTINENT-WIDE GENE-ENVIRONMENT RESOURCE

ADAMS L, PAMPHLETT R

University of Sydney, NSW, Australia

E-mail address for correspondence: [email protected]

Keywords: DNA bank, association study, gene-environment interaction

Background: Current thinking is that sporadic MND (SMND) is a complex disorder most likely due to a gene-environment interaction. Large numbers of cases and controls are needed for genetic association studies to provide sufficient power for reproducible results. In addition, environmental exposures or demographic factors may be risk factors for SMND. Until recently no DNA collections of SMND were available in the southern hemisphere. We therefore set up an Australia-wide DNA and environmental data resource for MND researchers.

Objectives: [1] To collect blood DNA samples and environmental data from sufficient numbers of people with SMND and controls to perform gene-environment studies. [2] To offer this material to both Australian and international researchers.

Methods: Since 2000, blood samples have been collected from donors throughout Australia, both in large population centres and in remote country areas. Donors fill out 8-page questionnaires on demographic and environmental factors. Clinical information on MND donors is received from treating neurologists.

Results: The Bank currently has 1,830 blood DNA samples with matching questionnaires. Of these, 32% have SMND, 24% have SALS, and 63% are controls. The proportion of donors from each state is: New South Wales 51%, South Australia 12%, Queensland 12%, Victoria 10%, West Australia 7%, ACT 5% and Tasmania 2%. The great majority of donors are Caucasian in origin. Only 4 samples are from indigenous Australians. Samples from the Bank have been used in a number of SMND genetic studies ([1], [2], [22–24]).

Discussion: This is the first attempt to collect DNA samples from people with SMND throughout a whole continent. We estimate that we collect each year about 30% of all cases of SMND in Australia. Strategies are in place to collect more samples from remote areas. The Bank could be particularly useful for replicating genetic or environmental studies with a southern hemisphere population. Further requests for samples are anticipated as the number of donors increases to our target of 1,000 SMND patients and 2,000 controls.

P78 GENOME-WIDE STUDY OF COPY NUMBER VARIATION IN THE IRISH AND DUTCH ALS POPULATIONS

CRONIN S¹, BLAUW HM², VELDINK JH², VAN ES MA², OPHOFF RA², BRADLEY DG³, VAN DEN BERG LH², HARDIMAN O³

¹Beaumont Hospital, Dublin, Ireland, ²Rudolf Magnus Institute of Neuroscience, University Medical Center Utrecht, Netherlands, ³Trinity College Dublin, Ireland

E-mail address for correspondence: [email protected]

Keywords: CNV,genome wide study,genetics

Background: Sporadic ALS is an unrelenting neurodegenerative condition characterized by progressive limb and bulbar weakness. Genetic risk factors have been implicated in its pathogenesis. Copy number variations (CNVs) have the capacity to alter gene dosage and CNVs in SMN are a precedent for a biological role for CNVs in ALS.

Objective: To investigate the contribution of CNVs to ALS in the Irish and Dutch populations.

Methods: We did a genome-wide analysis of CNVs by comparison of Illumina 550K SNP array data on 206 Irish patients with ALS and 202 Irish controls with Illumina 317K SNP array data on 445 Dutch patients with ALS and 423 Dutch controls. The total number and length of CNVs, association of common CNVs with ALS and the co-occurrence of novel ALS-specific CNVs was examined using the QuantiSNP algorithm and data from the Database of Genomic Variants build 36.

Results: We detected 4,987 CNVs using 550K data in 408 Irish individuals and 4,103 CNVs using 317K data in 868 Dutch individuals. Using pooled data, no SNP showed a significant copy number difference between ALS patients and controls after correction for multiple testing. We identified 16 genes bearing loss or gain of copy number exclusively among ALS patients replicating in both the Irish and Dutch analyses. Among these, there were three genes where the same copy number alteration was observed in 6 individual patients. These variants did not occur in controls and have not been reported in previous studies of CNVs.

Discussion: Common CNVs in the regions of the genome included on the current SNP arrays do not alter susceptibility to ALS. However, rare CNVs observed uniquely in ALS patients may contribute to ALS pathogenesis. Future work should seek to profile the contribution of CNVs located in regions not covered on the present SNP platforms.

P79 A GENETICAL GENOMICS APPROACH TO IDENTIFY CAUSAL GENE NETWORKS IN ALS

VELDINK JH¹, SARIS C¹, VERSTAETE E¹, FRANKE L¹, VAN ES MA¹, VAN VUGHT P¹, BLAUW H¹, OPHOFF R¹, WOKKE J¹, JANSEN R², VAN DEN BERG LH¹

¹University Medical Center Utrecht, Netherlands, ²Groningen Bioinformatics Centre, University of Groningen, Netherlands

E-mail address for correspondence: [email protected]

Keywords: genomics, networks, susceptibility

Background: Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease, characterized by the selective death of central and peripheral motor neurons. Mulitple genetic risk factors are considered to be implicated in disease susceptibility.

Genome wide studies that are aimed at identifying this genetic susceptibility typically involve arrays with many expression probes or genotype probes. Also in ALS, a number of genome-wide association studies have been performed. Due to the overwhelming number of hypotheses in relation to sample sizes, these studies are most often underpowered, while biologically relevant markers and probes are present in the final results (type II errors).

It has been shown that a large part of the expression of mRNA is under genetic control, also in humans. If a genetic marker is in or near an expression probe, the genetic effect is considered to be in “cis” with the probe, e.g. due to a promoter polymorphism that drives expression. If the marker is more distant from the probe, the genetic effect is considered to be in “trans” with the probe(s). The combined analysis of genotypic data and expression data in order to elucidate the genetic determinants of mRNA expression is known as “genetical genomics”.

Objective: To identify genetically determined differential mRNA expression in ALS versus healthy controls (cis-effects), and to identify gene networks that are implicated in ALS susceptibility (trans-effects).

Methods: 217 subjects, 108 patients with ALS, and 107 age-, and sex matched controls were included. These subjects were genotyped using the 317K Illumina Beadarray SNP chip, in combination with the Illumina human-8 refseq V1 and V2 expression chips. After quantile normalization and log transformation, we used a variance components approach with multivariate linear regression to determine the associations between SNPs and expression probes, correcting for batch effects, sex, and age. In addition ALS specific SNPs that were associated with expression probes (eSNPs) were identified by also including an interaction term between genotype and group status (ALS or control).

Results: After quality control of the SNP markers, 305837 SNPs were included (out of 317503), and 15,123 expression probes, that were shared between the V1 and V2 arrays and that had a >95% unique BLAT hit in the current reference genome. Using a false discovery rate of 0.05, we identified 580 cis eSNPs, and 250 trans eSNPs, with several eSNPs showing a relationship with multiple expression probes, corresponding to potential gene networks. We will determine which cis eSNPs are determinants of the differential mRNA expression, and which trans eSNPs are specific to ALS.

Discussion: Our preliminary results show that the combined analysis of genetic markers and expression probes is able to identify which differential mRNAs are causal, rather than reactive to ALS. In addition, the identified ALS specific gene networks are interesting candidates for further (functional) investigations.

P80 GENETIC POLYMORPHISMS IN ALS REVISITED: A SYSTEMS BIOLOGY APPROACH IN SILICO

MOUGEOT JL, BONKOVSKY H, BROOKS B

Carolinas Medical Center, Charlotte, North Carolina, United States

E-mail address for correspondence: [email protected]

Keywords: Genetics, SNPs, Systems Biology

Background: Amyotrophic lateral sclerosis (ALS) is a neurological disorder characterized by the progressive degeneration of motor neurons, muscle atrophy and eventually death. There is currently no cure for ALS. The etiology of ALS is not well understood and it is unclear how genetic regulations responsive to environmental factors might affect the irreversible and progressive loss of motor neurons.

Objectives: A growing number of genes with single nucleotide polymorphisms (SNPs) or mutations have been implicated in ALS in recent literature. In order to better characterize molecular pathways, critical genetic factors and predictive patterns in ALS, we conducted a comprehensive systems biology analysis using the Pathway Studio^TM program.

Methods: Pubmed scans using search terms in Pathway Studio^TM automatically extracting gene names from abstracts were used to create pathway maps connecting proteins, complexes, small molecules and other entities such as diseases and cell processes. These maps included genes with or without known ALS-related polymorphisms (ALS-SNP genes). In parallel, a list of ALS-SNP genes was manually constructed using full text articles found in the literature presenting such genes listed in tables or text. Molecular networks identifying direct or indirect interactions/regulations between ALS-SNP genes were established. Published data indicating SNP association strength with ALS were integrated within these networks. Common targets and regulators were identified and initial networks created by Pubmed scans were used for cross-validation and refinements.

Results: Five pathway maps were obtained using the search terms “Amyotrophic lateral sclerosis” alone and with “polymorphism”, “SNP”, “mutation”, or “gene” which consisted of 593 proteins/2368 regulations, 52/114, 13/16, 174/527, and 278/912, respectively. These maps included ALS-SNP genes and interacting genes with no known ALS-SNP. The list of ALS-SNP genes based on full text articles contained 85 genes. The minimal molecular network with most direct interactions included 115 proteins, 2,027 regulations, 26 cellular processes, 10 functional classes and 18 disease entities. Gene ontology groups with p < 0.0001 for this network included 1) angiogenesis, 2) synaptic transmission, 3) response to hypoxia, 4) heme binding, 5) superoxide release, 6) response to oxidative stress, 7) potassium channel activity, 8) viral genome replication, etc. Each ontology group corresponds to a submolecular network containing ALS-SNPs gene(s) that may significantly favor or inhibit these processes.

Conclusion: Using Pathway Studio^TM program we were able to highlight diverse regulations potentially affecting or affected by ALS-SNP genes. Our approach identified closely interconnected pathways involving many ALS-SNP genes that can be influenced by environmental factors or be controlled by tissue-specific regulations. Functional analysis of these pathways is critical for the development of new drugs and molecular-based early diagnostic methods for ALS. In conclusion, systems biology tools significantly enhance the power of analysis of genetic regulations with etiologic significance in ALS.

P81 ALS IN THE OFFSPRING OF A CONJUGAL ALS CASE

PAGEOT N¹, CORCIA P², MORALES R¹, VOURC'H P³, ANDRES C³, CAMU W¹

¹CHU de Montpellier ALS Center, Montpellier, France, ²CHU de Tours ALS center, Tours, ³INSERM U930, Tours, France

E-mail address for correspondence: [email protected]

Key words: conjugal ALS, genetics

Background: Conjugal ALS cases have been described for more than 20 years. As it appeared important to determine whether the incidence of ALS in the children of those patients was higher than normally expected, we followed the offspring of 9 couples with ALS.

Objectives: To describe a family with ALS in both parents and in which a son recently also developed ALS.

Material and Methods: The couple were described in our work published in Arch Neurol (Corcia et al., 2003, family 3). The woman had bulbar ALS in 1984 at age 60 and died 18 months later. The man had flail arm syndrome since 1997 at age 72, he died 48 months later. This couple had 4 children, 2 sons and 2 daughters.

Results: The oldest child, a man, was first seen in our center in October 2005. He complained of cramps since 2004 and noted a right foot drop since April 2005. ALS was diagnosed due to diffuse fasciculations, weakness of the right leg and diffuse and active denervation in the four limbs and the abdomen. His ALSFRS was rated 45/48 at that time.

Differential diagnoses were ruled out with MRI of the brain, comprehensive biological examinations and lumbar puncture. Sequencing of the 5 exons of the SOD1 gene was normal.

He significantly deteriorated in the following months with (in February 2008) severe tetraparesis with amyotrophy and fasciculations and pyramidal syndrome in the 4 limbs. He was treated with riluzole since 2005 and has been received lithium for one month. His ALSFRS is now rated 32/48. Further familial investigations did not find any other ALS case in the family.

Discussion: To our knowledge, this is the first description of an ALS case in the child of a couple suffering from conjugal ALS. SOD1 mutations were absent. The three patients had an onset of different sites. All had, at one time in the course of their disease, both upper and lower motor neuron signs. It is not possible to draw any conclusion regarding the genetic implications of such a case. It nevertheless seems important to follow up the offspring of the conjugal ALS cases.

P82 A MULTIPLEX ALS FAMILY WITH AUTOSOMAL DOMINANT D90A MUTATION

GUY N¹, CORCIA P², VOURC'H P³, CLAVELOU P¹, ANDRES C³, CAMU W⁴

¹CHU de Clermont-Ferrand ALS center, Clermont-Ferrand, France, ²CHU de Tours ALS center, Tours, France, ³INSERM U930, Tours, ⁴CHU de Montpellier ALS center, Montpellier, France

E-mail address for correspondence: [email protected]

Keywords: SOD1 mutations, familial ALS, D90A

Background: One of the most frequent SOD1 mutations in Europe is the homozygous D90A, of recessive influence. Despite some descriptions of ALS cases with heterozygous D90A, it has not yet been demonstrated that dominant cases may exist.

Objectives: To describe a multiplex family with heterozygous D90A mutation transmitted as a dominant trait.

Material and Methods: Samples from a family from Clermont-Ferrand were collected and the SOD1 gene sequenced for all exons in one patient and 11 other family members.

Results: This family is composed of 3 ALS cases. The proband had lower limb onset by the age of 62 in 2001 and is still alive. He had unequivocal upper and lower motor neuron signs. His father died of ALS in 1970 at the age of 71. He had lower limb onset and disease duration was 10 years. The proband's sister died of ALS at the age of 52. She had lower limb onset with a rapid bulbar involvement inducing death after 3 years of disease.

Results from SOD1 sequencing are featured on the pedigree. Our proband carried het-D90A mutation. This mutation was also present in 2 children of his sister and in one child of a brother who died of myocardic infarction at 62 years old. The father had one brother. This 91 years old man also carries het-D90A but does not complain of neurologic troubles.

Discussion: In this family het-D90A is present or deduced by pedigree analysis, in all 3 ALS patients. One subject did not develop ALS, even after 90 years. One brother carrying the mutation died at 62 yrs old of another cause.

As in the hom D90A cases, all the patients had lower limb onset. This family comes from the same area as another family with compound heterozygote D90A/D96N already published. This compound heterozygote mutation was ruled out here.

As far as we know, this pedigree is the first to be described in which the distribution of the D90A mutation is undoubtedly autosomal dominant. However, penetrance is incomplete.

P83 SOD1 IN SPORADIC AND FAMILIAL ALS IN THE NETHERLANDS

VAN ES MA¹, BIRVE A², BAAS F³, WOKKE J¹, VELDINK JH¹, OPHOFF R⁴, ANDERSEN P², VAN DEN BERG LH¹

¹Department of Neurology, Rudolf Magnus Institute of Neuroscience, University Medical Center Utrecht, Netherlands, ²Institute of Clinical Neuroscience, Umeå University, Sweden, ³Department of Neurogenetics, Academic Medical Center, Amsterdam, Netherlands, ⁴Department of Medical Genetics and Rudolf Magnus Institute of Neuroscience, University Medical Center Utrecht, Netherlands

E-mail address for correspondence: [email protected]

Keywords: SOD1, D90A

Background: SOD1 is mutated in about a quarter of families with ALS in the US, the UK, Germany, Sweden, and Belgium. In most western countries SOD1 mutations can also be found in approximately 5% of sporadic ALS patients. However, SOD1 mutations appear to be very rare in families with ALS in Portugal, Switzerland and The Netherlands.

Objectives: To determine the prevalence of SOD1 mutations in a Dutch population of familial and sporadic ALS patients.

Methods: 490 sporadic and 52 familial ALS patients were screened for mutation in SOD1.

Results: 2 mutations were identified in 2 patients with sporadic ALS; a homozygous D90A mutation and one novel mutation I99V. One mutation was identified in a familial patient; heterozygous D90A. In our population 0.004% of SALS patients and 1.9% of familial patients carried a SOD1 mutation.

Discussion: Mutations in SOD1 appear to be rare in The Netherlands compared to most western countries. The finding of both a homozygous and a heterozygous D90A mutation is very interesting, considering D90A mutations behave recessively in Sweden, but cause autosomal dominant ALS in Belgium.

P84 THE E22G MUTATION IN THE CU/ZN SUPEROXIDE DISMUTASE GENE PREDICTS A LONG SURVIVAL TIME. CLINICAL AND GENETIC CHARACTERIZATION OF A SIX-GENERATION ALS1 SPANISH PEDIGREE

GAMEZ J¹, SYRIANI E², CORBERA-BELLALTA M¹, MORALES M²

¹ALS Unit Neurology Department, Hospital Universitari Vall d'Hebron, Autonomous University of Barcelona (UAB), Barcelona, ²Institut d'Investigacions Biomèdiques August Pi i Sunyer, Department of Physiological Sciences I, Facultad de Medicina-University of Barcelona, Barcelona, Spain

E-mail address for correspondence: [email protected]

Keywords: Q21G, SOD1 gene, clinical-genetic characterization

Background: Despite the genetic heterogeneity reported in familial ALS (fALS), SOD1 gene mutations are the most frequent cause of fALS, accounting for around 20% of familial cases (ALS1) and isolated sporadic cases. Some mutations are associated with a long survival time, while others are linked to a very rapid progression. For this reason, clinical-genetic characterization of ALS1 families is important as it can provide information on the phenotype associated with a given mutation, the distribution of SOD1 mutations in different ethnic groups, and can clarify the genotype-phenotype correlation in patients with SOD1 gene mutations.

Objectives: To describe the phenotype linked to this previously reported SOD1 gene mutation, E22G (E21G in the old nomenclature), in a large ALS1 Spanish kindred. This mutation was previously reported in a sporadic ALS case and in a Canadian ALS family but no clinical information was available or reported.

Methods: Clinical characterization including gender, age at onset, site of onset and survival, were available from twelve affected members, belonging to a six-generation pedigree. The possibility of gender predominance or anticipation was also analyzed. DNA samples were available in three living symptomatic members. Informed consent for blood samples was obtained. Genomic DNA was extracted from whole blood lymphocytes using a QIAamp\ DNA blood midi-kit (QIAGEN, CA, USA). PCR fragments encompassing the entire SOD1 gene were amplified from the total DNA using 5 sets of 20-mer primers. Sequence analysis of these fragments was performed with an ABI Prism 310 automatic sequencer.

Results: An A-to-G transition at nucleotide position 65 (c.65A > G) leading to an E22G residue change was identified in the three affected ALS patients. The phenotype was similar in all affected members in our E22G family. Initial symptoms occurred in the distal limb muscles, predominantly in the legs, and there was a mean survival time of approximately 13.3 years. Mean age at onset was 50.6 years (SD 8.2). The prevalence in males and females was similar, with no difference in phenotype as regards gender. The age range for onset of symptoms was between 40 and 69 years of age, although 66% of the members presented symptoms before their fiftieth birthday. The information available for five affected parent/affected offspring pairs suggested no apparent anticipation, as has been described in some SOD1 mutations.

Conclusions:. E22G is the ninth SOD1 gene mutation reported in Spain, and the third of these associated with long survival (G38R- previously G37R, and D77V- previously D76V). Our results emphasize the importance of genetic and clinical characterization of ALS1 families around the world in understanding the genotype-phenotype relationships of each SOD1 mutant and their relative frequency in different ethnic groups worldwide.

Acknowledgements:. This work was supported FIS 02/648-03/1495 SAF V-2002-PN03517-O, BFI 2003–01190, and BFU2006–04169/BFI grants.

P85 RESPIRATORY ONSET IN AN ALS FAMILY WITH L144F SOD1 MUTATION

CORCIA P², PAGEOT N¹, MORALES R¹, VOURC'H P³, ANDRES C³, CAMU W¹

¹CHU de Montpellier ALS center, Montpellier, France, ²CHU de Tours ALS center, ³INSERM U930, Tours, France

E-mail address for correspondence: [email protected]

Key words: Familial ALS, respiratory onset, SOD1 mutation

Background: ALS cases with SOD1 mutations may sometimes present with unsual clinical features such as an absence of upper motor neuron signs (e.g. A4V, G93C).

Objectives: To describe a family with initial and predominant respiratory involvement in which an L144F SOD1 mutation was uncovered.

Results: A man was refered to our center in August 2007, at 55 years old, showing symptoms of intense fatigue and dyspnea since January 2007. At that time, he experienced difficulties breathing in a lying position. His clinical state required non invasive ventilation (NIV) from July 2007. Upper and lower motor neuron signs were noted on the right upper and lower limbs with bisk tendon reflexes, fasciculations and amyotrophy. Denervation was noted at EMG on all four limbs. ALSFRS was 32/48. In January 2008, he was clearly deteriorating with an ALSFRS rated 22/48.

The mother of this patient died at 83 years old in 2004. She had dyspnea since 1990 and NIV began in 1994. In 2000, she was well adapted to NIV and in all limbs weakness was present. She had to walk with a support. Weight loss was noted but not quantified. In 2002, hypercapnia was noted (pCO2 50mmHg) despite ventilation.

The maternal uncle of our patient died several years before at the age of 70. He had initial respiratory deficiency and diffuse amyotrophy. He lived in Yugoslavia.

Discussion: L114F SOD1 mutation was initially described by Deng et al in 1993. While atypical clinical features are not exceptional in ALS families with SOD1 mutations, to our knowledge this is the first family with a respiratory onset in the three affected members. Given this atypical onset, diagnosis was not ascertained until death in the mother and the maternal uncle. While a significant number of patients with SOD1 mutations have limb and particularly lower limb onset, our family underlines the fact that a respiratory onset in an ALS patient should not exclude the possibility of either both a familial case or a SOD1 mutation.

P86 HIGH RESOLUTION MELTING (HRM) ANALYSIS FOR MUTATION SCREENING OF THE CU/ZN SUPEROXIDE DISMUTASE (SOD1) GENE

KAWAMATA-AKIMOTO C, MORITA M, NAKANO I

Jichi Medical University,Yakushiji 3311-1 Yakushiji, Shimotsuke City, Tochigi Pref., Japan

E-mail address for correspondence: [email protected]

Keywords: High resolution melting (HRM) analysis, SOD1 gene

Background: Amyotrophic lateral sclerosis (ALS) is a neurodegenerative disorder selectively affecting motor neurons in the spinal cord, brain stem and cerebral cortex. Mutations in the Cu/Zn superoxide dismutase (SOD1) gene are the most frequent genetic defects known to underlie ALS, accounting for 20% of familial cases and 2 to 4% of apparently sporadic ALS (SALS). We tried to use high resolution melting (HRM) analysis as a rapid and easy tool to screen mutations in the SOD1 gene.

Objective: To screen mutations in the SOD1 gene we performed HRM analysis. Also, we applied this method to examine the prevalence of SOD1 mutations in Japanese SALS cases.

Method: With informed consent, genomic DNA was extracted from lymphocytes using standard procedures. We designed PCR primers for HRM analysis to screen all 5 exons of the SOD1 gene. First, we examined 19 known SOD1 mutations to determine the sensitivity of this method. Then we used this method for approximately 190 Japanese SALS cases who fulfilled the revised El Escorial criteria.

Results: HRM analysis could clearly distinguish 18 of 19 previously identified mutations in normal controls. Three different SOD1 mutations were found in the SALS cases. The mutations identified were Cys6Tyr (C6Y) and Gln22His (Q22H) in exon 1, and Ser134Thr (S134T) in exon 5.

Discussion and Conclusion: The prevalence of SOD1 mutations among Japanese SALS cases was 1.6%. HRM analysis is a sensitive method to screen mutations in the SOD1 gene.

P87 SCREENING OF SUPEROXIDE DISMUTASE 1 GENE MUTATION AND SNP GENOTYPING BY HIGH-RESOLUTION MELTING IN CHINESE AMYOTROPHIC LATERAL SCLEROSIS PATIENTS

LI X, ZHANG J, LIU M, CUI L

Peking Union Medical College Hospital, Beijing, China

E-mail address for correspondence: [email protected]

Keywords: Amyotrophic lateral sclerosis, single nucleotide polymorphisms, Chinese population

Background: Approximately 20% of FALS cases and 2% of overall ALS cases have more than 130 different identifiable mutations located in all five exons of Cu/Zn superoxide dismutase (SOD1) gene worldwide. We have reported our screening results of SOD1 gene mutations which identified by DNA sequencing in sporadic cases with five mutation found in 111 cases and 19 individuals from three ALS family, but routine screening for SOD1 mutation using DNA sequencing is expensive, more economic procedure needed. Sporadic ALS may be caused by the interaction of multiple environmental factors and previously unknown genes. A number of association studies have been performed in an effort to find genetic component, several specific SNPs seem to be strongly associated with susceptibility to SALS in different populations of European and American ancestry. To assess the role of those genetic variations in Chinese ALS, we performed a SNP genotyping.

Objectives: We developed a procedure for screening of SOD1 gene mutations and SNP genotyping using a PCR and high resolution melting analysis method.

Methods: The El Escorial ALS diagnostic criteria were used. Genomic DNA was prepared from ALS patients and health controls using standard procedures. PCR amplification of five exons of the SOD1 gene was performed using primers as previously described. Additional samples with known mutations were used as positive controls. Asymmetric PCR amplification of DPP6 and FLJ10986 was performed using self designed primers. After PCR, melting curves were generated by monitoring the fluorescence of a saturating dye on LightScanner. Statistical analysis was performed comparing the ALS and Control groups.

Results: A V47A mutation was found in exon 2 of SOD1, which was confirmed by DNA sequencing, with a different phenotype compared with the same allele locus mutation reported (V47F). We found a variant on 8,227 base of SOD1, shown to be a SNP. None of the suggested associations (DPP6 and FLJ10986) were found in Chinese ALS patients or healthy controls.

Discussion and Conclusions: This study reports a simple and economic procedure for screening SOD1 mutations and SNP genotyping in Chinese ALS patients. Identification of the V47A mutation increases the number of SOD1 mutation to 6 in Chinese ALS SOD1 mutation screening series. High resolution melting analysis shows excellent accordance with identified by DNA sequencing in the samples with some mutations reported as previously (V29A, H46R, N86 I, G72C and E133V). SNP variation in DPP6 and FLJ10986 failed to show a major effect on susceptibility to Chinese sporadic ALS in this study. Further confirmation is warranted in additional large sample size of the Chinese population.

P88 INVESTIGATION OF FAMILIAL ALS IN CHINA BASED ON A COUNTRY-WIDE DATABASE

DENG M, ZHANG J, ZHANG N, ZHANG H, ZHANG Y, FAN D

Department of Neurology, Peking University Third Hospital, Beijing, China

E-mail address for correspondence: [email protected]

Keywords: Chinese, pedigree, copper/zinc superoxide dismutase 1

Background and Objectives: Familial amyotrophic lateral sclerosis (FALS) is reported in approximately 5%-10% of ALS in the world. Inheritance is usually autosomal dominant with variable penetrance. Mutations in copper/zinc superoxide dismutase 1 (SOD1) are found in approximately 20% of FALS. As a huge country with a big population, however, China has no data of FALS so far. In this study, we try to investigate the number of Chinese patients with FALS based on our country-wide database.

Methods: Data from ALS patients included in the ALS Clinic database of Peking University Third Hospital from January 2005 to end of March 2008 were included. The diagnosis of ALS follows the El Escorial criteria (revised version). The geographical distribution of patients covers more than 93% of Chinese administrative provinces (excluding Taiwan and Tibet). Some of these families have been screened for SOD1 mutations.

Results: During the study period, a total of 749 individuals were diagnosed with probable or definite ALS. Among them, 11 pedigree patients from 10 different provinces all over the country were found to have a positive family history of ALS, with the bulbar-onset form accounting for 9.1% (1/11), and the rest 90.9% limb-onset form (10/11). Seven family pedigrees included from two to five patients with ALS (63.6%), while four pedigrees had more than 12 affected individuals (36.4%). Two of these families had pedigrees of more than 20 patients in 6 generations, with fully penetrant autosomal dominant inheritance (18.2%). The total FALS patients from 11 family pedigrees were 84 (accounting for 11.2% in our ALS database, 84/749), and the male-to-female ratio was 0.79 (37/47), with a median age of onset of 48.6 years. Following their diagnosis, a median lifespan was 5.2 years, ranging from 1 year to longer than 15 years. Most FALS patients show autosomal dominant inheritance (90.9%) except for one with autosomal recessive inheritance (9.1%). SOD1 analysis was performed in 5 of 11 FALS family pedigrees, and only two families (40%) were identified SOD1 mutations, in which one is a rare mutation Ser105Leu in exon 4, and another is a novel mutation Glu133Val in exon 5, respectively, both with quite different from those previously described.

Conclusions: In this study, we found the percentage of FALS patients is 11.2% in our country-wide database and these 11 families come from 10 Chinese administrative provinces. The survival of FALS cases seems longer, but the form of bulbar-onset and the difference of gender are lower than reported. The incidence and pattern of inheritance are similar to the data reported in the literatures. Our data is the first report of FALS in China, which will be important and helpful for researchers on ALS genetics.

P89 A CHROMOSOME 21-LINKED FAMILY WITH ALS BUT NO MUTATIONS IN THE SOD1 GENE

VALDMANIS PN¹, LEE J¹, BELZIL VV¹, ST. ONGE J¹, VERLAAN DJ¹, CAMU W², ROULEAU GA²

¹Center of Excellence in Neuromics, CHUM Notre Dame Hospital, Montreal, QC, Canada, ²Unité de Neurologie Comportementale et Dégénérative, Institute of Biology, Montpellier, France

E-mail address for correspondence: [email protected]

Keywords: SOD1, linkage analysis, mutation detection

Background: Mutations in the SOD1 gene have been well-established as a cause of approximately 15–20 percent of cases of familial ALS. We have identified a multigenerational family with ALS from France. No mutations were identified in the SOD1 gene after sequencing the complete gene in five affected individuals with ALS.

Objectives: Our objective is to determine the genetic cause for ALS in this family.

Methods: DNA from seven individuals (three affected, four unaffected) was sent for a 550-marker eight centiMorgan whole genome scan. This was followed by PCR amplification of the SOD1 gene using various sets of primers.

Results: The genome scan yielded only one locus with a LOD score above 1.5, specifically a region on chromosome 21 between markers D21S1432 and D21S266. This region contains the SOD1 gene. A follow-up PCR analysis was conducted with a second set of non-overlapping primers for SOD1 which spans the coding and UTR regions. In addition, a long-range PCR strategy was employed to test for the possibility that intronic insertions or deletions may be present. In both cases, no causative variants were detected.

Discussion: Several possibilities exist as to why no mutations were identified. A mutation may exist in a regulatory region, or deep within an intron. Alternatively, a larger and undetected copy number variant may be present at the SOD1 locus. Finally, another nearby gene which is causative for ALS may be mutated in this family. It was the examination of this last scenario for the MAPT locus on chromosome 17 which led to the discovery that mutations in the progranulin gene cause frontotemporal dementia. Thus, the identification of the mutation that is responsible for ALS in this family will enable an understanding of further genetic mechanisms for developing ALS and may help determine the genetic cause of ALS for a proportion of familial ALS patients.

P90 CHARACTERISATION OF THE PROPERTIES OF A PUTATIVE NEW DISEASE ASSOCIATED FAMILIAL ALS LOCUS ON CHROMOSOME 12: D-AMINO ACID OXIDASE

MITCHELL J, PRAVEEN P, CHEN JH, MORRIS A, DE BELLEROCHE J

Imperial College London, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: molecular genetics, amino acid metabolism, neurogenetics.

Background: Following the discovery of SOD1 mutations in familial Amyotrophic Lateral Sclerosis (FALS), two further genes have been identified causing classical ALS, VAPB and TDP-43. However, these three genes alone can account for no more than 30% of cases and the remaining causal genes remain to be elucidated.

Objectives: To carry out a genome screen using microsatellite markers in extended kindreds in order to identify potential new FALS loci

Methods: A genome screen was carried out in extended FALS kindreds which identified a 9.2 cM region on chromosome 12 linked to disease in a single family. A disease-associated mutation was detected following sequencing of candidate genes in this region and N-terminal GFP fusion constructs were generated and expressed in COS-7 and neuroblastoma (N2a) cell lines to evaluate the effect of this mutation.

Results: A putative locus was indicated on chromosome 12q22–23 from a maximum multipoint LOD score of + 3.33 at D12S1646 localising the region to a 9.2 cM interval between the markers D12S330 and D12S79. This region was screened for mutations in candidate genes and a point mutation, R199W, in the open reading frame of the D amino acid oxidase (DAO) gene was demonstrated in all affected individuals in this extended family. We screened for the R199W mutation in a total of 439 unrelated individuals (277 controls, 23 sporadic ALS cases and 139 FALS index cases, all of Caucasian origin residing in the UK) in order to determine its prevalence in the general population but no further individuals carried this mutation. To obtain insight into the effect of the R199W mutation, the DAO transcript and enzyme activity were examined in post mortem spinal cord and motor cortex tissue available from an individual with the R199W mutation and compared to control and sporadic ALS cases. The expression of GFP fusion DAO in COS-7 cells (R199W DAO and wild-type DAO) yielded abundant levels of protein (∼39kDa) corresponding to DAO, that localises appropriately to peroxisomes. Enzyme activity was severely impaired in the mutant form, obtained both from cell extracts and the autopsy case from the affected pedigree

Discussion and Conclusion: This is the first report of an ALS-associated mutation in DAO in FALS. DAO is abundant in spinal cord where it has been demonstrated to regulate the major excitatory neurotransmitter, glutamate, through its action on D-serine. DAO also has an important role in the inactivation of toxic D-amino acids, which accumulate in the central nervous system with age. A mutation at the active site of DAO is likely to lead to pathological consequences which could underlie ALS pathogenesis through the potentiation of glutamate-mediated neurotoxicity. Importantly, D-serine is now known to accumulate in the spinal cord in ALS and in the SOD1 mouse model of ALS indicating that this abnormality may represent a fundamental component of the disease process. Our long-term aim is to establish whether this DAO mutation is causal in ALS and whether modifying D-serine levels affects disease progression.

P91 MONOAMINE OXIDASE B GENE POLYMORPHISM IN PATIENTS WITH AMYOTROPHIC LATERAL SCLEROSIS

SAZCI A, YILMAZ M, IDRISOGLU H

¹University of Kocaeli, Faculty of Medicine, Department of Medical Biology and Genetics, Kocaeli, Turkey, ²University of Istanbul, Faculty of Medicine, Department of Neurology, Istanbul, Turkey

E-mail address for correspondence: [email protected]

Keywords: MAOB, Polymorphism, Association

Background: Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease affecting 1–3 in 100,000 worldwide. Its prevalence is somewhat more in men, and has an average onset age of 56 years, with onset ranging from 20 to 80 years. ALS is a sporadic disease with 10% familial. Death occurs within 3 to 5 years after the first symptoms from respiratory failure.

Monoamine oxidase B (MAO B) is a mitochondrial outer membrane flavoenzyme involved in the degradation of biogenic amines, and is a well-known target for anti-depressant and neuroprotective drugs. MAO B levels induce apoptosis in neuronal and kidney cells. Elevated MAO B levels have been also reported in plaque-associated astrocytes of brains from Alzheimer's patients. MAO B may also be important in pathological processes resulting from exposure to various xenobiological compounds. MPTP is biologically activated by MAO B which causes chemically induced Parkinson's disease. Components in tobacco smoke inhibit MAO B gene expression. The MAO B gene is located on Xp11.23, and the protein contains 520 amino acids. The crystal structure of MAO B reveals that it is dimeric.

Objectives: The fact that MAO B may play a role in the pathogenesis of neurodegenerative disorders, has been based on the assumption that this enzyme is responsible for neuron oxidative damage through the production of free radicals. We analysed a single base difference (A or G) in intron 13 of the MAO B gene in connection with sporadic amyotrophic lateral sclerosis (SALS).

Methods: We used a PCR-RFLP method to analyse the genotypes and SPSS to do the statistics.

Results: We studied 248 SALS patients (age range 48.40±16.840) and 116 (age range 52.67±15.522) controls. In this cohort study, the MAO B gene polymorphism was not associated with SALS (Chi2 = 2.231; P = 0.328). The distribution of the MAO B AA, AG, and GG genotypes was 44.0%, 19.0%, and 37.1% in controls and 50.8%, 19.8%, and 29.4% in cases respectively. The distribution of the MAO B A allele was 53.45% in controls and 60.69% in cases.

Conclusion: In conclusion, the MAO B intron 13 AG polymorphism was not associated with SALS. Therefore it is unlikely to play a role in the etiology of SALS.

P92 COMT VAL158MET POLYMORPHISM AND SUCCEPTIBILITY TO AMYOTROPHIC LATERAL SCLEROSIS

YILMAZ M¹, SAZCI A¹, IDRISOGLU H²

¹University of Kocaeli, Faculty of Medicine, Department of Medical Biology and Genetics,Umuttepe,, Kocaeli, Turkey, ²University of Istanbul, Faculty of Medicine, Department of Neurology,, Istanbul, Turkey

E-mail address for correspondence: [email protected]

Keywords: COMT gene; Polymorphism; Association

Background: Amyotrophic lateral sclerosis (ALS) is a multifactorial neurogenetic disorder characterized by the gene interactions and environmental factors. Epidemiologic studies on ALS reveal that the prevelance of ALS is 1 to 3 in 100,000 worldwide. The mean ALS onset age is approximately 56.3±11.1 years, with no gender difference.

Catechol-O-methyltransferase (COMT) is a ubiquitous enzyme that catalyzes the O- methylation of catechols and plays a role in the metabolism of neurotransmitters such as dopamine, noradrenaline and adrenaline. The COMT gene, located on chromosome 22q11, encodes both soluble and membrane-bound forms of this enzyme. A G-to-A transition at codon 158 of the COMT gene, resulting in the substitution of methionine for valine, is designated the L (low activity) allele, in contrast to the H ( high activity) allele.

Objective: The aim of the study was to determine whether the COMT Val158Met polymorphism was associated with ALS.

Methods: A PCR-RFLP method was used in analysing the genotypes. Statistics were calculated using the SPSS v 12.

Results: We studied 124 ALS patiens and 124 controls. The distribution of the COMT 158 H allele was 56.85% in cases and 52.42% in controls. The distribution of the COMT 158L allele was 43.15% in cases and 47.58% in controls.

Conclusions: The COMT Val158Met polymorphism was not associated with ALS (X²=2.606; P = 0.272). It seems that the COMT Val158Met polymorphism does not play a role in the etiopathogenesis of ALS.

P93 GENETIC SCREENING OF THE NRF2 ANTIOXIDANT PATHWAY IN AMYOTROPHIC LATERAL SCLEROSIS (ALS)

GOODALL E, SMITH W, KIRBY J, HARTLEY J, NIXON H, WOOD-ALLUM C, SHAW P

University of Sheffield, South Yorkshire, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: NRF2, Antioxidant pathway, Genetic screening

Background: Oxidative stress has been implicated in the pathogenesis of amyotrophic lateral sclerosis (ALS) and other neurodegenerative diseases. Antioxidant response elements (AREs) are transcription enhancers located in the promoters of antioxidant defence and phase II detoxification genes. ARE-containing genes are up-regulated via the redox-sensitive transcription activator nuclear factor erythroid-2 related factor 2 (NRF2). Under conditions of oxidative stress this protein translocates to the nucleus where it activates the transcription of ARE-containing genes. There is accumulating evidence to suggest that the NRF2/ARE pathway is disrupted in ALS. The expression of NRF2 and ARE-containing genes is decreased in mutant SOD1 cell culture models and familial ALS spinal cord motor neurones ([1],[2],[3]). Increased ARE-driven gene transcription in astrocytes is, moreover, protective of co-cultured neuronal cells suggesting that modulation of the NRF2/ARE pathway could be a useful therapeutic strategy in ALS ([4],[5]).

Objectives: To sequence genes from the NRF2/ARE pathway in ALS patients and controls in order to identify ALS associated genetic variants. The functional consequences of any disease associated genetic changes discovered will be investigated.

Methods: DNA was extracted from post-mortem brain tissue obtained from neuropathologically proven ALS patients and control subjects stored within the Sheffield (UK) brain tissue bank. Tissue was donated for the purpose of research and fully informed consent obtained. Exons of candidate genes were amplified using the polymerase chain reaction and sequenced using BigDye® terminator methodology. Sequence analysis was performed using Chromas and Sequencher™ software.

Results: To date we have screened the 5 exons and promoter region of NRF2, plus the protein coding exons of three ARE-containing genes in 126 ALS patients. A number of novel sequence variants have been identified. Those that change the amino acid sequence, and may therefore have a functional effect on the protein, have been prioritised for mutation screening in additional patient and control samples.

Discussion: This is the first genetic study to target the NRF2/ARE pathway in ALS patients. NRF2 is a master regulator of antioxidant defence. If function altering genetic changes are identified in ALS patients, this will provide further evidence of disrupted NRF2/ARE signalling in ALS and pave the way for further functional studies.

P94 GENETIC VARIATION IN DPP6 GENE IS NOT ASSOCIATED WITH SUSCEPTIBILITY TO AMYOTROPHIC LATERAL SCLEROSIS IN TWO EUROPEAN POPULATIONS

FOGH I¹, D'ALFONSO S², GELLERA C³, BELEZA A¹, RATTI A⁴, CEREDA C⁵, CORRADO L², PENCO S⁶, TICOZZI N⁴, SORARÙ G⁷, CASTELLOTTI B³, GAGLIARDI S⁵, MAZZINI L², COZZI L⁶, AL-CHALABI A¹, POWELL J¹, SILANI V⁴

¹Institute of Psychiatry King's College, London, United Kingdom, ²Università del Piemonte Orientale A. Avogadro, Novara, Italy, ³IRCCS Istituto Nazionale Neurologico Carlo Besta, Milano, Italy, ⁴IRCCS Istituto Auxologico Italiano, Milano, Italy, ⁵IRCCS Istituto Neurologico C. Mondino, Pavia, Italy, ⁶Ospedale Niguarda Ca'Granda, Milano, Italy, ⁷Università degli Studi di Padova, Padova, Italy

E-mail address for correspondence: [email protected]

Keywords: amyotrophic lateral sclerosis, genome wide association,dipeptidyl-peptidase 6

Background: Only 15% of amyotrophic lateral sclerosis (ALS) patients have a family history, the remaining cases occurring sporadically. While familial ALS is well characterized with several causative genes identified to date, the genetics of sporadic ALS is poorly understood. Last year three independent genome-wide association (GWA) studies ([1],[2],[3]) found no single gene strongly associated with susceptibility to ALS; suggesting that a complex interaction between environmental factors and many susceptibility genes of small effect best describes this disease. One such susceptibility gene of small effect recently nominated by a GWA study from different populations of European origin is dipeptidyl-peptidase 6 (DPP6) which slightly increases the risk of developing ALS (OR 1.3) ([4], [5]).

Aims: To confirm the association of DPP6 with ALS phenotype we tested the candidate polymorphism rs10260404 in two other European populations, Italian and British.

Methods: The Italian cohort included 980 cases and 1119 healthy controls while the British cohort consisted of 500 cases and 500 controls. The Italian cohort tested in this study was collected by the Italian ALS Consortium created from the collaboration of 7 different institutions located in North Italy. We are recruiting more cases in other Italian neurological institutions for the first large-scale Italian SALS Screen. The British cohort was collected at King's College London. Individuals were genotyped following standard procedures (TaqMan, Applied Biosystems) and allelic discrimination was scored using an ABI799HT.

Results: Assuming an odds ratio of 1.3 with a causative allele frequency of 0.4, as described by the authors, we have > 99% power to detect an association at p = 0.05. Preliminary data show no evidence of association with susceptibility to ALS (Fisher's exact chi-squared test p = 0.9). Genotype scores were tested for association with age at onset (AAO) on 700 cases of the Italian cohort and no significant association was detected.

Conclusion: The polymorphism rs10260404, as with other variants reported to be associated with sporadic ALS, was not replicated in a different population. These findings highlight the genetic heterogeneity of sporadic ALS even within European populations.

P95 ASSOCIATION OF THE H63D HFE ALLELE WITH AMYOTROPHIC LATERAL SCLEROSIS

MITCHELL R¹, TRAYNOR B², STEPHENS H¹, SIMMONS Z¹, CONNOR J¹

¹Penn State University Milton S. Hershey Medical Center, Hershey, PA, United States, ²Laboratory of Neurogenetics, National Institute on Aging, National Institutes of Health, Bethesda, MD, United States

E-mail address for correspondence: [email protected]

Keywords: genetics, HFE, iron

Background: Five independent groups have published the frequencies of HFE polymorphisms in ALS patients compared to various control groups. A meta-analysis of these studies shows that possession of at least one H63D HFE allele is associated with a 26% increased risk of developing ALS.

Objective: To determine if the haplotype containing the H63D HFE allele is associated with ALS and determine the linkage of this haplotype with other haplotypes.

Methods: DNA samples were obtained from 115 ALS patients and 121 normal controls from a single outpatient Neurology clinic. DNA samples were also obtained for 557 ALS patients and 519 normal controls from the NINDS Neurogenetics Repository at the Coriell Institute for Medical Research. We genotyped for the H63D allele and surrounding SNPs on chromosome 6 between positions 25,783,000 and 26,414,000 spaced approximately every 10 kb by sequencing. Additionally, we determined the linkage of the H63D HFE with nearby haplotype blocks.

Results: In the current study, the H63D HFE allele was not significantly associated with ALS (heterozygotes: OR 0.90, 95% CI 0.70–1.17; homozygotes: OR 1.69, 95% CI 0.87–3.28). Combining our results with previous association studies shows a dose-dependent association of the H63D HFE allele with ALS (heterozygotes: OR 1.18, 95% CI 1.03–1.34; homozygotes: OR 2.04, 95% CI 1.48–2.80). The common region containing the H63D allele spanned 44.3 kb including the genes HFE, HIST1H4C, HIST1H1T, HIST1H2BC, HIST1H2AC, and was the same between cases and controls.

Discussion and Conclusions: While the results of our current study show no significant association of the H63D HFE allele with sporadic ALS, our study may have been underpowered to detect a significant association, particularly for H63D homozygotes. The combined analysis of all available data, however, shows a significant association of this allele with ALS in a dose-dependent fashion. The region containing the H63D HFE allele includes several highly conserved genes that increases confidence in the association between the H63D variant and ALS. Furthermore, the HFE gene variants are known to have potentially broad-ranging effects resulting from iron dyshomeostasis. The high degree of linkage disequilibrium in the region containing the HFE gene does not exclude the contribution of other genetic variants to the risk of sporadic ALS. The partial linkage disequilibrium between this region and other surrounding regions further extends the number of loci that may be associated with ALS. The data support the need for functional analysis of the HFE genetic variants to contribute to our understanding of the roles of these variants to ALS pathogenesis.

P96 MUTATIONS OF THE ANGIOGENIN GENE IN FRENCH PATIENTS WITH SPORADIC AMYOTROPHIC LATERAL SCLEROSIS

PRALINE J¹, VOURC'H P², PAUBEL A³, MEININGER V⁴, CAMU W⁵, CORCIA P¹, ANDRES CR²

¹ALS Centre, CHRU de Tours, Tours, ²INSERM U930, Tours, ³Biochemistry and Molecular Biology Department, CHRU de Tours, Tours, ⁴ALS Centre, CHRU Pitié Salpétrière, Paris, France, ⁵ALS Centre, CHRU de Montpellier, Montpellier, France

E-mail address for correspondence: [email protected]

Keywords: Genetics, Pathophysiology, Angiogenesis

Background: The pathophysiology of amyotrophic lateral sclerosis (ALS) remains unknown. Some factors that play a role in the angiogenesis have been linked to ALS. Several studies on the vascular endothelial growth factor gene (VEGF) suggested a link between angiogenesis and the pathogenesis of ALS. Recently, seven missense mutations in a second hypoxia responsive gene, the ANG gene, have been identified. The ANG gene located in 14q11.2 encodes angiogenin, a 14.1 kD enzyme which belongs to the pancreatic ribonuclease A superfamily. Angiogenin, whose expression is induced by hypoxia, mediates neovascularization and has been recently involved in neurite pathfinding. The characterization of new mutations in the ANG gene will certainly help to elucidate the molecular mechanisms linking ANG to ALS.

Objectives: To assess the frequency of ANG gene mutations in 855 French sporadic ALS patients.

Methods: We analyzed the coding region of the ANG gene in a large French cohort of 855 sporadic ALS patients.

Results: We observed a previously identified mutation (pI46V) in two ALS cases with no known family history, and found a novel mutation (pR121H) in one patient who developed ALS with rapid progression. We did not find an association between ALS patients and the rs11701 polymorphism, as previously reported in certain ALS populations of other ethnic origins.

Discussion and Conclusion: Overall, our findings support the implication of ANG gene mutations as a rare but widespread aetiology of ALS. The implication of the mutations in the ANG gene in ALS will have to be assessed by biological assays on angiogenesis and neurite pathfinding, and molecular assays on nuclear transport, RNAse activity, and rRNA synthesis.

P97 MUTATION SCREENING OF ALS2/ALSIN IN PATIENTS SUFFERING FROM PRIMARY LATERAL SCLEROSIS (PLS) AND UPPER MOTOR NEURON DOMINANT AMYOTROPHIC LATERAL SCLEROSIS (UMD-ALS)

RAMAN R, SHAW P, KIRBY J, MCDERMOTT C, NIXON H, HARTLEY J

University of Sheffield, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: Alsin,mutation,PLS

Background: PLS is a rare, sporadic, adult-onset variant of motor neuron disease (MND) in which upper motor neurons (UMN) giving rise to the pyramidal tract progressively degenerate. However, clinically there exists an overlap between PLS and Amyotrophic Lateral Sclerosis (ALS) and there is a debate as to whether PLS is a distinct disease entity or at one end of a spectrum of MND. Mutations in ALS2 have been identified in a number of UMN disorders including Juvenile ALS, Juvenile PLS and Inherited Ascending Hereditary Spastic Paraperesis. Thus mutation screening of ALS2 in clinically diagnosed PLS and upper motor neurone dominant (UMN-D) ALS patients may reveal mutations or polymorphisms associated with the disorder

Objectives: To screen the 34 exons of ALS2 for mutations in a cohort of 95, clinically diagnosed PLS and UMN-D ALS patients.

Methods: 37 amplicons encoding the entire protein coding regions and intron/exon boundaries of ALS2 were amplified and bi-directionally sequenced at a Core Genomics Sequencing Facility. Sequence chromatograms were analyzed using FinchTV (Geospiza) and aligned with Sequencher (Genecodes)

Results: Twenty three nucleotide substitutions spanning the entire ALS2 gene have been identified. Out of these, 8 are previously unidentified variants. The variant c.1960A > G; pSer654Gly occurs at a highly conserved residue and the change of polar to hydrophobic amino acid may affect protein function. Variant c.1737 + 1A > G is predicted to abolish splicing of intron 7 and therefore likely to produce a mutant protein. Neither of these changes were found in 322 control chromosomes. The remaining 6 novel substitutions are situated in intronic sequences >50 bp from the nearest exon.

Conclusions: The ALS2 was screened in a large cohort of adult onset PLS and UMN-D patients. Of the 8 previously unidentified variants, 2 are predicted to alter the Alsin protein and may therefore represent factors conferring susceptibility to UMN degeneration.

P98 NOVEL MUTATIONS OF SENATAXIN GENE ASSOCIATED WITH SPORADIC CASES WITH AMYOTROPHIC LATERAL SCLEROSIS

WU ZY¹, ZHAO ZH¹, CHEN WZ², WANG N²

¹Huashan Hospital, Fudan University, Shanghai, China, ²First Affiliated Hospital, Fujian Medical University, Fuzhou, China

E-mail address for correspondence: [email protected]

Keywords: senataxin gene, mutation, sporadic amyotrophic lateral sclerosis

Background: Amyotrophic lateral sclerosis (ALS) is the commonest adult onset disorder of motor neurones, and among the most common of adult onset neurodegenerative diseases. Although enormous scientific progress has been made, its etiology and pathogenesis are still not well known, especially for sporadic ALS (SALS). FALS is caused by mutations of many genes including SOD1 (ALS1), ALSIN (ALS2), Senataxin (SETX, ALS4) and VAPB (ALS8). SETX is a related gene of ALS4. ALS4 is a rare autosomal dominant form of juvenile amyotrophic lateral sclerosis, characterized by distal muscle weakness and atrophy, normal sensation, and pyramidal signs. Patients with ALS4 usually have a juvenile onset of symptoms (ages < 25 years), a slow rate of progression with a normal life span. SOD1 mutations are found in some apparently SALS patients, so hypothesised that SETX mutations may be associated with SALS patients.

Objectives: 45 patients with SALS from the Han ethnic group in South China and 120 healthy controls were studied.

Methods: All 21 exons of SETX including the intron-exon boundaries were amplified in the subjects using primer combinations, based on the intronic sequences of SETX. The full length of exon10 is 4,176bp, thus 10 pairs of primers were designed flanking this area. PCR products were generated using a GeneAmp PCR system 9700 (Applied Biosystems, Foster City, CA, USA) using standard conditions. Amplified products were purified and subjected to direct sequencing. The sequences obtained were compared with genomic DNA sequence of SETX (NCBI Sequence Viewer NT_022184).

Results: All 21 exons of SETX were analyzed in 45 patients with SALS and 120 healthy controls. One missense mutation, one synonymous mutation and 19 novel polymorphisms were identified. The patients with mutations had different clinical phenotypes from ALS4.

Conclusions: Our study implied SETX may be associated with SALS as well as ALS4. Although the significance of the sequence variations is still unknown, the study broadened the mutation spectrum and polymorphism spectrum of SETX and expanded the clinical phenotype of SETX mutation. It might be productive to search in larger samples.

P99 INVESTIGATION OF SMN GENE IN ALS PATIENTS OF CHINESE ORIGIN

WANG N¹, SU ZQN¹, CHEN WJ¹, WU ZY²

¹First Affiliated Hospital, Fujian Medical University, Fuzhou, China, ²Huashan Hospital,Fudan University, Shanghai, China

E-mail address for correspondence: [email protected]

Keywords: SMN1, SMN2, sporadic ALS

Background: Amyotrophic lateral sclerosis (ALS) is a rapidly progressive, incurable and terminal disease, whose pathogenesis is not clear. For years, some researchers reported that survival motor neuron (SMN) which is the determining gene of spinal muscular atrophy (SMA) is related to ALS, but there is no consistency of results. Moulard B et al [1] considered SMN2 deletion to be a risk factor for the specific subset of LMN ALS. Veldink JH et al [2] reported SMN2 deletion as a risk and poor prognostic factor for sporadic ALS. Corcia and colleagues [3] reported that abnormal SMN1 gene copy number is risk factor for sporadic ALS. In 2005, Veldink JH et al [4] further reported that lower SMN2 copy number and lower level of estimated SMN protein increased susceptibility to and mortality rate of ALS. In 2006, Corcia P et al [5] further reported that abnormal SMN1 gene copy number is a genetic risk factor in sporadic ALS and there was no modulator effect of the SMN2 gene. There are no related reports of the Asian population, especially of Chinese origin.

Objective: To investigate the information of SMN1/SMN2 deletions and SMN2 copy number in sporadic ALS patients and normal controls of Chinese origin, to investigate the risk factors for ALS in Chinese.

Methods: 61 patients with sporadic ALS and 60 healthy individuals of the Chinese Han population were screened for SMN1 or SMN2 deletions using PCR-RFLP, the detailed procedure for this has been described previously [6]. All of the subjects were further investigated for the SMN2 copy number by real-time fluorescence quantitative PCR which has also been described previously [7].

Results: No SMN1 deletion has been detected in any of the ALS patients and healthy individuals. Two of the 60 healthy individuals were determined as having SMN2 deletions and the frequency was 3.3%, but none of the ALS patients have SMN2 deletions. The numbers of ALS patients who possess 1, 2, 3, 4 copies of SMN2 are 13 (21.3%), 30(49.2%), 16(26.2%) and 2(3.3%) respectively, the numbers of healthy individuals who posses 1, 2, 3, 4 copies of SMN2 are 18(30.5%), 25(41.7%), 12(19.4%) and 5(8.4%) respectively. Chi-square test comparing the SMN2 copy number distribution between ALS patients and healthy individuals showed no significant difference (p = 0.375).

Conclusions: Our results don't show the correlations of SMN1/SMN2 deletions and SMN2 copy number to ALS. As the case number of this report is small, it is necessary to collect more Chinese ALS patients and carry out further studies in the future.

P100 THE INITIAL GENE STUDY ON SPORADIC AND FAMILIAL JUVENILE MUSCULAR ATROPHY OF THE DISTAL UPPER LIMBS

HUANG X, LIU L, LI N, FENG C, CHEN Z, PU C

Chinese PLA General Hospital, Beijing, China

E-mail address for correspondence: [email protected]

Keywords: Hirayama disease, Inheritance, X-chromosome

Background: Hirayama disease (HD) is also called juvenile non-progressive amyotrophy. Its mechanism is not clear. The disease is usually sporadic. Familial occurrence is occasional. There are only two cases in one pedigree in the reported pedigrees. Research on the molecular genetics of the disease has so far been restricted by the lack of a large pedigree.

Objective: A pedigree and sporadic cases of HD were used to investigate the role of SOD1 and SMN genes in the pathogenesis of the disease. To investigate the gene location of a Chinese HD pedgree.

Methods: Blood samples were collected from 14 sporadic HD patients and 15 members of the familial HD in which there are 6 patients in 3 generations. Polymerase chain reaction and single-strand conformation polymorphism of the SOD1 gene and PCR- restriction enzyme digestion analysis of exon 7 and exon 8 in SMN gene were scanned in 14 sporadic HD and in the patients in the pedigree of HD. DNA samples of the 15 members of familial HD were analyzed with 18 polymorphic microsatellites distributed on the X chromosome on the Human MapPairs marker set at 10cM interval by PCR. The linkage between eighteen markers on chromosome X and Hirayama disease was tested by MLINK software in the package of LINKAGE (version 5.1).

Results: The 5 exons of SOD1 gene were normal and the exon 7 and 8 gene deletion of SMN were not found in the familial or sporadic HD patients. Fifteen samples from the family with HD were genotyped using 18 polymorphic microsatellites which cover the whole X chromosome. None of the Polymorphic microsatellites suggested linkage and yielded a Lod scores which were less than 0 by means of the MLIN K software respectively.

Conclusion: SOD1 and SMN genes may not contribute to the pathogenesis of HD. The disease gene for this familial HD is not located on X chromosome, which suggests an autosomal dominant pattern of inheritance.

P101 HETEROZYGOUS S44L MISSENSE CHANGE OF THE SPASTIN GENE IN AMYOTROPHIC LATERAL SCLEROSIS

MÜNCH C¹, ROLFS A², MEYER T³

¹Jewish Hospital Berlin, Berlin, Germany, ²University of Rostock, Rostock, Germany, ³Charité University Hospital, Berlin, Germany

E-mail address for correspondence: [email protected]

Keywords: spastin, missense S44L, ALS

Background: Amyotrophic lateral sclerosis (ALS) is a clinically and genetically heterogeneous group of disorders affecting the upper and lower motor neurons. There is increasing evidence that spastin is a candidate gene for ALS. Mutations in the spastin gene represent the most common cause of dominantly inherited hereditary spastic paraplegia (HSP). The majority of patients with spastin mutations develop uncomplicated HSP, but additional features are common.

Objectives: Here we report a patient with rapidly progressing adult-onset ALS. Molecular genetic analysis revealed a heterozygous missense change (S44L) of the spastin gene.

Methods: All 17 exons and flanking intronic sequences of the spastin gene were amplified by PCR and followed by single strand conformation analysis (SSCP) and direct DNA sequencing.

Results: Mutation analysis of the spastin gene identified a previously described heterozygous missense change (c.131C > L) located in exon 1 (RefSeq NM_014946). The sequence variant is predicted to replace serine by leucine in position 44 (S44L) of the putative spastin protein.

Conclusion: The single case of ALS with a spastin mutation presented here does not allow any conclusion about causality, but there is also no evidence against a monogenic cause of ALS by a spastin mutation. Alternatively, the spastin S44L change may serve as a previously unknown genetic risk factor for ALS. The present study suggests that spastin sequence changes may contribute to a wider range of phenotypes, including ALS. Given the common and perhaps universal finding of axonal dysfunction in both ALS and HSP, variants of the spastin gene may contribute to a common pathway in the complex pathogenesis of motor neuron disorders.

P102 SPG4 HEREDITARY SPASTIC PARAPLEGIA (HSP): THE INTRON 3 TAAT-DELETION IS UNLIKELY TO BE A PATHOGENIC MUTATION

PRUDLO J¹, ROEMER K², MEHRAEIN Y², SCHWAN A³, ARNING L⁴

¹University of Rostock, Rostock, Germany, ²Saarland University, Homburg/Saar, Germany, ³Dr. Eberhard & Partner Lab, Dortmund, Germany, ⁴Ruhr-University, Bochum, Germany

E-mail address for correspondence: [email protected]

Keywords: hereditary spastic paraplegia, SPG4, mutation

The Spastic paraplegia type 4 (SPG4) is the most common cause of the hereditary spastic paraplegias (HSP) and accounts for approximately 40% of the dominantly inherited forms. More than 150 disease-related spastin (SPG4 gene)-mutations have been described so far. Missense, nonsense, frameshift and splice-site mutations as well as gross deletions have been reported in SPG4, all presumably acting via haploinsufficiency.

Here, we describe the genetic findings in three affected siblings (44–61 years of age) of a HSP-family. The deceased father also suffered from spastic paraplegia. All three siblings show pure HSP with slow progressive lower limb spasticity and urinary incontinence but without dorsal column disturbances and extraspinal involvement. Disease onset was from the late third to the early fifth decade. After approximately 20 years of disease, the ability to walk is preserved but reduced to less than 500 m with support.

We found the siblings to be compound heterozygous for a sequence variation within a 5 splice site of SPG4 (1321 + 2T > G) and a previously described 4-base pair deletion (delTAAT) near the 3 splice-site of exon 3.

Further analyses have indicated that the base exchange c. 1321 + 2T > G results in abnormal splicing and skipping of exon 10. Notably, the c.586 + 9_12delTAAT deletion was also identified in 2 of 476 unaffected control blood donors, and no aberrant splicing pattern was found in these individuals, strongly suggesting that this deletion is an irrelevant rare variation rather than a pathogenic mutation.

These observations highlight the need for caution in the interpretation of likely splice-site mutations, especially when these variations occur outside the highly conserved donor and acceptor consensus dinucleotides.

P103 FUNTIONAL CHARACTERISATION OF A NEW AMYOTROPHIC LATERAL SCLEROSIS (ALS) ASSOCIATED MUTATION IN VESICLE-ASSOCIATED-MEMBRANE-PROTEIN-ASSOCIATED PROTEIN B (VAPB)

CHEN HJ, ANAGNOSTOU G, ADHIKAREE J, MORRIS A, DE BELLEROCHE J

Imperial College London, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: familial amyotrophic lateral sclerosis, VAPB, gene mutations

Background: ALS is a complex degenerative disease selectively affecting motor neurones. A point mutation (P56S) in VAPB has been identified in a large Brazilian pedigree which causes a wide spectrum of motor neurone degenerative disease phenotypes including atypical and typical ALS [1]. No other VAPB mutations associated with disease have been identified in other ALS populations [2], [3].

Objectives: To characterise a potential new disease-associated mutation found in the UK and to investigate its pathogenic mechanisms.

Methods: In the first part of this study, 107 non-SOD1 familial ALS (FALS) families were screened for mutations in known ALS-linked genes including VAPB. A point mutation causing an amino acid change from threonine to isoleucine at codon 46 (T46I) was identified in one FALS family but not in 119 controls. To evaluate the effect of this mutation, N-terminal GFP fusion constructs (VAPB, T46I VAPB and P56S VAPB) were generated and expressed in COS-7 and neuroblastoma (N2a) cell lines.

Results: The expression of GFP fusion VAPBs in COS-7 cells showed that the T46I point mutation caused an effect on VAPB protein distribution that was distinct from wild type VAPB and P56S mutant VAPB. Unlike the distribution of wild type VAPB which showed an ER-like localisation, both P56S and T46I showed different degrees of aggregation ranging from small granules to large coalesced structure, the later being more prominent in P56S. In N2a cells, a similar pattern was found, but in a proportion of cells the mutant protein retained an ER-like localization in combination with granules. In order to examine whether the abnormal VAPB distribution affected cell viability, we quantified cell death using flow cytometry. After correcting for transfection efficiency, there was no difference in cell death between cells expressing wild type or new paramutant VAPB.

Discussion and Conclusion: This is the first report of an ALS-associated mutation in VAPB found in a non-Brazilian population. Like P56S, T46I is located in a very conserved region in the N-terminal MSP domain. The distribution study showed that T46I caused a mild but distinct effect on VAPB protein localisation. Whether this effect will affect protein function is under investigation. Recently, evidence of ER stress in ALS patients and animal models has been reported [4] and VAPB was shown to interact with the ER stress transducer, ATF6 [5]. We are currently investigating the effect of T46I on the unfolded protein response pathway which is affected by P56S [6].

P104 FUNCTIONAL CHARACTERISATION OF PKD1L2, A CANDIDATE GENE IMPLICATED IN MUSCLE DENERVATION AND AXONAL REGENERATION

MACKENZIE F¹, ROMERO R¹, HILTON H¹, WONG F¹, ARKELL R¹, GILLINGWATER T², RIBCHESTER R³, BLANCO G¹

¹MRC Mammalian Genetics Unit, Harwell, United Kingdom, ²Centre for Neuroscience Research, University of Edinburgh, Edinburgh, United Kingdom, ³University of Edinburgh, Edinburgh, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: mouse, genetics, denervation

Background: We have exploited N-ethyl N-nitrosourea (ENU) mutagenesis to identify new mouse models and thus novel genes involved in motor neuron loss and muscular atrophy. Ostes is an ENU-induced genetic mouse mutant which displays complex neuromuscular phenotypes including denervation, polyinnervation, muscular atrophy and delayed axonal regeneration.

Objectives: We aim to identify the genetic defect and characterise the underlying neuromuscular pathology in ostes mice. This would further understanding of the genetic and physiological causes of MND and neuromuscular disorders, and could suggest new genetic pathways for therapeutic intervention.

Methods: cDNA mutation screening was carried out by amplification and sequencing of cDNA from wildtype and ostes/ostes mice. PKD1L2 ^-/- mice were generated by injection of a PKD1L2 gene-trapped ES cell line into blastocysts. Protein analysis was carried out by immunoprecipitation, Western blot, mass spectrometry and immunofluorescence. Microarray analysis was carried out on mRNA from wildtype and ostes/ostes skeletal muscle.

Results: Previously we presented our findings on the genetic mapping of ostes and the functional characterisation of a candidate gene, the polycystic kidney disease-1-like gene PKD1L2 (18^th Symposium on ALS/MND, Toronto 2007) including protein analysis and transgenic work. Following a round of genomic sequencing including all exons in the non-recombinant region which failed to identify any mutation, we are now undertaking a major cDNA sequencing effort to analyse the cDNA of the full complement of genes in the ostes non-recombinant region for mutations that affect mRNA splicing. To further assess the candidacy and function of PKD1L2, we have generated PKD1L2 ^-/- knockout mice using a gene-trapped ES cell line to assess if PKD1L2 deficiency recapitulates the ostes phenotype. Furthermore, we have also generated PKD1L2-/ ostes double heterozygotic mice to test for non-complementation. Lastly, we have further investigated the function of PKD1L2 and have identified fatty acid synthase (FASN) as an interacting partner of PKD1L2. Interaction has been confirmed by reciprocal co-immunoprecipitation and co-localization experiments. These experiments suggest a role for PKD1L2 in the regulation of fatty acid or lipid metabolism. Consistent with these findings, expression profiling of muscles from ostes/ostes mice demonstrates significant expression changes of genes involved in lipid metabolism.

Conclusion: Our results indicate that defective or misregulated PKD1L2 causes complex neuromuscular defects in mice. We suggest that PKD1l2 is essential for the regulation of muscle innervation and function and is therefore a candidate gene for uncharacterised forms of motor neuron disease. The direct interaction between FASN and PKD1L2 and the secondary increase of FASN protein observed in ostes/ostes mice suggest that dysregulation of lipid metabolism may play a role in the pathology of this disease.