Amyotrophic Lateral Sclerosis and Frontotemporal Degeneration

C1 ALS IN A WORLD OF MULTIPLE PHENOTYPES

Appel SH

Beers DR

Zhao W

Methodist Neurological Institute, Houston, Texas, USA

Email address for correspondence: [email protected]

Keywords: neuroinflammation, activated microglia, regulatory T lymphocytes

The last several decades have witnessed an exponential growth in the clinical phenotypes labelled as ALS, fuelled by the discovery of mutant genes, now numbering almost 30, which present as ALS. Each mutated gene appears to have a distinctive signature altering a complex pathway leading to motor neuron injury and death. Such pathways include compromise of cytoskeletal function and axonal transport, misfolded aggregated proteins and their impaired digestion by proteosomes and autophagy, and impaired RNA biology. Despite the diversity of molecular pathologies, different mutant genes can produce a single clinical phenotype, and a single genetic mutation can produce multiple clinical phenotypes. These familial ALS phenotypes are indistinguishable from sporadic ALS phenotypes which are also heterogeneous, with markedly variable sites of onset, ages of onset and rates of disease progression. More recently the heterogeneity of ALS has expanded with the recognition that approximately 50% of sporadic ALS patients posses the behavioural and cognitive impairment of frontotemporal dysfunction, and 15% manifest frank frontotemporal dementia. Thus ALS results from compromise of multiple molecular pathways and can present with multiple clinical phenotypes. An important goal is to identify factors that contribute to the diverse clinical expressions of this syndrome.

In affected members from a single family with the same mutation, age of onset and disease severity and duration can be quite variable. Thus other factors, possibly genetic or environmental, modify the expression of ALS. Potential factors can be identified with the use of transgenic mouse models of ALS, where neurons do not die alone; neuronal injury is non-cell- autonomous and depends on well-orchestrated dialogues involving motor neurons, glia and T cells. The immunological responses of glia and T cells are not merely the passive consequences of injury, but actively influence and significantly contribute to the balance of neuroprotection and neurotoxicity and thereby mediate neuronal injury and neuronal viability. Regardless whether mutations in a specific gene initiate a cascade of intraneuronal injury in inherited disease, or whether the cause of the initial neuronal injury is undefined as in sporadic disease, a similar inflammatory response ensues. In both inherited and sporadic cases of ALS, differing temporal and mechanistic compromise of intraneuronal organelles may contribute to variations in immune responsiveness and disease progression. Thus the innate and adaptive immune systems contribute to phenotypic heterogeneity and represent potentially important therapeutic targets for modifying expression of disease.

Supported by Grants from the NIH and the Muscular Dystrophy Association.

C2 MECHANISMS UNDERLYING SELECTIVE NEURONAL VULNERABILITY IN ALS

Caroni P

Friedrich Miescher Institute, Basel, Switzerland

Email address for correspondence: [email protected]

Keywords: excitability, ER stress, endogenous neuroprotection

Delaying clinical disease onset would greatly reduce the burden of neurodegenerative diseases, but the mechanisms that influence early preclinical progression are poorly understood. Here we show that in mouse models of familial motor neuron (MN) disease SOD1 mutants specifically render vulnerable MNs dependent on endogenous neuroprotection signaling involving excitability and mTOR. The most vulnerable low-excitability FF MNs already exhibited evidence of pathology and endogenous neuroprotection recruitment early postnatally. Enhancing MN excitability promoted MN neuroprotection and reversed misfolded SOD1 (misfSOD1) accumulation and MN pathology, whereas reducing MN excitability augmented misfSOD1 accumulation and accelerated disease. Modulating excitability and/or alpha-MN mTOR activity had comparable effects on the progression rates of motor dysfunction, denervation, and death. Therefore, excitability and mTOR are key endogenous neuroprotection mechanisms in motor neurons to counteract clinically important disease progression in ALS.

C3 SIZE-DEPENDENT AXON LOSS IN THE CORTICOSPINAL TRACT IN ALS PATIENTS WITH UPPER MOTOR NEURON SIGNS

Song F^1,2

Liu J¹

Ravits J³

Loeb J^1,2

^aHiller ALS Clinic and Research Center, Department of Neurology

^bCenter for Molecular Medicine and Genetics; Wayne State University, Detroit, Michigan, USA

^cALS and Neuromuscular Translational Research, University of California, San Diego, California, USA

Email address for correspondence: [email protected]

Keywords: size dependent axon loss, corticospinal tract, upper motor neuron

Background: Amyotrophic lateral sclerosis (ALS) is a progressive neurodegenerative disease that selectively involves both upper and lower motor neurons (UMN and LMN). The mechanism by which both systems are involved is unknown. However, a common pathological finding in the spinal cord is corticospinal tract (CST) degeneration. We recently found a combination of myelin loss together with axonal loss in the tract in patients (1). Interestingly, axons of different diameters are also affected.

Objectives: In order to better understand the pathology, we compared the degree of axonal degeneration in the CSTs in the presence or absence of UMN symptoms in well-characterized postmortem cervical, thoracic, and lumbar spinal cords regions to determine the anatomical relationships of the connection between upper and lower motor neuron systems in ALS patients.

Methods: We systemically observed myelin loss in three regions of spinal cord. The degree of loss of axons of different diameters was measured and quantified in the lateral and ventral CSTs and dorsal column regions in the different levels of spinal cords from ALS patients with and without clinical UMN symptoms as well as from control patients.

Results: Only patients with clinical upper neuron symptoms showed lateral and ventral CSTs degeneration (myelin and axon losses). We found that all three spinal cord regions were affected in the lateral CSTs but mainly two regions were affected in the ventral CSTs. Quantification of axon density showed a loss of axons of both small and large diameters in the lateral and ventral CSTs in ALS patients with UMN signs. Comparing to ALS patients without UMN signs, we see a unique gradient of smaller diameter axon loss from lumbar to thoracic and cervical in ALS patients with UMN sign. In ALS patients without UMN signs, we only see some loss of larger diameter axons in the lateral CST at all three levels.

Discussion and conclusion: Our current findings (1) show an ‘All or None’ effect of myelin–axon loss at all spinal cord levels of lateral CST and (2) appear to be a ‘Dying Back’ of only small but not large diameter axons in patients with UMN signs. Our results suggest that selective vulnerability of axonal loss depends on fiber size and should be considered in interpretation of pathology of corticospinal tracts in ALS patients with UMN signs. We hypothesize that aberrant glioaxonal interactions in the spinal cord could contribute into this process.

Reference:

• Song F, Chiang P, Wang J et al. J Neuropathol Exp Neurol. 2012;71:104–15.
   PubMed Web of Science ®Google Scholar

C4 AXON DEGENERATION AND AXON PROTECTION IN ALS

King A

Southam K

Ditmann J

Vickers J

University of Tasmania, Hobart, Australia

Email address for correspondence: [email protected]

Keywords: axonopathy, microtubule, excitotoxicity

Background: Axon degeneration is a key pathological feature of ALS although the cause remains uncertain. Studies have shown that axon degeneration is a process separate from cell body apoptosis and protection from cell death may not prevent the axon from degenerating. In this respect, axon protection may be an attractive target for therapeutic intervention in ALS. Although mechanisms of axon degeneration following axon transection have been well characterized, little is known about mechanisms of axon degeneration in neurological disease and whether they share similarities with axonal transection. Our investigations have demonstrated that axon degeneration can result from excitotoxic insult in both cortical and lower motor neurons grown in culture. Excitotoxicity is a key pathogenic process implicated in ALS leading us to propose that excitotoxic damage may contribute to axonal pathology in ALS.

Objectives: We have investigated axonopathy following excitotoxin exposure. Our focus has been the role of caspase activation, the cytoskeleton and the therapeutic potential of cytoskeletal stabilization to prevent axon degeneration. Current studies include development of animal models of excitotoxin induced axon degeneration.

Methods: Cortical neurons were cultured from C57Bl/6 mice and grown in compartmented microfluidic chambers to examine the role of axon and soma. Neurons at 10 days were exposed to 100μM kainic acid (18hours) in the absence or presence of taxol in the axon or soma compartment. Axonal fragmentation was determined from phase contrast images of axons. Immunohistochemical analysis was performed using antibodies to active caspase-3, neurofilament and MAP2. (n = 5 repeats from three separate cultures). To determine if excitotoxin induced axon degeneration shares mechanisms with axonal transection, cultured neurons were transfected with a construct that encodes the Wlds fusion protein (a gift from Michael Coleman), which protects axons from degeneration induced by transection. Transfected neurons were exposed to 100μM kainic acid.

Results: Kainic acid applied to soma induced a 41.3% (± 8.5 SEM) increase in axon degeneration in the axon compartment, which was associated with axonal caspase-3. Pretreatment of axonal and somal compartments with taxol reduced subsequent kainic acid-induced caspase activation and axonopathy, with a more marked effect following taxol applied to the axonal compartment (12.2± 2.3% fragmentation, p < 0.05) as compared to the somal compartment (25.8± 5.2% fragmentation). Expression of Wlds provided axonal protection from degeneration.

Conclusion: Excitotoxin induced axon degeneration, unlike axonal transection involves activation of caspases. However, the protective effect of the Wlds construct suggests that excitotoxin induced axonopathy and also shares some similarities with axon transection. These studies highlight the importance of investigating disease specific mechanisms of axon degeneration and suggest that microtubule stabilization may be a viable therapeutic option for ALS, either alone or in combination with other therapies.

C5 INHIBITORY LOSS OR DYSFUNCTION: A PRIMARY MECHANISM IN ALS?

Clark R¹

Fielder T¹

King A²

Dickson TC¹

^dMenzies Research Institute Tasmania

^eWicking Dementia Research and Education Centre, School of Medicine; University of Tasmania, Hobart, Tasmania, Australia

Email address for correspondence: [email protected]

Keywords: interneurons, pathology, transgenics

Background: In ALS, increased excitability of circuitry precedes motor neuron degeneration, suggesting that ALS results from disturbances in regulation of cell excitability. However, the mechanism of presymptomatic excitability remains unknown. There is strong clinical evidence, in both cortical and spinal regions, of reduced inhibition implicating this as a potential primary mechanism. We examined the motor and somatosensory cortex of SOD1^G93A and non-transgenic mice for expression of interneuron-specific calcium binding and neuropeptide protein markers. Studies were also performed in human ALS and control brain tissue, investigating the presence of pathological changes within interneuron populations.

Objectives: To characterise the pathological alterations to cortical interneuron subpopulations in ALS mouse models compared to that present in ALS tissue.

Methods: Cortical tissue from presymptomatic (8 week) and end-stage (20 week) SOD1^G93A and ALS human cortex were serially sectioned (40um), alongside age-matched controls, and immunohistochemically labelled with antibodies against calretinin (CR), parvalbumin (PV), somatostatin (SOM), Neuropeptide Y (NPY) and Vasoactive Intestinal protein (VIP). Qualitative analysis of cell soma appearance was performed alongside with quantitation of cell number and soma area. The presence of ubiquitinated and SOD protein inclusions was also considered. Analysis was performed on confocal images of supragranular and infragranular motor and somatosensory cortex lamina.

Results: Analysis of presymptomatic SOD1^G93A animals found CR⁺ interneurons to be unchanged (n = 4, 22.218 ± 3.607) relative to that of controls (n = 4, 27.660 ± 3.404), despite a 24% reduction in SOD1^G93A CR⁺ cells in this region. Analysis of end-stage SOD1^G93A cortical regions, however, found the number of CR⁺ interneurons in the supragranular motor cortex lamina (I–IV) was significantly decreased (P < 0.05) by 37% (n = 4, 12.53 ± 2.137 SEM), as compared with wild type (n = 4, 19.59 ± 2.452 SEM) (Two-way ANOVA, multiple comparisons Bonferroni Test). Analysis of end-stage SOD1^G93A NPY⁺ interneuron numbers in the infragranular motor cortex (V–VI) identified a significant increase (P < 0.05) by 40% (11 ± 1.493 SEM), as compared with wild type (6.592 ± 0.879 SEM). Analysis of VIP labelling revealed a significant decrease in SOD1^G93A area of labelling in the infragranular somatosensory cortex lamina (V–VI) by 54% (n = 4, 33.799 ± 6.80), as compared with wild type (n = 4, 73.243 ± 6.072).

Discussion and conclusion: These findings demonstrate pathological alterations to inhibition in the end-stage SOD1^G93A mouse model of ALS, specifically implicating CR, NPY and VIP interneuron populations in cortical dysfunction. Furthermore, results indicate variable susceptibility of interneuron populations, with varying cortical regions and functional domains affected. This suggests specific populations may be differentially implicated in the disease. Ongoing investigations utilising human ALS tissue and other ALS transgenics will be important for further interpreting this data and determining whether inhibitory loss could be a potential mechanism underlying ALS.

Acknowledgements

MNDRIA- Zo-èe Research Grant, and s'ship top-up to RC.

C6 PERIPHERAL NERVOUS SYSTEM DYSFUNCTION IN A RAT MODEL AND IN HUMAN MOTOR NERVE BIOPSIES OF AMYOTROPHIC LATERAL SCLEROSIS

Riva N¹

Chaabane L¹

Peviani M²

Ungaro D¹

Domi T¹

Dina G¹

Spano G²

Cerri F¹

Corbo M¹

Del Carro U¹

Nobile-Orazio E³

Comi G¹

Bendotti C²

Quattrini A¹

^fSan Raffaele Scientific Institute, Department of Neurology and INSPE, Milan, Italy

^gLaboratory of Molecular Neurobiology, Department of Neuroscience, Mario Negri Institute for Pharmacological Research, Milan, Italy

^hNeurology, IRCCS Istituto Clinico Humanitas, Milano University, Milan, Italy

Email address for correspondence: [email protected]

Keywords: peripheral nerve, motor nerve biopsy, magnetic resonance imaging

Background: Despite the fact that the loss of peripheral axons is a major cause of disability in amyotrophic lateral sclerosis and motor axons are involved in the pathogenesis of this disease, the peripheral nervous system (PNS) involvement in amyotrophic lateral sclerosis (ALS) has not been extensively studied so far.

Objectives: To characterize in detail the natural history of the PNS damage in a hSOD-1^G93A ALS rat model using both in vivo and ex vivo readouts and compare such results with histopathological findings in diagnostic motor nerve biopsies from ALS patients.

Methods: Longitudinal magnetic resonance imaging, neurophysiological, and histological investigations were employed to monitor the extents of PNS damage in a hSOD-1^G93A ALS rat model. Light microscope and immunocytochemical analysis of human motor nerve biopsies were performed.

Results: By in vivo magnetic resonance imaging, follow-up of the sciatic nerve allowed to define the imaging signature of the disease. Initial abnormalities within sciatic nerve were detected by an increase of T2 relaxation time, before symptom onset. In addition, diffusion magnetic resonance imaging acquired during disease course showed a progressive increase of mean diffusivity (MD) and radial diffusivity (l_⊥) within the sciatic nerve which was associated with reduction of fractional anisotropy (FA) at advanced stage of disease. Neurophysiological examination showed a gradual reduction of sciatic nerve distal compound motor action potential amplitude during disease course and concomitant signs of active denervation at needle examination. Histology showed early impairment of the blood-nerve barrier, endoneurial oedema and acute axonal degeneration associated with signs of neuroinflammation, such as a macrophage response in the motor nerve compartments, before the appearance of symptoms. Progressive axonal degeneration and motor nerve fiber loss were observed, correlating with changes in MRI and neurophysiological studies.

Histopathologic studies of human diagnostic motor nerve biopsies confirmed the presence of signs of acute axonal degeneration, associated with signs of neuroinflammation, predominantly represented by endoneurial macrophages and occasional epineurial inflammatory infiltrates.

Discussion and conclusion: The functional and morphological platform established here could be used to follow disease progression in the amyotrophic lateral sclerosis rat model and to evaluate possible therapies relevant to the disease. Moreover, the study of the peripheral nervous system involvement could also shed new lights in the ALS disease pathogenesis.

References:

• Bilsland LG, Sahai E, Kelly G et al Proc Natl Acad Sci USA 2010;107:20523–8.
   PubMed Web of Science ®Google Scholar
• Coleman MP, Perry VH. Trends Neurosci 2002;25: 532–37.
   PubMed Web of Science ®Google Scholar

C7 TO TEST OR NOT TO TEST, THAT IS THE QUESTION

Hardiman O

Trinity College Dublin, Dublin, Ireland

Email address for correspondence: [email protected]

Keywords: genetics, epidemiology, risk factors

ALS physicians and those affected by the condition are faced with a series of difficult dilemmas generated to a large extent from advances in genetics over the past decade.

We know that some forms of ALS are familial, and that the condition is linked in some cases to frontotemporal dementia, and also to a wider range of brain disorders including schizophrenia, major psychosis and possibly suicidal behaviour.

Familial disease implies that some people with ALS harbour one or possibly more genes that have greatly increased their risk of developing ALS. In others, the condition has developed following exposure to certain environmental factors, the nature of which are unknown, that may have interacted with a more complex array of genetic factors, many of which have yet to be determined.

The current EFNS guidelines state that testing should only be performed in patients with a known family history of ALS, and following appropriate genetic counselling.

The first dilemma is to establish a universal definition of familial disease. Surprisingly, there is no consensus for this at present. Familial disease can be difficult to detect in small families, but the risks of two people in a family developing ALS by chance increases in larger kindreds.

The second dilemma is whether to perform any genetic testing, given that there is currently no treatment. A decision is then required as to whether to look for all known variants in all known genes, or for variants one gene at a time. This is important for three reasons. Firstly because reported variants may not always be truly pathogenic, secondly because there is evolving evidence that ALS may be ‘oligogenic’, implying that variants in different disease-associated genes occur with greater frequency that would be expected by chance in people with apparently familial ALS. And thirdly the frequency of pathogenic variants differs across populations. There is also a cost implication if many genes are to be tested at the same time.

The third dilemma relates to the implications of testing for other family members. There are major ethical and legal implications of genetic testing, as once a disease variant is identified in an affected family member, all other members of the family are at risk, and decisions must be made regarding testing asymptomatic relatives, and whether pre-natal testing should be available. This is important because there are no preventative treatments at present, and because the penetrance of various pathogenic genetic variants has not been adequately established.

Ultimately, genetic testing for ALS will need to be standardized by modifying existing guidelines established for other diseases, supported by ongoing research in clinical and genetic epidemiology and governed by a robust ethical framework.

C8 ALS CLINICS AND THE EMERGING CHALLENGE OF GENETICS: A WORLDWIDE SURVEY

Rudnicki S¹

Shoesmith C²

Chiò A³

ⁱUAMS, Little Rock, AR, USA

^jWestern Universtiy, London, Ontario, Canada

^kUniversity of Turin, Torino, Italy

Email address for correspondence: [email protected]

Keywords: genetic testing, survey, genetic counselling

Background: As more ALS genes have been identified, and phenotypic variability has expanded, how best to address the genetics of the disease in the clinic has become increasingly complex.

Objective: To determine how ALS clinics in North America, Europe and Israel address the genetics of motor neuron disease.

Methods: We sent out emails to 203 ALS/MND clinics with a link to the survey web site (Survey Monkey.com).

Results: Eighty-seven responses were received, for a return rate of 43%. 81% of the respondents were in an academic setting. The majority did not have a genetic counsellor or geneticist (GC/G) in their clinic but could readily make an appointment locally (74.4%), 19.8% had a GC/G at selected clinics and 5.8% had them available at every clinic. Time to get an appointment with a GC/G was less than 2 months for 85.7%. Overwhelmingly, neurologists were willing to discuss and send genetic testing on patients without first getting a genetics consult (82.9%). The most common reasons for sending someone for a genetics consult were an ALS patient with a family history of ALS, and an ALS patient or their family member who specifically requests the referral. For patients with a family history of ALS (gene unknown), 30.3% test commercially for all genes available, 25% do selective testing commercially, 23.7% would do base testing on results of a genetics consult, 14.5% would test for research purposes only, and 6.6% would not do testing. When a pathological mutation is identified, 54.8% refer the patient for genetic counselling, the neurologist does the genetic counselling in 39.3%, and 25% refer family members for counselling. When recommendations are made to get genetic testing in patients with a family history of ALS, 33.7% find that none refuse, 47.0% report fewer than 1/3 refuse, 16.9% report between 1/3 and 2/3 refuse, and 2.4% report more than 2/3 refuse. Uncertainty about what positive results mean in the presence of a negative family history was listed as the first or second biggest challenge regarding genetic testing by 50% of respondents. However, among US neurologists, 74% chose cost as the first or second biggest challenge regarding genetic testing while this was true for only 13% of non-US neurologists.

Discussion and conclusion: Practice patterns regarding genetic testing vary widely in ALS clinics. This survey emphasizes the need for developing guidelines regarding such testing, though differences in insurance coverage need to be taken into account when these are developed.

C9 DEVELOPING A MODEL OF PATIENT-CENTRED DECISION-MAKING FOR AMYOTROPHIC LATERAL SCLEROSIS MULTIDISCIPLINARY CARE

Hogden A¹

Greenfield D¹

Nugus P^1,2

Kiernan MC³

^lCentre for Clinical Governance Research, Australian Institute of Health Innovation, University of New South Wales, Sydney, NSW, Australia

^mCentre for Medical Education and Department of Family Medicine, McGill University, Montreal, Quebec, Canada

ⁿPrince of Wales Clinical School, University of New South Wales, and Neuroscience Research Australia, Sydney, NSW, Australia

Email address for correspondence: [email protected]

Keywords: patient decision-making, multidisciplinary care, decision-making model

Background: Amyotrophic lateral sclerosis (ALS) patients continually make decisions for symptom management and quality of life as their condition deteriorates. Health professionals aim to support patient autonomy through specialised ALS multidisciplinary care delivery. Even so, established models of patient-centred decision making do not account for the complex and changing needs of ALS patients throughout the disease trajectory.

Objectives: To investigate ALS patient decision making and to derive a decision-making model for specialised ALS multidisciplinary care.

Methods: Fifty-four respondents (32 health professionals, 14 patients and 8 carers) from two specialised ALS multidisciplinary clinics participated in semi-structured interviews between April 2011 and May 2012. Interview topics were derived from the patient decision-making literature body, refined in reference to ALS. Audio recordings of interviews were transcribed, coded and analysed thematically.

Results: Comparison of health professional, patient and carer perspectives revealed broad agreement on decision making for ALS symptom management and quality of life. Six factor domains were reported to influence patient- centred decision making. These were the process of decision making; use of a patient-centred focus; timing and planning; information sources; engagement with specialised ALS services; and access to non-specialised services. In addition, psychosocial factors and continually changing symptoms, including physical, cognitive and behavioural deterioration, impacted on patients’ capacity to participate. Participants agreed that specialised ALS multidisciplinary clinics offered an optimal setting for decision making. Nevertheless, issues of timing of evidence-based care delivery and the role of carers were contentious.

Stakeholders perceived ALS decision making as a collaborative, complex and cyclical process. The derived model is embedded in the decision-making environment of the specialised ALS multidisciplinary clinic. Health professionals, patients and carers form a decision-making triad, and move through a cycle of four interlinked stages. Patients move within and between each stage of the model until ready to proceed. The first stage, ‘Patient Engagement’, identifies the participants and establishes their values, preferences and expectations. In stage two, ‘Option Information’, information and guidelines on the available management options are determined, including the optimal timing for implementation of each choice. During the third stage, ‘Deliberation’, patients weigh up the risks and benefits, and decide between proceeding with an option, deferring their decision, or choosing to do nothing. The final ‘Implementation’ stage results once an option is chosen.

Discussion and conclusion: Participant engagement in ALS patient-centred decision making is tested by the dynamic nature of the disease, limited treatment options and patient and family distress. The roles and expectations of stakeholders influence the decision-making process. Respect for patient autonomy may conflict with delivery of well-timed, evidence-based care. This empirically derived model captures these complexities and offers a framework for health professionals, researchers and policy makers in this challenging environment.

C10 QUALITY OF LIFE, DEPRESSION AND PERCEIVED SOCIAL SUPPORT IN THE COURSE OF ALS

Lulé D¹

Sorg S¹

Nonnemacher S²

Kübler A³

Birbaumer N^2,4

Ludolph AC¹

^oUniversity of Ulm, Ulm, Germany

^pUniversity of Tübingen, Tübingen, Germany

^qUniversity of Würzburg, Würzburg, Germany

^rOspedale San Camillo, IRCCS, Venice, Italy

Email address for correspondence: [email protected]

Keywords: quality of life, depression, social support, progression

Background: ALS is associated with severe physical function loss and major burden. However, there is evidence that hedonic quality of life (QoL), which refers to transient feelings, such as life satisfaction and happiness is known to be good in a majority of ALS patients. Depression rate is often low. QoL and depression are usually not associated with loss of physical function in cross-sectional analysis.

Objectives: The aim of the study is to determine the dynamics of hedonic QoL and depression in association with perceived social support over time.

Methods: ALS patients (n = 93) were interviewed in a prospective longitudinal approach thrice in the course of one year. QoL was assessed with the anamnestic comparative self-assessment (ACSA, range −5 for as bad as possible and +5 as good as possible), subjective QoL was assessed with the schedule for the evaluation of subjective quality of life (SEIQoL, range 0 for as bad as possible and 100 as good as possible). Depression was measured with the ALS depression inventory (ADI-12, range: 12–48; > 28 indicates clinically relevant depression). Perceived social support was evaluated using the emotional scale of the social support scale (SOZU K 22, range: 22–110). Physical function of ALS patients was measured using the ALS functional rating scale revised form (ALS-FRS-R, range: 0–48).

Results: Indicators of good psychosocial adaptation such as high SEIQoL (score > 70 on a scale of 100 indicates a good QoL) and neutral to positive global QoL (ACSA), and low depression (ADSK) were stable throughout the study. Furthermore, perceived emotional social support was stable despite significant physical function decline.

Discussion: Subjective QoL was good and depression was low and stable throughout the study despite significant decrease in physical function. This is in line with the literature on stable and good subjective QoL in the course of ALS. Therefore, hedonic QoL is independent of physical function in ALS and differences in QoL between patients are likely due to pre-existing individual differences (Roach et al., 2009). Good and stable emotional social support is a prerequisite for good QoL and low depression rate.

Conclusion: The stability of QoL is a highly important finding as an anticipated poor QoL was one of the most important reasons for requesting physician assisted suicide in Oregon (Ganzini et al., 2009) and probably in other locations as well.

C11 UNDERSTANDING QUALITY OF LIFE IN MOTOR NEURONE DISEASE: QUALITATIVE EXPLANATIONS FROM THE TRAJECTORIES OF OUTCOME IN NEUROLOGICAL CONDITIONS STUDY (TONIC)

Ando H^1,2

Cousins R¹

Young C²

^sLiverpool Hope University, Liverpool, UK

^tThe Walton Centre NHS Foundation Trust, Liverpool, UK

Email address for correspondence: [email protected]

Keywords: quality of life, psychological well-being, TONiC study

Background: Previous studies have reported that motor neurone disease (MND) may affect perceived quality of life (QoL). Quantitative relationships between QoL and other factors (eg physical deterioration and social support) have been reported, while qualitative researchers have examined conceptualisation of QoL. Although these studies provide insight, more work is needed on contributory factors and their relationships.

Objectives: The aim of this study was to identify factors that affect perceived QoL and to understand how they function.

Methods: Semi-structured interviews were conducted with 19 MND patients (male = 13, bulbar onset = 8, mean age = 62.7 yrs). The mean duration since the diagnosis was 22.4 months (range: 1–110 months). The interviews were audio recorded and transcribed verbatim. Thematic analysis was employed in which inductive coding systems were developed to identify themes.

Results: The analysis identified four influential factors for QoL: Social Factors, Spirituality, Personal Qualities and Coping Strategies. Social Factors entailed four subthemes which were having significant others, getting understanding from others, receiving support, engaging with social activities, and making contributions to others. Positive outcomes from these were empowerment, enjoyment and existential meaning for life, while negative outcomes were psychological distress and behavioural withdrawal. The data also showed that some individuals are more psychologically independent than some others to whom empowerment from external support was vital for perceived QoL. Spirituality was found to have positive influence in giving individuals peace and purpose of life. Personal Qualities included positive attitude, locus of control and self-efficacy. While presence or absence of positive attitude was found to be consistent throughout an interview, degree of self-efficacy was observed to shift depending on situations as well as illness severity. Similarly, both internal and external locus of control were occasionally displayed by the same individual. Coping Strategies consisted of both emotion-focused coping and problem-focused coping. Among the four main themes, Social Factors, Spirituality and Personal Qualities appeared to be more influential than Coping Strategies to QoL. Coping strategies were perceived to be rather influenced by other three factors. In addition, Social Factors and Spirituality were found to influence Personal Qualities, yet this varied between patients.

Discussion and conclusion: QoL of individuals with MND was explored using a qualitative approach. The analysis showed that Social Factors, Spirituality, Personal Qualities, and Coping Strategies are influential to QoL. The impact of the Social Factors and Personal Qualities in particular may be stressed for their influence on QoL. Individual differences observed suggest the importance of the patient-centred approach. Specifically, study suggests the importance to recognise those who may benefit from receiving support and to understand the shifts in their sense of locus of control and self-efficacy over time.

Acknowledgements:

This study was supported by Motor Neurone Disease Association, UK.

C12 HNRNP A3 BINDS TO GGGGCC REPEATS OF PATIENTS WITH C9ORF72 MUTATIONS: CONSEQUENCES FOR RAN TRANSLATION

Mori K¹

Haass C^1,2

^uAdolf-Butenandt-Institute, Munich, Germany

^vDZNE - German Center for Neurodegenerative Diseases, Munich, Germany

Email address for correspondence: [email protected]

Keywords: C9orf72, RAN translation, hnRNPA3

Genetic analysis revealed the hexanucleotide repeat expansion GGGGCC within the regulatory region of the gene C9orf72 as the most common cause of familial amyotrophic lateral sclerosis and the second most common cause of frontotemporal lobar degeneration. Since repeat expansions might cause RNA toxicity via sequestration of RNA binding proteins, we searched for proteins capable to bind to GGGGCC repeats. In vitro transcribed biotinylated RNA containing hexanucleotide GGGGCC or, as control, AAAACC repeats were incubated with nuclear protein extracts.Using stringent filtering protocols 20 RNA binding proteins with a variety of different functions in RNA metabolism, translation and transport were identified. A subset of these proteins was further investigated by immunohistochemistry in human autopsy brains. This revealed that hnRNP A3 formed neuronal cytoplasmic and intranuclear inclusions in the hippocampus of patients with C9orf72 repeat extensions. Confocal microcopy showed that these inclusions belong to the group of the so-far enigmatic p62 positive/TDP-43 negative inclusions characteristically seen in autopsy cases of diseased C9orf72 repeat expansion carriers. Thus we have identified one protein component of these pathognomonic inclusions. We are now investigating if hnRNPA3 affects transport or translation of the GGGGCC-containing repeat RNA. Moreover, we will present data on the potential of primary cells from patients with C9orf72 mutations as simple models for the disease pathology. Repeat associated non-ATG repeat translation can occur not only from sense-strand repeat RNA transcripts but also from antisense transcript. We have therefore raised antibodies against the potential reverse strand specific di-peptide repeat (DPR) proteins poly-AP and poly-PR. Both antibodies specifically recognize a portion of characteristic p62-positive inclusions in the hippocampi and cerebella of C9orf72 repeat expansion carriers. Thus multiple DPR proteins are deposited in brains of patients with C9orf72 repeat extensions. Finally, we will present data on the potential of primary cells from patients with C9orf72 mutations as models for the disease pathology. We have investigated iPS cells from human patients with C9orf72 repeat extensions. Cells were so far grown for 80 days after differentiation into neurons. We searched for deposition of DPR proteins translated directly from the C9orf72 repeat extensions as well as for p62 pathology. However, so far no depositions were found. Cells growing for longer time points are currently under investigation.

C13 THE ROLE OF RNA BINDING PROTEIN HNRNP K IN ALS AND FTD

Moujalled D¹

James J¹

Yang S²

Turner B³

Blair I²

White A¹

^wThe University of Melbourne, Melbourne, VIC, Australia

^xANZAC Research Institute, Concord Hospital, Sydney, NSW, Australia

^yFlorey Neuroscience Institutes and Centre for Neuroscience, University of Melbourne, Melbourne, VIC, Australia

Email address for correspondence: [email protected]

Keywords: TDP-43, FUS, hnRNP K

Background: Tar DNA binding protein 43 kDa (TDP-43) has been identified as the major pathological protein of ALS and FTLD-U. In pathological brain and spinal cord tissue, one of the hallmarks of the diseases is the relocalisation of TDP-43 from the nucleus to the cytoplasm, where it undergoes various post-translational modifications. Alterations in neuronal RNA processing are characteristics of many neurodegenerative disease states. Heterogeneous nuclear ribonucleoprotein K (hnRNP K) is a RNA-binding protein that is implicated in apoptosis, neurodegeneration and is upregulated in various cancers. Preliminary results suggest that TDP-43 and hnRNP K are closely related. The clinical and pathological commonality between ALS and FTD suggests these diseases share underlying mechanisms that constitute to the diseases. This is reinforced by TDP-43 positive inclusions being a signature feature in both ALS and FTLD-U cases.

Objective: The objective of the present study is to decipher the molecular mechanism of TDP-43 cytoplasmic protein accumulation and to elucidate why greater toxicity and proteinopathies are associated with mutated TDP-43 in ALS and FTD patients.

Methods: NSC-34 (mouse hybridoma line as a motor neuron-like model) cells stably transfected to express either normal (wild-type, WT) or mutant (A315T or Q331K) TDP-43 labeled with the fluorescent cherry tag, human fibroblasts derived from an ALS patient harboring an M337V mutation in TDP-43 and control human fibroblasts with normal TDP-43 were employed. For knock down experiments, SH-SY5Y cells were transiently transfected with hnRNP K siRNA at 10nM using DharmaFECT reagent.

Results: In response to stress induced by sodium arsenite, NSC-34 cells expressing WT TDP-43 showed robust phosphorylation of cytosolic, but not nuclear hnRNP K, while this was not evident in the mutant cell lines. Under the same conditions, in the cytosol of cells containing the Q331K mutation, there was almost a complete loss of expression of RNA binding proteins FUS and hnRNP K with lesser changes to expression induced by A315T or WT TDP-43. Fibroblasts taken from a patient with ALS displayed dramatically reduced expression of hnRNP K compared to control fibroblasts. In parallel studies, we found that knockdown of hnRNP K by siRNA in SH-SY5Y cells attenuated stress granule (SG) formation and TDP-43 accumulation.

Discussion and conclusion: The data demonstrate that disease-causing mutations in TDP-43 are capable of inhibiting a key stress-associated change to cytosolic hnRNP K and FUS. Due to loss of hnRNP K and FUS in cells expressing mutant TDP-43, the data may suggest that WT TDP-43 can regulate the stability of hnRNP K and other RNA binding proteins, while the mutations inhibit this. This reinforces a key role for altered hnRNP K processing in TDP-43-mediated disease. Finally, this research provides further support for an important and complex interaction between TDP-43 and hnRNP K.

C14 TDP-43'S NEUROTOXICITY IS MEDIATED BY FRAGILE X PROTEIN AND SPECIFIC MRNA TARGETS

Coyne A¹

Yamada S¹

Boehringer A¹

Estes P¹

Lockwood D¹

Hart M²

Freibaum B³

Cassel J⁴

Reitz A⁴

Taylor JP³

Gitler A²

Zarnescu D¹

^zUniversity of Arizona, Tucson, AZ, USA

^aaStanford University, Palo Alto, CA, USA

^abSt. Jude Children's Research Hospital, Memphis, TN, USA

^acALS Biopharma, Doylestown, PA, USA

Email address for correspondence: [email protected]

Keywords: TDP-43, RNA dysregulation, FMRP

Background: Recent findings have demonstrated that defects in RNA processing are at the heart of pathophysiological mechanisms leading to neuronal dysfunction and death (1). Several RNA-binding proteins have been implicated in ALS both at the level of pathology and as causative agents (2). Of these, TDP-43 is a member of the hnRNP A/B family and harbors two RRM domains, an NLS, an NES and a prion-like C terminus domain. TDP-43 is localized to the nucleus under physiological conditions but associates with cytoplasmic RNA granules during stress (3).

Objectives: We used a Drosophila model of ALS (4, 5) to test the hypothesis that by associating with cytoplasmic RNA granules, TDP-43 leads to the sequestration of specific RNA binding proteins and subsequent dysregulation of their mRNA targets, which in turn may provide the basis for neural toxicity.

Methods and results: In a genetic screen for candidate RNA-binding proteins that modulate TDP-43 phenotypes, we have identified Fragile X Mental Retardation protein (FMRP), which has an established role in RNA transport and translation. FMRP and TDP-43 form a complex in mammalian cells and bind in vitro. Genetic interactions in Drosophila show that loss of FMRP enhances TDP-43 toxicity in neurons. In contrast, FMRP overexpression alleviates TDP-43 phenotypes both in the retina and motor neurons. To further evaluate the significance of FMRP as an effector of TDP-43 toxicity, we tested whether profilin and futsch, two established FMRP targets, can also modulate TDP-43 phenotypes. Interestingly, mutations in either profilin or futsch can alleviate TDP-43 neurodegeneration in the retina but only futsch rescues locomotor defects caused by TDP-43 overexpression.

Discussion and conclusion: These findings support our model of TDP-43-induced RNA dysregulation and point to specific mRNA targets regulated by FMRP as effectors of TDP-43 toxicity in vivo, in a tissue specific manner.

Acknowledgements:

This work was supported by the Jim Himelic Foundation, MDA and NIH.

References:

• Li YR, King OD, Shorter J et al. The Journal of cell biology 2013;201:361.
   PubMed Web of Science ®Google Scholar
• Lagier-Tourenne C, Cleveland DW. Cell 2009;136:1001.
   PubMed Web of Science ®Google Scholar
• McDonald KK et al. Human molecular genetics 2011; 20:1400.
   PubMed Web of Science ®Google Scholar
• Estes PS et al. Disease models and mechanisms 2013;6: 721–55.
   PubMed Web of Science ®Google Scholar
• Estes PS et al. Human molecular genetics 2011;20:2308.
   PubMed Web of Science ®Google Scholar

C15 MISSENSE MUTATIONS IN DIFFERENT DOMAINS OF THE MOUSE TDP43 GENE CAUSE DIVERSE EFFECTS ON RNA METABOLISM

Fratta P¹

Ricketts T²

Oliveira H²

Collins T¹

Wang E³

Housman D³

Greensmith L¹

Plagnol V¹

Acevedo-Arozena A²

Fisher EM¹

^adUniversity College London, London, UK

^aeMRC Mammalian Genetics Unit, Oxfordshire, UK

^afKoch Institute, MIT, Cambridge, USA

Email address for correspondence: [email protected]

Keywords: TDP43, RNA, model

Background: TDP43 is a ubiquitously expressed prevalently nuclear protein involved in RNA splicing, RNA stability and miRNA processing. Post-mortem analysis of patients with amyotrophic lateral sclerosis (ALS) has shown that TDP43 is depleted from the nucleus and accumulates in cytoplasmic neuronal inclusions, which are the pathological hallmark of the disease. Mutations in TDP43 have been found to be causative of a proportion of ALS familial cases reinforcing the primary importance of this molecule in the disease pathogenesis.

The pathogenic mechanism by which TDP43 acts is unclear, and both loss of nuclear function (LOF) and gain of function (GOF) mechanisms have been proposed.

TPD43 null mice have not proved a helpful tool to study TDP43 LOF due to early embryonic lethality, and transgenic mice overexpressing both wild type and mutant TDP43 have been reported to develop to neurotoxicity making the mechanistic dissection of mutant toxicity challenging.

Objectives: Here we characterize two novel mouse TDP43 mutant lines, carrying ENU-induced point mutations in the mouse endogenous Tardbp gene in order to study the effects of TDP43 mutations expressed at physiological levels in the mammalian central nervous system.

Results: The two mutations lay in two different domains of the TDP43 protein: (a) the F210I mutation is located in the RRM2 domain, involved in RNA binding, and indeed strongly reduces the RNA-binding capacity of the protein; and (b) the M323K mutation is in the glycine-rich C-terminal domain where the majority of pathogenic mutations are found. Both mutations are very disruptive and in homozygosity cause either late foetal or early post-natal lethality. We analyse TDP43 splicing target transcripts to show that the two mutations mainly have opposing effects. We perform RNAseq on embryo brains of homozygous, heterozygous and control animals from both lines and find that the two mutations have very different effects on RNA expression and splicing.

Conclusions: Our results underline the importance of studying models with physiological expression levels of TDP43 mutations and shed light on the different effects on RNA metabolism caused by the TDP43 loss and gain of function.

C16 SYSTEMIC DYSREGULATION OF TDP-43-BINDING MICRORNAS IN AMYOTROPHIC LATERAL SCLEROSIS

Freischmidt A

Müller K

Ludolph AC

Weishaupt JH

Ulm University, Ulm, Germany

Email address for correspondence: [email protected]

Keywords: microRNA, TDP-43, epigenetic

Background: Amyotrophic lateral sclerosis (ALS) is a classical neurodegenerative disease affecting primarily motor neurons. Central aspects of this disease are pathological aggregates of the TARDBP-coded protein TDP-43 in the vast majority of ALS patients. TDP-43 binds to and is involved in processing of both coding RNAs and a small subset of microRNAs (miRNAs), which are key epigenetic regulators of transcriptome plasticity and suspected to play a role in the pathogenesis of neurological diseases.

Methods: We therefore hypothesized that nine recently identified TDP-43-binding miRNAs might be dysregulated in ALS patients and quantified their levels in cerebrospinal fluid (CSF), serum and immortalized lymphoblast cell lines derived from ALS patients and healthy controls.

Results: We found that five of the TDP-43-binding miRNAs were dysregulated in the CSF and six in the serum of sporadic ALS cases (in at least 22 per experimental group). Differentially altered miRNAs together with a poor correlation between CSF and serum levels indicate a systemic dysregulation of miRNA biogenesis or degradation also outside the CNS, in line with the ubiquitous expression of TDP-43. The most downregulated miRNA-132 could be confirmed in lymphoblast cell lines from sporadic, TDP-43, FUS and C9orf72, but not SOD1 mutant patients. This parallels the TDP-43 pathology found in most ALS cases, but usually not in SOD1 mutant ALS patients.

Conclusion: Taken together, we report a systemic and genotype-dependent dysregulation of TDP-43-binding miRNAs in human biomaterial that might reflect an easily accessible biological measure of TDP-43 dysfunction in ALS.

C17 STRESS GRANULE (SG) DYNAMICS IS REGULATED BY AUTOPHAGIC MACHINERY IN FUS-RELATED ALS

Daigle G¹

Lanson N¹

Casci I¹

Monaghan J¹

Nichols C¹

Kryndushkin D^1,2

Shewmaker F^1,2

Pandey U¹

^agLouisiana State University Health Sciences Center, New Orleans, LA, USA

^ahUniformed Services University of the Health Sciences, Bethesda, MD, USA

Email address for correspondence: [email protected]

Keywords: FUS, stress granules, RNA-binding protein

FUS is a DNA-/RNA-binding protein found to be mutated in some cases of both sporadic and familial forms of ALS. It is still not clear how ALS-causing mutations in FUS leads to motor neuron degeneration. Here, we exploited a Drosophila model and mammalian neuronal cell lines to elucidate the role of the RNA-binding ability of FUS in regulating FUS- mediated toxicity. To determine the role of the RNA-binding ability of FUS in ALS, we mutated FUS RNA-binding sites (F305, F341, F359, and F368) to leucines and generated RNA-binding-incompetent mutants (4F-L) with and without ALS causing mutations R518K or R521C. We found that mutating 4F to L residues makes FUS RNA-binding-incompetent. We observed that ectopic expression of RNA-binding- incompetent FUS in fly brain, eyes, and motor neurons strongly blocks neurodegenerative phenotypes as compared to RNA-binding-competent FUS carrying ALS causing mutations. Interestingly, RNA-binding deficient FUS strongly localized to the nucleus of Drosophila motor neurons and mammalian neuronal cells, whereas FUS carrying ALS linked mutations was distributed to the nucleus and cytoplasm.

Importantly, we found that incorporation of mutant FUS into stress granules (SG) is dependent on the RNA-binding ability of FUS. SGs are dynamic aggregates composed of proteins and RNA that are formed when cells are under a variety of stresses. We observed that normally cytoplasmic SGs rapidly disassemble when stress conditions end, whereas cytoplasmic SGs formed in ALS patient cells having a FUS mutation fail to disassemble. This suggests that mutant FUS sequesters proteins and RNAs important for cellular homeostasis and the defect in disassembly of cytoplasmic SGs contributes to ALS. Interestingly, we found that induction of autophagy by rapamycine was sufficient to accelerate disassembly of cytoplasmic stress granules in ALS-patient cells. Furthermore, we observed that ectopic expression of atg1 strongly suppressed mutant FUS-related neurodegenerative phenotypes in our fly model of ALS. We strongly believe that these findings suggest potential therapeutic targets for ALS.

C18 SCREENING FOR COGNITIVE AND BEHAVIOUR CHANGE IN ALS

Abrahams S

Newton J

Niven E

Foley J

Bak TH

University of Edinburgh, Edinburgh, UK

Email address for correspondence: [email protected]

Keywords: cognition, behaviour, screen

Background: Despite the increase in awareness of ALS as a multisystem disorder, the cognitive status of the majority of ALS patients attending clinics remains unknown. Standard assessments for the detection of dementia are of limited use due to the range of physical problems in ALS with difficulties speaking, drawing and writing. Other cognitive screening tools fall short of a comprehensive assessment, measuring a single cognitive domain (executive functions) which is therefore not sensitive to the heterogeneity of cognitive change in ALS.

Objectives: This study presents the new Edinburgh Cognitive and Behaviour ALS Screen (ECAS), specifically developed for ALS patients with either limb or speech physical disability for use by health care professionals within the clinic. This multi-domain screen consists of a 15- to 20-minute interview and separate carer behaviour scale and is designed to detect the specific profile of cognition and behaviour changes in ALS and to differentiate it from other disorders.

Methods: Forty-eight ALS patients (none with evident dementia), 40 healthy controls and 20 carers were recruited. The ECAS, includes an ALS-Specific score (executive functions including social cognition; fluency; and language); an ALS Non-specific score (memory and visuospatial functions); a carer behaviour screen of five domains characteristic of frontotemporal dementia (FTD).

Results: Data from healthy controls produced abnormality cut-offs of 77/100 ALS-Specific score; 24/36 ALS non-specific score; 105/136 ECAS Total. 29% of patients showed abnormal ALS-Specific Scores, and 6% also showed abnormal ALS Non-specific scores. The most prevalent deficit occurred in language functions (35%) followed by executive functions and fluency (23% each). 40% of carers reported behaviour change in at least one domain, while 15% met criteria for possible FTD. The most prevalent symptoms were apathy, loss of sympathy/empathy and change in eating behaviour.

Discussion and conclusion: The ECAS is an effective brief assessment for ALS which determines the presence, severity and type of cognitive and/or behavioural changes. This is an essential first step to managing such symptoms and will enable streamlining of care into appropriate pathways and tailoring intervention.

C19 HIGH RATES OF COGNITIVE AND BEHAVIORAL IMPAIRMENT IN A LARGE PROSPECTIVE ALS STUDY

Murphy J¹

Factor-Litvak P²

Goetz RR³

Hupf J⁴

Woolley SC⁵

Lomen-Hoerth C¹

Mitsumoto H⁴

Als Cosmos Study Group⁴

^aiUniversity of California at San Francisco, San Francisco, CA, USA

^ajDepartment of Epidemiology, Mailman School of Public Health, New York, NY, USA

^akNew York State Psychiatric Institute, New York, NY, USA

^alColumbia University, New York, NY, USA

^amForbes Norris Center at California Pacific medical Center, San Francisco, CA, USA

Email address for correspondence: [email protected]

Keywords: FTD, cognitive, behavior

Background: Patients with ALS develop frontotemporal dementia (FTD) and more subtle cognitive impairment (ALSci) and behavioral impairment (ALSbi) (1). To recognize this extramotoneuronal involvement is essential to understand the pathogenesis of ALS. The development of screening tests to identify cognitive impairment is an important advancement in the field, to provide clinicians with effective tools for use in busy clinics.

Objectives: To investigate the cognitive and behavioral functioning of patients prospectively enrolled in the longitudinal ALS COSMOS study, and to identify the relationship between cognitive and behavioral symptoms and clinical variables among patients.

Methods: We have completed enrollment of 364 patients with probable or definite ALS based on El Escorial criteria, 258 of whom had completed baseline data for inclusion in this study. Patients were administered three well-recognized screening cognitive-behavioral measures: the ALS-Cognitive Behavioral Screen (ALS-CBS) (2). Frontal Behavioral Inventory-ALS version, and the Abrahams Written Fluency Test (3). Cognitive and behavioral impairment was defined based on screening tests and not a full exam, using the ALS-CBS.

Results: The sample was 42.9% female, with a mean age of 60.9 (SD = 9.9), mean education of 15.0 years (SD = 3.0), mean FVC of 79.7 (SD = 22.1) and mean ALSFRS-R of 36.3 (SD = 6.3). 9.8% of the sample met criteria for FTD with cognitive impairment and 18.4% of the sample met criteria for FTD with behavioral impairment. An additional 64.4% and 13.4% met ALSci and ALSbi criteria, respectively. While verbal fluency was not correlated with FVC or bulbar involvement, impaired performance on concentration and mental tracking subtests was associated with reduced FVC and reduced bulbar functioning. Increased behavioral impairment was associated with reductions in FVC, bulbar functioning, and ALSFRS-R and increased PBA symptoms.

Discussion: These diagnostic cut-offs were established using the ALS-CBS screening exam and are not to be confused with a formal neuropsychological diagnosis. Concentration and mental tracking were associated with FVC and bulbar impairment, with verbal fluency being unassociated with these clinical traits.

Conclusions: Approximately 10–15% of patients met FTD criteria, with larger proportions of patients meeting criteria for ALSci and ALSbi. Behavioral impairment correlated with more bulbar dysfunction, more PBA, and lower socioeconomic status. Illness duration did not correlate with ALSbi or ALSci, suggesting that extramotor involvement may be independent of the primary motor disease process.

Acknowledgments:

Grant support (HM): NIEHS, R01ES016348 and MDA.

References:

• Lomen-Hoerth C, Murphy J, Langmore S et al. Neurology 2003;60:10947.
   Google Scholar
• Woolley-Levine S, York M, Haring K et al. A L S 2007; 8:102.
   Google Scholar
• Abrahams S, Leigh PN, Harvey A et al. Neuropsychologia. 2000;38:73447.
   Google Scholar

C20 NEUROPSYCHIATRIC SYMPTOMS APPEAR VERY EARLY IN ALS – AND DO NOT AFFECT SURVIVAL

Mioshi E^1,2

Caga J¹

Hsieh S¹

Lillo P^1,3

Ramsey E¹

Devenney E¹

Hornberger M^1,2

Hodges JR^1,2

Kiernan MC^1,2

^anNeuroscience Research Australia, Sydney, Australia

^aoSchool of Medical Sciences, UNSW, Sydney, Australia

^apUniversity of Chile, Santiago de Chile, Chile

Email address for correspondence: [email protected]

Keywords: neuropsychiatric symptoms; behavioural changes; survival

Background: While the presence of cognitive deficits has been well-established in ALS, the identification and assessment of neuropsychiatric symptoms is less developed to date. Apathy has been recognised but presence of other challenging behaviours such as disinhibition and stereotypical behaviour, common to FTD, is less understood. Finally, concomitant ALS-FTD is associated with worse prognosis but there is little understanding if milder degrees of dysfunctional behaviour also affect survival in ALS.

Objectives: (1) To investigate patient susceptibility to neuropsychiatric symptoms and classic motor symptoms in ALS; and (2) To examine the impact of neuropsychiatric symptoms on survival.

Methods: Two hundred and ninty-nine patients (Limb onset = 159 and bulbar onset = 60) were included, following current diagnostic criteria for ALS. Behavioural symptoms were measured via a short version of the Cambridge Behavioural Inventory (CBI-R) and classic ALS symptoms via the ALSFRS-R.

For the analysis of symptom susceptibility (behavioural vs classic motor), a Rasch analysis was employed (n = 219). For the survival analysis, Cox proportional hazard regression models were applied (n = 159 patients with complete information), which included classic motor symptoms and neuropsychiatric symptoms.

Results: The Rasch analysis demonstrated that behavioural symptoms appear earlier than the development of classic motor symptoms in ALS. The differences in behavioural scores did not affect survival; patients with abnormal scores in neuropsychiatric domains did not have a different rate of survival than those without (Chi square: 3.447, p = 0.328, −2 log likelihood 377.341).

Discussion: The notion of neuropsychiatric symptoms appearing prior to classic motor symptoms in ALS is novel and potentially controversial, given that it brings ALS even closer to FTD, while confirming that ALS is a multisystem disorder, affecting different areas concomitantly. The marked and early presence of the neuropsychiatric symptoms are likely to affect clinical decisions (eg, PEG insertion; compliance in use of respiratory devices, etc.). The fact that neuropsychiatric symptoms alone do not affect survival confirms their pervasiveness, and only the co-occurrence of ALS and FTD seems to affect prognosis.

Conclusions: Neuropsychiatric symptoms are an early feature in ALS, but have not been thoroughly evaluated to date. They appear earlier than classic motor symptoms. Importantly, these behavioural symptoms alone do not seem to affect survival in ALS, which in turn confirms their pervasive nature in ALS.

C21 BEHIND THE CURTAIN OF DYSARTHRIA: THE NATURE OF LANGUAGE IMPAIRMENT IN MND

Rewaj P

Abrahams S

Ladd DR

Bak TH

University of Edinburgh, Edinburgh, UK

Email address for correspondence: [email protected]

Keywords: language, speech, cognition

Background: While aphasic-type impairments have been documented in MND since the early 1990s (1), the common clinical view is still that communication difficulties can be attributed solely to dysarthria.Yet recent evidence raises the possibility of central processing impairments. Naming deficits have been frequently reported, and comprehension of grammatical contrasts has also been shown to be impaired in some patients (2). However, there is also increasing evidence of spelling errors, which could suggest contribution of a more phonologically based impairment. Japanese authors have reported MND patients with selective errors writing phonologically based kana characters, but not semantically based kanji (3), while others have reported substitution and transposition of letters in patient writing (4).

Objectives: To examine the nature of speech and language deficits in people with MND and the extent to which: (1) expressive communication impairment cannot be solely explained by dysarthria; (2) receptive language deficits impact upon communication; and (3) clinical management should be adapted to meet the communication needs of patients.

Methods: Twenty-five MND patients from across Scotland with changes in speech and/or language were tested using a battery of experimental and standardised tests of naming, spelling, grammatical comprehension, prosody and phonological and orthographical awareness. Patients were screened for levels of dysarthria and hearing impairment and results were compared to those of 25 age-, sex- and education-matched controls.

Results: Of the 25 MND patients, 13 (52%) performed two or more SD below the control mean on at least one linguistic assessment. As a group, MND participants performed significantly worse than controls on measures of naming (p = 0.001), spelling (p = 0.006), grammatical comprehension (p = 0.001), emotional, but not acoustic prosody (p = 0.002) and orthographical awareness (p = 0.003). However, the pattern of impairment was not global, with five patients showing dissociation of performance on naming and grammatical comprehension and six between naming and spelling.

Discussion and conclusion: Communication impairment in MND cannot solely be attributed to dysarthria in every patient. Dissociation in performance for some patients between linguistic measures suggests that there may be multiple impairment profiles, and that naming assessment alone cannot give an accurate measure of linguistic abilities. This study highlights the importance of multidimensional assessment of language for clinical management, particularly with regard to AAC strategies.

Acknowledgements:

This study was funded by MND Scotland.

References:

• Caselli RJ, Windebank AJ, Petersen RC et al. Ann Neurol 1993;33(2):200–207.
   Google Scholar
• Bak TH, O’Donovan DG, Xuereb JH et al. Brain 2001; 124(Pt 1):103–120.
   PubMed Web of Science ®Google Scholar
• Ichikawa H, Hieda S, Ohno H et al. European Neurology 2010;64:148–155.
   PubMed Web of Science ®Google Scholar
• Zago S, Poletti B, Corbo M et al. Behavioural Neurology 2008;19(4):169–175.
   Google Scholar

C22 EFFICACY OF HYPNOSIS-BASED TREATMENT IN ALS AND ITS EFFECT ON THE CAREGIVER: RESULTS OF A SIX-MONTH LONGITUDINAL STUDY

Palmieri A¹

Calvo V¹

Kleinbub J R¹

Pagnini F²

Sambin M¹

D'Ascenzo C³

Querin G³

Barilaro P¹

Sorarù G³

^aqDepartment of Philosophy, Sociology, Pedagogy and Applied Psychology. University of Padua, Padua, Italy

^arDepartment of Psychology. Catholic University of Milan, Milan, Italy

^asDepartment of Neurosciences. University of Padua, Padua, Italy

Email address for correspondence: [email protected]

Keywords: psychological intervention, hypnosis, quality of life

Background: Amyotrophic lateral sclerosis (ALS) has a catastrophic psychological impact on affected patients and their caregivers. Previous research has shown that psychological welfare and quality of life (QoL) are crucial factors for the patient’s prognosis. However, although would be strongly needed, there is a lack of research on the efficacy of psychological intervention and by now there are no clinical intervention guidelines.

Objectives: Following the promising results of a pilot research (1), the aim of the study is to investigate the long term effects of an hypnosis-based intervention on psychological health and perception of secondary physical symptoms in patients and the indirect effect on their caregivers.

Methods: Fifteen typical ALS patients, (8 females and 7 males) and their closest caregiver (10 females and 5 males) participated in the study. The patients were treated with a domiciliar hypnosis intervention and self-hypnosis training protocol lasting 1 month conducted by a trained clinician. Anxiety and depression levels were measured with the Hospital Anxiety and Depression Score in patients and caregivers before treatment (T0) and in three follow-up phases: immediately following the treatment (T1) and after 3 (T2) and 6 months (T3). QoL and perceived physical symptoms changes were investigated in patients, in every study phase, with ALS Assessment Five Items Questionnaire and the ALS Specific Quality of Life-Revised questionnaire. Test–retest analyses were conducted with Wilcoxon signed-rank tests and the Cliff’s delta was used to estimate effect sizes.

Results: One month pre-post treatment improvement in depression, anxiety, and physical and affective facets of QoL were found significant. Decreases in perception of physical symptoms such as cramps, sleep disorders, emotional lability, and fasciculations were reported by our patients. Reduced anxiety and improvements in QoL were maintained at the follow-ups at 3 and 6 months. Significant improvements in caregiver’s anxiety and depression values were measured at T2 and maintained at T3.

Discussion: The collected data clearly showed a positive effect of the hypnosis-based intervention on the patients and indirectly on their caregivers. The preservation and improvement in some aspects of anxiety, depression and QoL after 6 months appears as the most relevant clinical result of our study.

Conclusions: To the best of our knowledge, this is the first report on psychological intervention protocol effectiveness on ALS patients. The findings provide initial support for using hypnosis to manage ALS physical consequences and mainly to cope its dramatic psychological implications for patients. Hypnosis could represent an eligible treatment where classical psychological colloquy may result impossible in severe bulbar symptomatology or locked-in condition.

Reference:

• Palmieri A, Kleinbub JR, Calvo V et al. Front Psychol. 2012;3:465.
   PubMed Web of Science ®Google Scholar

C23 DIGNITY THERAPY: A PSYCHOTHERAPEUTIC INTERVENTION TO ENHANCE THE END OF LIFE EXPERIENCE FOR PEOPLE WITH MOTOR NEURONE DISEASE AND THEIR FAMILY CARERS

Bentley B¹

Aoun S¹

O’Connor M¹

Breen L¹

Chochinov HM²

^atCurtin University, Perth, Western Australia, Australia

^auUniversity of Manitoba, Winnipeg, Manitoba, Canada

Email address for correspondence: [email protected]

Keywords: psychosocial support, dignity therapy, quality of life

Background: Quality of life for people with MND is primarily dependent on spiritual, existential, relationship and support factors. There have been numerous calls for psychosocial interventions which address hope, meaning and existential distress, though psychotherapies addressing these needs are rare. Studies have also documented the substantial burden and distress experienced by MND family carers, as well as the absence of targeted interventions to alleviate this distress.

Dignity Therapy is a brief psychotherapy designed for people with terminal diagnoses to enhance a sense of purpose, meaning, and overall quality of life. Dignity Therapy has proven successful at relieving psychosocial distress in patients, as well as providing comfort to family members, though most patients in previous studies had cancer diagnoses.

Objectives: This study examined whether Dignity Therapy enhances the end of life experience for people with MND and their family carers. Feasibility and acceptability were assessed, and any required modifications to the therapy or other special considerations were examined.

Method: Twenty-six people with MND and 17 family carers from Western Australia participated in the study, which adopted a pre-post design to investigate the impact of Dignity Therapy on both people with MND and their family carers. Secondary outcome measures were also obtained 1 week after the therapy using feedback questionnaires completed by patients and family carers.

Results: Dignity Therapy was beneficial to people with MND, especially in the areas of encouraging acceptance (70%), strengthening identity (70%), helping to address unfinished business (65%) and helping to feel they could still play an important role (65%). Acceptability was high, with 84% reporting the therapy was helpful to them and 92% reporting the therapy was satisfactory. Acceptability was also high with family carers, with 94% finding it helpful to their family member and 75% recommending it to others. Seventy-five percent of family carers believed the dignity therapy document would be a comfort during bereavement. Low base rates of distress precluded being able to demonstrate significant post-intervention differences on measures of distress. These results mirror those found in previous studies of dignity therapy with cancer patients and their carers.

Mild cognitive decline and pseudo-bulbar affect were found to have minimal impact on the acceptability and feasibility of the psychotherapy. Dysarthria (speech impairment) did not impact acceptability or feasibility and the therapy was successfully completed using assisted communication devices.

Discussion and conclusion: Dignity Therapy is acceptable to people with MND and their family carers, who report numerous benefits from the therapy. This psychotherapy can be successfully delivered using assisted communication devices.

Acknowledgements:

This research is supported by a grant from the Australian Research Council and funding from the Motor Neurone Disease Association of Western Australia.

C24 SMALL-MOLECULE SCREENING FOR NEUROPROTECTIVE AGENTS

Finkbeiner S

University of California, San Francisco and the Gladstone Institutes, California, USA

Email address for correspondence: [email protected]

Keywords: therapeutics discovery, human induced pluripotent stem cells, automated imaging

Effective therapies for motor neuron disease are desperately needed. Unfortunately, clinical trials of compounds that showed some efficacy in preclinical models have thus far largely failed. Although the precise reasons for these failures are debated, the need for preclinical models that better predict the results of clinical trials seems clear. Moreover, the majority of patients with motor neuron disease have no identifiable genetic cause. How to develop preclinical models that faithfully represent patients with sporadic motor neuron disease has been unclear. To help address these unmet needs, we developed primary neuron models of motor neuron disease that exhibit several disease-relevant phenotypes. Of particular interest are motor neuron disease models developed from human motor neurons and astrocytes differentiated from patient-derived induced pluripotent stem cells (iPSC) reprogrammed from fibroblasts of patients who have genetic and sporadic motor neuron disease. To develop human iPSC models of motor neuron disease, we applied high-throughput automated single- cell longitudinal analysis, which uncovered several disease-relevant phenotypes. In turn, we are using these primary murine and human neuron models of motor neuron disease to screen for genetic and pharmacological modifiers that mitigate disease phenotypes and show therapeutic potential and to help medicinal chemistry efforts to optimize leads from primary screens.

C25 RNA-INDUCED TOXICITY FROM THE C9ORF72 ALS/FTD REPEAT EXPANSION IS MITIGATED BY ANTISENSE INTERVENTION

Donnelly C¹

Haeusler A¹

Pham J¹

Vidensky S¹

Hoover B¹

Daley E¹

Zhang P¹

Poth E¹

Rigo F²

Wang J¹

Bennett F²

Traynor BJ³

Blackshaw S¹

Sattler R¹

Rothstein J¹

^avJohns Hopkins University, Baltimore, MD, USA

^awIsis Pharmaceuticals, Carlsbad, CA, USA

^axNational Institutes of Health, Bethesda, MD, USA

Email address for correspondence: [email protected]

Keywords: C9ORF72, antisense, RNA toxicity

Background: A hexanucleotide ‘GGGGCC’ repeat expansion in the noncoding region of the C9ORF72 gene has recently been identified in ˜ 30% of familial and up to 10% of sporadic ALS cases (1–3) and is therefore the most common genetic abnormality associated with ALS to date. Since the function of the C9ORF72 protein is unknown and a C9ORF72 rodent model has not yet been generated, few methodologies exist to begin to elucidate the pathogenicity behind this repeat expansion. However, repeat expansions in non-protein coding regions are the known cause of other neuromuscular disorders (eg DM1/2) and pathogenesis is, in part, due to the accumulation of cis acting expanded repeat-containing RNA that sequester trans acting RNA binding proteins (RBP).

Objectives: To generate and characterize iPS neurons and astrocytes from C9ORF72 ALS patient fibroblast, compare with human autopsied CNS tissue, and develop antisense therapeutics (ASO) that target toxic RNAs to rescue aberrant iPS phenotypes.

Methods: Using high-throughput screenings, we have profiled the transcriptome of C9ORF72 patient-derived iPS motor neurons and astrocytes as well as human autopsy tissue. Proteome arrays and biochemical assays were employed to identify GGGGCC RNA-binding partners. RNA FISH-IF was utilized to validate the presence and co-localization of toxic RNA foci and (GGGGCC) exp RNA-binding partners in vitro and in vivo. RAN translation products were detected with monospecific antibodies. Glutamate toxicity studies were performed using propidium iodide imaging and LDH assays.

Results: We have identified RNA-induced toxicity through accumulation of intranuclear (GGGGCC) exp RNA, an aberrant transcriptome, RBP sequestration, and dose dependent susceptibility to excitotoxicity in patient-derived C9ORF72 ALS iPS neurons. RAN translation products were present in iPS neurons – but did not contribute to toxicity. Importantly, ASO treatment rescues the C9ORF72 phenotype.

Discussion and conclusion: Using iPS-differentiated neurons, we have 1) identified intranuclear (GGGGCC)exp RNA foci, 2) described dysregulated gene expression in C9ORF72 ALS tissue that match iPS cell lines, 3) have identified (GGGGCC)exp RNA binding partners, and 4) determined that C9ORF72 iPS neurons are highly susceptible to excitotoxicity. Importantly, all of these pathogenic characteristics are mitigated with antisense therapeutics to the C9ORF72 transcript or repeat expansion despite the presence of repeat associated non-ATG translation (RAN) products. Taken together, these data indicate a toxic RNA gain-of-function mechanism as a cause of C9ORF72 ALS. These studies also provide candidate antisense therapeutics as well as human pharmacodynamic biomarkers for drug actions.

Acknowledgements:

NIH, Robert Packard Center, Ansari Center, ALSA, Adams Foundation, MDA, and the Maryland Stem Cell Research Fund.

References:

• DeJesus-Hernandez M, Mackenzie IR, Boeve BF et al. Neuron 2011;72:245–256.
   PubMed Web of Science ®Google Scholar
• Renton AE, Majounie E, Waite A et al. Neuron 2011;72: 257–268.
   PubMed Web of Science ®Google Scholar
• Majounie E, Ranton AE, Mok K et al. Lancet Neurol 2012; 11:323–330.
   PubMed Web of Science ®Google Scholar

C26 COMPARISON OF DISEASE MECHANISMS AND THERAPEUTIC INTERVENTIONS IN PRIMARY CULTURE MODELS OF MULTIPLE FAMILIAL FORMS OF ALS/MND

Tran L

Tibshirani M

Gentil B

Durham H

McGill University, Montreal, QC, Canada

Email address for correspondence: [email protected]

Keywords: calcium channel blocker, heat shock proteins, glutamate receptors

Background: Culture and animal models of familial ALS due to mutations in SOD1 (ALS1) have been used extensively for preclinical studies of candidate therapies for ALS. The discovery of additional genes linked to ALS provides the opportunity to compare disease mechanisms and therapeutic efficacy in multiple models. Our laboratory has established primary culture models of ALS1, ALS6 (FUS mutations) and ALS10 (TARDP mutations). A common factor is increased propensity of ALS-causing mutant proteins to aggregate and form inclusions in motor neurons; however, the pathways disrupted and factors contributing to cell-type vulnerability could vary, conferring differential sensitivity to intervention. This study is comparing how mutants of SOD1 and of the RNA-binding proteins, FUS and TDP43, exert toxicity in motor neurons.

Objectives: (1) Previous studies established a role for protein misfolding, glutamate receptor activation, calcium dysregulation, mitochondrial dysfunction and ER stress in the toxicity of mutant SOD1. The aim is to determine if these factors apply to mutant FUS and TDP43. (2) Upregulation of protein chaperones protects motor neurons from mutant SOD1. The aim is to determine if increasing expression of heat shock proteins prevents FUS and/or TDP43 from mislocalizing and aggregating in the cytoplasm, as well as downstream consequences in motor neurons.

Methods: Dissociated cultures of murine spinal cord-DRG are matured for 3–6 weeks. Human WT or ALS-causing mutants are expressed in motor neurons by intranuclear microinjection of expression plasmids along with a fluorescent dextran marker. Cytosolic Ca²⁺ is measured using fura-2. Mitochondrial morphology is visualized by co-expression of mitochondrially targeted eGFP/dsRed. Intracellular localization is assessed by indirect immunocytochemistry or by incorporating an eGFP tag in the sequence.

Results: (1) In contrast to SOD1^G93A, neither FUS nor TDP43 mutants increased cytosolic Ca²⁺. Mitochondrial length was decreased in motor neurons expressing mutant FUS, but not to the severe extent as SOD1^G93A. Although the calcium channel inhibitor, lomerizine, the AMPA glutamate receptor antagonist, CNQX, and riluzole-inhibited mutant SOD1 toxicity, results to date indicate that they do not affect cytoplasmic accumulation or formation of inclusions of mutant FUS or TDP43. (2) Upregulation of HSPs prevents multiple aspects of mutant SOD1 toxicity, despite only partially limiting Ca²⁺ dysregulation. Experiments with FUS and TDP43 mutants are ongoing.

Discussion and conclusion: Many factors could contribute to the failure of clinical trials of compounds showing efficacy in models of ALS1, one being that ALS is a syndrome with multiple causes. This study points to early contribution of calcium dysregulation in toxicity of mutant SOD1, but not FUS or TDP43. The identification of multiple genes and pathways linked to familial ALS provides an opportunity for enhanced preclinical testing and stratification of patients in trials.

Acknowledgements:

This work was funded by ALS Canada, ALSA and MDA.

C27 TARGETING RNA FOCI SHOWS A THERAPEUTIC EFFECT IN IPSC-DERIVED MOTOR NEURONS FROM C9ORF72 REPEAT PATIENTS

Sareen D¹

Gire O’rourke J¹

Muhammad AK¹

Grant S¹

Simpkinson M¹

Bell S¹

Carmona S¹

Ornelas L¹

Sahabian A¹

Baughn M²

Ravits J²

Harms MB³

Rigo F⁴

Bennett F⁴

Svendsen CN¹

Baloh RH¹

^ayCedars-Sinai Medical Center, Los Angeles, CA, USA

^azUCSD, San Diego, CA, USA

^baWashington University, St. Louis, MO, USA

^bbIsis Pharmaceuticals, Carlsbad, CA, USA

Email address for correspondence: [email protected]

Keywords: iPSC, antisense

Background: Expansions of a hexanucleotide repeat (GGGGCC) in the noncoding region of the C9ORF72 gene were recently found to be the most common cause of familial ALS (C9-ALS) and also frontotemporal lobar degeneration, and are also present in other neurological diseases. While the mechanism of repeat toxicity remains unclear, accumulations of RNA GGGGCC rich foci are a hallmark of pathological changes seen in these conditions.

Objectives: To report a cellular model of C9-ALS using patient-derived induced pluripotent stem cells (iPSCs), and use it as a tool to understand the mechanism of the disease.

Results: The absolute number of motor neurons generated from C9-ALS and control lines was similar, suggesting no overt neurodegeneration occurred. Interestingly, the expansion showed somatic instability in patient-derived neuron cultures, and transcription of the repeat was paradoxically enhanced, leading to accumulations of GGGGCC repeat containing RNA foci selectively in C9-ALS patient cells. Repeat containing RNA foci co-localized with hnRNPA1, suggesting they are functional structures that can alter splicing and transcription to influence disease pathogenesis. Accordingly, C9-ALS patient motor neurons showed a distinct transcriptional profile with changes in several genes including DPP6, previously implicated in ALS genome wide association studies. Finally, we show that antisense oligonucleotides targeting the C9ORF72 transcript can suppress RNA foci, and reverse gene expression alterations in patient cells.

Discussion and conclusion: These data support the concept of a toxic RNA gain of function mechanism in C9-ALS patients. Furthermore, using patient-specific motor neurons as a model, we suggest that ASOs are a promising therapeutic strategy for C9ORF72 expansion diseases.

Acknowledgements:

Burroughs-Wellcome Fund, NIH/NINDS

C28 THERAPY DEVELOPMENT FOR ALS/MND AND FRONTOTEMPORAL DEMENTIA WITH C9ORF72 EXPANSION: ANTISENSE OLIGONUCLEOTIDE MEDIATED REDUCTION IN NUCLEAR RNA FOCI

Lagier-Tourenne C^1,2

Baughn M¹

Rigo F³

Sun S^2,4

Markmiller S⁴

Watt A³

Sun Y^2,4

Polymenidou M^2,4

Qiu J⁴

Li H⁴

Xue Y⁴

Chun S³

Liu P⁴

Katz M³

Freier S³

Fu XD⁴

Yeo GW⁴

Bennett F³

Cleveland DW^2,4

Ravits J¹

^bcNeurosciences, UCSD, La Jolla, CA, USA

^bdLudwig Institute for Cancer Research, La Jolla, CA, USA

^beISIS Pharmaceuticals, Research, Carlsbad, CA, USA

^bfCellular and Molecular Medicine, La Jolla, CA, USA

Email address for correspondence: [email protected]

Keywords: C9orf72, RNA foci, antisense oligonucleotides

Background: Expanded hexanucleotide GGGGCC repeats in a non-coding region of the C9orf72 gene were recently identified as the most common genetic cause of amyotrophic lateral sclerosis and frontotemporal dementia (C9-ALS/FTD). The pathogenic mechanisms of this expansion are not understood but initial observations point to either a loss of function of the endogenous C9orf72 gene, or a toxic gain of function of the expanded RNA, mediated either by sequestration of RNA-binding proteins or by production of aberrant polypeptide(s) through repeat-associated non-ATG-dependent (RAN) translation.

Objectives: Our objectives were to determine the contribution of loss of function versus toxic gain of function in neurodegeneration linked to C9orf72 expansion and to test the efficacy and tolerability of therapeutic strategies lowering C9orf72 RNA.

Methods: We used locked nucleic acid probes to identify expanded RNA foci by in situ hybridization in peripheral cells and autopsies from C9-ALS/FTD patients. We determined the sensitivity of foci to RNAse H-dependent antisense oligonucleotides (ASOs) versus cytoplasmically acting siRNAs, and we used genomic approaches to define RNA profiles linked to C9orf72 expansion. Tolerability to C9orf72 depletion in the adult nervous system was determined by behavioral testing of normal mice after intraventricular delivery of a mouse C9orf72-specific ASO.

Results: We identified accumulation of expanded RNAs into nuclear foci in multiple cell types including spinal motor neurons, cortical neurons, Purkinje cells and glial cells from C9-ALS/FTD patients. Such foci are not seen in nervous systems from sporadic ALS, Parkinson's disease, MAPT mutation or non-neurologic controls.

The presence of RNA foci in patients supports a toxic gain of function that may be tackled by therapeutic strategies lowering the production of abnormal RNAs. We demonstrated the efficiency of nuclear RNase H-dependent ASOs to target C9orf72 RNA transcripts and reduce the formation of foci. Importantly, isoform-specific ASOs lowering only RNAs that contain the expansion reduced foci without altering overall C9orf72 transcript levels. By comparison, cytoplasmically acting siRNAs against C9orf72 reduced overall RNA levels, but did not reduce foci.

Finally, RNA alterations, relative to unaffected controls or sporadic ALS samples, were identified in fibroblasts and spinal cords from C9-ALS/FTD patients by genome wide RNA analysis approaches. Depletion of C9orf72 in control fibroblasts and in spinal cords from normal adult mice resulted in gene expression changes that do not significantly overlap with the C9orf72 RNA signature, thereby providing evidence that C9orf72 RNA alterations are not due to C9orf72 loss of function. We determined that reducing endogenous C9orf72 for several months in the central nervous system of normal mice was well-tolerated, a crucial step towards the development of ASO therapy in C9-ALS/FTD.

Conclusion: A toxic gain of function is likely fundamental in C9-ALS/FTD pathogenesis. Strong evidence supports that ASOs targeting C9orf72 will be a powerful therapeutic strategy.

C29 ADVANCE CARE PLANNING IN ALS – THE ROLE OF THE PHYSICIAN

Borasio GD

CHUV, University of Lausanne, Lausanne, Switzerland

Email address for correspondence: [email protected]

Keywords: advance care planning, palliative care, end of life decisions

The importance of advance care planning (ACP) in ALS has been increasingly recognized in recent years. ACP is much more than completing advance directives and nominating a health care proxy, important as though these elements are. ACP is an ongoing process that starts when the diagnosis is communicated and continues throughout the course of the illness.

As ALS is an illness with a relatively predictable course in most patients, prevention is the key to successful disease management and palliative care. Several triggers have been identified by expert consensus (1) for starting end of life discussions in ALS:

• The patient or family asks – or 'opens the door’ – for end-of-life information and/or interventions;

• Severe psychological and/or social or spiritual distress or suffering;

• Pain requiring high dosages of analgesic medications;

• Dysphagia requiring feeding tube;

• Dyspnea or symptoms of hypoventilation, a forced vital capacity of 50% or less; and

• Loss of function in two body regions.

Importantly, end-of-life decisions in ALS need regular revisiting, as ALS patients have been shown to change their priorities and wishes for the end-of-life, sometimes dramatically, within short periods of time.

The role of the physician in this process is manifold:

• To be aware of the legal regulations governing ACP and end-of-life decisions, which vary significantly between countries.

• To be attentive to clues indicating the patients desire to discuss ACP.

• To actively inquire for the presence of the trigger points described above.

• To ensure inclusion of the relatives in ACP, with the patient’s consent.

• To alert the patient early enough of impending treatment decisions, so as to allow time for reflection and discussions within the family.

• To bring the patient in contact with other patients who had to face similar decisions (eg PEG) in order to facilitate the decision process.

• To respect the patients’ decisions even when they appear irrational and/or self-harming.

At its core, ACP is an ongoing communication process which rests in particular on the active listening skills on the physician’s side. As Søren Kierkegaard said:

“If we want to help somebody, we must first find out where he stands. This is the secret of all caring. Those who cannot do this, are stuck with an illusion if they think they can help others. In order to really be able to help somebody, I must understand more than he does - but first and foremost I must understand what he understands.” (2).

References:

• Mitsumoto H et al. ALS 2005;6:145–154.
   Google Scholar
• Søren Kierkegaard, Synspunkter for min Forfatter Virksomhet. In: SK, Die Schriften über sich selbst, Diederichs Verlag, Regensburg 1951; 38–39.
   Google Scholar

C30 THE ISSUES OF END-OF-LIFE CARE PLANNING – THE FINAL STAGES

Oliver D^1,2

^bgWisdom Hospice, Rochester, Kent, UK

^bhUniversity of Kent, Chatham, Kent, UK

Email address for correspondence: [email protected]

Keywords: advance care planning, end of life, conflicts

In the care of people with ALS/MND, there is a need to prepare for end-of-life care early in the disease progression. This is particularly the case if there are communication difficulties, due to bulbar disease, or evidence of cognitive change. It can be argued that preparation for end-of-life care starts at diagnosis and certainly how the diagnosis is given and the information provided at this time can greatly influence the attitudes of the patient and family and their concerns as the disease progresses.

Advance care planning is widely advocated across the world, although the exact procedures will vary from country to country. However, the overall aims will be to allow discussion of the patient’s wishes while they can clearly express them, so that all involved, patient, family and health and social care professionals, can act appropriately at the end of life, when the patient is unable to express their wishes. However, there are often barriers to these discussions:

• From the person with ALS/ MND;

  • ○ Fear of disease progression

  • ○ Magical thinking – “if I talk about dying I will die”

  • ○ Concerns for the family/carers

• From the family;

  • ○ Fear of the unknown;

  • ○ Not wanting to upset the patient; and

  • ○ Fearing that discussion of dying will reduce hope and affect the disease progression.

• From the professionals;

  • ○ Patients will become upset

  • ○ Families do not want these discussions;

  • ○ Concerns about giving up hope;

  • ○ It is too hard to talk about these issues.

However, there is the need to face these issues and help patients and families express their wishes, and hopes, and facilitating them in achieving their wishes if possible, such as visiting friends or family or significant places. If the wishes are not known this can lead to conflict and increased issues for all involved in the care of the patient. These conflicts may include the disagreements within families, with varying views of the patient’s wishes, and conflicts within teams as they all have different ideas about what the patient would have wished and fail to be objective as they may bring their own views into these discussions.

Thus careful discussion throughout the disease progression will allow the views and wishes of the patient, in collaboration with their family and professional carers, to be expressed and recorded. In this way the patient’s care can be co-ordinated effectively and then all can be supported – patient, family and professionals.

C31 TARGETING IMMUNE RESPONSES IN NEURODEGENERATIVE DISEASE

Rivest S

CHU de Quebec, Quebec, Canada, Laval University, Quebec, Canada

Email address for correspondence: [email protected]

Keywords: monocytes, brain diseases, innate immunity

The concept of the central nervous system (CNS) as an immune privileged organ has led to a common misunderstanding that it is not an active immunological organ, guarded from its surroundings by the blood-brain barrier (BBB). Recent advances in this field clearly demonstrate that the CNS is a highly immunologically active organ, with complex immune responses mostly based on innate immune processes. Such responses implicate a continuum of heterogenous cell types both inside the CNS, in the periphery and at their interface, the BBB. We recently found that the progressive cognitive decline and decrease in expression of numerous synaptic markers and neurotrophins in the brain of mouse models of Alzheimer’s disease (AD) correlated with major changes in the proportions of peripheral blood monocyte subsets when compared with age-matched controls. Indeed, there is a defect in the production of circulating M1 monocytes in APP/PS1 mice, whereas the population of M2 monocytes remains normal in this mouse model of AD. In this regard, low levels of macrophage colony-stimulating factor (M-CSF) were recently measured in patients with presymptomatic AD or mild cognitive impairment (MCI), which together with low levels of other haematopoietic cytokines predicted the rapid evolution of the disease towards a dementia diagnosis 2–6 years later. In vitro, exposure of mouse microglia to M-CSF enables the acidification of their lysosomes and, subsequently, the degradation of internalized Aβ. Injecting M-CSF to transgenic mice that spontaneously develop AD on a weekly basis prior to the appearance of learning and memory deficits prevented cognitive loss. The treatment also restored the population of M1 monocytes in the circulation and greatly decreased Aβ levels. In addition, M-CSF treatment resulted in the stabilization of the cognitive decline state in transgenic mice that already had Aβ pathology. These results are quite encouraging as they suggest that stimulating circulating M1 cells may have a great therapeutic potential for AD. This could explain why recent clinical trials in AD have revealed that anti-inflammatory drugs not only failed to improve cognitive functions, but worsened disease in some patients. It is therefore likely that stimulating monocytic cells may be a new therapeutic avenue for treating brain diseases, such as AD. In this presentation, will show new data regarding the potent effects of new molecules to stimulate innate immune cells as a preventive and curative treatment for brain diseases. We will also show the central role of the neurovascular unit in diseases of the CNS and how it can be targeted for novel therapeutic strategies.

Acknowledgements:

The Fonds de la Recherche du Québec – Santé (FRQS), Canadian Institutes in Health Research (CIHR) and the Multiple Sclerosis Scientific Research Foundation of Canada support this research.

C32 ACTIVATION OF THE BRAIN'S CHOROID PLEXUS FOR LEUKOCYTE TRAFFICKING AS A THERAPEUTIC APPROACH FOR ALS

Baruch K

Kunis G

Schwartz M

Weizmann Institute of Science, Rehovot, Israel

Email address for correspondence: [email protected]

Keywords: vaccination, immunomodulation, choroid plexus

Background: Circulating CD4⁺ T cells were shown to play a beneficial role in controlling the local neuroinflammation in Amyotrophic lateral sclerosis (ALS); in the mutant superoxide dismutase 1 G93A (mSOD1) murine model of ALS, T-cell deficiency contributes to increased neuronal loss, while boosting T cell levels reduces it (1).

Objectives: We recently identified the brain’s choroid plexus (CP) as a compartment in which CNS-specific CD4⁺ T cells reside (2), and as a trafficking gate for “healing” leukocytes to the damaged CNS (3). Based on these findings, we hypothesized that the activation of this compartment for leukocytes trafficking is dysregulated in ALS, resulting in poor immune cell recruitment to the CNS.

Methods: Adult C57BL/6J wild-type (WT) and mSOD1 mice were immunized with MOG_35–55 or ovalbumin peptide. Cerebrospinal fluid, CP and spinal cords were collected and analyzed by flow cytometry, qPCR, immunohistochemistry and Luminex systems. CP epithelial cells were cultured in vitro.

Results: We found that activation of the CP for leukocyte trafficking is dependent on IFN-g production by T cells residing in the CP stroma. In mSOD1 mice, starting from the asymptomatic stage, the CP showed lower numbers of IFN-g-producing CD4⁺ T cells, lack of activation for leukocyte trafficking, and poor recruitment of CD4⁺ T cells to the cerebrospinal fluid and the spinal cord. Active immunization of mSOD1 mice with a CNS-derived peptide, at the asymptomatic stage of the disease, activated their CP to facilitate leukocyte trafficking, and led to the specific accumulation of Foxp3⁺ regulatory T cells in the spinal cord gray matter, adjacent to the motor neurons – reaching more than 40% of the CD4⁺ T cell population 28 days post immunization. Importantly, immunized mSOD1 mice showed increased average survival of 15 days (from 144.2 ± 3.8 to 159.5 ± 2.2 days); at a stage at which 90% of the vaccinated mSOD1 mice were still alive, only 10% of the untreated mice survived.

Discussion and conclusion: Taken together, we propose that activation of the CP for leukocyte trafficking is beneficial in ALS, yet do not spontaneously occur along disease course. It is therefore possible that immunization with a self-peptide, which initiates an immune cascade leading to recruitment of inflammation-resolving immune cells to disease sites, may lead to the development of hitherto unexplored therapeutic vaccines for fighting ALS, and perhaps other neurodegenerative diseases.

Acknowledgements:

This study was funded by the Israel Science Foundation (ISF)-Legacy-Bio-Med program, grant No. 710312020, and was also supported by the IsrA.L.S Research Fund.

References:

• Appel SH, Beers DR, Henkel JS. Trends Immunol 2010; 31:7–17.
   PubMed Web of Science ®Google Scholar
• Baruch K, Ron-Harel N, Gal H et al. Proc Natl Acad Sci USA 2013;110:2264–9.
   PubMed Web of Science ®Google Scholar
• Shechter R, Miller O, Yovel G et al. Immunity 2013; 38:555–69.
   PubMed Web of Science ®Google Scholar

C33 RECOMBINANT HUMAN-DERIVED MONOCLONAL ANTIBODIES TARGETING MISFOLDED SOD1 AS NOVEL THERAPEUTICS FOR THE TREATMENT OF ALS

Maier M¹

Welt T²

Montrasio F¹

Wirth F²

Brännström T³

Andersen PM³

Weber M⁴

Nitsch RM²

Grimm J¹

^biNeurimmune AG, Schlieren, Switzerland

^bjUniversity of Zurich, Zurich, Switzerland

^bkUmeå University, Umeå, Sweden

^blNeuromuscular Diseases Unit/ALS Clinic, Kantonsspital St. Gallen, St. Gallen, Switzerland

Email address for correspondence: [email protected]

Keywords: SOD1, human antibodies, immunotherapy

Background: Misfolded superoxide dismutase 1 (SOD1) accumulates in both familial and sporadic ALS, suggesting a generalized role as a drug target for the treatment of ALS. Conformational neo-epitopes within misfolded SOD1 can trigger immune responses in human subjects leading to B-cell memory for misfolded SOD1.

Objectives: To identify, clone and recombinantly express human monoclonal antibodies selectively targeting misfolded SOD1 with high affinity, and to evaluate their pharmacological efficacy in transgenic mouse models of ALS.

Methods: Immune repertoires from cohorts of healthy elderly human donors without clinical signs of neurodegenerative disease were screened for memory B cells against misfolded SOD1. High-affinity and selective antibodies were cloned, recombinantly expressed in CHO cells and characterized both in vitro and on post-mortem spinal cord sections from ALS patients. Pharmacological efficacy in vivo was tested in two independent transgenic mouse lines overexpressing human mutant SOD1.

Results: We cloned, expressed and characterized high- affinity human antibodies that selectively target misfolded SOD1 while not binding to physiological SOD1 dimers. In spinal cord tissue sections obtained at autopsy from patients with ALS or human control subjects, the lead antibody selectively bound to misfolded SOD1 in motor neurons of ALS patients. In transgenic mice, chronic antibody treatment significantly reduced SOD1 pathology and rescued spinal cord motor neurons resulting in significantly reduced muscle atrophy and better motor functions. The lead antibody increased survival times by up to 2 months, was efficacious in two independent SOD1 transgenic mouse lines, and was effective through both peripheral and intracerebroventricular administration routes.

Discussion and conclusion: High-affinity, recombinant human-derived antibodies against conformational epitopes in misfolded SOD1 selectively bind pathological SOD1 deposits in spinal cord sections obtained from ALS patients while sparing physiological SOD1 dimers in human control subjects. At therapeutic doses, these antibodies are effective in independent transgenic mouse models of ALS. The lead antibody is a promising candidate for the development of a safe and efficacious immunotherapy for ALS.

C34 NANOBODY AGAINST SOD1 REDUCES IN VITRO AGGREGATION, RESCUES SOD1-INDUCED AXONOPATHY AND EXTENDS SURVIVAL IN ALS MODELS

Hernandez S^1,2

Laird A^1,2

Rudyak S¹

Boeynaems S^1,2

Van Damme P^1,2

Van Den Bosch L^1,2

Rousseau F¹

Schymkowitz J¹

Robberecht W^1,2

^bmKU Leuven, Leuven, Belgium

^bnLIND, Leuven research Institute for Neuroscience & Disease, Leuven, Belgium

Email address for correspondence: [email protected]

Keywords: ALS, SOD1, nanobody

Background: Many neurodegenerative diseases such as Alzheimer’s and Parkinson’s disease, Huntington’s disease and amyotrophic lateral sclerosis (ALS) are caused by a mutant protein which, through the presence of this mutation, gains a function that is toxic for the cell. Reducing the levels of the pathogenic mutant protein is an obvious strategy to treat patients suffering from these disorders. ALS is an adult-onset fatal neurodegenerative disease that is familial in about 10% of patients. In about 20% of patients with familial ALS (FALS) the disease is caused by gain-of-toxic-function mutations in the gene encoding Cu/Zn-superoxide dismutase (SOD1). Reducing the levels of the pathogenic SOD1 is an interesting strategy to treat patients with mutant SOD1- associated familial ALS, and maybe also of patients with sporadic ALS.

It has been described that camelids and some sharks produce functional antibodies devoid of light chains, where the single N-terminal domain was fully capable of antigen binding. This so-called VHH domain can be cloned and the product, called nanobody, shows high-target specificity and affinity, and low inherent toxicity. Furthermore nanobodies are highly soluble, extremely stable, have refolding capacity, can recognize hidden antigenic sites, can be administered by means other than injection and are easy to manufacture. Nanobodies may have applications as diagnostic markers and as therapeutics in many fields of medicine. Furthermore, nanobodies have very limited immunogenicity due to high sequence similarities with human VH family III. Finally, and particularly important for their potential use in neurodegenerative disorders, some nanobodies have been shown to cross the blood-brain barrier efficiently.

Objectives: To reduce the toxic effect of mutant SOD1 in the motor neurons, we have raised anti-SOD1 heavy chain antibodies in dromedaries and alpacas, and cloned its N-terminal antigen binding VHH region, coding for an anti-SOD1 nanobody (SOD1 nanobody). We have investigated the therapeutic potential of this nanobody, by studying its effect on mutant SOD1-induced toxicity in vitro and in vivo.

Results and discussion: We have used two different approaches: transfection or injection of constructs that express the SOD1 nanobody and also the administration of the SOD1 nanobody itself. In vitro, we have demonstrated that SOD1 nanobody has high affinity for SOD1 and blocks its fibril formation. It also prevents mutant SOD1 aggregation when different cell lines are transfected with mutant SOD1. We have found this effect when the cells are co-transfected with mutant SOD1 and the SOD1 nanobody and when the cells are treated with the SOD1 nanobody. In vivo, its injection and its administration rescues mutant SOD1-induced axonopathy in a zebrafish model for ALS. It also delays onset and extends lifespan of transgenic mice that overexpress SOD1^G93A, murine model for ALS, when they are treated regulary with the SOD1 nanobody intracerebroventriculary. This positive effect is also significant once the neurodegenerative symptoms have started. We have also demonstrated that it binds to SOD1 in human ALS spinal cord. Our data demonstrate the potential use of the SOD1 nanobody as a possible therapy for ALS.

C35 AAV9-MEDIATED SOD1 DOWNREGULATION AS A FUTURE THERAPY FOR AMYOTROPHIC LATERAL SCLEROSIS

Likhite S^1,2

Foust K²

Salazar D³

Ferraiuolo L¹

Ditsworth D³

Schmelzer L¹

Braun L¹

Cleveland DW³

Kaspar B^1,2

^boResearch Institute at Nationwide Children’s hospital, Columbus, OH, USA

^bpThe Ohio State University, Columbus, OH, USA

^bqLudwig Institute for Cancer Research, La Jolla, CA, USA

Email address for correspondence: [email protected]

Keywords: gene therapy, non-human primates, motor neurons

Background: In vivo transgenic removal of mutant SOD1 from motor neurons as well as glial cells can significantly delay the disease progression and improve survival in ALS mouse models. Moreover, misfolding of SOD1 in sporadic ALS patients has been reported, thus opening the potential of targeting SOD1 for all ALS patients. Therapeutic strategies to reduce SOD1, including antisense oligonucleotides and viral-delivered RNA interference (RNAi) have been attempted with varying success, and to date, only antisense oligonucleotides have been tested clinically.

Objectives: Here we investigate the feasibility and efficacy of post-natal downregulation of SOD1 using a novel approach of AAV9-mediated shRNA delivery in the SOD1G93A mice. The ultimate goal of this study is to determine the safety of this approach and to define guidelines for a clinical trial by testing its efficacy in non-human primates (NHP).

Methods: SOD1G93A mice, overexpressing human mutant SOD1, were intravenously injected at P1, P21 or P85 with 3 × 10¹² viral particles of scAAV9-SOD1 shRNA or scAAV9-GFP virus. 3 NHP were injected intrathecally at the sacral level with 1 × 10¹³ vg/Kg scAAV9-GFP-shSOD1 to determine biodistribution and SOD1 suppression level.

Results: P1 injected mice showed widespread spinal cord transduction of scAAV9-SOD1 shRNA virus targeting both neuronal and non-neuronal cells, while P21- and P85-injected mice showed predominant astrocytic transduction. Importantly, the transduced cells persisted throughout the life span of the treated animals and reduced levels of mutant SOD1 protein were detected. P1- and P21-treated mice showed improved performance on behavioral tasks. P1-injected mice showed significant delay in the disease onset, whereas P21- and P85-injected mice showed delay in disease progression. Importantly, all three treatments significantly improved the median survival of SOD1G93A mice between 30 and 50 days. Excitingly, intrathecal administration of scAAV9-SOD1 shRNA in NHP also resulted in efficient spinal cord transduction and significant SOD1 reduction. We also determined the safety of scAAV9-SOD1 shRNA administration by injecting wild-type mice at P1 or P21 and monitoring them up to 6 months of age. Injected wild-type mice showed no difference in their motor performance compared to controls. Hematology, serum profiles and histopathology studies showed no significant alterations.

Conclusions: Taken together, we report here one of the longest survival extensions in two regularly used als mouse models using scaav9 sod1 shrna. Thus, the success of post-natal suppression of SOD1 toxicity by AAV9-mediated shrna delivery in ALS mouse models provides a sound basis for its clinical translation. Also, the efficient knockdown of SOD1 in non-human primates and long-term safety assessment in wild-type mice further strengthen the suitability of this approach, thus setting the stage for human clinical trials.

C36 ENDEMIC ALS: IS THERE ANYTHING WE CAN LEARN FROM CLUSTERS?

Beghi E

Istituto di Ricerche Farmacologiche “Mario Negri”, Milano, Italy

Email address for correspondence: [email protected]

Keywords: genetics, epidemiology, risk factors

Clusters are groupings of health-related events having temporal and/or geographical correlation. Spatial and temporal clusters in amyotrophic lateral sclerosis (ALS) have been reported, implying a genetic predisposition of the affected population and/or an environmental influence. A classical example of clustering in ALS is represented by the Western Pacific form (the ALS-Parkinson-Dementia-Complex, PDC) observed in the island of Guam, the Kii Peninsula of Japan, and in Western New Guinea. In these areas, the incidence of the disease has been found, in the fifties, about 50- to 100-fold higher than elsewhere in the world. Since then, the decreased incidence and the changed phenotype suggested that the disease could be caused by an environmental factor. Beta-N-methyamino-L-alanine (BMAA), a cyanobacterial-derived neurotoxin found in cycad flour and shown in the brains of ALS patients, has been implicated as a possible cause. At the end of the ALS-PDC outbreak, several other environmental factors have been postulated to trigger ALS, mostly in the context of occupational exposures. These include pesticides, metals, solvents, electromagnetic fields, head trauma, physical exercise, and smoking. However, none of these factors has been unequivocally found to increase the risk of ALS to a great extent. In recent years, a higher than expected risk of ALS was reported among veterans of the First Gulf War. This risk was apparently limited to the decade following the war, which again suggests an environmental exposure. These findings are consistent with other reports and support a possible correlation between ALS and military service. Exposure to toxic agents, traumatic brain injury (TBI), manual work, and strenuous physical activity have been all implicated. Another cohort recently found at higher risk of ALS is represented by professional soccer players. This cohort has some exposures in common with the veterans (TBI, physical exercise) and, interestingly, is likewise characterized by a frequent onset of the disease under the age of 45 years. This latter finding can be explained by a possible interaction between genetic and environmental factors.

Despite being useful instruments for the investigation of risk factors for ALS, at present studies on the temporal or geographical variation of the frequency of the disease have failed to provide evidence of a strong causal relationship between ALS and several environmental factors. Even BMAA has not been definitely found to represent a significant risk factor, as cyanobacteria are ubiquitous and increasingly prevalent, and good animal models are still lacking. Nevertheless, in light of the recent advances in the genetics of ALS, the study of clusters should continue and the role of environmental factors should be assessed in combination with the investigation of the genetic background of the affected individuals.

C37 FEASIBILITY ASSESSMENT OF AN EPIDEMIOLOGIC STUDY OF ELECTROCONVULSIVE THERAPY AND MOTOR NEURON DISEASE

Mezei G¹

Vergara X¹

Feychting M²

^brElectric Power Research Institute, Palo Alto, California, USA

^bsKarolinska Institute, Stockholm, Sweden

Email address for correspondence: [email protected]

Keywords: magnetic fields, electric shocks, electroconvulsive therapy

Background: Electrical environments, mostly in occupational settings, have been linked to development of neurodegenerative diseases including motor neuron disease (MND). In occupational epidemiologic studies, the association for MND appears to be stronger with occupational titles designating electrical occupations than with estimated levels of magnetic field (MF) exposure. It has been suggested that exposure to electric shocks (ES) in these occupations rather than the MF exposure may be the causal factor in these associations. It is difficult to separate occupational exposure scenarios for ES and MF as they frequently occur together. Electroconvulsive therapy (ECT), a relatively commonly used treatment for some psychiatric conditions, presents a scenario where ES occurs without significant exposure to MF.

Objectives: To evaluate the feasibility of a large-scale international epidemiologic study of ECT and MND.

Methods: We sought information on treated person rate (TPR) for ECT and frequency of ECT utilization among psychiatric patients in several counties. We also estimated MND incidence and estimated the anticipated number of MND patients among patients receiving ECT.

Results: In Western countries, the TPR tends to range between roughly 2 and 5 per 10,000 residents per year. It has been estimated that ECT might be used in about 5% of hospitalizations for psychiatric diseases in Denmark; the same proportion is estimated at 2% in Sweden. It was also estimated that between 1987 and 2011, approximately 190 patients with depression or schizophrenia died from MND in Sweden. Under the assumption of no association, five of these patients would be expected to have received ECT. In the US Veterans Health Administration, among almost 190,000 patients with major depression, 307 (0.16%) received ECT in a 5-year period.

Discussion and conclusion: ECT represents a unique exposure scenario where the potential effect of ES on MND risk may be evaluated uncoupled from the potential effects of MF exposure. To further assess the feasibility of an epidemiologic study of this kind, the availability of data on ECT will be required in several countries, along with the availability of additional medical records information to evaluate if early symptoms of MND may potentially confound the association. Due to rarity of both the exposure (ECT) and the disease (MND), an international collaboration will be required to pool data from several countries. An epidemiologic study investigating the ECT–MND association will contribute to our understanding whether the consistently reported association between electrical occupations may be potentially explained by the etiologic role of ES in MND development.

References:

• Vergara et al. J Occup Environ Med 2013;55:135.
   PubMed Web of Science ®Google Scholar
• Leiknes et al. Brain Behav 2012;2:283.
   PubMed Web of Science ®Google Scholar
• Pfeiffer et al. J Affect Disord 2011;130:21.
   Google Scholar

C38 N-3 AND N-6 POLYUNSATURATED FATTY ACID INTAKE AND RISK OF AMYOTROPHIC LATERAL SCLEROSIS: POOLED RESULTS FROM FIVE COHORT STUDIES

Fitzgerald K¹

O’Reilly E^1,2

Falcone G^1,3

McCullough M⁴

Le March, L⁵

Ascherio A^1,2

^btHarvard School of Public Health, Boston, MA, USA

^buChanning Division of Network Medicine, Harvard Medical School, Boston, MA, USA

^bvDepartment of Neurology, Massachusetts General Hospital, Boston, MA, USA

^bwEpidemiology Research Program, American Cancer Society, Atlanta, GA, USA

^bxEpidemiology Program, Cancer Center, Honolulu, HI, USA

Email address for correspondence: [email protected]

Keywords: diet, polyunsaturated fatty acids, epidemiology

Objective: Prior research has suggested the possible role of oxidative stress and inflammation in the pathogenesis of amyotrophic lateral sclerosis (ALS). Dietary n-3 and n-6 polyunsaturated fatty acids (PUFA) may affect oxidative stress and inflammation and could thus contribute to determine ALS risk. Therefore, using a prospective design, we examined whether the fatty acid composition of the diet is related to ALS risk.

Methods: Risk of ALS associated with total dietary fat and fat subtypes as well as individual PUFAs was investigated in five prospective cohorts: the Cancer Prevention Study II Nutrition Cohort (CPS-II), the Multiethnic Cohort (MEC), the National Institutes of Health-Association of American Retired Persons Diet and Health Study (NIH-AARP), the Health Professionals Follow-up Study (HPFS), and the Nurses Health Study (NHS). Diet was assessed via food frequency questionnaires. Incident cases of ALS (NHS and HPFS) and ALS deaths (all cohorts) were identified by biennial follow-up questionnaires or from the National Death Index and confirmed by review of the medical records or death certificates. A total of 994 ALS cases occurred among 995,755 participants (476,980 women and 522,775 men) during a median follow-up of 12.0 years. We applied Cox proportional hazards regression to calculate cohort-specific multivariable-adjusted risk ratios (RR) across quintiles of intake of the dietary variables, and pooled the cohort- specific results using random-effects methods.

Results: In analyses adjusted for age, gender, smoking, and body mass index, a greater n-3 PUFA intake was associated with a reduced risk of ALS (RR for the highest versus the lowest quintile: 0.63; 95% CI: 0.50–0.79; P trend < 0.001). An inverse association was observed for both α-linolenic acid (ALA; 18:3 n-3) and long-chain marine n-3 fatty acids – for ALA, the RR for a 1 g/day increment (corresponding to about 1 tablespoon of canola oil per day) was 0.65 (95% CI: 0.48–0.88; P = 0.005), and for marine n-3 the RR for a 200 mg/day increment (about two servings of fatty fish per week) was 0.86 (95% CI: 0.75–0.99; P = 0.03). Total fat and other types of dietary fats, including n-6 PUFA, were not associated with ALS risk.

Discussion: The results of this large longitudinal study suggest that consumption of foods high in n-3 PUFAs may reduce ALS risk.

C39 INTERACTION BETWEEN HFE POLYMORPHISMS AND CUMULATIVE LEAD EXPOSURE ON THE RISK OF AMYOTROPHIC LATERAL SCLEROSIS

Eum K¹

Seals R¹

Grespin M¹

Umbach D²

Sandler D²

Hu H³

Kamel F²

Weisskopf M¹

^byHarvard School of Public Health, Boston, MA, USA

^bzNIEHS, Raleigh, NC, USA

^caDalla Lana School of Public Health, University of Toronto, Toronto, Canada

Email address for correspondence: [email protected]

Keywords: oxidative stress, iron metabolism, hemochromatosis

Background: We previously found an association between cumulative lead (Pb) exposure and risk of amyotrophic lateral sclerosis (ALS). Polymorphisms in the hemochromatosis (HFE) gene-involved in regulating iron uptake-have previously been associated with ALS. Oxidative stress may contribute to ALS pathophysiology, and excess iron can amplify toxic effects of Pb related to oxidative stress. No study has examined the possible modifying role of HFE on the association of cumulative Pb exposure with ALS risk.

Objectives: To determine whether HFE polymorphisms alter the association of cumulative Pb exposure with ALS risk.

Methods: We measured bone Pb – a biomarker of cumulative Pb exposure-using K-shell-X-Ray Fluorescence in 100 neurologist-confirmed ALS cases and 192 controls from New England. We genotyped H63D and C282Y HFE polymorphisms; participants were considered variant carriers or not for each. We estimated odds ratios (ORs) and 95% confidence intervals (CIs) for ALS using logistic regression adjusted for age, sex, education, and smoking. To assess effect modification by HFE polymorphism, we included in one model terms for Pb (continuous), H63D and C282Y genotypes (separately), and separate cross-product terms between Pb and the polymorphisms.

Results: We found an elevated OR for ALS among H63D variant carriers (OR: 5.30; 95% CI: 1.81–15.52), but not C282Y carriers, as seen in other studies. The cross-products terms indicated that the OR for association of patella lead with ALS among C282Y variant carriers was 15 times larger (15.48; 95% CI: 1.54–155.9) than that among those without C282Y variants, while the OR among H63D variant carriers was about half as large (0.43; 95% CI: 0.20–0.93) as that among those without H63D variants.

Discussion and conclusion: Our results suggest that the H63D and C282Y HFE genotypes modify the association of Pb with ALS risk, but in opposite directions. Thus effects on iron are unlikely to explain the findings since both H63D and C282Y increase labile iron.

C40 ALS MULTICENTER COHORT STUDY OF OXIDATIVE STRESS (ALS COSMOS): THE STUDY METHODOLOGY, RECRUITMENT, AND BASELINE DEMOGRAPHICS AND DISEASE CHARACTERISTICS

Mitsumoto H¹

Factor-Litvak P²

Andrews H³

Goetz RR.⁴

Andrews L²

Rabkin J⁴

McElhiney M⁴

Nieves J^2,5

Santella RM⁶

Nagy PL⁷

Murphy J⁸

Hupf J¹

Singleton J¹

Merle D³

Kilty M²

Heitzman D⁹

Bedlack R¹⁰

Katz J¹¹

Forshew DA¹¹

Barohn RJ¹²

McVey A¹²

Dimachkie M¹²

Sorenson EJ¹³

Oskarsson B¹⁴

Fernandes Filho JAM¹⁵

Kasarski E¹⁶

Lomen-Hoerth C⁷

Mozaffar T¹⁷

Rollins YD¹⁸

Nations SP¹⁹

Swenson A J²⁰

Shefner JM²¹

Andrews JA²²

Koczon-Jaremko A²²

Als Cosmos Study Group¹

^cbEleanor and Lou Gehrig MDA/ALS Research Center, Department of Neurology

^ccDepartment of Epidemiology, Mailman School of Public Health

^cdMailman School of Public Health Biostatistics Department Columbia University, New York, NY, USA

^ceDepartment of Psychiatry, New York State Psychiatric Institute & Columbia University, New York, NY, USA

^cfDepartment of Nutrition, Helen Hays Hospital, NY, West Haverstraw, NY, USA

^cgDepartment of Environmental Health, Mailman School of Public Health

^chDepartment of Cell Biology and Pathology; Columbia University, New York, NY, USA

^ciDepartment of Neurology, University of California, San Francisco, San Francisco, CA, USA

^cjTexas Neurology, PA, Dallas, TX, USA

^ckDepartment of Neurology, Duke University, Durham, NC, USA

^clForbes Norris ALS Center, California Pacific Medical Center, San Francisco, CA, USA

^cmDepartment of Neurology, University of Kansas, Kansas City, KS, USA

^cnMayo Clinic, Rochester, NY, USA

^coDepartment of Neurology, University of California, Davis, CA, USA

^cpDepartment of Neurological Sciences, University of Nebraska Medical Center, Omaha, NE, USA

^cqDepartment of Neurology, University of Kentucky, Lexington, KY, USA

^crDepartment of Neurology, University of California, Irvine, CA, USA

^csDepartment of Neurology, University of Colorado, Aurora, CO, USA

^ctUniversity of Texas - Southwestern, Dallas, TX, USA

^cuDepartment of Neurology, University of Iowa, Iowa City, IA, USA

^cvDepartment of Neurology, SUNY - Upstate Medical University, and, Syracuse, NY, USA

^cwDepartment of Neurology, Hospital for Special Care, New Britain, CT, USA

Email address for correspondence: [email protected]

Keywords: ALS disease progression, epidemiology, oxidative stress

Background: Although the majority of patients die within 3–4 years, the rates of disease progression and survival time are highly variable in ALS.

Objective: In a multicenter study design of newly diagnosed ALS patients without a family history, we prospectively investigated whether oxidative stress (OS) is associated with disease progression.

Methods: All de-identified patients seen at each study site were evaluated for study eligibility and enrollment. For enrolled patients, detailed clinical assessments and cognitive screening tests for frontotemporal dementia were performed. Extensive structured telephone interviews ascertained clinical, environmental, lifestyle, dietary, and psychological risk factors. Data collection was performed at baseline and at 3- to 6-month intervals using modified questionnaires for 30 months or until death or study drop out. We developed a biorepository including urine, DNA, plasma and skin biopsies for initial analyses of OS and lipid biomarkers and genetic screening of C9ORF72.

Results: A total of 356 patients were enrolled during a 36-month-enrollment period by 16 sites. The most common reasons that eligible patients were not enrolled into the study were (1) the study was not discussed, (2) patients refused to participate, and (3) patients were overwhelmed by the ALS diagnosis. Differences in insurance carriers/coverage were noted between study-eligible/non-enrolled and enrolled patients. The latter were more likely to have private insurance and less likely to be on Medicare (P = 0.004). Otherwise, there were no differences in the demographic and diagnostic features between these two groups. The enrolled patients were arbitrarily further grouped into the single high enrollment site (n = 139), median enrollment sites (> 0.6 patients/month; n = 160), and lower enrollment sites (< 0.5 patients/month; n = 57) to investigate differences in demographics and disease characteristics based on enrollment. Among these three groups, the high enrollment site and low enrollment sites had more neurologist-referred patients (p = 0.007). For diagnostic categories, the high enrollment site had more possible ALS and PMA diagnoses, but fewer definite ALS diagnoses compared to other sites (P < 0.001). The remaining demographic and disease characteristics did not differ by enrollment group.

Discussion: Although some minor differences were found between study-eligible/non-enrolled and enrolled patients and among different study sites, our study population is likely to represent the patient population seen at all study sites.

Conclusion: To our knowledge, this is the first prospective, interdisciplinary, in-depth multicenter epidemiological investigation of OS related to ALS progression. At the time of the presentation, we will show the results of the analysis of the baseline questionnaires and OS/lipid biomarkers. We believe data generated from our study will expand the understanding of the mechanisms involved in ALS prognosis (Grant support by NIEHS, R01ES016348, and MDA).

C41 A MITOCHONDRIAL ETIOLOGY OF METABOLIC AND DEGENERATIVE DISEASES

Wallace D

Center of Mitochondrial and Epigenomic Medicine, Philadelphia, USA

Email address for correspondence: [email protected]

Keywords: mtDNA, adaptation, neurodegenerative disease

For half a millennium Western medicine has focused on anatomy and for the past century on nuclear DNA (nDNA), Mendelian, genetics. While these concepts have permitted many biomedical advances, they have proven insufficient for understanding the common “complex” diseases. In addition to anatomy, life requires energy and about 90% of energy comes from the mitochondrion. The mitochondrial genome consists thousands of copies of the maternally inherited mitochondrial DNA (mtDNA) plus between 1000 and 2000 nDNA genes. The mtDNA has a very high mutation rate, but the most deleterious mutations are removed by an ovarian prefertilization selection system. Hence, functional mtDNA variants are constantly being introduced into the human population, and the sequential accumulation of these mutations has permitted people to adapt to different regional environments as they migrated out-of-Africa to populate Eurasia and the Americas. As human migrations progressed, adaptive mtDNA variants became enriched in local environments giving rise regional groups of related haplotypes, or haplogroups, but these same adaptive variants can be maladaptive in alternative environments. A tRNA^Gln nt A4336G variant arose in between 8500 and 14000 years in Europe and persists today in 0.4% of European mtDNAs, yet this variant accounts for 3% late-onset Alzheimer's Disease (AD), 5% Parkinson Disease (PD), and 7% AD+ PD. A ND1 T3397C missense mutation (M31V) has arisen on the 4336 lineage but also independently and in both cases was found in AD+ PD patients. A missense mutation in the codon adjacent to the T3397C mutation, ND1 T3394C (Y30H), increases the penetrance of Leber Hereditary Optic Atrophy (LHON) mutations on macrohaplogroup N mtDNAs, but is adaptive for high altitude in Tibetans on a macrohaplogroup M. Haplogroup H mtDNAs, when combined with a homozygous frame shift mutation in the nDNA ADP/ATP translocase isoform 1 gene, are associated with mild cardiomyopathy, but when the ANT1 mutation is paired with haplogroup U mtDNAs the heart disease is severe and life-threatening. The phenotypic effects of mtDNA haplogroups can be augmented by nDNA variation and by deleterious germline or somatic mtDNA mutations. Confirmation that mtDNA variation contributes to common disease risk has been obtained by introducing mtDNA mutations into the mouse using mouse female embryonic stem cells (mfESCs). Mice harboring a COI T6859C V421A missense mutation develop a myopathy and cardiomyopathy while mice harboring a ND6 G13997A P25A missense mutation develop a LHON-like optic neuropathy. Simply mixing two normal mtDNAs from 129 and NZB mice together is sufficient to cause marked neuropsychiatric symptoms including reduced activity, heightened excitability, and severe learning defects. Hence, the primary genetic factors for predisposition to complex diseases may more often be regional mitochondrial variants rather than global anatomical gene variants.

C42 MUTATED SOD1 CAUSES REGION SPECIFIC DIFFERENCES IN CA2+ CYT DEPENDENT PROPERTIES OF MITOCHONDRIA FROM CNS OF SOD-1G93A MICE AND CA2+ DYSHOMEOSTASIS IN FIBROBLASTS OF FALS PATIENTS

Gellerich FN^1,2

Gizatullina Z¹

Debska-Vielhaber G¹

Zuschratter W²

Gainutdinov T¹

Kunz WS.³

Schwalenstöcker B⁴

Ludolph AC⁴

Vielhaber S¹

^cxUniversity of Magdeburg, Department of Neurology, Magdeburg, Sachsen/Anhalt, Germany

^cyLeibniz Institute for Neurology, Magdeburg, Sachsen/Anhalt, Germany

^czDepartment of Epileptology, University of Bonn, Bonn, Germany

^daDepartment of Neurology, University Ulm, Ulm, Germany

Email address for correspondence: [email protected]

Keywords: mitochondrial function, Ca²⁺-homeostasis, fibroblasts

Background: The reasons why in experimental ALS models mitochondrial dysfunction and impaired Ca²⁺ homeostasis occur are not clear. It is also not understood why specific neuronal populations are selectively vulnerable in ALS.

Objectives: The aim of our project is to study the interrelation between mitochondrial dysfunction and the cytosolic Ca²⁺ concentration (Ca²⁺_cyt) in mitochondria isolated from different CNS regions as well as in fibroblasts from fALS patients with SOD1 mutations.

Methods: Mitochondria isolated from spinal cord (SCM), brain stem (SM), and cerebellum (CM) of SOD1^G93A-mice were respirometrically and fluorimetrically investigated at varied Ca²⁺_cyt in comparison to wild type. Mitochondrial function was also investigated in permeabilized and intact fibroblasts (FM) of fALS patients compared to controls.

Results: CNS mitochondria of SOD1^G93A-mice exhibited mitochondrial impairments (diminished state 3_{glutamate/malate/pyruvate} but unchanged state 3_succinate, increased flux control of complex I, decreased Ca²⁺ uptake rates and Ca²⁺ uptake thresholds, increased ROS formation). Also the stability of respirometric properties against Ca²⁺ stress (addition of 10 μM Ca²⁺_cyt) were significantly reduced compared to wild type. Largest and most significant changes were observed in SCM > SM > CM of 130 days old SOD1^G93A mice. Similar changes were detectable also at 90 days old mice, but in a lesser extent. Since the Ca²⁺_cyt dependent stimulation of state 3_{glutamate/malate} was found to be a typical property of neuronal mitochondria the extent of this parameter depends on the relative content of neuronal mitochondria (Gellerich unpublished). Whereas no difference was observed between SOD1^G93A and wild-type CM the Ca^{2 +} _cyt stimulation was diminished by 23% in SOD1^G93A SCM.

Also in FM of fALS^SOD1 patients functional impairments were detected. Whereas the state 3_{glutamate/malate/pyruvate} was not significantly altered the flux control coefficients of complex I increased by + 26%, (p < 0.05) and the activity ratio of complex I/complex IV tended to be decreased by 35 %. Morphologically mitochondria appeared to be swollen with fragmented or missing christae. Stationary Ca²⁺_cyt concentrations were decreased in intact fibroblasts of fALS^SOD1 patients compared to control cells −37% at endogenous conditions and −14 % in 2mM Ca²⁺ containing medium. Treatment of ALS fibroblasts with 5 μM CoQ and 300 μM Trolox normalized in the most cases the elevated flux control coefficient for complex I.

Discussion and conclusion: We have shown in SOD1^G93A mice and in fibroblasts of fALS patients that mutated SOD1 causes impairments of OXPHOS and mitochondrial Ca²⁺ metabolismus with consequences for an impaired Ca²⁺-homeostasis. Our data support the view that ALS has also a generalized component potentially allowing the use of cultivated fibroblasts from ALS patients as a biomarker and for therapeutic studies.

C43 MITOCHONDRIAL METABOLIC MARKERS IN ALS FIBROBLASTS

Kirk K¹

D’Aurelio M¹

Tadesse S²

Hupf J²

Hirano M²

Mitsumoto H²

Manfredi G¹

Study Group Als Cosmos²

^dbBrain and Mind Research Institute, Weill Medical College of Cornell University, New York, NY, USA

^dcDepartment of Neurology, Columbia University, New York, NY, USA

Email address for correspondence: [email protected]

Keywords: skin fibroblasts, mitochondria, oxidative stress

Objective: We performed metabolic screening by medium throughput assays on fibroblasts from a cohort of ALS patients without family history and healthy controls. Our goal was to identify metabolic signatures that correlate with disease and to determine if metabolic abnormalities are linked to the clinical phenotype, establishing proof of concept that such alterations could represent viable biomarkers.

Background and significance: A large body of evidence links energy metabolism abnormalities to both sporadic and familial forms of ALS. We predicted that genetic and environmental factors could affect energy metabolism and predispose individuals to develop ALS and modulate the disease course. Different metabolic backgrounds could also affect the response to therapeutics. Thus, the failure to find effective compounds to treat the metabolic component of ALS may in part be attributable to the lack of understanding of energy metabolism as a disease modifier.

Methods: We surveyed a series of randomly selected 50 primary skin fibroblast lines from more than 170 skin biopsies of well-characterized ALS patients in the prospective ALS COSMOS study and 50 gender- and age-matched healthy controls. All patients were diagnosed and followed at multicenter study sites and had precise clinical and electrophysiological records. ALS and control skin fibroblasts were collected by skin biopsy, expanded, and frozen in liquid nitrogen after the first passage. For all assays, fibroblasts were used from early (2–3) passages. Metabolic assays were performed on fibroblasts, in triplicates. Mitochondrial membrane potential in medium containing glucose, pyruvate, and glutamine was measured with the potentiometric dye TMRM in an optimized plate reader assay. Background fluorescence was subtracted after addition of the uncoupler FCCP. Mitochondrial mass was estimated on the same cells with the membrane potential-independent dye mitoTracker green. Total cellular content of GSH was measured in a 96-well assay by the conversion of monochlorobimane (MCB). All results were normalized by cell protein content.

Results: We found a significant increase in the average membrane potential of mitochondria from ALS fibroblasts. In the same cells, we found a decrease in the average mitochondrial mass. Furthermore, ALS fibroblasts had a significantly higher average content of GSH. We are currently in the process of analyzing the data to determine if correlations between the bioenergetic, mitochondrial, and redox state parameters exist with specific clinical subgroups of patients.

Conclusions: Our results disclosed changes in bioenergetic parameters in ALS fibroblasts that will be further investigated to assess the underlying mechanisms. Slow ADP phosphorylation could result in a build up of membrane potential and a downregulation of mitochondrial biogenesis. These changes may underscore a global shift of metabolism in ALS. The ongoing correlation studies will determine if there are metabolic signatures that distinguish different clinical disease parameters. (Grant support: NIH R01ES-16348 and MDA)

C44 ALTERED GROWTH HORMONE/INSULIN BALANCE IN HSOD1G93A MICE: IMPLICATIONS FOR INSULIN RESISTANCE IN AMYOTROPHIC LATERAL SCLEROSIS

Steyn F¹

Lee K¹

Veldhuis J²

McCombe P^3,4

Chen C¹

Ngo S^1,3

^ddThe University of Queensland, Brisbane, Queensland, Australia

^deMayo Clinic, Rochester, Minnesota, USA

^dfThe University of Queensland Centre for Clinical Research, Brisbane, Queensland, Australia

^dgRoyal Brisbane & Women’s Hospital, Brisbane, Queensland, Australia

Email address for correspondence: [email protected]

Keywords: metabolism, growth hormone, insulin resistance

Background: A significant metabolic component underlies amyotrophic lateral sclerosis ALS pathogenesis; GH deficiency (1), insulin resistance (2), glucose intolerance (3), and mitochondrial dysfunction (4) are observed in ALS. While defective metabolic balance negatively impacts ALS survival, the mechanism by which this contributes to ALS pathogenesis is unknown. To investigate the causes and consequences of altered metabolic balance in ALS, we conducted metabolic assessment in hSOD1G93A mice throughout disease progression.

Objective: Assess the expression of endocrine factors involved in the regulation of metabolic homeostasis in hSOD1G93A mice.

Methods: Male wild-type and hSOD1G93A transgenic mice were studied at four stages of disease; pre-symptomatic, onset of symptoms, mid-stage of disease, and end-stage of disease. We assessed pulsatile growth hormone (GH) secretion using a tail-tip method, and an in-house GH ELISA. Properties associated with the pulsatile pattern of GH secretion were analyzed by deconvolution analysis (5). Insulin tolerance was assessed via insulin tolerance test. Circulating levels of insulin, muscle glycogen content, and levels of non-esterified free fatty acids (NEFAs) were determined using commercial assays. Expression of mitochondrial respiratory chain complexes was determined by western blot.

Results: We report a dynamic GH secretory profile during ALS disease progression; we observe a significant increase in GH secretion at the onset of disease symptoms, and a decrease in GH secretion by the end-stage of disease (n ≥ 8 wild-type, n ≥ 7 SOD1G93A). We observe insulin resistance in hSOD1G93A mice by the mid-stage of disease, a time after the significant elevation in GH secretion. This coincides with a decrease in the expression of circulating insulin, decreased muscle glycogen, and an increase in the expression of NEFAs, and mitochondrial complexes II and IV in skeletal muscle (n ≥ 4/group).

Discussion and conclusion: We report the first definitive account of dynamic GH secretion in a transgenic mouse model of ALS. The development of insulin resistance, and changes in expression of circulating insulin is reflective of altered GH/insulin balance, and suggests that hypersecretion of GH at the onset of disease symptoms in hSOD1G93A mice underlies the development of insulin resistance in ALS. Moreover, our results imply that GH-induced insulin resistance causes a reduction in mitochondrial function, resulting in a decreased ability for muscle to utilise glucose whilst also driving accumulation of fat into skeletal muscle. Thus, GH-induced insulin resistance may underlie mitochondrial dysfunction in ALS.

Acknowledgements:

This research was supported by the NHMRC, MNDRIA and UQ. S.T.N. is a recipient of a Bill Gole Fellowship from the MNDRIA.

References:

• Morselli LL et al. Clin Endocrinol 2006;65(3):385–8.
   Google Scholar
• Reyes ET et al. J Neurol Sci 1984;63(3):317–24.
   Google Scholar
• Pradat PF et al. A L S 2010;11(1–2):166–71.
   Google Scholar
• Afifi AK et al. Neurology 1966;16(5):475–81.
   Google Scholar
• Steyn FJ et al. Endocrinology 2012;153(8):3735–3746.
   Google Scholar

C45 MECHANISMS OF ER-GOLGI TRANSPORT INHIBITION IN AMYOTROPHIC LATERAL SCLEROSIS

Soo KY¹

Farg M¹

Halloran M¹

King A^2,3

Southam K^2,3

Parakh S¹

Sundaramoorthy V¹

Horne M^4,5

Atkin J^1,4

^dhDepartment of Biochemistry, La Trobe Institute of Molecular Science, Bundoora, Victoria, Australia

^diMenzies Research Institute, Hobart, Tasmania, Australia

^djUniversity of Hobart, Hobart, Tasmania, Australia

^dkDepartment of Florey Neuroscience, University of Melbourne, Parkville, Victoria, Australia

^dlDepartment of Neurology, St Vincent Institute, Fitzroy, Victoria, Australia

Email address for correspondence: [email protected]

Keywords: ER-Golgi transport, COPII, microtubules

Background: Previously we showed that ALS-mutant forms of SOD1, TDP43 and FUS inhibit ER-Golgi transport in cellular models of ALS, triggering ER stress and apoptosis by the accumulation of secretory proteins within the ER. Functional ER–Golgi transport requires the co-ordinated action of multiple proteins including COPII, which forms the coat protein surrounding vesicles as they bud from the ER. ER–Golgi transport also requires the presence of stabilized microtubules over which the vesicles are carried.

Objectives: The objectives of this study were to (i) determine the molecular and cellular mechanisms by which SOD1, TDP-43 and FUS impair secretory protein transport between the ER and Golgi apparatus, and (ii) to determine if inhibition of ER-Golgi transport can be detected in vivo.

Methods: VSVG^ts045 is a widely used marker to quantify ER–Golgi transport. Neuro2a cells were co-transfected with SOD1-EGFP, EGFP-TDP43 or HA-FUS and VSVG^ts045-mCherry, and budding of COPII vesicles from the ER was examined using an in vitro assay. The presence of stabilised microtubules was also examined by immunocytochemistry using antibodies against acetylated microtubule. Primary cortical neurons and motor neurons from SOD1 transgenic mice and non-transgenic mice at E13.5 were isolated and transfected with VSVG^ts045-mCherry.

Results: ER-derived vesicles obtained from cells expressing either ALS mutant TDP43 or FUS were depleted of COPII, suggesting that COPII vesicle formation is dysfunctional in these cells. In contrast, in cells expressing mutant SOD1, COPII vesicle formation appeared normal, but a high proportion of cells with destabilised microtubules was detected compared to controls. Inhibition of ER–Golgi transport was detected in both primary E13.5 motor neurons and cortical neurons obtained from SOD1^G93A mice, thus validating inhibition of ER–Golgi transport in vivo very early in pathology.

Discussion and conclusion: These findings show that dysfunction of ER–Golgi transport is a pathogenic mechanism shared by SOD1, TDP-43 and FUS. However, transport inhibition occurs by different mechanisms in cells expressing mutant SOD1 compared to mutant TDP43 or FUS. Furthermore, inhibition of ER–Golgi transport was present in embryonic motor neurons in SOD1 mice, implying that it occurs very early in pathogenesis.

Acknowledgements:

We thank Dr Jennifer Lippincott-Schwartz and Dr George Patterson for VSVG-ts045-mCherry construct, Dr Dorothee Dormann for HA-FUS constructs, and Professor Benjamin Wolozin for GFP-TDP43 constructs.

C46 MUTANT TDP-43 LEADS TO PATHOLOGICAL ACCUMULATION OF SMN AND ITS NUCLEAR COMPLEXES IN MOTOR NEURONS

Perera N¹

Sheean R¹

White A²

Crouch P²

Turner B¹

^dmFlorey Institute of Neuroscience and Mental Health, Melbourne, Australia

^dnUniversity of Melbourne, Melbourne, Australia

Email address for correspondence: [email protected]

Keywords: TDP-43, SMN, FTLD-U

Background: Abnormal copy number of SMN1 and SMN2, which encode survival motor neuron (SMN) protein, are associated with ALS and recent evidence suggests that SMN1 duplications confer increased risk of ALS. SMN normally interacts with TDP-43 at transcriptional and post- translational levels with TDP-43 regulating SMN2 splicing and assembly of SMN nuclear complexes or gems in motor neurons. However, the effect of TDP-43 pathology on SMN expression and gem formation in ALS is unclear. We therefore examined SMN expression and nuclear complexes in transgenic TDP-43^A315T mice.

Objectives: To investigate SMN expression and nuclear complex formation in motor neurons of transgenic TDP-43^A315T mice at different disease stages. To determine whether SMN is a modifier of mutant TDP-43 mediated ALS, we also crossed TDP-43^A315T and SMN transgenic mice and analysed double transgenic mice.

Methods: SMN expression was analysed in presymptomatic (P30, 60) and symptomatic (P90) TDP-43^A315T mice and age-matched wild-types using real-time PCR and immunoblotting. SMN nuclear complexes were also counted in spinal motor neurons of mice. TDP-43^A315T mice were intercrossed with mice overexpressing human SMN driven by the prion promoter (PrP-SMN). Double transgenic TDP-43^A315T; PrP-SMN mice and control genotypes TDP-43^A315T and PrP-SMN were examined for weight loss, motor function and survival. Spinal cords and brains were analysed by motor neuron counts, SMN expression and nuclear complexes.

Results: SMN protein expression was increased by 1.5-fold and 2-fold, respectively, in spinal cords of presymptomatic and symptomatic TDP-43^A315T mice. There was a corresponding accumulation of nuclear and cytoplasmic SMN complexes in spinal motor neurons of TDP-43^A315T mice. Transgenic overexpression of SMN accelerated weight loss, motor deficits and death in double transgenic TDP-43^A315T;PrP-SMN mice compared to TDP-43^A315T controls. This resulted from increased severity of motor neuron loss and excess nuclear and cytoplasmic SMN complexes in spinal cords.

Discussion and conclusion: Our results demonstrate that mutant TDP-43 triggers pathological accumulation of SMN and its nuclear complexes in spinal motor neurons, consistent with recent genetic association data showing that SMN1 duplications increase susceptibility to ALS. Furthermore, forced overexpression of SMN compounds the disease phenotype of TDP-43^A315T mice, suggesting that SMN is an enhancing modifier of TDP-43 pathology.

C47 CYCLOPHILIN A INTERACTION NETWORK PERTURBATION IS A CONVERGING PATHO-MECHANISM IN DIFFERENT FORMS OF AMYOTROPHIC LATERAL SCLEROSIS

Lauranzano E¹

Riccardo S¹

Pasetto L¹

Pozzi S¹

Massignan T¹

Nardo G²

Lunetta C³

Corbo M³

Mora G⁴

Bendotti C²

Bonetto V¹

^doDulbecco Telethon Institute

^dpDepartment of Neuroscience; IRCCS – Istituto di Ricerche Farmacologiche Mario Negri, Milano, Italy

^dqNEuroMuscular Omnicentre (NEMO), Niguarda Ca’ Granda Hospital, Milano, Italy

^drIRCCS - Fondazione Salvatore Maugeri, Milano, Italy

Email address for correspondence: [email protected]

Keywords: TDP-43, SOD1, hnRNP

Background: Familial and sporadic amyotrophic lateral sclerosis (ALS) phenotypes are clinically indistinguishable. Cyclophilin A (CypA) is a translational biomarker of ALS: it is a hallmark of disease in mutant SOD1 (mSOD1) animal models and in sporadic patients (1-2). Moreover, CypA is enriched in the spinal cord aggregates of mSOD1 mice and sporadic ALS patients (3). CypA is an ubiquitous protein with multiple functions relevant to the central nervous system, where it is abundantly expressed.

Objectives: The aim of our study is to provide insights into the molecular function of CypA in familial and sporadic forms of ALS. In particular, starting from apparently unrelated forms of disease we provided the molecular basis for a converging clinical phenotype and offer hints for therapeutic strategies.

Methods: The CypA interactors were identified by a co- immunoprecipitation approach combined with 2D gel electrophoresis and MALDI TOF/TOF mass spectrometry in HEK293 cells stably expressing mSOD1. We validated the results in the SOD1-G93A mouse model and in lymphomonocytes from sporadic ALS patients. We tested the contribution of CypA to disease mechanism(s) performing functional analyses in vitro and in vivo.

Results: The proteomic analysis of CypA interacting proteins revealed that CypA functionally associates with different protein networks. In particular, it extensively binds proteins regulating RNA metabolism, including several hnRNPs and TDP-43. TDP-43 and CypA interact in the nucleus, in an RNA-dependent way. CypA has a key role in the stabilization of TDP-43/hnRNP A2/B1 interaction, and TDP-43-mediated HDAC6 expression regulation, properties impaired in TDP-43 ALS-mutants, possibly because of a loss-of-interaction with CypA. Interestingly, an altered CypA/TDP-43 interaction was observed also in lymphomonocytes of sporadic ALS patients, indicating that CypA interaction network perturbation is relevant also in sporadic forms, where it is aberrantly post-translationally modified. CypA interacts also with mSOD1, and mice expressing mSOD1 and lacking CypA show increased levels of insoluble mSOD1 and hyperphosphorylated TDP-43 in the spinal cord.

Discussion and conclusion: This work shows that CypA has a protective role in ALS: as a chaperone for mSOD1 and in maintenance of multi-protein (TDP-43/hnRNPs) complex dynamic nature. Regardless of the cause of the disease, CypA interaction network is perturbed and CypA may be sequestered in proteinaceous aggregates, altering its protective activities. The net effect is the development of TDP-43 proteinopathy that may lead to a compromised RNA metabolism. CypA being a key interacting partner of both mSOD1 and TDP-43 can represent the 'missing link’ of seemingly divergent patho-mechanisms in ALS and an interesting target for therapeutic interventions.

Acknowledgement:

This work was supported by Telethon-Italy Foundation.

References:

• Massignan T, Casoni F, Basso M et al. BBRC 2007; 353:719–725.
   PubMed Web of Science ®Google Scholar
• Nardo G, Pozzi S, Pignataro M et al. PlosOne 2011; 6(10):e25545.
   PubMed Web of Science ®Google Scholar
• Basso M, Massignan T, Samengo G et al. JBC 2009; 281(44):33325–35.
   Google Scholar

C48 NEUROIMAGING IN ALS: CAN WE SEE MORE CLEARLY?

Filippi M

University Vita-Salute San Raffaele, Milan, Italy, Scientific Institute San Raffaele, Milan, Italy

Email address for correspondence: [email protected]

Keywords: MRI, diffusion tensor imaging, resting stat functional MRI

The use of conventional magnetic resonance imaging (MRI) of the brain and spinal cord in patients suspected of having a motor neuron disease (MND) is recommended to exclude other causes of signs and symptoms of MN pathology. The detection of corticospinal tract (CST) hyperintensities on conventional MRI and a T2-hypointense rim in the precentral gyrus can support a pre-existing suspicion of MND. However, the specific search of these abnormalities for the purpose of making a firm diagnosis of MND is not recommended.

Over the past 10 years, there have been significant advances in the identification of advanced neuroimaging patterns in MND. A significant cortical thinning of the precentral gyrus, damage to the CST and corpus callosum assessed using diffusion tensor (DT) MRI, and altered N-acetylaspartate levels in the primary motor cortex and CST hold promise for assessing the upper motor neuron involvement before clinical symptoms become apparent. Importantly, DT MRI measures of the CST have a prognostic value in amyotrophic lateral sclerosis patients. Furthermore, patterns of brain damage are emerging to identify patients prone to develop dementia or have a rapid progression. It is strongly advisable to incorporate measures derived from advanced MRI techniques into new clinical trials as exploratory outcomes to gain additional insights into the value of these techniques in the assessment of MND in the individual patient and to validate them further in the context of multicenter, longitudinal studies.

C49 THE NEUROIMAGING SIGNATURE OF THE C9ORF72 HEXANUCLEOTIDE REPEAT IN AMYOTROPHIC LATERAL SCLEROSIS – A MULTIMODAL MRI STUDY

Bede P^1,2

Bokde A¹

Byrne S¹

Elamin M¹

McLaughlin R¹

Kenna K¹

Fagan A¹

Pender N²

Bradley D¹

Hardiman O^1,2

^dsTrinity College Dublin, Dublin, Ireland

^dtBeaumont Hospital Dublin, Dublin, Ireland

Email address for correspondence: [email protected]

Keywords: C9orf72 hexanucleotide repeat expansion, MRI, diffusion tensor imaging

Background: The main pathological and clinical features of the C9orf72 hexanucleotide repeat expansion in Amyotrophic lateral sclerosis are currently being established. Population-based studies indicate that the clinical phenotype is associated with fronto-temporal dysfunction, younger age of onset and worse survival outcomes compared to C9orf negative ALS.

Objectives: To characterise the distinctive cortical and subcortical features of the C9orf72 genotype in ALS in comparison to C9orf72 negative ALS patients, C9orf72 negative ALS-FTD patients and healthy controls.

Methods: A prospective, single-centre, single-protocol, grey and white matter magnetic resonance imaging study was undertaken with 30 C9orf72 negative ALS patients, nine ALS patients carrying the C9orf72 hexanucleotide repeat expansion, eight C9orf72 negative ALS-FTD patients and 44 healthy controls. Groups were carefully matched for disease duration, age and handedness. Voxel-based morphometry, cortical thickness analyses and tract-based spatial statistics were carried out for axial diffusivity, radial diffusivity, mean diffusivity and fractional anisotropy. All patients underwent comprehensive neuropsychological profiling and were screened for other mutations in genes previously implicated in ALS such as FUS, OPTN, SOD1, TARDBP, GRN, ANG, ATXN2.

Results: A congruent pattern of cortical and subcortical involvement was identified in those with the C9orf72 genotype, affecting fusiform, thalamic, supramarginal, orbitofrontal regions and Borca's area. The C9orf72 negative cohort demonstrated the “classical” distribution of ALS pathology affecting primarily the motor cortex, motor pathways and cerebellar regions with strikingly limited extra-motor expansion. While the body of the corpus callosum and superior motor tracts were affected in both ALS genotypes, the anterior commissure and the genu of the corpus callosum showed C9orf72 specific vulnerability.

Conclusions: Extensive cortical and subcortical frontotemporal involvement was identified in association with the C9orf72 genotype in ALS, compared to the relatively limited extra-motor pathology in C9orf72 negative patients. The distinctive, genotype-specific pathoanatomical patterns are consistent with the neuropsychological profile of the two ALS cohorts. Our findings also suggest that previously described extra-motor changes in ALS could be partially driven by the inclusion of patients with the C9orf72 genotype. Given the very strong neuropathological signature of the C9orf72 genotype, we suggest that future imaging, neuropsychology and pharmaceutical studies in ALS be stratified for the presence of this mutation.

References:

• Bede P, Bokde ALW, Byrne S et al. Neurology 2013; in press.
   Google Scholar
• Byrne S, Elamin M, Bede P et al. Lancet Neurol 2012;11: 232–40.
   PubMed Web of Science ®Google Scholar

C50 A VISUAL MRI ATROPHY SCALE FOR THE ALS-FTD CONTINUUM

Devenney E

Ambikairajah A

Mioshi E

Tan R

Hodges JR

Kiernan MC

Hornberger M

Neuroscience Research Australia, Sydney, New South Wales, Australia, University of New South Wales, Sydney, New South Wales, Australia

Email address for correspondence: [email protected]

Keywords: Visual Rating Scale, ALS-FTD continuum, clinical tool

Background: ALS and FTD overlap with some patients exhibiting clinical features of both diseases (ALSFTD). Early identification of ALSFTD remains challenging; however, recent imaging findings suggest that ALSFTD patients show significantly more cortical atrophy than ALS but less atrophy than bvFTD. Still these studies employed complex imaging techniques which are not suitable for a clinical setting.

Objective: To distinguish ALS, ALSFTD and bvFTD via a novel visual MRI cortical atrophy scale which can be employed in a clinical setting.

Methods: MRI images of 71 participants (22 ALS, 11 ALSFTD, 24 bvFTD and 14 controls) were rated in four brain areas such as orbitofrontal cortex, anterior temporal lobe, anterior cingulate and motor cortex. Areas of atrophy were rated on a 5-point Likert scale by two raters blinded to the diagnosis.

Results: A continuum of atrophy scores emerged with bvFTD patients exhibiting the highest level of atrophy, while ALS patients had the lowest atrophy scores. ALSFTD patients had a higher motor cortical atrophy rating compared to ALS patients with statistical trends for more atrophy in ALSFTD for the anterior cingulate and anterior temporal lobe. ALSFTD patients could be distinguished from bvFTD patients on orbitofrontal cortex atrophy, with bvFTD being more severely affected.

Conclusion: Our study demonstrates that a simple visual MRI rating scale can reliably distinguish ALS, ALSFTD and bvFTD atrophy patterns in a clinical setting. Motor cortex, anterior cingulate and anterior temporal atrophy emerged as good diagnostic markers for ALSFTD, whereas orbitofrontal cortex atrophy was specific for bvFTD. Employment of this MRI rating scale can complement clinical diagnostics of patients in the ALS-FTD continuum.

Acknowledgements:

The authors are very grateful for the support of the Motor Neurone Disease Research Institute of Australia.

C51 CEREBELLAR SUBSTRUCTURE INTEGRITY IN AMYOTROPHIC LATERAL SCLEROSIS AND BEHAVIOURAL VARIANT FRONTOTEMPORAL DEMENTIA

Tan R¹

Devenney E¹

Hodges JR^1,2

Kiernan MC^1,2

Hornberger M^1,2

^duNeuroscience Research, Randwick, NSW, Australia

^dvUniversity of New South Wales, Randwick, NSW, Australia

Email address for correspondence: [email protected]

Keywords: cerebellar substructures, voxel-based morphometry (VBM)

Background: Emerging evidence implicating the human cerebellum in intact cognitive and behavioural processes has awakened a wide interest in this region, which was previously associated only with sensorimotor control. Recent structural findings across the ALS-FTD continuum found gross cerebellar atrophy in both ALS and bvFTD, suggesting a crucial role of this region to the overlapping cognitive and motor deficits that are increasingly recognised in ALS and bvFTD, respectively. However, there have been no studies to date assessing which cerebellar subregions are affected in ALS and bvFTD.

Objective: To contrast the grey matter integrity of cerebellar subregions in ALS and bvFTD.

Methods: Seventy-nine participants with ALS (n = 25) and behavioural variant FTD (n = 17) as well as controls (n = 37) underwent structural imaging using voxel-based morphometry (VBM) in FSL. Grey matter differences in the cerebellar lobules, vermis and crus were established using a region of interest (ROI) imaging approach of the probabilistic cerebellar atlas (1). All results are reported at p < 0.05 corrected for family-wise error (FWE) or false discovery rate (FDR) multiple comparison correction.

Results: Patients with ALS demonstrate widespread bilateral atrophy in the posterior lobules and only minor atrophy in the left crus in comparison to controls. By contrast, bvFTD patients revealed only minor atrophy changes in the right anterior lobules, but substantial bilateral atrophy in the crus. Direct comparisons between patient groups confirmed this dissociation, with more severe atrophy in the posterior lobules in ALS, and significant changes in the crus and anterior lobules in bvFTD. The vermis was not affected in either disease.

Discussion: We demonstrate here for the first time a structural dissociation between cerebellar subregions affected in ALS and bvFTD, with posterior lobules being affected in ALS and anterior lobules and crus being affected in bvFTD. These results are also corroborated by functional neuroimaging in the healthy showing the posterior lobule being critical for motor control (2), while the right anterior lobule and bilateral crus are activated for behavioural and emotions correlates (3–5). Interestingly, the cerebellar vermis, recently described as being a recipient of dense motor cortical afferents (6), was not affected in either ALS or bvFTD. Overlap in cerebellar substructures may underlie the cognitive and motor deficits observed across the ALS-FTD continuum, which future studies need to investigate.

Conclusion: Grey matter integrity in cerebellar substructures distinguishes between ALS and bvFTD.

Acknowledgements:

The authors are very grateful for the support of MNDRIA, ARC and NHMRC.

References:

• Diedrichsen J et al. Neuroimage 2009;46(1):39–46.
   Google Scholar
• Mottolese C et al. Brain 2013;136(Pt 1):330–42.
   Google Scholar
• Kipps CM et al. Brain 2009;132(Pt 3):592–603.
   PubMedGoogle Scholar
• E KH et al. Hum Brain Mapp 2012.
   Google Scholar
• Baumann O, Mattingley JB Neuroimage 2012;61(4): 805–11.
   Google Scholar
• Coffman KA, Dum RP, Strick PL Proc Natl Acad Sci USA 2011;108(38):16068–73.
   Google Scholar

C52 PROTON MRSI OF CEREBELLUM IN ALS

Sharma K¹

Sheriff S¹

Casanova R²

Govind V¹

^dwUniversity of Miami, Miami, Florida, USA

^dxWake Forest University Health Sciences, Winston-Salem, North Carolina, USA

Email address for correspondence: [email protected]

Keywords: cerebellum, whole-brain MRS, upper motor neuron (UMN)

Background: ALS is a neurodegenerative disorder that primarily affects the motor system, which is composed of motor and supplementary motor cortices, spinal cord, basal ganglia and cerebellum. The pathological (1,2) and functional MRI studies (3) have demonstrated the involvement of cerebellum in patients with ALS. A limited number of in vivo studies of DTI (4,5) and MRS (6) in patients with ALS have revealed variable results of significant (5) to no involvement of cerebellum (4,6).

Objectives: To assess the involvement of cerebellum in patients with ALS using proton MRS method.

Methods: Thirty-seven definitive-ALS patients (50 ± 7 (SD) years) and 37 age-matched controls were scanned on a 3T scanner using a whole-brain MRSI sequence (FOV: 280 × 280× 180 mm³, 50 × 50× 18 phase encodes, slab thickness of 135 mm, TR/TE = 1710/70 ms, and 26 min acq time). The details of data acquisition and processing are published (7). The data from cerebellar hemispheres were obtained to evaluate N-acetyl aspartate (NAA), creatine (Cre), choline (Cho), and ratios among them. The normalized metabolite data were compared using ANCOVA and a p-value of less than 0.05 was considered significant. None of the patients had clinical evidence of extra-motor involvement. The clinical neurological assessments included percentage maximum forced vital capacity (FVC), ALSFRS-R, quantification of upper motor neuron function (foot tap, finger tap, lip and tongue movement rate (7)).

Results: The major findings included lower NAA (13213 ± 181 (SE) vs. 13771 ± 181; p = 0.03), higher Cho (3074 ± 67 vs 2801 ± 67; p = 0.005), lower Cre (13265 ± 212 vs 14434 ± 212; p = < 0.0001), lower NAA/Cho (4.66 ± 0.15 vs 5.13 ± 0.15; p = 0.028), lower NAA/Cre (0.95 ± 0.02 vs 1.11 ± 0.02; p = 0.028) in patients compared to controls. There were mild to moderate correlations (r = 0.3–0.5; p = 0.04–0.003) between the metabolite and various clinical measures.

Conclusions: The lower NAA, Cre, NAA/Cho and NAA/Cre, and higher Cho in the cerebellum of patients with ALS are indicative of neuronal loss or dysfunction and alterations in the choline containing membranes.

Acknowledgements:

NIH grant # R01 NS060874.

References:

• Brownell B, Openheimer DR, Hughes J J Neurol Neurosurg Psychiatry 1970;33:338–357.
   PubMed Web of Science ®Google Scholar
• Swash M, Scholtz C, Vowels G et al. J Neurol Neurosurg Psychiatry 1988;51:785–789.
   PubMed Web of Science ®Google Scholar
• Agosta F, Valsasina P, Absinta M et al. Cereb Cortex 2011; 21:2291–2298.
   PubMed Web of Science ®Google Scholar
• Iwata N, Aoki S, Okabe S et al. Neurology 2008;70: 528–532.
   PubMed Web of Science ®Google Scholar
• Keil C, Prell T, Peshel T et al. BMC Neuroscience 2012; 13:141.
   PubMed Web of Science ®Google Scholar
• Gredal O, Rosenbaum S, Topp S et al. Neurology 1997; 48:y878–881.
   PubMed Web of Science ®Google Scholar
• Govind V, Sharma K, Maudsley A et al. PLoS ONE 2012; 7:e35607.
   PubMed Web of Science ®Google Scholar

C53 DISCRIMINANT VALUE OF 18FDG-PET IN AMYOTROPHIC LATERAL SCLEROSIS

Chiò’ A¹

Pagani M^2,3

Oberg J³

Nobili F⁴

Calvo A¹

Moglia C¹

Fania PC⁵

Morbelli S⁶

Valentini C⁷

Cistaro A⁵

^dy'Rita Levi Montalcini’ Department of Neuroscience, University of Turin, Turin, Italy

^dzInstitute of Cognitive Sciences and Technologies, Rome, Italy

^eaDepartment of Nuclear Medicine, Karolinska Hospital, Strockolm, Sweden

^ebDepartment of Neurosciences, Ophthalmology and Genetics, Genoa, Italy

^ecPositron Emission Tomography Center IRMET S.p.A, Turin, Italy

^edDepartment of Internal Medicine, Nuclear Medicine Unit, University of Genoa, Genoa, Italy

^eeDepartment of Neuroradiology, Azienda Ospedaliera Citte’ della Salute e della Scienza, Turin, Italy

Email address for correspondence: [email protected]

Keywords: PET, sensitivity, specificity

Background: Amyotrophic lateral sclerosis (ALS) is characterized by the loss of spinal and bulbar motor neurons and corticospinal tracts degeneration. A disease marker of upper motor neuron (UMN) degeneration is still lacking. Studies on ¹H-MR spectroscopy and DTI hold promise for detecting and quantifying subclinical UMN damage. Few data have been reported on cerebral ¹⁸FDG-PET.

Objective: To assess the discriminant value of ¹⁸FDG-PET at rest in the largest series of ALS patients investigated so far.

Methods: A total of 195 ALS patients (77 women and 118 men, mean age: 63 (SD 12)) and 40 control subjects (11 women and 29 men, mean age 62 (SD 14)) underwent brain ¹⁸FDG-PET. Twenty-six cortical and sub-cortical brain regions were segmented by the Pick Atlas tool in SPM2 and relative metabolic uptakes individually normalized by whole brain values. Factorial (FA) and Discriminant (DA) analysis were performed using the 52 bilateral regions as well as age and sex. The strict statistical constraint of post-hoc cross-validation was applied.

Results: FA identified eight factors in CTRL and 10 factors in ALS explaining 90% and 87% of total variance, respectively. As compared to CTRL, ALS showed factors with stronger laterality, networking separately left and right temporal lobes and left somatosensory and superior parietal cortex. In ALS FA gathered also selectively insulas and central structures (caudate and thalamic nuclei). DA performed on all 235 subjects showed following cross-validation an accuracy of 88% (sensitivity 89% and specificity 83%) when all 52 regions were taken into account and of 80% (equal sensitivity and specificity) when analyzing the discriminant value of the 10 factors.

Conclusions: In ALS different cortical and subcortical networks were found as compared to normal controls. When 52 functional regions were submitted to discriminant analysis the overall accuracy was 88% with a very high sensitivity of 89%. These findings show for the first time the usefulness of ¹⁸FDG-PET in differentiating ALS patients from normal controls and pave the way to a larger implementation of such methodology in ALS diagnosis.

C54 DEVELOPMENT OF A PET RADIOLIGAND FOR THE NON-INVASIVE IMAGING OF CANNABINOID TYPE 2 RECEPTOR

Mu L¹

Bieri D²

Slavik R²

Drandarov K¹

Mueller A²

Cermak S²

Weber M³

Schibli R^1,2

Kraemer SD²

Ametamey SM²

^efUniversity Hospital Zürich, Zurich, Switzerland

^egETH Zurich, Zurich, Switzerland

^ehKantonsspital St. Gallen, St. Gallen, Switzerland

Email address for correspondence: [email protected]

Keywords: cannabinoid receptor type 2 ligand, Positron Emission Tomography

Background: In amyotrophic lateral sclerosis (ALS) post-mortem tissue and in mice the Cannabinoid type 2 receptor (CB2) receptor is upregulated. Manipulation of the endocannabinoid system via activation of the CB2 in ALS animal models has consistently shown neuroprotective effects (1–3). However, so far no radioligand is available which would enable to study this upregulation in vivo by Positron Emission Tomography (PET).

Objectives: To develop a selective CB2 radioligand, we selected one of 4-oxoquinoline derivatives (designated KD-2) as a potential PET tracer for imaging CB2.

Methods: KD-2 and its corresponding precursor for radiolabeling have been synthesized according to the published procedure (4). The radiosynthesis of [¹¹C]KD-2 was accomplished in a one-step reaction sequence starting from a phenolic precursor and [¹¹C]methyl iodide. In vitro studies included transport experiments across a blood-brain barrier model (P-glycoprotein-transfected MDCK cells) and autoradiography with rodent spleen samples, a tissue with high CB2 levels and post-mortem spinal cord of ALS patient. [¹¹C]KD-2 was further evaluated in vivo in the rat by PET under baseline and blocking conditions.

Results: [¹¹C]KD-2 (ca. 3-5 GBq) was obtained in 99% radiochemical purity after semi-HPLC purification. In vitro barrier permeation of [¹¹C]KD-2 was in the range of blood-brain barrier (BBB) permeating compounds. No efflux by P-glycoprotein was detected. In vitro autoradiography with rat and mouse spleen slices demonstrated high-specific binding towards CB2. High spleen uptake of [¹¹C]KD-2 was observed by PET with Wistar rats and its specificity was confirmed by displacement with the selective CB2 agonist GW405833. A pilot autoradiography study with post mortem spinal cord tissues from ALS patients showed specific binding of [¹¹C]KD-2, suggesting the presence of significant levels of CB2 in spinal cord of ALS patients.

Discussion and conclusion: [¹¹C]KD-2 shows good in vitro and in vivo properties as a potential PET tracer for CB2. CB2 imaging by PET may become of interest for diagnosis and monitoring of disease progression and therapy success in ALS.

Acknowledgments:

This work was partially funded by the Swiss ALS Foundation.

References:

• Ashton JC, Glass M. Curr Neuropharmacol 2007;5: 73–80.
   PubMed Web of Science ®Google Scholar
• Ramirez SH, Hasko J, Skuba A et al. J Neurosci 2012;32: 4004–4016.
   Google Scholar
• Shoemaker JL, Seely KA, Reed RL et al. J Neurochem 2007;101:87–98.
   PubMed Web of Science ®Google Scholar
• Pasquini S, De Rosa M, Pedani V et al. J Med Chem 2011;54:5444–5453.
   PubMed Web of Science ®Google Scholar

C55 REDUCED C9ORF72 GENE EXPRESSION IN C9FTD/ALS IS CAUSED BY TRIMETHYLATION OF HISTONE H3K9

Belzil V

Bauer P

Van Blitterswijk M

Bieniek K

Murray M

Boylan K

Rademakers R

Dickson D

Petrucelli L

Mayo Clinic, Florida, USA

Email address for correspondence: [email protected]

Keywords: C9ORF72, epigenetic modifications, methylation

Individuals carrying expanded repeats in the C9ORF72 gene represent a significant portion of patients suffering from amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). Elucidating how these expanded repeats cause “C9FTD/ALS,” and the associated reduction in C9ORF72 transcript expression, has since become an important goal of the field. As such, we sought to investigate whether epigenetic changes, such as methylation of histones and repeat sequences, and hypermethylation of neighboring islands rich in cytosine-phosphate-guanine (CpG) which lead to gene silencing and consequently loss of function in other repeat disease play a similar role in C9FTD/ALS. We treated fibroblasts obtained from C9FTD/ALS patients with 5-aza-2- deoxycytidine (5-Aza), and compared C9ORF72 mRNA and protein expression levels in treated and untreated cells. We also performed chromatin immunoprecipitation (ChIP) as well as bisulfite modification of genomic DNA to respectively evaluate histone and CpG island methylation. Moreover, we analyzed mRNA and protein expression levels as well as DNA and histone methylation in the frontal cortices and cerebella of postmortem C9FTD/ALS cases. After confirming reduced levels of C9ORF72 mRNA and protein levels in C9FTD/ALS, we showed that the 5-Aza treatment increased expression. We also demonstrated that CpG islands flanking the repeat expansion were not methylated in fibroblasts and brain tissues. More important, we discovered that histone H3 at lysine 9 (H3K9) was trimethylated, an event known to repress gene expression, in all pathogenic repeat carriers. Taken together, our results clearly demonstrate that trimethylation of histone H3K9 is the mechanism involved in reducing the expression of C9ORF72 mRNA and protein in expanded repeat carriers, and may be a promising target for therapeutic intervention. Given that therapies targeting DNA/histone methylation are currently being developed for use in cancer patients, the application of such therapies, in combination with histone deacetylase inhibitors, may be a promising therapeutic strategy for C9FTD/ALS patients.

Acknowledgements:

We would like to thank the patients involved in this study as well as acknowledge the technical support of Kristin Staggs, Matthew C. Baker, Karen R. Jansen-West, Patricia H. Brown, Luc Pregent, Caroline T. Stetler and Mercedes Prudencio. This work was supported by Mayo Clinic Foundation (LP), National Institutes of Health/National Institute on Aging (R01AG026251 (LP)), National Institutes of Health/National Institute of Neurological Disorders and Stroke (R01 NS 063964-01 (LP), R01 NS077402 (LP)), Amyotrophic Lateral Sclerosis Association (LP), the Department of Defense (W81XWH-10-1-0512-1 (LP) and W81XWH-09-1-0315AL093108 (LP)), and the Canadian Institutes of Health Research (VVB).

C56 EXTENSIVE SOUTHERN BLOT STUDY OF C9ORF72 EXPANSION CARRIERS

van Blitterswijk M¹

Dejesus-Hernandez M¹

Niemantsverdriet E¹

Murray M¹

Heckman M²

Brown P¹

Baker M¹

Finch N¹

Bauer P¹

Serrano G³

Beach T³

Josephs K⁴

Knopman D⁴

Petersen R⁴

Boeve B⁴

Graff-Radford N⁵

Boylan K⁵

Petrucelli L¹

Dickson D¹

Rademakers R¹

^eiDepartment of Neuroscience, Mayo Clinic, Jacksonville, FL, USA

^ejSection of Biostatistics, Mayo Clinic, Jacksonville, FL, USA

^ekBanner Sun Health Research Institute, Sun City, AZ, USA

⁴Department of Neurology, Mayo Clinic, Rochester, MN, USA

^elDepartment of Neurology, Mayo Clinic, Jacksonville, FL, USA

Email address for correspondence: [email protected]

Keywords: C9ORF72, repeat expansion size

Background: Hexanucleotide repeat expansions in chromosome 9 open reading frame 72 (C9ORF72) are the major genetic cause of frontotemporal dementia (FTD) and motor neuron disease (MND). These expansions can harbor hundreds to thousands of GGGGCC repeat units. Repeat-primed PCR methods are able to assess the presence or absence of C9ORF72 expansions; however, they are unable to determine the actual repeat size, as opposed to Southern blotting techniques. To date, large scale Southern blot studies that investigate C9ORF72 repeat sizes in brain tissue have not been reported, and hence, it is currently unknown whether these repeat sizes affect disease severity or phenotypes.

Objectives: To determine repeat sizes in a large cohort of C9ORF72 expansion carriers and to correlate repeat sizes with patient characteristics.

Methods: We performed an extensive Southern blot characterization study in a cohort of 84 C9ORF72 expansion carriers, including FTD patients (n = 35), FTD/MND patients (n = 16), MND patients (n = 30), and unaffected subjects (n = 3). More than 200 independent DNA samples obtained from frontal cortex, temporal cortex, parietal cortex, occipital cortex, spinal cord, cerebellum, blood, skin-derived fibroblasts, spleen, heart, muscle, pancreas, liver, and testes were investigated. A total of 7–10 μg of genomic DNA was used for Southern blotting.

Results: Repeat lengths in the cerebellum were significantly smaller (median ˜ 1667 repeat units) than in the frontal cortex (median ˜ 5250 repeat units, p < 0.001), or in blood (median ˜ 2717 repeat units, p < 0.001). Within these tissues, there was no significant difference in repeat length between disease subgroups, nor did we detect associations with gender, TDP-43 type, family history, or size of the wild-type allele. Individual patients did demonstrate substantial variation in repeat length across tissues, such as blood and fibroblasts. In the frontal cortex of FTD patients, age at onset strongly correlated with repeat length (r = 0.63, p = 0.003) and smear size (r = 0.66, p = 0.002). Finally, in the cerebellum, survival after disease onset was poorer in patients from our overall cohort with repeat lengths greater than 1467 repeat units (RR 3.28, p = 0.010), corresponding to a decrease in survival of ˜ 2.5 years.

Discussion: C9ORF72 repeat expansions cause a range of neurodegenerative phenotypes. At the moment, it is impossible to predict disease onset, progression, and/or manifestations of C9ORF72 expansion carriers. Our present study has revealed significant variability in C9ORF72 repeat sizes across tissues and patients. In the frontal cortex of FTD patients, repeat sizes were relatively unstable and age-dependent; long repeat sizes in the cerebellum conferred an important survival disadvantage. Repeat sizes, however, did not predict disease phenotype, and were not associated with other clinical or pathological features. Based on our findings, only repeat sizes in the cerebellum, a region unaffected by neuronal loss, could provide reliable information on disease severity, which is highly relevant for genetic counseling.

C57 C9ORF72 GGGGCC EXPANDED REPEATS PRODUCE SPLICING DYSREGULATION WHICH CORRELATES WITH DISEASE SEVERITY IN AMYOTROPHIC LATERAL SCLEROSIS (ALS)

Cooper-Knock J¹

Walsh M¹

Higginbottom A¹

Highley JR¹

Bury J¹

Rattray M²

Heath P¹

Kirby J¹

Hautbergue G¹

Shaw PJ¹

^emSheffield Institute for Translational Neuroscience, Sheffield, UK

^enLife Sciences, Manchester, UK

Email address for correspondence: [email protected]

Keywords: transcriptome, C9ORF72, splicing

Background: Expanded GGGGCC repeats in intron 1 of the C9ORF72 gene represent the most common cause of familial amyotrophic lateral sclerosis (ALS). Other neurodegenerative diseases result from expanded repeat sequences in non-coding regions raising the possibility of a common pathogenic mechanism. In particular, toxic gain-of-function by sequestration of RNA-binding proteins (RBPs) is a major factor underlying myotonic dystrophy types 1/2 (DM1/2). If C9ORF72 expansions mediate pathogenesis by a similar mechanism then a global disruption of splicing might be expected.

Objective: To utilise an exon level transcriptome analysis to characterise the effect of C9ORF72 expansion on global splicing patterns.

Methods: Lymphoblastoid cell lines derived from ALS patients (n = 54) and neurologically normal controls (n = 15) were obtained. Extracted RNA was assessed using Human Exon 1.0ST GeneChip^® microarrays. Data were analysed using the ‘finding isoforms using robust multichip analysis (FIRMA) package. Highly negative or positive values of the FIRMA score are indicative of alternative exon skipping or inclusion, respectively. Consistency of splicing within a sample group such as patients or controls was evaluated by comparing the number of splicing events which occurred in 1, 2, 3, … n samples within the group. To allow comparison between groups, reference was made to a model in which exons are spliced in or out at random.

Results: There was no difference in the total number of splicing events observed in lymphoblastoid cell lines derived from C9ORF72+ patients, C9ORF72- patients and controls. However, the nature of those splicing events was significantly different. It is expected that functionally appropriate splicing would be similar in samples of a particular group. Consistency was significantly reduced in the C9ORF72+ group compared to C9ORF72- patients and controls. In addition, splicing was less consistent in C9ORF72+ patients who lived less than 2 years compared to those that lived more than 5 years.

Discussion and conclusion: These data are consistent with sequestration of RBPs by the C9ORF72 expansion which would significantly impact the splicing machinery of the cell. We suggest that reduction of splicing consistency reflects an increased splicing error rate which may, over time, lead to a crucial pathogenic splicing event(s). This might explain both the variability and late age of onset of C9ORF72-ALS. Consistent with this those patients with the lowest consistency of splicing had more severe disease.

Acknowledgements:

This work was supported by an EU Framework 7 (Euromotor No259867) grant to PJS and JK. PJS is supported as an NIHR Senior Investigator. JRH and JCK are supported by MND Association/Medical Research Council Lady Edith Wolfson Fellowship awards [G0 800380] and [MR/K003771/1]. Samples used in this research were obtained from the UK National DNA Bank for MND Research, funded by the MND Association and the Wellcome Trust. We would like to thank people with MND and their families for their participation.

C58 MOTOR NEURON SPECIFIC TRANSLATIONAL PROFILING IN SOD1G93A TRANSGENIC MICE

Zhao B^1,2

Robertson J^1,2

Rogaeva E^1,2

Zinman L³

Keith J³

Heiman M^4,6

Dougherty J^5,6

^eoUniversity of Toronto, Toronto, ON, Canada

^epCentre for Research in Neurodegenerative Diseases, Toronto, ON, Canada

^eqSunnybrook Health Sciences Centre, Toronto, ON, Canada

^erMassachusetts Institute of Technology, Cambridge, MA, USA

^esWashington University in St. Louis, St. Louis, MO, USA

^etHoward Hughes Medical Institute, Chevy Chase, MD, USA

Email address for correspondence: [email protected]

Keywords: translational profiling, TRAP, motor neuron

Background: Previous profiling studies in cell and transgenic mouse models of ALS have provided large lists of genes and pathways that are potentially relevant to the disease mechanisms causing motor neuron degeneration. However, the mRNA isolation techniques used in these studies have been hampered by challenges such as lack of cell specificity or disruption of native cellular environment.

Objectives: To overcome these disadvantages and determine the translational changes occurring specifically in the spinal cord motor neurons of ALS, we employed a novel technique called Translating Ribosome Affinity Purification (TRAP).

Methods: TRAP utilizes BAC transgenic mice expressing an EGFP-tagged ribosomal protein, L10a, in genetically targeted cell populations via use of cell-specific gene promotors. TRAP facilitates immunoaffinity purification of EGFP-tagged polysomes and bound mRNAs from genetically determined cell populations, thereby combining coincident detection of cell-type specificity and all translated mRNAs in vivo.

SOD1^G93A (±) mice were crossed with ChAT bacTRAP (±) mice, which express EGFP-L10a exclusively in cholinergic neurons, including motor neurons. Spinal cords were dissected from SOD1^G93A (±): ChAT bacTRAP (±) mice and control ChAT bacTRAP (±) littermates at pre-symptomatic stage (10 weeks). mRNAs isolated using TRAP were identified on an Agilent G3 Mouse GE 8 x 60K microarray.

Results: Bioinformatics analyses revealed 75 transcripts differentially expressed in SOD1^G93A (±):ChAT bacTRAP (±) mice compared to control. Of 51 known genes 29 were up-regulated and 22 down-regulated. These genes have been documented to be involved in numerous biological processes. Immunohistochemistry confirmed two genes, 3-phosphoglycerate dehydrogenase (Phgdh) and anaphase-promoting complex subunit 1 (Anapc1), displayed the most profound changes at the protein levels. Phgdh, an enzyme critical for serine biosynthesis, was dramatically induced in the degenerating motor neurons of SOD1^G93A (±) mice in contrast to control where it was absent in motor neurons, consistent with previous observations in SOD1^G37R and SOD1^G85R transgenic mice using LCM (1). Anapc1, a component of the cycosome regulating mitosis progression, displayed a subcellular mislocalization from predominantly nuclear to cytoplasmic in motor neurons of SOD1^G93A (±) mice. A phosphorylated Anapc1, pSer377-Anapc1, displayed similar changes in degenerating motor neurons of SOD1^G93A (±) mice as well as all familial and sporadic ALS cases examined.

Discussion and conclusion: Together, using the TRAP technique, we have for the first time identified specific translational changes occurring in spinal cord motor neurons of 10-week SOD1^G93A transgenic mice. Our findings of Phgdh and Anapc1 protein expression changes in ALS-linked degenerating motor neurons have advanced our understanding of previously less explored pathways, ie serine biosynthesis and cyclosome function, which may be involved in the mechanisms causing ALS. We expect this work will ultimately lead to the development of biomarkers and/or effective therapeutics for this devastating disease.

Reference:

• Lobsiger C, Boillée S, Cleveland D. PNAS 2007;104: 7319–7326.
   PubMed Web of Science ®Google Scholar

C59 MORE EVIDENCE SUPPORTING PERTURBATION IN EXTRACELLULAR AND TRANSMEMBRANE DOMAINS AND OF PROTEIN SIGNALING BY TRANSCRIPTOME ANALYSIS OF MOTOR NEURONS FROM SPORADIC ALS SPINAL CORDS

Hutt K¹

Rabin S²

Baughn M¹

Libby R²

Markmiller S¹

Hoon S¹

Yeo GW¹

Ravits J^1,2

^euUniversity of California, San Diego, La Jolla, CA, USA

^evBenaroya Research Institute, Seattle, WA, USA

Email address for correspondence: [email protected]

Keywords: genomics, pathology, transcriptome

Background: While many of the ALS genes play an important role in RNA-related processes, how mutations lead to temporally spatially progressive motor neuron degeneration remains unclear. We have previously shown that the focal onset and regional spread of motor neuron dysfunction can leave motor neuron-rich regions in post-mortem spinal cords and that transcriptome signatures can be clearly resolved. Here we continue this work with RNA-seq technologies and new bioinformatic approaches.

Objectives: Our objectives were to illuminate key transcriptional signatures from residual motor neurons in neuron-rich regions using RNA-seq and to seek biological enrichments.

Methods: We used used laser-capture microdissection to selectively enrich RNA pools with motor neurons in 13 sALS and nine control spinal cords. cDNA libraries were generated from each spinal cord and RNA-seq was performed to obtain a whole-transcriptome dataset of gene expression levels from each replicate. We used BiomarkTM HD System from Fluidigm^® for high-throughput qPCR validation.

Results: Traditionally, disease samples would be compared to control samples in a pairwise fashion, providing a list of significantly changed genes for each comparison. Owing to the high level of variation found not just between individual patients, but also because of the unknown shared (or unshared) genetic factors underlying sALS patients, a unique analysis method was devised to focus on commonly changing genes despite the high level of noise, termed Median Percentile Rank (MPR). This analysis, meant to avoid biases in normalizing ratios (Buck, Lieb 2004), works as follows. Each of the 117 lists of gene expression fold-changes is sorted, and a rank is assigned. The percentile rank is calculated from a gene’s position in this list, and the median of a gene’s 117 percentile ranks is calculated, giving the MPR statistic. A histogram can then be constructed from these MPR values, and a clear threshold can be drawn near the two edges of the histogram if there is enrichment above expected.

From this analysis, 259 to more than 2,346 and 286 to more than 955 genes (strict to non-strict cutoffs) were found to be up- and down-regulated, respectively. Gene set enrichment analysis revealed results for the up-regulated genes, with enrichment for GO terms extracellular region (p < E-59), signal peptide (p < E-59), disulfide bond (p < E-68), glycoprotein (p < E-62), and membrane-related functions (p < E-24). In addition, defense/inflammatory response (p < E-30) was identified. Interestingly, the down-regulated set showed no confident results. To validate these changes, 25 up-regulated and 15 down-regulated genes were chosen to be analyzed across all 22 samples (13 sALS and 9 controls).

Conclusion: These results identify significantly perturbed transcriptional programs related to extracellular and membrane domains and to protein signaling, either contributing to or resulting from motor neuron degeneration in sALS. They are identified by unbiased discovery methods and have been relatively unemphasized.

C60 TRANSLATIONAL STUDY OF POTENTIAL PROGNOSTIC AND DIAGNOSTIC BIOMARKERS TO HUMAN SAMPLES

Calvo AC^2,1

Torre-Merino P¹

Juarez-Rufian A¹

Atencia G¹

Cordero-Vazquez P¹

Martin-Casanueva MA³

Muñoz-Blanco JL⁴

Galan L⁵

Esteban-Perez J¹

Osta R²

Garcia-Redondo A¹

^ewNeurology department – ALS Unit. CIBERER U-723. Health research Institute, October 12th University Hospital, MADRID, Spain

^exLAGENBIO-I3A, Aragonese Institute of Health Sciences (IACS), Faculty of Veterinary, University of Zaragoza, ZARAGOZA, Spain

^eyBiochemistry department. CIBERER U-723. Health research Institute, October 12th University Hospital, MADRID, Spain

^ezNeurology department – ALS Unit. Gregorio Marañón Hospital, MADRID, Spain

^faNeurology department – ALS Unit. Clinico San Carlos Hospital, MADRID, Spain

Email address for correspondence: [email protected]

Keywords: muscle biopsy, lymphocyte, prognostic factors

Background: The search of biomarkers in ALS is being carried out on post-mortem patient’s samples, such as brain or spinal cord, or on samples invasively obtained, such as cerebrospinal fluid (CSF) and more recently mesenchymal stem cells from bone marrow (1). Growing tendency relies on the study of new tissues that can be analyzed in a less invasive way (2). Previous studies in our group suggested five genes, Mef2c, Gsr, Col19a1, Calm1 and Snx10, as potential genetic biomarkers of longevity in the animal model for ALS (3). We translated this study to human samples from skeletal muscle and blood to validate the potential nature of these biomarkers.

Objectives: The main aim of this study was to identify potential prognostic and diagnostic biomakers in muscle biopsies and blood samples from ALS patients.

Methods: Muscle biopsies proceeded from brachial biceps. The lymphocyte fraction from total blood was isolated using Ficoll gradient. Real time PCR and western blot were used to analyze all the samples in the 17 genes previously selected. ANOVA and Kruskal Wallis test were used to compare means. ROC curves were calculated to study the diagnostic criterion.

Results: In muscle biopsies, the results suggested that transcriptional and transductional levels of Col19a1 could be used as a diagnostic biomarker for ALS. Regarding lymphocyte samples, transcriptional levels of Col19a1 could also be used as a diagnostic biomarker. Moreover, Col19a1, Impa1, Mef2c, Nogo A, Snx10, Gsk3 and Gsr could be considered as potential prognostic biomarkers of the disease.

Discussion: In this study, translational research work has been made from transgenic SOD1^G93A mice to muscle biopsies and lymphocytes from ALS patients. Seventeen genes were validated in human samples. Interestingly, transcriptional and transductional expression of Col19a1 was significantly increased and specifically linked to ALS patients, yielding a positive predictive value of 100% and a negative predictive value of 99,998%. Regarding lymphocyte samples, transcriptional expression of Col19a1 significantly increased and was specifically linked to ALS patients, considering this gene as a potential diagnostic biomarker. Regarding the rest of genes, Impa1, Mef2c, Nogo A, Snx10, Gsk3, Gsr and also Col19a1 were identified as potential prognostic biomarkers of the disease in lymphocytes from ALS patients.

Acknowledgements:

This work was supported by grants PI10/00092, PI10/01787 and EC08/00049 from the Instituto de Salud Carlos III (ISCIII) and the support of the Spanish Foundation for the development of ALS research (FUNDELA).

References:

• Nachmany H, Wald S, Abekasis M et al. Disease Markers 2012;32:211–220.
   PubMed Web of Science ®Google Scholar
• Pradat PF, Bruneteau G, González de Aguilar JL et al. Ann. Neurol 2007;62:15–20.
   PubMed Web of Science ®Google Scholar
• Calvo AC, Manzano R, Atencia-Cibreiro G et al. PLoS One 2012;7:e32632.
   Google Scholar

C61 SERUM CREATININE, A BIOMARKER FOR MUSCLE MASS IN AMYOTROPHIC LATERAL SCLEROSIS(ALS), PREDICTS LOSS OF AMBULATION MEASURED BY ALS FUNCTIONAL RATING SCALE-REVISED WALKING ITEM SCORE(ALSFRSW)

Fischer MP¹

Brooks BR¹

Lucas NM¹

Smith NP¹

Nichols MS¹

Belcher SL¹

Story JS¹

Desai UG¹

Bockenek WL²

Lindblom SS³

Paccico TJ³

Sanjak MS^1,4

Bravver EK¹

Langford VL¹

Ward AL^1,5

Wright KA¹

Henderson AM^1,2

Holsten SE¹

Corey QD¹

Russo PC¹

^fbCarolinas Medical Center Neurology, Carolinas HealthCare System

^fcCarolinas Rehabilitation, Carolinas HealthCare System

^fdCarolinas Medical Center, Internal Medicine, Carolinas HealthCare System; University of North Carolina School of Medicine - Charlotte Campus, Charlotte, NC, USA

^feUniversity of North Carolina - Charlotte - Kinesiology, Charlotte, NC, USA

^ffCabbarus College of Health Sciences - Occupational Therapy, Concord, NC, USA

Email address for correspondence: [email protected]

Keywords: biomarker, creatinine, disease progression

Background: Serum creatinine has been identified as a muscle mass biomarker related to the diagnosis and rate of progression of ALS and bulbar spinal muscular atrophy (Kennedy’s disease) in French, Japanese and American ALS patients (1–5). Recent analysis of clinical staging in ALS patients has identified increased burden of leg involvement in El Escorial Criteria-Clinically Definite ALS patients (6).

Objectives: To define the relationship between serum creatinine and ALSFRSRw cross-sectionally and longitudinally over time in a clinic-based ALS patient population at Carolinas Neuromuscular/ALS-MDA Center.

Methods: Serum creatinine at first and subsequent clinic visits was related to ALSFRSR total score and ALSFRSRw from 2010 to 2012 in 311 ALS patients (183M and 128F; 55.7 ± 12.7 (SD) years). Cross-sectional analysis of all data as well as change in serum creatinine and ALSFRSRw between first and subsequent visits was analyzed with statistical software (MedCalc Software, Ostend, Belgium www.medcalc.org/).

Results: Serum creatinine cross-sectionally decreased significantly with decreasing ALSFRSw category (Creatinine = 0.54 (95% CI: 0.49-0.58)+ 0.09(95% CI: 0.07–0.11) • ALSFRSRw (R^{2 =} 0.2082: P < 0.001)). Moreover, in individual ALS patients, change in serum creatinine decreased proportionate to decrease in ALSFRSRw (DeltaCreatinine = −0.11 (95% CI: −0.16 to 0.05) +0.08(95% CI: 0.04– 0.12)• ALSFRSRw (R^{2 =} 0.1506: P < 0.001)).

Discussion and conclusion: Decrease in serum creatinine significantly predicted change in ambulation measured by ALSFRSRw longitudinally. Further analysis is required to determine whether leg function alone and leg muscle mass are the major determinants of serum creatinine in ALS patients over the course of the disease.

Acknowledgements:

ALS and Neuromuscular Garden-Carolinas Garden of Hope Funds, Carolinas ALS Research Fund, Pinstripes ALS Foundation, Edwin C Holt Communications Laboratory Fund, Mike Rucker ALS Care Fund, Carolinas HealthCare Foundation, North Carolina Jim “Catfish” Hunter Chapter - ALS Association, Muscular Dystrophy Association - ALS Division.

References:

• Paillisse C et al. Amyotroph Lateral Scler Other Motor Neuron Disord 2005;6(1):37–44. (DOI:10.1080/14660 820510027035)).
   Google Scholar
• Ikeda K et al. J Neurol Sci 2009;287:294. (DOI:10.1016/ j.jns.2009.08.008).
   PubMed Web of Science ®Google Scholar
• Ikeda K et al. Intern Med 2012;51:1501–1508. (DOI:10.2169/internalmedicine.51.7465).
   PubMed Web of Science ®Google Scholar
• Hashizume A et al. Brain 2012;135:2838–2848. (DOI:10.1093/brain/aws170).
   PubMed Web of Science ®Google Scholar
• Miller RG, Katz J, Moore D ALS 2012;13(S1):56–57. (DOI:10.3109/17482968.2012.721231/094).
   Google Scholar
• Brooks BR et al. ALS 2012;13(S1):184. (DOI:10.3109/ 17482968.2012.721231/339).
   Google Scholar

C62 MISFOLDED SOD1 IN BLOOD PLASMA IS AN ANTIBODY-ACCESSIBLE BIOMARKER FOR SPORADIC ALS

Uger M¹

Grad L²

Chen H¹

Karaskov E¹

Lee V¹

Mutukura T¹

Plawinski E¹

Syam S¹

Lee P³

Liu X³

Luk M⁴

Mackenzie I⁴

Cashman N^1,2

^fgAmorfix Life Sciences Ltd., Mississauga, ON, Canada

^fhBrain Research Centre, University of British Columbia, Vancouver, BC, Canada

^fiEpitomics, Inc., Burlingame, CA, USA

^fjDepartment of Pathology and Laboratory Medicine, University of British Columbia, Vancouver, BC, Canada

Email address for correspondence: [email protected]

Keywords: SOD1, biomarker, diagnosis

Background: ALS pathology is often associated with protein misfolding and aggregation. Clinical and pathological similarities between all forms of ALS suggest the existence of a common pathogenic pathway. Misfolded Cu/Zn superoxide dismutase (SOD1) has been increasingly identified in SALS and non-SOD1 FALS (1–3). There is increasing evidence that wild-type (wt) SOD1 may play a role in SALS, which in its misfolded form acquires many of the same cytotoxic properties as mutant SOD1 (4–6). Thus, misfolded SOD1 is a likely candidate for a common ALS diagnostic biomarker.

Objective: To determine if misfolded SOD1 is detectable in blood plasma from ALS patients using a novel diagnostic testing platform.

Methods: The mouse monoclonal antibody (mAb) 3H1 specifically binds to misfolded SOD1 and was used to perform immunoprecipitation experiments from 27 SALS and 27 normal control plasma samples. There was a significantly higher level of misfolded SOD1 detected in ALS compared to normal plasma samples (p = 0.0001333). However, immunoprecipitation is a cumbersome technique that is not easily used for diagnostic testing. We have developed a simple dual-bead immunoassay that detects misfolded SOD1 in spinal cord from SOD1 G93A mutant mice and from two individuals with SALS, but not from a neurological control.

Results: Levels of misfolded SOD1 are approximately 35-times higher in G93A spinal cord than in SALS spinal cord. An anti-SOD1 rabbit mAb was generated with EC₅₀ = 35 pM against misfolded SOD1, which is 10–100 fold higher than observed with mouse mAbs. This antibody specifically detects misfolded SOD1 in G93A mouse and human ALS spinal cord by immunohistochemistry, and control plasma spiked with G93A mouse spinal cord homogenate via the dual-bead immunoassay.

Conclusions: We have developed a mouse mAb that specifically detects misfolded SOD1 in SALS patient plasma via immunoprecipitation. Application of this antibody in a dual bead immunoassay specifically detects misfolded SOD1 in both ALS mouse and human spinal cord homogenate. A high-affinity rabbit mAb based on the same disease-specific epitope detect misfolded SOD1 in normal plasma spiked with G93A spinal cord homogenate. Further immunoassay development continues with ALS patient tissue and plasma samples.

Acknowledgments:

We acknowledge use of tissues procured by the National Disease Research Interchange (NDRI) with support from NIH grant 5 U42 RR006042.

References:

• Forsberg K, Jonsson PA, Andersen PM et al. PLoS ONE 2010;5:e11552.
   PubMed Web of Science ®Google Scholar
• Bosco DA, Morfini G, Karabscak NM et al. Nat Neurosci 2010;13:1396–403.
   PubMed Web of Science ®Google Scholar
• Pokrishevsky E, Grad LI, Yousefi M et al. PLoS ONE 2012;7:e35050.
   PubMed Web of Science ®Google Scholar
• Rakhit R, Cunningham P, Furtos-Matei A et al. J Biol Chem 2002;277:47551–6.
   PubMed Web of Science ®Google Scholar
• Rakhit R, Crow JP, Lepock JR et al. J Biol Chem 2004; 279:15499–504.
   PubMed Web of Science ®Google Scholar
• Casoni F, Basso M, Massignan T et al. J Biol Chem 2005; 280:16295–304.
   PubMed Web of Science ®Google Scholar

C63 PROTON NMR SPECTROSCOPY METABOLOMICS IN SERUM AND CSF

Gray E¹

Larkin J²

Claridge T³

Talbot K¹

Sibson N²

Turner M¹

^fkNuffield Department of Clinical Neuroscience, University of Oxford, Oxford, UK

^flCR-UK/MRC Gray Institute for Radiation Oncology and Biology, University of Oxford, Oxford, UK

^fmDepartment of Chemistry, University of Oxford, Oxford, UK

Email address for correspondence: [email protected]

Keywords: biomarker, metabolomics, spectroscopy

Background: Candidate ‘wet’ biomarkers are urgently sought in ALS. Results from agnostic proteomic approaches have not yet been replicated independently. Metabolomic studies using the non-selective technique ¹H NMR spectroscopy have reported serum glutamate as a marker of disease duration (1) and markers in CSF consistent with altered glucose metabolism (2).

Objectives: To determine whether there is a ‘signature’ of serum or CSF metabolites common to a large group of ALS patients.

Methods: The Oxford Study for Biomarkers in MND (BioMOx) obtained baseline serum and CSF samples in up to 70 patients across a range of initial disability and progression rates, with six-monthly longitudinal collection where possible, for up to two years. Samples were processed within one hour of extraction and centrifuged prior to storage in polypropylene at −80°C. Aliquots (100μL CSF or 150μL ultracentrifuged serum) were mixed with phosphate buffer in D₂O containing 1mM TSP as an internal standard (final volume 0.6mL). ¹HNMR spectra were acquired using a 700MHz Bruker spectrometer. Partial least squares discriminant analysis (PLS-DA) was used to determine differences between the spectra from patients and control volunteers, and according to progression rate. q² values above 0.4 were deemed to be significant.

Results: When comparing CSF spectra from age-matched control volunteers with the most advanced of the longitudinal samples (≥ 12 months from study enrolment), a significant separation was seen (q² = 0.49; 20 ALS and 17 control cases). This separation was still evident when less advanced patient samples were included (≥ 6 months from study enrolment, q² = 0.45; 26 ALS and 17 control cases), but not when the most advanced samples from each patient were modelled against the control cohort (i.e. including some baseline samples, q² = 0.22; 47 ALS and 17 control cases). No significant separation was seen with serum.

Discussion and conclusion: CSF proton NMR spectroscopy has potential for distinguishing ALS patients from healthy controls. The process of defining the key metabolites responsible for this separation is underway, and it is possible they will reflect systemic processes associated with the later stages of disease. Further model-building using multilevel PLS-DA will be used in an attempt to phenotypically stratify patients.

Acknowledgements:

This work was funded by the Medical Research Council and Motor Neurone Disease Association UK Lady Edith Wolfson Fellowship, and the Fondation Thierry Latran.

References:

• Kumar A et al. International journal of clinical chemistry 2010;411:563–7.
   Google Scholar
• Blasco H et al. PloS one 2010;5:e13223.
   PubMed Web of Science ®Google Scholar

C64 BETA-BAND INTERMUSCULAR COHERENCE AS A BIOMARKER OF UPPER MOTOR NEURON DYSFUNCTION IN MOTOR NEURON DISEASE

Jaiser S^1,2

Williams T^1,2

Baker S¹

Baker M^1,2

^fnNewcastle University, Newcastle upon Tyne, UK

^foRoyal Victoria Infirmary, Newcastle upon Tyne, UK

Email address for correspondence: [email protected]

Keywords: biomarker, upper motor neuron, coherence

Background: It has previously been demonstrated that beta-band intermuscular coherence (IMC) is a potential biomarker of upper motor neuron dysfunction (1), and does not change significantly during healthy adult life (2).

Objective: To investigate beta-band IMC in patients with motor neuron disease (MND) at first presentation to a specialist MND service.

Methods: We recruited 70 patients with possible, probable and definite MND (according to Awaji criteria) from the local tertiary MND service. Lower limb IMC was measured between a calf muscle and an intrinsic foot muscle during an unrestrained ankle dorsiflexion task. Upper limb IMC was similarly estimated between a forearm muscle and an intrinsic hand muscle during performance of an auxotonic precision grip task. Average beta-band IMC was computed for each limb and subject. Control data from 92 normal volunteers (age: 22–77) were available from our previous study. The probability distributions of MND and control data were modelled by variable kernel density estimates, and these estimates used to construct receiver operator characteristic (ROC) curves.

Results: The cumulative probability distribution of beta-band IMC in MND was similar in shape to that in normal controls but was shifted towards lower IMC values. The area under the ROC curve was approximately 75% for each limb.

Conclusion: Beta-band IMC represents an easily tolerated and inexpensive method for electrophysiological assessment of patients with MND. Our previous work suggests that the observed abnormalities are explicable in terms of upper motor neuron dysfunction. Upper motor neuron abnormalities have thus far proven difficult to detect in early MND, and beta-band IMC has potential as a semi-quantitative, clinically applicable biomarker with early sensitivity.

Acknowledgements:

This work was supported by the Wellcome Trust [089893/Z/09/A] and the National Institute for Health Research (NIHR).

References:

• Fisher KM, Zaaimi B, Williams TL et al. Brain 2012; 135(9):2849–64.
   PubMedGoogle Scholar
• Jaiser SR, Barnes JD, Baker MR et al. Society for Neuroscience 2012.
   Google Scholar

C65 TRANSGLUTAMINASE 6 ANTIBODIES IN THE SERUM OF PATIENTS WITH ALS – IS GLUTEN SENSITIVITY INVOLVED IN MOTOR NEURON DEGENERATION?

Nefussy B¹

Gadoth A¹

Bleiberg M¹

Klein T^2,3

Artmann I¹

Abraham A^1,3

Drory V^1,3

^fpTel-Aviv Medical Center, Tel-Aviv, Israel

^fqRabin Medical Center, Petach-Tiqva, Israel

^frSackler Faculty of Medicine, Tel-Aviv University, Tel-Aviv, Israel

Email address for correspondence: [email protected]

Keywords: gluten sensitivity, celiac, transglutaminases

Background: Celiac is an autoimmune disease in which the immune system attacks the inner wall of the intestine by a hypersensitivity reaction to gluten in genetically predisposed individuals. Transglutaminase 2 (TG2) is the primary autoantigen and IgA anti-TG2 production is used for diagnosis, as well as antibodies to deamidated gliadin peptide (DPG) and endomysium. Almost all patients with celiac disease present specific HLA genotypes (HLA -DQ2 encoded by DQA1*05 and DQB1*02 alleles or DQ8 encoded by DQB1*03:02).

Gluten sensitivity can cause extra-intestinal manifestations including neurological syndromes, commonly ataxia and neuropathy, with or without gastrointestinal symptoms. Most patients with gluten ataxia produce antibodies towards the newly identified neuronal transglutaminase 6 (TG6).

Two independent case reports described patients initially diagnosed with amyotrophic lateral sclerosis (ALS) and ultimately with celiac. Both started a strict gluten-free diet with improvement of symptoms during the following months.

Objectives: To evaluate the incidence of celiac-related and TG6 antibodies and HLA genotypes in ALS patients and controls, in order to evaluate whether a neurological presentation of gluten sensitivity mimicking ALS might occur occasionally.

Methods: We measured serum levels of IgA antibodies to TG2 and endomysium, IgG antibodies to DGP, IgA and IgG antibodies to TG6 in a cohort of patients with ALS and a group of healthy individuals of similar age and gender.

Results: We examined 149 patients (98 men, age 61.7 ± 12.3 years) and 114 controls (80 men, age 61.05 ± 11.5). All patients and controls were negative to IgA antibodies to endomysium and TG2. Sixty patients and 24 controls were tested for DGP antibodies and all except one were negative. Remarkably, 23 (15%) patients were positive to TG6 IgA antibodies as compared to only 5 (4%) controls (p = 0.0037). Three patients had borderline levels of TG6 antibodies.

The seropositive TG6 patients showed a classical picture of ALS with 75% males, age at disease onset 57.6 ± 12.5 years, 3 (13%) with bulbar onset, 19 (83%) patients had both upper and lower motor neuron involvement and a usual rate of progression, eight patients died or performed tracheostomy 24.5 ± 25 (range: 7–86) months after disease onset.

Fifty-one patients were tested for association with celiac specific HLA alleles. 73% of the TG6 IgA positive patients were positive to DQB1 as compared to only 44% of the TG6 IgA negative patients. The DQA1 alleles were carried by 74% of the TG6 IgA positive patients and 60% of the TG6 IgA negative patients. HLA typing in control individuals of similar ethnic origin is ongoing and will be reported at the meeting.

Conclusion: These preliminary data indicate that in certain cases ALS might be related to autoimmunity and gluten sensitivity. As gluten sensitivity is potentially treatable, this diagnostic challenge should not be overlooked.

C66 EFFECT OF LIPID PROFILE ON PROGNOSIS IN THE PATIENTS WITH AMYOTROPHIC LATERAL SCLEROSIS

Rafiq M¹

Lee E²

Bradburn M²

McDermott C¹

Shaw PJ¹

^fsSheffield Institute for Translational Neuroscience, Sheffield, UK

^ftThe School of Health and Related Research (ScHARR), Sheffield, UK

Email address for correspondence: [email protected]

Keywords: lipids, prognosis, high fat diet

Background: Patients with amyotrophic lateral sclerosis (ALS) are particularly predisposed to malnutrition for a variety of reasons which include dysphagia, fear of choking and aspiration, inability to feed themselves and high resting metabolic rate. Hence various defence mechanisms are likely to be activated in such patients to provide energy substrates, for example, gluconeogenesis, lipolysis and ketogenesis. Higher weight in ALS is associated with a better outcome and it has been reported that patients with raised LDL/HDL ratio have a significantly improved survival. However, an opposing view is that body mass index and not dyslipidaemia is an independent predictor of survival in ALS. In common with the normal population of a similar age, an abnormal lipid profile is commonly seen in patients with ALS. An obvious explanation might relate to the mounting of a defence mechanism to provide energy substrates. Another possible explanation is that it might result from metabolic dysregulation or mitochondrial and/or endoplasmic reticulum (ER) stress. Mitochondrial and ER stress have been implicated in the pathogenesis of ALS. If it is a defence mechanism, then patients with a raised lipid profile may demonstrate a better prognosis. If it is a reflection of mitochondrial/ER dysfunction, then it could be associated with a poor outcome.

Objectives: To determine: (1) the prevalence (by gender) of raised lipid profile (cholesterol, LDH and triglycerides) in a large cohort of patients with ALS. (2) The relationship of lipid profile with the body mass index (BMI) through the ALS disease course. (3) Whether hyperlipidaemia develops with the progression of the disease or may be an early observation? (4) The implications (if any) of a raised lipid profile (cholesterol, LDH and triglycerides) on disease outcome or prognosis?

Methods: This is a prospective observational cohort study consisting of 512 ALS patients, recruited for the TRO19622 (Olesoxime) investigational medicinal product trial. Fasting serum concentrations of total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C) and high-density lipoprotein cholesterol (HDL-C) were measured at baseline and during follow-up.

Results: 73% of the participants had hypercholesterolaemia on the screening visit. The prevalence of hypercholesterolaemia decreased with time and was 64% at 18 months follow- up. On univariate analysis total cholesterol, LDL-C and LDL/HDL ratio had a statistically significant effect on survival (p = 0.015, 0.003 and 0.027, respectively). On multivariate analysis, however, none of the lipids were found to have a statistically significant effect on survival.

Conclusions: This study does not provide evidence for the lipid profile to be an independent prognostic factor in ALS.

Acknowledgements:

We are thankful to TROPHOS for providing the data of TRO19622 trial.

C67 GENETIC BACKGROUND EFFECTS ON LIFESPAN OF SOD1 MOUSE MODELS OF ALS

Sher R¹

Heiman-Patterson T²

Blankenhorn E²

Alexander G²

Deitch J²

Cox G³

^fuUniversity of Maine, Orono, ME, USA

^fvDrexel University College of Medicine, Philadelphia, PA, USA

^fwThe Jackson Laboratory, Bar Harbor, ME, USA

Email address for correspondence: [email protected]

Keywords: modifiers, genetic background, mouse

Background: Dominant SOD1 mutations account for ˜20% of familial forms of ALS. There is wide heterogeneity in age of onset and symptom severity within families carrying the same SOD1 mutation, suggesting that modifier genes significantly impact the disease. Similarly, there is variation in onset and severity in hSOD1Tg mice on genetically heterogeneous backgrounds.

Objectives: The goals of our studies were (1) to test the hypothesis that genetic modifiers can significantly affect the onset or progression of ALS symptoms in G93A mutant SOD1 transgenic mice, (2) to identify QTL loci associated with longevity, and (3) to identify candidate genes affecting the ALS phenotype.

Methods: We developed a range of inbred strains containing the SOD1-G93A mutation with varying lifespans. Long-lived and short-lived strains were used in reciprocal backcrosses for QTL analysis of modifier loci. Reciprocal congenics were used to narrow the QTL interval.

Results: We identified three inbred strains (ALR/LtJ, SJL/J, and NOD/LtSz-Rag1tm1Mom) that significantly accelerate disease, and three (C57BL/6J, DBA/2J, and BALB/cByJ) that significantly delay disease. Through reciprocal backcrosses between B6 & ALR and B6 & SJL lines, we have mapped a major QTL on Chr 17 (LOD 11.99) that significantly modifies the lifespan of G93A SOD1 mice.

Additionally, we crossed B6.SOD1-G93A with a B6.NOD-Chr17 congenic covering the entire QTL, and with a more distal B6.SJL-Chr17 congenic line. Lifespan for SOD1 NOD/B6 mice was 149.8 ± 9.0 d, and for SOD1 NOD/NOD mice was 136.4 ± 7.0 d, a statistically significant difference between groups and from the B6.SOD1 line (all p < 0.001). The decrease in lifespan is regulated in a dose-specific manner, with one NOD copy resulting in an ˜ 7.1% lifespan reduction, and two NOD copies resulting in an ˜ 15.4% reduction. We have also found that crossing the B6.NOD-Chr17 congenic to the B6.SOD1-G37R line decreases lifespan in a dose-regulated manner. In contrast, crossing to the distal B6.SJL-Chr17 congenic does not affect lifespan. A similar result was found when crossing the short-lived NOD.SOD1-G93A to a distal NOD.B10- Chr17 congenic, which does not increase lifespan. Interestingly, crossing to the distal B6.SJL-Chr17 congenic does decrease age of onset, indicating that the proximal region is critical for both onset and disease duration. We have resequenced the entire Chr 17 region for mutations, and in the proximal region we have identified 10 major genes with non-synonymous coding changes.

Discussion and conclusion: We have demonstrated that genetic background of hSOD1-G93A transgenic mice significantly affects lifespan, and that a region of Chr17 has a major dose-dependent effect on lifespan. The proximal region of this chromosome impacts lifespan for SOD1 mutant mice, while the distal region contains a modifier of onset.

Acknowledgments:

We gratefully acknowledge support from the ALS Hope Foundation and The Muscular Dystrophy Association.

C68 ATXN2 CAG REPEAT EXPANSIONS INCREASE THE RISK FOR CHINESE ALS PATIENTS

Liu X

Lu M

Tang L

Zhang N

Chui D

Fan D

Peking University Third Hospital, Beijing, China

Email address for correspondence: [email protected]

Keywords: ATXN2, polyglutamine, mainland of China

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder with unclear etiology. Recently, intermediate CAG repeat expansions in ATXN2, the gene responsible for spinocerebellar ataxia type 2 (SCA2), have been identified as a possible genetic risk factor for ALS. In this study, we analyzed the ATXN2 CAG repeat length in Chinese patients with ALS to evaluate the relationship between the genotype and phenotype. We studied 1,067 patients with ALS and 506 controls from mainland China (excluding Tibet). We collected clinical data and analyzed fluorescent PCR products to assess ATXN2 CAG repeat length in all of the samples. We observed that intermediate CAG repeat expansions in ATXN2 (CAG repeat length > 30) were associated with ALS (P = 0.004). There was no significant difference in clinical characteristics between the groups with and without intermediate CAG repeat expansions in ATXN2. Our data indicate that, for ALS patients from mainland China, intermediate CAG repeat expansions in ATXN2 increase the risk of ALS but have no effect on disease phenotype.

Acknowledgments:

This study was supported by grants from the National Natural Sciences Foundation of China (81030019), the Beijing Natural Science Foundation (7102161), Doctoral Fund of Chinese Ministry of Education (20100001110084), and the Major Projects of the National Science and Technology of China (2011ZX09307-001-07).

C69 GENOME-WIDE ASSOCIATION ANALYSES IN HAN CHINESE IDENTIFY TWO NEW SUSCEPTIBILITY LOCI FOR AMYOTROPHIC LATERAL SCLEROSIS

Deng M¹

Wei L²

Zuo X³

Wang K²

Ju X⁴

^fxMedical Research Center,Peking University Third Hospital, Beijing, 100191, China

^fyDepartment of Neurology,Anhui Medical University, No.1 Hospital, Hefei, Anhui, 230022, China

^fzState Key Laboratory Incubation Base of Dermatology, Ministry of National Science and Technology, Hefei, Anhui, 230032, China

^gaInstitute of Sports Medicine, Peking University Third Hospital, Beijing,100191, China

Email address for correspondence: [email protected]

Keywords: GWAS, new susceptibility loci, genetic

To identify susceptibility genes for ALS, we conducted a GWAS in 533 patients with sporadic amyotrophic lateral sclerosis (ALS) and 1,892 controls of Chinese Han. Ninety top SNPs suggested by the current GWAS and six SNPs identified by previous GWA studies were analyzed in an independent cohort of 706 ALS patients and 1,777 controls of Chinese Han. We discovered two new susceptibility loci for ALS at 1q32 (CAMK1G, rs6703183, P_combined = 2.92 × 10²⁸, OR = 1.31) and 22p11 (CABIN1 and SUSD2, rs8141797, P_combined = 2.35 × 10²⁹, OR = 1.52). These two loci explain 12.48% of the overall variance of disease risk in the Chinese Han population. We found no association evidence for the previously reported loci in the Chinese Han population, suggesting the genetic heterogeneity of disease susceptibility for ALS between ethnic populations. Our study discovered new genetic susceptibility factors and suggested new biological mechanisms of ALS.

Acknowledgments:

We thank all participants in this study and all neurologists at relevant hospitals for their help in the recruitment of subjects, This study was funded by National Natural Science Foundation of China (81072374, 31171048, 30973043, 30700906), the Key Project of the National Natural Science Foundation of China (91232717), the National Basic Research Program of China (2011CB707805), the Science and Technology New Star Funds of Beijing (2007A008 and 2009A04), the Beijing Science Foundation (7112146 and 7102159).

Reference:

• Deng M, Wei L, Zuo X et al. Nat Genet 2013;45(6): 697–701.
   PubMed Web of Science ®Google Scholar

C70 A GENOME-WIDE ASSOCIATION META-ANALYSIS IDENTIFIES A NOVEL LOCUS AT 17Q11.2 ASSOCIATED WITH SPORADIC AMYOTROPHIC LATERAL SCLEROSIS

Fogh I^1,2

Ratti A^2,3

Gellera C⁴

Sorarù G⁵

Cereda C⁶

Robberecht W⁷

Chiò A⁸

Meininger V⁹

Hardiman O¹⁰

Andersen PM¹¹

Glass JD¹²

Veldink JH¹³

Brown RH¹⁴

Landers JE¹⁴

Comi GP^3,15

D’Alfonso S¹⁶

van den Berg LH¹³

Al-Chalabi A¹⁷

Powell J¹

Silani V

The Slagen Consortium^2,3

^gbDepartment of Neuroscience, Institute of Psychiatry, King's College London, London, UK

^gcDept. of Neurology and Lab. of Neuroscience, IRCCS Istituto Auxologico Italiano, Milano, Italy

^gdDept Pathophysiol &Transplant, “Dino Ferrari” Center, University of Milan, Milano, Italy

^geUnit of Genetics of Neurodegenerative and Metabolic Diseases, Fondazione IRCCS Istituto Neurologico “Carlo Besta”, Milano, Italy

^gfDept. of Neurosciences, University of Padova, Padova, Italy

^ggLab. of Experimental Neurobiology, IRCCS ‘C. Mondino’, Pavia, Italy

^ghDepartment of Neurology, University Hospital Leuven, Leuven, Belgium

^giDepartment of Neuroscience, University of Torino, Torino, Italy

^gjUniversité Pierre et Marie Curie, Hôpital de la Salpêtrière, Paris, France

^gkTrinity Biomedical Sciences Institute, Trinity College Dublin, Dublin, Ireland

^glDepartment of Clinical Neuroscience, Umeå University, Umeå, Sweden

^gmDepartment of Neurology, Emory University, Atlanta, USA

^gnDepartment of Neurology,University Medical Centre Utrecht, Utrecht, The Netherlands

^goDepartment of Neurology, University of Massachusetts Medical School, Worcester, USA

^gpNeurologic Unit, IRCCS Foundation Ca’ Granda Ospedale Maggiore Policlinico, Milano, Italy

^gqInterdisciplinary Research Center of Autoimmune Diseases, ‘A. Avogadro’ University, Novara, Italy

^grDept.of Clinical Neuroscience, Institute of Psychiatry, King's College London, London, UK

Email address for correspondence: [email protected]

Keywords: genome-wide association study, meta-analysis, heritability

Background: While the genetic architecture of familial amyotrophic lateral sclerosis (ALS) is well characterized, that of the more common sporadic form is poorly understood with only the locus on chromosome 9p21 reliably replicated. At this locus expanded repeats in C9ORF72 gene have been identified as the main causative mutation in familial (23–47%) and sporadic (˜ 5%) ALS cases. As there is evidence for a strong genetic component in sporadic ALS with heritability estimated to be 0.61 in a recent study of 171 ALS twin pairs (1), estimation of additive genetic variance explained by common SNPs can contribute to explain the complex interactions between genes and environment.

Objectives: The major objective of this project was to identify new loci in sporadic ALS. We designed the largest GWAS meta-analysis study to date in ALS combining Italian and international genotype data. A second aim was to estimate the additive genetic variance explained by common SNPs.

Methods: The Italian SLAGEN Consortium collected a novel cohort of 3,959 individuals, while the international ALSGEN Consortium collected GWAS data from 11,611 individuals worldwide (2). After stringent quality controls on individuals and markers, genotype data was analysed for population stratification (EIGENSTRAT). Cleaned genotype data of each study were imputed genome wide (IMPUTE.v2) using the 1,000 Genomes Pilot project (June 2011) as reference panel. Single statistic tables were combined in a final meta-analysis weighting by the inverse of β-coefficients (METAL). In each study, heritability was estimated using the Genome-wide Complex Trait Analysis software that quantifies the additive genetic variance explained by all SNPs.

Results: We analyzed almost 7 million variants in 13,225 individuals (6,100 cases; 7,125 controls). We confirmed the previously reported association at 9p21.2 (rs3849943, P = 7.69 × 10⁻⁹) and identified a novel locus with genome-wide significance at 17q11.2 (P = 1.11 × 10⁻⁸) as well as suggestive evidence for a second locus at 18q11.2 (P = 7.67 × 10⁻⁸). Functional variants in LD with the lead SNPs were investigated by eQTLs analysis. The contribution of common variation to heritability was ˜12% (95% CI: 0.11–0.13) (data submitted).

Discussion and conclusion: We have identified a novel locus for sporadic ALS risk at 17q11.2, as well as suggestive evidence for a second locus at 18q11.2 and confirmed the association at 9p21. For the first time we have estimated heritability of sporadic ALS. In contrast with twin studies, polygenic variation attributable to common variation does not exceed 0.12. This difference suggests a substantial role for variation not captured by genome wide association studies that can be fulfilled by the detection of rarer variants.

References:

• Al-Chalabi A, Fang F, Hanby MF et al. J Neurol Neurosurg Psychiatry 20;81:1324–1326.
   Google Scholar
• ALSGEN Consortium et al. Neurobiol Aging 2012; 34:357.
   Google Scholar

C71 EXOME SEQUENCING TO IDENTIFY DE NOVO MUTATIONS IN SPORADIC ALS TRIOS

Chesi A

Gitler A

Stanford University, Stanford, CA, USA

Email address for correspondence: [email protected]

Keywords: exome sequencing, de novo mutation, chromatin

There have been several recent advances in defining the genetic landscape of ALS. These include discoveries of mutations in TARDBP, FUS/TLS, VCP, OPTN, UBQLN2, C9ORF72, and PFN1 as new ALS disease genes. Together with mutations in SOD1, the causes of over 50% of familial ALS cases have now been elucidated. Despite these extraordinary advances, all together mutations in these genes explain only a small percentage of sporadic cases (< 10%). A possible genetic mechanism for sporadic disease is de novo mutation – a mutation that arises spontaneously in the germline of one of the unaffected parents. Indeed, de novo mutations have recently emerged as contributors to neurodevelopmental disorders such as autism spectrum disorders, schizophrenia, and mental retardation. We performed the first systematic analysis of ALS trios (ALS patient and both unaffected parents). Because ALS is a late onset disease, trios for which DNA samples are available for patients and their parents are much rarer than for childhood disorders like autism. Nevertheless, we were able to assemble a collection of 50 ALS trios and we performed whole exome sequencing on all 150 individuals (50 × 3 = 150 exomes). To our knowledge, this is the largest collection of ALS trios assembled. We discovered a significant enrichment in de novo mutations in genes encoding chromatin regulators, including a de novo nonsense mutation in a neuronal chromatin remodeling complex component, SS18L1, and provide evidence that this mutation profoundly affects dendrite outgrowth when expressed in primary neurons. Resequencing this gene in an independent FALS pedigree identified an additional variant, which segregated with disease. Drugs modulating histone acetylation have shown protective effects in ALS mouse models and patient iPSC-derived motor neurons and have undergone phase 2 clinical studies in ALS subjects. Our results now reveal potential genetic connections to ALS as well. These results provide the first systematic analysis of de novo mutations in ALS (or any late-onset neurodegenerative disease) and reveal genes encoding chromatin regulators as new candidates for ALS genetic contributors. We propose that the specific genes we identify here, as well as their network of interacting partners (genetic and physical interactions), especially the other components of the SS18L1-containing chromatin remodeling complex are now candidates for evaluation in larger ALS patient cohorts.

C72 USING PUBLIC DATABASES OF GENETIC VARIATION TO TEST THE PATHOGENICITY OF REPORTED ALS MUTATIONS

Kenna K

McLaughlin R

Bradley D

Hardiman O

Trinity College, Dublin, Ireland

Email address for correspondence: [email protected]

Keywords: high penetrance, non-pathogenic, mutation

Background: Over 400 potential ALS mutations have been reported in the literature but the pathogenicity of most is uncertain.

Objective: To determine whether previously reported ALS mutations occur with too high a frequency among the general population to be disease causing.

Methods: Genomic coordinates and alternate alleles were established for 342 nonsynonymous mutations reported in the Amyotrophic Lateral Sclerosis Online genetics Database (ALSoD). This information was used to screen dbSNP, the 1000 genomes project and the NHLBI Exome Sequencing Project (ESP) for matching records. Where matches were identified, the associated database content was used to evaluate the potential for high penetrance disease causing effects.

Results: One hundred and forty-six of the mutations were identified within at least one of the three reference databases (dbSNP 142; 1000 genomes 28; ESP 56). Given the published lifetime risk of ALS and reported patient carrier frequencies, we determined that 51 mutations occurred too frequently within the 1000 genomes/ESP cohorts to cause ALS with high penetrance. Twenty-two of these mutations related to ‘causative’ ALS genes (ANG, DAO, DCTN1, FIG4, FUS, NEFH, OPTN, SETX, SOD1, TARDBP, and TAF15) while 29 related to tentative ALS genes (CDH13, CDH22, CRIM1, DIAPH3, FEZF2, GRB14, LUM, NETO1, OMA1, SOX5, SQSTM1, and SYT9). Notably, our results challenged the pathogenicity of the DAO:c.595C>T(p.Arg199Trp) mutation, suggesting that DAO may not represent a Mendelian ALS gene. We also found that six mutations mapped to 1000 genomes/ESP records could not be excluded as high penetrance ALS variants at the specified type I error rate (a = 0.05). The majority of mutations observed within dbSNP but not within the 1000 genomes/ ESP, had been curated based solely on the observation of carriers among individuals afflicted with ALS and/or other disease phenotypes.

Discussion: Our results cast serious doubt over the pathogenicity of 51 mutations previously associated with ALS. They also highlight the importance of allowing for variable expressivity and the chance inclusion of mutation carriers when using reference populations to evaluate variant pathogenicity. This has important implications for the conduct of patient resequencing studies.

C73 THE EFFECT OF TIRASEMTIV ON FUNCTIONAL STATUS IN PATIENTS WITH ALS

Shefner JM¹

Barragan D²

Bian A²

Andrews J²

Meng L²

Watson M¹

Lee J²

Wolff A²

Als Study Group Benefit²

^gsSUNY Upstate Medical University, Syracuse, NY, USA

^gtCytokinetics, Inc, South San Francisco, CA, USA

Email address for correspondence: [email protected]

Keywords: skeletal muscle activator, clinical trial, functional benefit

Background: Tirasemtiv is a fast skeletal muscle activator that sensitizes the sarcomere to calcium and increases the force of muscle contraction at submaximal simulation frequencies. In both single and multiple dose studies, it has been well tolerated in ALS patients, and dose-dependent improvements in measures of skeletal muscle strength and endurance were suggested.

Objectives: This study was designed to determine the safety, tolerability, and efficacy of tirasemtiv administered at up to 500 mg daily on patients with ALS.

Methods: Up to 500 patients with ALS will be enrolled from 73 centers in North America and Europe. Eligibility criteria include a slow vital capacity of greater than 50% of predicted, at least one moderately weak handgrip, and intermediate scores on at least four items in the ALSFRS-R. Patients are randomized to receive either placebo or tirasemtiv in a dose escalation protocol up to 500 mg daily given as 250 mg BID for a total of 12 weeks. Prior to randomization, all patients receive open label tirasemtiv 125 mg BID for 1 week to ensure that dose is tolerated and to allow adverse events to abate with continued treatment. Patients who tolerate tirasemtiv are randomized and begin a flexible 3-week dose escalation to each patient's maximum tolerated daily dose up to 500 mg. Efficacy measures include ALSFRS-R and measures of extremity and respiratory muscle strength and endurance obtained at 4, 8 and 12 weeks during double-blind treatment and at 1 and 4 weeks afterwards. Placebo-treated patients taking riluzole receive 50 mg BID; tirasemtiv-treated patients taking riluzole receive a reduced dose of 50 mg daily to account for the previously described increase in riluzole concentration caused by tirasemtiv.

Results: All patients will be enrolled in this study by July 2013, with last patient completing double-blind treatment by October 2013.

Discussion and conclusion: This study tests the hypothesis that tirasemtiv administered at tolerable doses for 12 weeks can increase skeletal muscle performance resulting in meaningful functional improvements in patients with ALS.

Acknowledgement:

We gratefully acknowledgement the participation of the Benefit ALS study management team and the study sites.

C74 EFFICACY OF ERYTHROPOIETIN IN AMYOTROPHIC LATERAL SCLEROSIS: A MULTICENTRE, RANDOMIZED, DOUBLE BLIND, PLACEBO-CONTROLLED, PHASE III STUDY (EPOS TRIAL)

Lauria G¹

Borghero G²

Capasso M³

Caponnetto C⁴

Chiò A⁵

Corbo M⁶

Eleopra R⁷

Fazio R⁸

Filosto M⁹

Giannini F¹⁰

Granieri E¹¹

La Bella V¹²

Logroscino G¹³

Mandrioli J¹⁴

Mazzini L¹⁵

Monsurrò MR¹⁶

Mora G¹⁷

Morino S¹⁸

Pietrini V¹⁹

Quatrale R²⁰

Rizzi R²¹

Salvi F²²

Siciliano G²³

Sorarù G²⁴

Volanti P²⁵

Filippini G for the Epos Trial Study Group¹

^guFondazione IRCCS Istituto Neurologico “Carlo Besta”, Milano, Italy

^gvU.O. Neurologia, Policlinico di Monserrato, Cagliari, Italy

^gwClinica Neurologica, Ospedale SS.Annunziata, Chieti, Italy

^gxDip. Neuroscience, Riabilitazione, Oftalmologia, Genetica, Salute Materno-Infantile, IRCCS A.O.U. San Martino IST, Genova, Italy

^gyCentro SLA, Dip Neuroscienze, A.O.U. Le Molinette, Università di Torino, Torino, Italy

^gzCentro Clinico NEMO, Milano, Italy

^haU.O. Neurologia, A.O.U. S.Maria della Misericordia, Udine, Italy

^hbDip. Neurologia IRCCS Ospedale San Raffaele, Milano, Italy

^hcClinica Neurologica, Università di Brescia, Brescia, Italy

^hdDip Scienze Mediche, Chirurgiche e Neuroscienze, Università di Siena, Siena, Italy

^heClinica Neurologica, Università di Ferrara, Ferrara, Italy

^hfCentro di Ricerca SLA, BioNeC, Università di Palermo, Palermo, Italy

^hgDip Neurologia e Psichiatria, Università di Bari, Bari, Italy

^hhDip. Neuroscienze, Ospedale S.Agostino-Estense, Modena, Italy

^hiCentro SLA, Dip Neurologia, Ospedale Maggiore della Carità, Università di Novara, Novara, Italy

^hjClinica Neurologica, Seconda Università di Napoli, Napoli, Italy

^hkFondazione Salvatore Maugeri IRCCS, Milano, Italy

^hlU.O Neurologia, Ospedale S. Andrea, Roma, Italy

^hmDip. Neuroscienze, Unità di Neurologia, Università di Parma, Parma, Italy

^hnU.O. Neurologia, Ospedale dell’Angelo, Mestre, Italy

^hoU.O. Neurologia, Arcispedale S: Maria Nuova, Reggio Emilia, Italy

^hpIRCCS Istituto Scienze Neurologiche, Bologna, Italy

^hqDip di Medicina Clinica e Sperimentale, U.O. Neurologia, Università di Pisa, Pisa, Italy

^hrDip Neuroscienze, Università di Padova, Padova, Italy

^hsU.O. Neuroriabilitazione Intensiva, Fondazione S. Maugeri Mistretta, Mistretta (ME), Italy

Email address for correspondence: [email protected]

Keywords: erythropoietin, clinical trial, treatment

We designed an independent, multicentre, phase III RCT to assess the efficacy of rhEPO 40.000 UI administered i.v. fortnightly as add on treatment to riluzole 100 mg daily over a 12-month period.

Inclusion criteria were laboratory-supported, probable, or definite ALS according to El Escorial revised criteria, age between 18 and 75 years, sVC 70% or higher, onset 18 months or less. Patients were randomized 1:1 to rhEPO or placebo based on a permuted-block design stratified to ALSFRS-R score (cut-off 33) and onset (spinal or bulbar). Primary outcomes were survival, tracheotomy or more than 23-h NIV for 14 consecutive days and adverse events causing withdrawal. Secondary outcomes were ALSFRS-R decline, sVC decline, and quality of life (ALSAQ-40). Sample size was estimated on 0.8 power (alpha 0.01, two tails).

Analyses were performed by intention-to-treat and per protocol. Twenty-five Italian centers participated in the study. We screened 545 and randomized 208 patients. One patient in rhEPO and seven patients in placebo were lost at enrolment, therefore 103 patients in rhEPO and 97 patients in placebo entered the study. At baseline, arms (rhEPO and placebo, respectively) were balanced as to gender (55 men and 50 men), mean age (59.4 ± 10; 58.6 ± 10), onset (spinal: 73.8%, 74.2%; bulbar: 26.2%, 25.8%), median ALSFRS-R score (40 (21–48); 39 (20–48)), median sVC (87% (37–110), 86% (23–114)), median ALSAQ40 (98 (43–200); 97 [52–161]), riluzole treatment (97%; 95%). At 12-month follow-up, 38 patients dropped-out and 1 was lost in rhEPO arm, whereas 39 patients dropped-out and 2 were lost in placebo arm. Overall rates for survival (9.7% and 7.2%) and tracheotomy or more than 23-h NIV (14.6% and 15.5%) were non- significantly different between rhEPO and placebo, even after stratification by onset and ALSFRS-R score at baseline. ALSFRS-R decline did not significantly change in rhEPO compared to placebo (−2, 95% CI 0–4), even after stratification by onset and ALSFRS-R score at baseline.

The proportion of adverse events causing drop-out was non-significantly different between rhEPO and placebo arms (16.5% and 8.3%, respectively). Only three events were considered probably associated to rhEPO treatment. At 18-month follow-up, the overall rates for survival (19% and 11.8%) and tracheotomy or more than 23-h NIV (22% and 21.5%) remained non-significantly different between rhEPO and placebo. In conclusion, add-on treatment with i.v. rhEPO 40.000 UI fortnightly did not reduce survival, tracheotomy or NIV rate, neither change the course of ALS over a 12-month period. Treatment was safe and well tolerated.

Funded by IRCCS Foundation “Carlo Besta” Neurological Institute, Milan, Italy.

EudraCT No: 2009–016066-91.

C75 ADDITIONAL FOLLOW-UP AND BIOMARKER DATA FROM A PHASE II SAFETY AND PRELIMINARY EFFICACY TRIAL OF NP001: A NOVEL IMMUNE REGULATOR FOR SLOWING PROGRESSION OF ALS

Miller RG¹

Block G²

Gopalakrishnan V²

McGrath M³

Study Group Np001 Phase II²

^htCalifornia Pacific Medical Center, San Francisco, CA, USA

^huNeuraltus Pharmaceuticals, Inc, Palo Alto, CA, USA

^hvUniversity of California, San Francisco, San Francisco, CA, USA

Email address for correspondence: [email protected]

Keywords: clinical trial, therapeutic treatment, neuroinflammation

Background: Abnormal inflammatory macrophages (AIM), systemically and locally in the CNS, are implicated in ALS progression. AIM activation is related to rate of disease progression suggesting that ongoing CNS inflammation may contribute to neuronal death. CNS AIM fuel the ongoing pathogenesis through production of cytokines that attract and drive further AIM migration into the CNS. NP001 is a novel immune regulator that lowers ALS-associated markers of AIM in vitro and in vivo. In a single-dose phase I trial in ALS patients, NP001 reduced blood AIM inflammatory biomarkers in a dose-dependent manner. NP001 is hypothesized to slow the progression of ALS by modulation of CNS inflammation.

Objectives: To assess the safety and preliminary efficacy of NP001 with additional, recently aquired biomarker and follow-up data, from a phase II trial.

Methods: One hundred and thirty-six patients were enrolled in a randomized, double-blind, placebo-controlled study. Patients met key entry criteria: FVC: > 70%, and onset of weakness < 3 years. Patients were randomized 1:1:1 to receive 6 months of NP001 1mg/kg/dose, 2 mg/kg/dose or placebo intravenously. Study drug was given as an induction cycle of five consecutive daily doses followed by five monthly cycles of three consecutive daily doses. Patients were seen monthly for 3 months post-dosing to assess durability of effect. The primary and secondary efficacy outcomes were ALSFRS-R slope, with and without matched historical placebo controls, and change from baseline over the 6-month treatment period. Additionally, a post-hoc assessment of non-progressors over the 6-month treatment period was conducted. Safety assessments were conducted throughout the trial. Compared with an earlier presentation, additional follow-up data and the blood inflammatory biomarkers, wrCRP and MCP-1, were assessed before and during the treatment and during a follow-up period without treatment.

Results: One hundred and thirty-six patients were randomized and 115 completed treatment. NP001 was generally safe and well-tolerated. NP001 2mg/kg showed a consistent pattern of slowing progression as assessed by ALSFRS-R slope or change from baseline by 13–21%. Patients with wrCRP greater than median at baseline had greater slowing of disease. NP001 2mg/kg halted disease progression in 27% of patients versus 11% on placebo. Additionally, the majority of responders remained stable during the follow-up period of 3 months without treatment.

Discussion: NP001 2 mg/kg had a modest clinical benefit on slope and unexpectedly halted disease progression in 2.5X as many patients compared to placebo. The absence of progression in most responsive patients, during a 3-month follow-up without treatment, suggests a long-acting effect. Trends in wrCRP support the anti-inflammatory mechanism of NP001. The unprecedented finding that NP001 2 mg/kg halted disease in a subset of patients and the overall benefit-risk support further development.

C76 IDENTIFICATION OF IMPROVED CLINICAL OUTCOMES AND CREATININE-SPARING EFFECT OF DEXPRAMIPEXOLE BASED ON SIGNIFICANT INTER-STUDY DIFFERENCES IN THE PHASE 2 AND PHASE 3 (EMPOWER) CLINICAL TRIALS IN ALS

Archibald D⁷

Brooks BR³

Mitsumoto H²

Rudnicki S⁶

Cudkowicz M⁴

van den Berg LH⁵

Moore D¹

Zhang B⁸

Mather J⁷

Petzinger T⁷

Bozik M⁷

^hwCalifornia Pacific Medical Center, San Francisco, CA, USA

^hxColumbia University, New York, NY, USA

^hyNeurology Carolinas Medical Center - University of North Carolina School of Medicine - Charlotte Campus, Charlotte, NC, USA

^hzMassachusetts General Hospital, Boston, MA, USA

^iaUniversity Medical Center Utrecht, Utrecht, The Netherlands

^ibUniversity of Arkansas Medical Center, Little Rock, AR, USA

^icKnopp Biosciences Inc., Pittsburgh, PA, USA

^idMacroStat, Inc., Hockessin, DE, USA

Email address for correspondence: [email protected]

Keywords: dexpramipexole, clinical, biomarker

Background: The recent apparent failure of dexpramipexole in the Phase 3 (EMPOWER) trial has heightened questions about strategies for drug development in amyotrophic lateral sclerosis (ALS). Critically examining the differences in ALS Phase 2 and Phase 3 study designs and conduct may identify factors underlying late-stage drug development failures as well as subpopulations more likely to benefit from particular interventions.

Objective: To determine if significant inter-study differences were present in a post-hoc analysis of the Phase 2 and Phase 3 dexpramipexole trials and whether such differences might account for conflicting trial results and lead to identification of responder subgroups for future clinical trials.

Methods: Baseline characteristics in the Phase 2 and EMPOWER trials were compared to determine whether significant differences were present. Significant differences in baseline characteristics between studies were assessed for their effect on the EMPOWER primary outcome, a Combined Assessment of Function and Survival (CAFS); on the components of CAFS (function, as measured by the ALSFRS-R change, and survival, as measured by the hazard for mortality), and their potential for identifying responder subgroups.

Results: Significant baseline differences were present in Phase 2/EMPOWER riluzole use (61%/75%, p = 0.002), El Escorial Criteria (EEC) definite ALS participants (46%/ 32%, p = 0.005), and symptom duration (14.0 months/ 15.2 months, p = 0.037). Participants with EEC definite ALS had significantly worse CAFS outcomes (468.0 vs. 405.9, p = 0.014) compared with not-definite participants among EMPOWER placebo-treated subjects.

EEC definite ALS (p = 0.013) and symptom duration (p < 0.001), but not riluzole use (p = 0.139), were significant predictors of CAFS outcomes. In the EMPOWER subgroup (n = 147) defined by riluzole use, EEC definite ALS, and short symptom duration (< 18 months), participants receiving dexpramipexole (n = 74) versus placebo (n = 73) had improved outcomes on CAFS (416.7/347.7, p = 0.059), ALSFRS-R slopes (−1.24/−1.67, p = 0.006), and mortality (H.R. 0.55, p = 0.080). Treatment with dexpramipexole also significantly reduced the decline from baseline (time averaged difference) in plasma creatinine over 12 months (4.71 μm/l, p < 0.001). The significance of this creatinine-sparing effect increased after adjusting for weight change in dex-treated and placebo-treated participants.

Discussion and conclusion: Significant differences were present in the baseline characteristics of participants enrolled in the Phase 2 and Phase 3 ALS trials of dexpramipexole. In a post-hoc analysis of EMPOWER subgroups selected for these differences, statistically significant benefits of dexpramipexole on ALSFRS-slope and creatinine-sparing and near significant benefits on CAFS and mortality were identified in the subgroup of riluzole-treated, short-symptom duration participants with definite ALS. This subgroup may represent ALS patients with more treatment-responsive disease-related events (faster ALSFRS-R decline/higher mortality) observed over the EMPOWER study period than ALS patients not meeting these criteria. These findings support additional therapeutic trials of dexpramipexole in ALS in a target population enriched for these characteristics, incorporating creatinine as a potential biomarker, and studied for more than 12 months.

C77 FETAL NEURAL STEM CELLS TRANSPLANTATION IN ALS: PRELIMINARY RESULTS OF A PHASE 1 CLINICAL TRIAL

Mazzini L¹

Gelati M^2,3

Profico D²

Sgaravizzi G²

Projetti Pensi M²

Muzi G²

Ricciolini C²

Rota Nodari L⁴

Carletti S⁵

Giorgi C⁵

Spera C⁵

Frondizzi D⁵

Bersano E¹

Nasuelli N¹

Querin G⁶

Masiero S⁶

Cantello R¹

Sorarù G⁶

Boulis NM⁷

Vescovi AL^3,4

^ieDepartment of Neurology, Eastern Piedmont University, Maggiore della Carità Hospital, Novara, Italy

^ifLaboratorio Cellule Staminali, Cell Factory e Biobanca, Terni Hospital, Terni, Italy

^igIRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo, Italy

^ihBiotechnology and Bioscience Department Bicocca University, Milano, Italy

ⁱⁱDepartment of Neurosurgery, “Santa Maria” Hospital, Terni, Italy

^ijDepartment of Neuroscience, University of Padova, Padova, Italy

^ikDepartment of Neurosurgery Emory University, Atlanta, USA

Email address for correspondence: [email protected]

Keywords: phase 1 clinical trial, foetal neural stem cells, transplantation

Background: There is no cure for ALS. Stem cell therapy represents a promising, perspective therapeutic option. Recently, a meta-analysis of 11 independent studies demonstrated that transplanted neural stem cells (NSCs) can slow both the onset and the progression of clinical signs and prolong survival in ALS mice. Our group has previously documented the integration capacity and potential therapeutic efficacy of hNSC in preclinical studies.

Objectives: We report on an ongoing Phase I trial, aimed at testing safety and feasibility of intraspinal injection of “clinical grade” (produced following the Good Manufacturing Guidelines in a pharmaceutical grade authorized facility) neural stem cells from natural miscarriages into a cohort of 18 ALS patients using a validated surgical apparatus and injection procedures.

Methods: The clinical GMP-status of the NSCs used in this study was granted by the Italian Medicines Agency (AIFA) and the clinical study was approved by the Italian Institute of Health (ISS) as well as by all of the competent ethical committees. An independent Safety Monitoring Board of multidisciplinary experts was also nominated and periodically reviewed and evaluated the accumulated study data.

Until now, six non-ambulatory patients have been recruited whom, following a 3 months observation period, received either unilateral (n = 3) or bilateral NSCs microinjections.(n = 3) into the lumbar spinal cord. For the latter cohort, a maximum of six microinjections was performed. NSCs concentration was 50,000 cells/microliter, for a total of 750,000 cells per injection site. All patients were treated with oral prednisone over 1 month and Tacrolimus in BID oral dosing, over 6 months. Patients are currently being monitored monthly by both standardized clinical and radiological assessment. All the clinical data were recorded in the national database for gene and somatic therapy of the ISS.

Results: No patients manifested severe treatment-related adverse events. The only treatment adverse event was pain noted immediately after surgery, which was confined to the injection sites and the corresponding dermatomes. This was reversible and disappeared after an average 6 days after surgery. Clinical assessments ranging from 6 to 10 months after transplantation demonstrated no evidence of acceleration of disease progression due to the treatment.

Discussion and conclusion: Our preliminary data confirm the procedural safety of this surgical procedure and show no evidence of immediate or delayed toxicity related to human NSC lines from the brain tissue of single fetuses, established under GMP guidelines and in the absence of ethical concerns, due to the origin of the tissue, derived from fetuses deceased by natural death, and its procurement according to the same international guidelines adopted for organ transplantation. We are now broadening the import of this trial, by testing intraspinal injections into the cervical spinal cord (C5–C6 level), of 12 ambulatory patients.

C78 ANALYSIS OF PATIENTS WITH AMYOTROPHIC LATERAL SCLEROSIS (ALS) TREATED WITH AUTOLOGOUS DIFFERENTIATED MESENCHYMAL STEM CELLS: A PHASE I/II AND IIA CLINICAL TRIAL

Karussis D¹

Petrou P¹

Offen D²

Argov Z¹

Gotkine M¹

Levi Y³

Vaknin Dembisnky A¹

Kassis I¹

Ben Hur T¹

Melamed E⁴

Gothelf Y³

^ilDepartment of Neurology & Laboratory of Neuroimmunology, at the Hadassah Hebrew University Medical Center, Jerusalem, Israel

^imFelsenstein Medical Research Center, Tel Aviv, Israel

ⁱⁿBrainstorm Cell Therapeutics Ltd, Petach Tikva, Israel

^ioRabin Medical Center, Tel Aviv University, Israel

Email address for correspondence: [email protected]

Keywords: mesenchymal, stem, cells

Objectives: To evaluate the safety and tolerability of treatment with autologous mesenchymal stem cells differentiated to secrete neurotrophic factors (‘MSC-NTF’) in ALS patients utilizing intramuscular (IM) and intrathecal (IT) administration.

Background: A previous pilot study from our group at Hadassah has shown the safety of IV/IT administration of unmodified MSC in ALS patients. The neuroprotective effects of MSC-NTF have been demonstrated in various animal models of neurodegenerative diseases, including ALS. In our clinic we are currently conducting the second part of two sequential clinical trials to evaluate the safety and tolerability of autologous MSC-NTF cells in ALS patients.

Methods: In our recently completed Phase I/II clinical study, MSC were isolated from the bone marrow of 12 ALS patients, expanded ex-vivo and induced to secrete neurotrophic factors such as GDNF and BDNF using BrainStorm’s NurOwn^TM technology. These autologous MSC-NTF cells were transplanted by IM (at 24 sites: 2 × 10⁵ cells per site) or IT (1 × 10⁶ cells/kg) injections to patients with early (ALSFRS score of > 30; n = 6) or advanced ALS (ALSFRS: 15–30; n = 6), respectively. All patients were followed up clinically on a monthly basis for a pre-treatment period of 3 months and for 6 months post-transplantation. Respiratory function tests, 3D-MRI of the muscles and compound muscle action potential amplitudes at three sites were used as additional surrogate markers of disease activity.

Results: During the six-month follow-up of the 12 transplanted patients, no serious treatment-related adverse events were observed, indicating short-term treatment safety. The clinical follow-up revealed a change in the rate of clinical progression (ALSFRS) and respiratory function (FVC) in favor of the IT-treated patients during the 6 months following treatment, as compared to the 3 months preceding treatment.

Conclusions: This first pilot trial in 12 patients with ALS showed that intrathecal or intramuscular injection of MSC-NTF is safe and revealed some indications of clinical beneficial effects. In the second part of the ongoing Phase IIa dose-escalating trial in our Center, 12 additional ALS patients are currently receiving combined IM and IT treatment with escalating doses of MSC-NTF cells, up to twice those administered in the Phase I/II trial. Initial observations from the first treated patients do not show any serious adverse events up to date. More detailed and updated data from this trial will be presented.

Trial registration: ClinicalTrials.gov (ClinicalTrials.gov Identifier: NCT01051882). The study is sponsored by BrainStorm Cell Therapeutics Ltd.

C79 OLIGODENDROCYTES: FROM BIOLOGY TO DISEASE

Casaccia P

Icahn School of Medicine at Mount Sinai, New York, USA

Email address for correspondence: [email protected]

Keywords: epigenetics, glia, chromatin

Recent advances have suggested an important role of myelin in plasticity and axonal survival. Myelin is formed by mature oligodendrocytes, which derive from oligodendrocyte progenitors. This presentation will address the progression from progenitors to myelinating cells, by reviewing the integration of extracellular factors with transcriptional networks and epigenetic modifiers. Epigenetic regulation of oligodendrocyte differentiation includes the study of histone-specific enzymatic activities, DNA methylation, and microRNAs. Epigenetic changes reflect the effect of environmental components on gene expression and as such, they play an important role in development and pathology.

C80 OLIGODENDROCYTES FROM THE ALS MOUSE MODEL AND ALS PATIENTS ARE TOXIC TO MOTOR NEURONS IN VITRO

Ferraiuolo L

Meyer K

Miranda C

Braun L

Kaspar B

Research Institute at Nationwide Children’s Hospital, Columbus, Ohio, USA

Email address for correspondence: [email protected]

Keywords: motor neurons, oligodendrocytes, co-culture

Background: In the past few years, it has become clear that non-neuronal cells are major players in dictating disease progression rate in ALS (1). It has been very recently reported that oligodendrocytes are pivotal in providing metabolic support to MN apart from regulating ion conductance through the axon (2).

Two studies so far have reported that the spinal cords of ALS patients as well as the mouse model are affected by morphological changes in grey matter oligodendrocytes that ultimately die during disease progression (2, 3). New oligodendrocyte precursor (NG2+) cells proliferate, but fail in reaching full maturation, thus leaving MN un-myelinated and deprived of metabolic support.

In this context, we have successfully developed in vitro systems to study oligodendrocyte differentiation and toxicity to MN from both the ALS mouse model and human samples.

Methods: Primary mouse NG2 + cells were isolated from the cortex of neonate (P2) SOD1^G93A and wild-type mice and either used 24h after isolation for co-culture with HB9:GFP+ MN or differentiated for 7 days into MBP+ oligodendrocytes and then co-cultured with HB9:GFP+ MN.

Human skin fibroblasts were converted to induced pluripotent stem (iPS) cells and these were then differentiated to tripotent neural progenitor cells (NPCs). NPCs were subsequently differentiated into MBP+/GalC+ oligodendrocytes by supplementing the medium with PDGF-AA and IGF-1 at different concentrations for one month.

Results: Data analysis showed that MBP+ oligodendrocytes from SOD1^G93A mice, but not NG2 + cells, are toxic to wild-type mouse MN after 6 days in co-culture, resulting in 40% decrease in MN survival and 50% decrease in axonal length.

Strikingly, human MBP+/GalC+ oligodendrocytes from ALS patients cause a dramatic decrease in HB9:GFP+ MN survival with only 50% of the cells surviving after 48h from plating and only 20% after 72h. Moreover, one of the most compelling characteristics of this co-culture system is the axonal phenotype displayed by MN plated onto ALS samples with marked axonal beading or retraction.

Finally, we developed a reliable tool to visualize mature oligodendrocytes in vivo using an AAV9 vector expressing green fluorescent protein (GFP) under the myelin basic protein (MBP) promoter in order to monitor oligodendrocyte fate during disease progression.

Conclusions: Our data confirm and build upon the present knowledge that oligodendrocyte cells are dysfunctional in ALS. Moreover, this study provides the first in vitro model to investigate the toxic properties of human ALS oligodendrocytes.

References:

• Ilieva H et al. J Cell Biol 2009.
   Google Scholar
• Lee Y et al. Nature 2012.
   Google Scholar
• Philips T et al. Brain 2013.
   Google Scholar

C81 ALTERED ASTROCYTIC EXPRESSION OF TDP-43 DOES NOT INFLUENCE MOTOR NEURON SURVIVAL

Haidet-Phillips AM

Gross Sk

Williams T

Tuteja A

Sherman A

Ko M

Jeong YH

Wong PC

Maragakis NJ

Johns Hopkins University, Baltimore, MD, USA

Email address for correspondence: [email protected]

Keywords: TDP-43, astrocytes, non-cell autonomous

Background: Recent studies have highlighted a role for glial cells in ALS pathogenesis. Nevertheless, the majority of studies implicating glia have focused on transgenic mouse models of mutant superoxide dismutase (SOD1) expression. Recently, mutations in tar DNA-binding protein 43 (TDP-43) have been linked to ALS. It is still unknown whether TDP-43 mutations cause ALS through a gain or loss of function mechanism, however, studies have indicated that regulation of TDP-43 expression is critical for maintaining normal function. Rodents with altered levels of TDP-43 expression develop an array of ALS phenotypes, but the contribution of astrocytes to TDP-43-linked ALS has not been investigated in these models.

Objectives: To determine whether astrocytes either overexpressing mutant TDP-43 or lacking TDP-43 can damage wild-type (WT) motor neurons (MNs) either in vitro in a co-culture system or in vivo after transplantation into the spinal cord of WT rats.

Methods: Glial-restricted precursors (GRPs) were isolated from mice overexpressing TDP-43^A315T or from Ubiquitin-ER-Cre;TDP-43^flox/- mice to use as a model for TDP-43 knockout. WT littermate and SOD1^G93A mice were used as controls. The GRPs were differentiated to astrocytes in vitro or transplanted to the spinal cord of WT rats for in vivo astrocyte differentiation. The effects of the astrocytes on WT MN survival were determined in vitro using a co-culture system or in vivo by examining host MNs at 3 months post-transplantation.

Results: Astrocytes overexpressing TDP-43 or lacking TDP-43 were cultured with WT MNs in vitro. Decreased MN survival was observed with SOD1^G93A astrocyte co-culture, but no change in MN survival was noted between WT astrocytes and astrocytes with alterations in TDP-43 levels. To examine the effects of TDP-43 alterations in astrocytes on WT MNs in vivo, WT, TDP-43^A315T, Ubi-ER-Cre;TDP-43^flox/- and SOD1^G93A GRPs were transplanted to the cervical spinal cord of WT rats, where they differentiate to astrocytes. To induce TDP-43 knockout in the transplanted Ubi-ER-Cre;TDP-43^flox/- cells in vivo, rats were injected with tamoxifen. Rats receiving SOD1^G93A astrocytes showed a marked decline in forelimb grip strength over time which correlated with a loss of cervical MNs. However, no MN loss or behavioral deficits were detected after transplantation of WT or TDP-43^A315T astrocytes or after knockout of TDP-43 in engrafted Ubi-ER-Cre;TDP-43^flox/- astrocytes in vivo.

Conclusions: Our data show that altering the levels of TDP-43 expression by either knocking out TDP-43 or overexpressing an ALS-linked TDP-43 mutant does not cause MN degeneration either in vitro or in vivo, implying that astrocytes may not be involved in human ALS caused by TDP-43 mutations. Our study highlights the probable heterogeneity in ALS disease mechanisms and underscores the importance of evaluating subsets of ALS patients for differences in disease pathways as novel genetic contributors are uncovered.

C82 MUTANT TDP-43 TRIGGERS ASTROCYTIC ACTIVATION AND IMPAIRED GLUTAMATE TRANSPORT IN PRIMARY CULTURES DERIVED FROM TDP-43A315T MICE

Lau C¹

Turner B¹

Muyderman H²

Beart P¹

^ipFlorey Institute of Neuroscience and Mental Health, University of Melbourne, Parkville, Victoria, Australia

^iqDiscipline of Medical Biochemistry and Centre for Neuroscience, School of Medicine, Flinders University, Adelaide, Australia

Email address for correspondence: [email protected]

Keywords: astrocytes, glutamate, TDP-43

Background: Cytoplasmic accumulation of pathological TDP-43 occurs in motor neurons and glial cells in ALS. The presence of TDP-43 pathology in astrocytes and astrocytic activation in ALS suggests non-cell autonomous effects of mutant or misfolded TDP-43 on motor neuron degeneration, as previously described for mutant SOD1-mediated ALS. Astrocytes can mediate injury to motor neurons by secretion of proinflammatory factors, reduced neurotrophic factor support and glutamate-mediated excitotoxicity. To address whether mutant TDP-43 action within astrocytes is harmful to motor neurons, we established and characterised primary astrocyte cultures expressing ALS-linked mutant TDP-43.

Objectives: To generate and characterise cell-autonomous effects of mutant TDP-43 on astrocytes and non-cell autonomous effects on motor neurons using primary cultures derived from transgenic TDP-43^A315T mice.

Methods: Primary astrocyte and motor neuron cultures were established from cortices and spinal cords of neonatal C57BL/6 or transgenic TDP-43^A315T mice. Astrocytes were cultured for 2 weeks and TDP-43 pathology, solubility and subcellular distribution was analysed using Western blotting and immunocytochemistry. Astrocyte morphology, proliferation and cytoskeleton were analysed using immunocytochemistry. Functional activity of astrocytes was examined using [³H]D-aspartate uptake assay. Astrocyte and motor neuron co-cultures were established to test non-cell autonomous effects of mutant TDP-43.

Results: Western blotting analysis showed similar levels of endogenous and mutant TDP-43 accumulation in both soluble and insoluble fractions of astrocyte cultures and no evidence for abnormal phosphorylation, ubiquitination or truncation of mutant TDP-43. Mutant TDP-43 was also predominantly localised to the nucleus in transgenic TDP-43^A315T astrocyte cultures. However, GFAP expression was significantly increased and [³H] D-aspartate uptake activity was significantly diminished in transgenic TDP-43^A315T astrocytes, compared to WT cultures. Astrocyte activation was not due to changes in microfilaments determined by F-/G-actin staining in TDP-43^A315T astrocytes.

Discussion and conclusion: Our findings demonstrate activation and impaired glutamate uptake in astrocytes derived from transgenic TDP-43^A315T mice, consistent with cell- autonomous effects of mutant TDP-43 on astrocytes. The toxic potential of TDP-43^A315T astrocytes to motor neurons is currently under investigation.

C83 HUMAN SPORADIC ALS AND RODENT FAMILIAL ALS PRIMARY ASTROCYTES ARE SELECTIVELY TOXIC TO SPINAL MOTOR NEURONS THROUGH THE SAME DEATH PATHWAY

Le Verche V^1,2

Re D B^1,2

Amoroso M W^1,3

Hoffman L¹

Kariya S^1,2

Ikiz B^1,2

Phani S^1,2

Nagy PL^1,2

Politi K^1,2

Lo Grasso P⁴

Wichterle H^1,3

Henderson C E^1,2

Mitsumoto H¹

Kariya S^1,2

Koolen M^1,2

Papadimitriou D^1,2

Nagata T^1,2

Przedborski S^1,2

^irCenter for Motor Neuron Biology and Disease

^isDepartment of Pathology and Cell Biology, Columbia University, New York, USA

^itProject ALS/Jenifer Estess Laboratory for Stem Cell Research, New York, USA

^iuDepartment of Molecular Therapy, Scripps Research Institut, Jupiter, USA

Email address for correspondence: [email protected]

Keywords: non-cell autonomous toxicity, astrocytes, motor neuron death

Although 90% of ALS cases are sporadic, the lion’s share of attention in ALS research has been paid, so far, to the remaining 10% that are inherited and present mutations in different genes. Rodents expressing human mutations for superoxide dismutase-1 (SOD1), the most commonly mutated gene in ALS families, develop a paralytic disorder that emulates clinical and pathological hallmarks of ALS. The cellular mechanisms leading to motor neuron death remain, however, largely unknown. The objectives of our research are to decipher the cellular and molecular mechanisms leading to motor neuron death in ALS and to identify new valuable therapeutic targets for the treatment of this debilitating disease.

Previously, we have shown in vitro that primary astrocytes produced from mutant SOD1 rodent models of ALS are selectively killing motor neurons through the release of soluble toxic factors (Nagai et al 2007). Recently, we successfully produced primary astrocytes from sporadic ALS (n = 6) and age-matched control (healthy (n = 7), chronic obstructive pulmonary disorders (n = 6) and Alzheimer’s (n = 3)) patient brain tissues. Here, we show that human sporadic ALS and control astrocytes exhibit a series of classical astrocyte markers, and a similar morphology and index of viability. However, human sporadic ALS astrocytes are selectively toxic to co-cultured mouse or human stem cell-derived motor neurons compared to astrocytes produced from control patients. Our data also suggest that both rodent and human ALS astrocytes recruit the same caspase- independent, but Bax/JNK3-dependent death pathway in motor neurons with features of necrosis. Furthermore, we provide evidence that this form of programmed necrosis relies on the receptor-interacting protein 1 RIP1 and mixed lineage kinase domain-like protein MLKL suggesting that necroptosis is the motor of cell death in both familial and sporadic ALS models.

Taken together, our findings suggest that astrocytes may also have a crucial role in motor neuron death in human ALS and, more importantly, that diseased astrocyte toxicity may be relevant to both the familial and the sporadic forms of ALS. JNK3, which is predominantly expressed in the central nervous system, is a promising molecular target for ALS therapy and is currently tested in a preclinical study in SOD1^G93A mutant mice.

C84 ALS/MND as a disease spectrum: time to leave the lumpers behind?

Strong M

Robarts Research Institute, London, Ontario, Canada, Western University, London, Ontario, Canada

Email address for correspondence: [email protected]

Keywords: frontotemporal dementia, motor neuron phenotype, genetics

While ALS/MND (ALS) has been traditionally described as being a single disease process, it is clear that this has never really been the case. Hyperendemic foci such as that described amongst the Western Pacific and Kii Peninsula and familial clustering long ago supported the concept that as a clinical disease entity, hence ALS must represent only the final expression of a broad range of pathological processes. Clinicians accept that any given individual will manifest with a largely predictable disease course, but that there is little consistency amongst individuals. We accept that ALS can be highly restricted and present with pure lower motor neuron, pure upper motor neuron, bulbar or neurocognitive deficits, and that these presentations are prognostically relevant. We further accept that there are individuals who will have a rapidly progressive disease course with survival of less than a year (a ‘malignant’ variant), a more typical course with survival of 3–5 years, or a markedly prolonged course with survival of greater than 10 years. The suggestion that young males with hand onset symptoms will be more likely find themselves in this latter, more benign category, suggests that the use of such clinical differentiators has important pathological correlates.

This thesis is further supported by the presence of frontotemporal dysfunction in a significant proportion of patients in whom disruption of higher cognitive networks finds expression in a broad array of neurocognitive deficits ranging from a frontotemporal dementia, a cognitive or behavior syndrome, or deficits in Theory of Mind (and thus reflective of more mesial frontal involvement). The critical point here is that none of this is reflective of pure motor involvement, but rather other, presumably equally discrete, cellular pathology.

The era of modern cellular and molecular biology further illustrates that ALS cannot be considered as a single disease process. The current genetic taxonomy of ALS highlights two broad pathways of disease process that include fundamental alterations in the processing of RNA, or in protein folding and degradation. Even within these two broad categories, the molecular biology is complex and again suggestive of discrete biologies, often associated with specific alterations in the expression of disease process as is illustrated by the spectrum of clinical manifestations even within a single mutated gene (SOD1).

In the end, the concept that ALS should be considered as a spectrum of biological processes of limited phenotypic expression, as we first proposed over 15 years ago still holds true and is now an inescapable conclusion. The next challenge will be to translate this knowledge into meaningful clinical trial design.

Acknowledgements:

Research supported by the Canadian Institutes of Health Research, the Michael Halls Endowment, and the Ontario Brain Institute.

C85 BEING PRO-ACTIVE - WHAT A CLINICAL TRIALS DATABASE CAN REVEAL ABOUT ALS

Zach N²

Kueffner R³

Shui A⁴

Sherman A⁵

Walker J⁵

Sinani E⁵

Katsovskiy I⁵

Schoenfeld D⁴

Stolovitsky G⁶

Norel R⁶

Atassi N^7,5

Berry J^7,5

Cudkowicz M^7,5

Leitner M¹

^ivPrize4Life, Cambridge, MA, USA

^iwPras L’Chaim, Tel Aviv, Israel

^ixLudwig-Maximilians University, Munich, Germany

^iyMGH Biostatistics Center, Boston, MA, USA

^izNeurological Clinical Research Institute, MGH, Charlestown, MA, USA

^jaIBM T.J. Watson Reseach Center, Yorktown Heights, NY, USA

^jbMassachusetts General Hospital, Harvard Medical School, Boston, MA, USA

Email address for correspondence: [email protected]

Keywords: clinicial trials, patient data, disease progression

Large datasets are critical for identifying statistically significant and biologically relevant observations, especially in rare disease like ALS. The Pooled Resource Open-access ALS Clinical Trials (PRO-ACT) platform provides an unprecedented opportunity to increase our understanding of the ALS patient population. The database consists of over 8500 ALS patients who participated in 17 clinical trials funded by industry and the non-profit and government sectors.

Data include demographic, family history, vital sign, clinical assessment, lab data, medication, and survival information. The PRO-ACT database was made open-access to researchers worldwide at the 2012 International Symposium on ALS/MND, and in the first 4 months the data have been downloaded by more than 125 different researchers from 18 different countries.

Analysis of the PRO-ACT data has revealed several novel and important findings which will be presented. The first of these is the identification of several novel baseline (ie prognostic) variables that significantly correlate with ALSFRS slope in a multivariate analysis (controlling for time from onset, age, gender, and baseline functional measures). The second is the demonstration of unexpected differences between active and placebo groups, in spite of well-matched demographics, riluzole use, and other enrollment criteria. In addition, several initial stratification models will also be presented.

Given the open-access nature of the PRO-ACT data, in addition to an internal analysis we undertook an innovative crowdsourcing initiative to shed light on the difficult challenge of ALS prognosis. The DREAM- Phil Bowen ALS Prediction Prize4Life was launched to incentivize the development of improved methods to accurately predict future change in ALSFRS at the individual patient level. This program brought in over 1000 solvers from around the world and led to the development of several valuable algorithms to predict the progression of ALS, with potential to aid both clinicians and future ALS clinical trials.

The challenge also led to the identification of new features predictive of ALSFRS progression that have now been verified using the full PRO-ACT dataset as well as in a recent large independent Phase III dataset, and these data will be presented.

These early results demonstrate the value of large datasets for developing a better understanding of ALS natural history, prognostic factors, and disease variables. More sophisticated and targeted analyses will continue to reveal new insights into this disease, which has for so long defied our understanding.

C86 ISOMETRIC MUSCLE TESTING USING HAND-HELD DYNAMOMETRY (HHD) IN A MULTICENTER ALS TRIAL

Shefner JM¹

Watson M¹

Schoenfeld D²

Hayden D²

Yu H²

Shui A²

Titus S²

Conwit R³

Cudkowicz M²

^jcSUNY Upstate Medical University, Syracuse, NY, USA

^jdMassachusetts General Hospital, Boston, MA, USA

^jeNational Institutes of Health, Bethesda, MD, USA

Email address for correspondence: [email protected]

Keywords: outcome measures, strength, clinical trial

Background: The decline in muscle strength is an important determinant of disability in ALS. Strength has been employed as an outcome measure in clinical trials for many years, with strength measured both qualitatively (MRC grading) or quantitatively with strain gauges. We have developed a method of assessing strength of multiple muscles using a hand-held dynamometer (HHD), and have assessed strength using HHD in the recently completed trial of ceftriaxone in ALS.

Objectives: To study changes in muscle strength assessed using HHD of ALS patients enrolled in the ceftriaxone study over the course of 1 year, and to compare HHD with other outcome measures employed in this trial.

Methods: Isometric muscle strength was measured over time in nine upper and lower extremity muscles bilaterally, for 513 subjects enrolled. All measurements were obtained with the subjects seated in a chair, with standard positions defined for each muscle for both evaluator and subject. Formal training of evaluators was conducted, and a criterion value of test–retest reliability on four normal subjects was required prior to patient testing. All muscles were tested to “break”, that is, the force required to induce movement in the muscle studied was recorded as the maximum force. Subjects were studied every 3 months for the duration of the study, but 1 year results are reported here.

Results: Data were evaluated using absolute force measurements, and Z scores were constructed for each value using both data from 240 normal subjects and from the screening visits for ALS subjects participating in the study. For each muscle, slope of decline over 1 year was determined, and the coefficient of variation (CoV) for rate of change calculated. The same measures were obtained for muscles combined as megascores from upper and lower extremities as well as a total body megascore. CoV for individual muscles ranged from −1.01 to −1.61. Megascore CoVs were −0.96 for upper extremities, −0.92 for lower extremities, and −0.85 for all muscles combined. Decline in muscle strength was highly correlated within a limb, and within body segment. Arm and leg muscles were less well correlated.

Discussion and conclusion: With formal training and validation, HHD provides a highly reliable method of measuring muscle strength in ALS patients. HHD is highly correlated with both ALSFRS-R and VC, and is comparable with these measures with respect to CoV for rate of change. Combinations of multiple muscles using megascores reduces CoV, and the high correlation of muscle strength within a limb allows for the use of a more restricted number of muscles than the 18 separate muscles evaluated here.

Acknowledgements:

We gratefully recognize the contributions of the NEALS project managment team and the NEALS study sites that participated in the Ceftriaxone in ALS clinical trial.

C87 BULBAR ALS: PREDICTING SURVIVAL FROM PHYSIOLOGICAL MEASURES OF SPEECH

Yunusova Y^1,2

Green JR³

Rong P³

Shellikeri S¹

Pattee G⁴

Zinman L⁵

^jfDepartment of Speech-Language Pathology, University of Toronto, Toronto, ON, Canada

^jgSunnybrook Research Institute and Health Science Centre, Toronto, ON, Canada

^jhInstitute for Health Professions, Massachusetts General Hospital, Boston, MA, USA

^jiMunroe-Meyer Institute, University of Nebraska Medical Center, Omaha, NE, USA

^jjDepartment of Neurology, University of Toronto, Toronto, ON, Canada

Email address for correspondence: [email protected]

Keywords: Bulbar, speech subsystems, survival

Background: Bulbar ALS is a fast progressing and perhaps the most devastating form of ALS. Only 25% of patients initially present with bulbar signs and symptoms, yet 75% of patients show bulbar disease as ALS progresses (1). The bulbar system is anatomically diverse and composed of the laryngeal, velopharyngeal, facial and oral articulatory subsystems. These subsystems are differentially impaired with previous reports suggesting that the tongue might be the earliest and most severely involved bulbar component (2). Disease survival has been estimated in ALS based on various factors. The prognostic value of bulbar decline for predicting survival has not been addressed with the exception of tongue weakness (3).

Objectives: The primary aim of this study was to identify the measures of bulbar function that affect survival across all speech subsystems.

Methods: One hundred and forty-six participants diagnosed with ALS (Mean age = 59.02 years, SD = 10.3) were recorded every 3 months for the average duration of 22.11 months (SD = 16.74). Thirty-four individuals presented with bulbar onset ALS, the remaining individuals presented with the spinal onset. The average ALSFRS score at the first session was 36.84 (SD = 6.56). The protocol and measurements are described in detail elsewhere (4). Survival was estimated as the number of months from symptom onset to death, and was available for 60% of participants. Patients those remained alive were censored at the time of the last recording session. The prognostic value of each predictor variable was estimated using the Cox proportional hazard analyses.

Results: The results showed that bulbar onset and %FVC were significant predictors of survival. Speech measures, including intelligibility and speaking rate, as well as indicators of velopharyngeal incompetence (eg nasal flow during syllable /pa/) and tongue dysfunction were able to predict survival.

Discussion and conclusion: Objective physiological measurements of bulbar dysfunction were able to predict survival in ALS. They are important to monitor at the time of diagnosis and as disease progresses. The ultimate goal of this work is to predict, based on the instrumental bulbar assessments, the course of bulbar disease progression for individuals and cohorts in clinical trials.

Acknowledgements:

This research was supported by NIH-NIDCD Grant#1R01DC009890-01A1, ALS Society of Canada Bernice Ramsay Discovery Grant, and the University of Toronto Connaught New Staff Matching Grant and the Barkley Trust at the University of Nebraska-Lincoln.

References:

• Millul A, Beghi E, Logroscino, G et al. Neuroepidemiology 2005;25:14–119.
   Google Scholar
• DePaul R. and Brooks BR. JSLHR 1993;36:1158–1167.
   Google Scholar
• Weikamp JG, Schelhaas HJ, Hendriks JC et al. J Neurol 2012;259:2360–5.
   Google Scholar
• Yunusova Y, Green J, Wang J et al. J. Vis. Exp. 2011;48.
   Google Scholar

C88 THE SOLEUS H-REFLEX DELINEATES UPPER MOTOR NEURONE PATHOPHYSIOLOGY IN AMYOTROPHIC LATERAL SCLEROSIS

Simon N^1,2

Lee M¹

Howells J³

Vucic S⁴

Lin C¹

Kiernan MC^1,2

^jkPrince of Wales Clinical School, University of New South Wales, Randwick, NSW, Australia

^jlNeuroscience Research Australia, Randwick, NSW, Australia

^jmSydney Medical School, The University of Sydney, NSW, Australia

^jnWestmead Clinical School, The University of Sydney, NSW, Australia

Email address for correspondence: [email protected]

Keywords: electrophysiology, biomarker, upper motor neuron

Background: The H-reflex has been used extensively to interrogate mechanisms of UMN dysfunction in neurological disorders, but its value in patients with ALS has not been established.

Objectives: The present study was undertaken to characterise the pathophysiological changes in the soleus H-reflex in amyotrophic lateral sclerosis (ALS), to relate these changes to the clinical phenotype, and to evaluate its potential use as an objective measure of upper motor neuron (UMN) dysfunction.

Methods: Recruitment curves of bilateral soleus H-reflex and M-wave were recorded with an automated system (QTRAC) in 28 patients with ALS and 15 age-matched control subjects. Analysed parameters included H_max:M_max ratio (H_max/M_max), and the minimum intensity required to produce a H-reflex and M-wave (H_thresh and M_thresh, respectively). The slope angles of the linear ascending portion of the H and M recruitment curves (H_θ and M_θ, respectively) were calculated and H_θ/M_θ was derived. In ALS patients, clinical UMN dysfunction was assessed with a quantitative scale (UMN Score, UMNS) grading hyperreflexia at the knee and ankle, and the presence of the Babinski response and ankle clonus. Additional clinical assessments included grading of thigh, leg and foot wasting (Wasting Score, WS), and a composite lower limb muscle strength score graded using the Medical Research Council grading system. The UMNS was incorporated into separate, preplanned, multiple linear regression models for H_max/M_max and H_θ/M_θ.

Results: H reflexes were identified in 92% of limbs of ALS patients, with absent H-reflexes only associated with marked reduction of M_max (0.11–1.30mV). Significant differences in H-reflex parameters were noted in the ALS group with reduced M_max (p = 0.001), H_max (p < 0.05), and M_θ (p < 0.05), and increased H_θ/M_θ (p < 0.05). Unexpectedly, H_max/M_max and H_θ were similar between groups. Both H_max/M_max and H_θ/M_θ were strongly predicted by clinical UMN dysfunction (p < 0.001), but H_max/M_max was also noted to decrease significantly with age (p = 0.001). Further analysis identified that H_max/M_max in ALS patients was predicted by the position of the H-recruitment curve relative to the M-recruitment curve.

Discussion: Parameters derived from the soleus H-reflex are closely linked with UMN dysfunction in ALS. Based on analysis of the relationships between H_θ, M_θ and other H-reflex parameters, it is hypothesised that alterations in lower motor neurone excitability following UMN injury may contribute to the changes in the H-reflex pathway in ALS. In addition, H_max/M_max may be more susceptible to the influence of age, and collision of the H-reflex with antidromically conducted impulses arising from directly stimulated motor axons in patients with ALS, and thus may not be considered the optimum measurement.

Conclusions: The present study demonstrated that measurement of the H-reflex provided insights into the pathophysiology of ALS and established the H-reflex as a robust objective measure of UMN dysfunction.

C89 MECHANISMS OF PRION-INDUCED TOXICITY

Aguzzi A

Institute of Neuropathology, University Hospital Zurich, Zurich, Switzerland

Email address for correspondence: [email protected]

Keywords: prions, mice, toxicity

Prions cause neurodegeneration in vivo, yet prion-infected cultured cells are asymptomatic. This has hampered mechanistic studies of prion-induced neurodegeneration. We have found that prion-infected cultured organotypic cerebellar slices (COCS) experienced progressive spongiform neurodegeneration closely reproducing prion disease, with three different prion strains giving rise to three distinct patterns of prion protein deposition. Neurodegeneration did not occur when the Prnp gene was genetically removed from neurons, and a comprehensive pharmacological screen indicated that was abrogated by compounds known to antagonize prion replication. Prion infection of COCS led to enhanced fodrin cleavage, suggesting the involvement of calpains or caspases in pathogenesis. Accordingly, neurotoxicity and fodrin cleavage were prevented by calpain inhibitors, but not by caspase inhibitors, whereas prion replication proceeded unimpeded. Hence, calpain inhibition can uncouple prion replication from its neurotoxic sequelae. These data validate COCS as a powerful model system that faithfully reproduces most morphological hallmarks of prion infections. The exquisite accessibility of COCS to pharmacological manipulations was instrumental in recognizing the role of calpains in neurotoxicity, and significantly extends the collection of tools necessary for rigorously dissecting prion pathogenesis.

The cellular prion protein PrPC consists of a globular domain (GD) hinged to a long N-proximal flexible tail (FT). We found rapid neurodegeneration in mice and in COCS exposed to holoantibodies, monovalent F(ab)1 fragments, or single-chain miniantibodies targeting the α1 and α3 helices of the GD. Degeneration was prevented by interstitial deletions within the FT and by treatment with various FT ligands, indicating that GD ligand toxicity was executed by the FT. Antibodies to the FT also prolonged the life of mice expressing a toxic PrPC mutant (PrPΔ94–134). These data uncover an essential role for the FT in two models of prion-related toxicity, and indicate that the FT triggers shared downstream effectors of neurodegeneration.

C90 IN VIVO PROPAGATION OF HUMAN WILD-TYPE SOD1 MISFOLDING IN A TRANSGENIC MOUSE MODEL

Grad L¹

Wang J¹

O’Neill M^1,3

Semenchenko V²

Wishart D⁴

Roskams J³

Cashman N¹

^joBrain Research Centre, University of British Columbia, Vancouver, BC, Canada

^jpNational Institute for Nanotechnology, Edmonton, AB, Canada

^jqDepartment of Zoology, University of British Columbia, Vancouver, BC, Canada

^jrDepartments of Biological Sciences and Computing Science, University of Alberta, Edmonton, AB, Canada

Email address for correspondence: [email protected]

Keywords: superoxide dismutase 1, propagated protein misfolding, mouse model

Background: One of the mechanisms by which a mutant or wild-type (Wt) protein can dominate pathogenesis of phenotypically diverse diseases is by propagated protein misfolding, such as that underpinning the prion diseases, which has been increasingly implicated in other neurodegenerative and systemic disorders (1, 2). A role for propagated protein misfolding in ALS is supported by the prion-like spatiotemporal progression of disease through the neuroaxis (3). Prion-like activity has been described for the cell-to-cell transmission of misfolding of mutant SOD1 (4), and we have reported that mutant SOD1 can confer its misfold on HuWtSOD1 in a species-dependent manner; however, mutation of misfolded SOD1 at residue tryptophan-32 (W32) abolishes this process (5). Here we report that propagated SOD1 misfolding can occur in an in vivo mouse model system.

Methods: Purified mutant SOD1 proteins G127X and G127X/W32S were injected into the left and right brain hemispheres (cortex area-primary motor cortex region), respectively, of human wild-type (HuWt) SOD1 transgenic mice and non-transgenic littermates. Tissue from the injection site was then analyzed by immunohistochemistry at 7 days, 1 and 2 months post-injection in order to detect if misfolded HuWt SOD1 is present, indicative of a propagated protein misfolding event induced by truncated SOD1 mutant G127X.

Results: Staining of misfolded HuWtSOD1 in mouse spinal cord was detected 1 and 2 months post-injection with purified G127X-SOD1 protein in cells surrounding the injection site using our SOD1 disease-specific antibodies. In contrast, no misfolded SOD1 staining was observed away from the injection site in tissue injected with G127X/W32S protein at the corresponding time-points. Likewise, misfolded HuWtSOD1 was not observed away from the injection site in non-transgenic mice as no endogenous HuWtSOD1 is available as a substrate for propagated protein misfolding.

Conclusion: Mutant misfolded SOD1 can impart its misfold on HuWtSOD1 in an in vivo mouse model. This process is species-specific as human G127X SOD1 does not propagate misfolding of endogenous mouse SOD1 and is restricted to the availability of residue W32, as previously described in vitro. Our results support the notion that cell-to-cell spread of SOD1 misfolding contributes to the systematic spread ALS pathology through the neuroaxis.

References:

• Guest WC et al. J Toxicol Environ Health A 2011;74(2–4): 1433–59.
   Google Scholar
• Prusiner SB. Science 2012;336(6088):1511–3.
   Google Scholar
• Ravits JM, La Spada AR. Neurology 2009;73(10): 805–11.
   PubMed Web of Science ®Google Scholar
• Munch C, O’Brien J and Bertolotti A Proc Natl Acad Sci USA, 2011;108(9):3548–53.
   Google Scholar
• Grad LI et al. Proc Natl Acad Sci USA 2011;108(39): 16398–403.
   Google Scholar

C91 OVEREXPRESSION OF HUMAN WILD-TYPE SOD1 HASTENS DISEASE ONSET AND INDUCES EARLIER PRESENCE OF MUTANT SOD1 AGGREGATES IN A MOUSE MODEL OF ALS

Tokuda E¹

Birve A²

Andersen PM²

Brännström T¹

Marklund SL¹

^jsDepartment of Medical Biosciences, Umeå University, Umeå, Sweden

^jtDepartment of Clinical Neuroscience, Umeå University, Umeå, Sweden

Email address for correspondence: [email protected]

Keywords: double transgenic mice, human wild-type SOD1, SOD1 aggregates

Background: Mutations in superoxide dismutase-1 (SOD1) cause familial amyotrophic lateral sclerosis (ALS). Since SOD1 mutations are mostly dominantly inherited, SOD1 mutant and wild-type SOD1 (SOD1^WT) are co-expressed in the patients. When both proteins are co-expressed in double transgenic mice, both mutant and SOD1^WT form aggregates, and SOD1^WT exacerbates disease course of mutant SOD1 mice. However, direct in vivo molecular behaviors of both proteins remain elusive.

Objectives: The aim of the present study was to determine direct molecular behaviors of both SOD1^WT and SOD1 mutant during disease course of double transgenic mice.

Methods: To allow distinction between SOD1^WT and SOD1 mutant in double transgenic system, we selected SOD1^G127insTGGG (SOD1^G127X) mice. Unique sequences of SOD1^G127X enable us separate both proteins even on conventional immunoblotting using mutually exclusive antibodies: one directed against the five non-native amino acids following Gly127 in SOD1^G127X, the other directed against the C-terminal 23 amino acids of SOD1^WT truncated in the mutant.

To generate double transgenic mice, homozygous SOD1^G127X mice were crossed with hemizygous SOD1^WT mice (line N1029). Disease onset was defined as the time when mice reached peak body weight. The end-point was defined as the age at which a mouse was unable to right itself within 5s after being pushed onto its side. Disease duration was determined as the period from disease onset to the end-point.

Spinal cords were harvested at three different stages of disease including presymptomatic, symptomatic, and terminal stage. For analysis of SOD1 aggregates, detergent-insoluble fractions were extracted from spinal cords, and the fractions were investigated using immunoblotting.

Results: The disease onset in double transgenic mice was significantly earlier than in SOD1^G127X mice by 49% from 377 ± 26 days to 192 ± 3.8 days. The lifespan was also markedly shortened from 422 ± 33 days to 238 ± 5.6 days, representing a decrease of 43%. There was no difference in the disease duration. The SOD1^G127X aggregates in double transgenic mice were significantly increased even at a presymptomatic stage. These aggregates were accumulated in a disease stage-dependent manner. SOD1^WT also formed aggregates in double transgenic mice, but they were found after the onset of symptoms. Insoluble murine SOD1 was not detected in double transgenic mice in all tested disease stages.

Conclusion: SOD1^WT aggregated in the double transgenic mice and induced earlier presence of SOD1^G127X aggregates. This resulted in acceleration of the disease onset.

C92 PROTEIN STABILITY AND NEURODEGENERATIVE DISEASE

Wright GSA

Antonyuk SV

Austin JA

Strange RW

Hasnain SS

Molecular Biophysics Group, Department of Structural & Chemical Biology, University of Liverpool, Liverpool, UK

Email address for correspondence: [email protected]

Keywords: protein stability, SOD1, TDP-43

Background: Protein stability and aggregation are an often recurring theme in the aetiology of neural disease. A synergistic interaction between mutant SOD1 aggregation propensity and instability strongly influences patient survival time after disease onset. In the final stages of SOD1 maturation, the copper chaperone for SOD1 (CCS) transfers copper and a disulphide bond to SOD1 leaving a stable and active enzyme. Complex formation between hCCS and SOD1 is mediated by the zinc-binding SOD1-like hCCSD2. Without effective post-translational modifications (PTMs), nascent SOD1 is structurally compromised. PTMs also play a large part in TDP-43 stability. Increased in vivo half-life is a common feature mutant and modified wild-type TDP-43. This characteristic shows an inverse correlation with patient age at onset of disease symptoms, however, its molecular cause is not known.

Objective: Determine how missense mutations and drug molecules affect the structure and integrity of TDP 43, SOD1, hCCS and the heterodimeric SOD1-hCCS complex.

Methods and Results: Using X-ray crystallography, we have discovered a novel binding site for catecholamine neurotransmitters in the SOD1 β-barrel at loop 2. We found cisplatin and bismaleimidoethane increase mutant SOD1 stability and completely inhibit aggregation, respectively. Using small angle X-ray scattering with online SEC, we have been able to describe wild-type hCCS, an R163W mutant hCCS and also the hCCS-SOD1 complex. In this last case, inclusion of ALS SOD1 mutations appears to maintain the overall conformation of the complex. However, we find that the disease-related R163W hCCS mutation reduces zinc binding by hCCSD2 and causes normally stable dimeric hCCS to monomerise. This mutant protein is also susceptible to thermal unfolding and is aggregation prone. Like Zn-apo wild-type, R163W hCCS cannot form the critical heterodimeric complex with SOD1. By contrast, we find that ALS and FTLD mutations in the TDP-43 nucleic acid binding domains increase the protein’s stability. This is maintained and elevated when DNA bound.

Discussion and conclusion: Amelioration of instability and aggregation is considered the target for SOD1 directed therapeutics. Cisplatin and bismaleimidoethane have previously been shown to inhibit aggregation and stabilise SOD1, respectively. We have shown that both characteristics are rescued by these compounds in a manner which is equal but opposite to the detrimental effects of SOD1 mutations. Our work with R163W hCCS indicates that it recapitulates the molecular phenotype of ALS SOD1 mutants; dimer instability and monomerisation leading to aggregation. These observations propose the question: Could SOD1 ALS be a result of ineffective interactions with hCCS? In contrast, TDP-43 displays a gain in stability conferred by ALS and FTLD-U mutations which do not cause unusual inter-domain interactions or unfolding, but may reduce protein turnover by inhibiting degradation.

C93 STUDYING AGGREGATION AND DISTRIBUTION OF TDP-43 IN MAMMALIAN CELLS USING BIARSENICAL LABELLING

Ng JSW¹

Barros TP¹

Esbjorner EK²

Christodoulou J³

Kumita JM¹

Luheshi LM¹

Dobson CM¹

^juUniversity of Cambridge, Cambridge, UK

^jvChalmers University of Technology, Gothenburg, Sweden

^jwUniversity of London, London, UK

Email address for correspondence: [email protected]

Keywords: TDP-43, protein aggregation, biarsenical labelling

Background: The aggregation and deposition of transactivation response DNA-binding protein 43 (TDP-43) in motor neurons is a key pathological feature in nearly all amyotrophic lateral sclerosis (ALS) cases (1). Despite this, relatively little is known about the mechanism of TDP-43 aggregate formation. To date, a detailed analysis of the dynamic processes leading to such aggregate deposits in live cells has not been feasible.

Objectives: We used the small tetracysteine tag (TC-tag) and corresponding biarsenical dyes (FlAsH and ReAsH), along with intermolecular FRET-based microscopy to analyse the protein dynamics of TDP-43 inside live mammalian cells.

Methods: Human neuroblastoma SH-SY5Y cells are transiently transfected with the full length (FL) or C-terminal fragment (CTF, a.a. 274–414) of TDP-43 bearing a TC-tag. These cells are then processed for microscopic (biarsenical dye staining, immunostaining and FRET) and biochemical analyses monitoring samples at different time points (24, 48 and 72 hours).

Results: At 24 hrs, FL-TC-FlAsH is distributed uniformly in the cell nucleus. However, after 48 hrs, FlAsH and FRET signals are detected in the cytoplasm, thus suggesting that FL-TC may be mis-localised and accumulates in the cytoplasm when it is overexpressed. Similar to the full-length protein, CTF-TC-FlAsH is initially distributed uniformly in the cytoplasm at 24 hrs, but higher FlAsH signal and FRET signal are detected after 48 hrs. Interestingly, different levels of FRET signal are observed when TDP accumulates, suggesting possible changes in the interactions between the molecules.

Discussion and conclusion: We demonstrate that the TC-tag and biarsenical-labelling technique enables the visualisation of the TDP localization for both the monomers and aggregated forms within live mammalian cells; in particular, we can monitor the changes in distribution and aggregation as it progresses over time. Such model will prove useful towards investigating how changes in the structure of TDP-43 influence its localisation and aggregation under both physiological and cell stress conditions, and how these processes may relate to the pathogenicity and progression of ALS.

Acknowledgement:

We cordially thank the Biotechnology and Biological Sciences Research Council (BBSRC), the Engineering and Physical Sciences Research Council (EPSRC) and the Wellcome Trust for funding the research at University of Cambridge and the Wenner-Gren Foundations and The Swedish Research Council for funding the microscopic experiments performed at Chalmers University of Technology. JSWN is funded by Cambridge Trust. JSWN also acknowledges support from Dr. J.Gregory and Dr. J.Yerbury.

Reference:

• Kabashi E et al. Nat Genet 2008;40:572–574.
   PubMed Web of Science ®Google Scholar

C94 STAGES OF PTDP-43 PATHOLOGY IN AMYOTROPHIC LATERAL SCLEROSIS

Brettschneider J^1,4

Del Tredici K⁴

Toledo J^1,2

Robinson J^1,2

Irwin D^1,2

Grossman M³

van Deerlin V³

McCluskey L³

Elman L³

Feldengut S⁴

Ludolph A⁵

Lee V^1,2

Braak H⁴

Trojanowski J^1,2

^jxCenter for Neurodegenerative Disease Research (CNDR), University of Pennsylvania School of Medicine, Philadelphia, PA, USA

^jyDepartment of Pathology and Laboratory Medicine, University of Pennsylvania School of Medicine, Philadelphia, PA, USA

^jzDepartment of Neurology, University of Pennsylvania School of Medicine, Philadelphia, PA, USA

^kaClinical Neuroanatomy Section, Department of Neurology, Center for Biomedical Research, Ulm, Germany

^kbDepartment of Neurology, University of Ulm, Ulm, Germany

Email address for correspondence: [email protected]

Keywords: TDP-43, neuropathology, spreading of pathology

Objective: To see if the distribution patterns of phosphorylated 43-kDa TAR DNA-binding protein (pTDP-43) intraneuronal inclusions in amyotrophic lateral sclerosis (ALS) permit recognition of neuropathological stages.

Methods: pTDP-43 immunohistochemistry was performed on 70 μm sections from ALS autopsy cases (N = 76) classified by clinical phenotype and genetic background.

Results: ALS cases with the lowest burden of pTDP-43 pathology were characterized by lesions in the agranular motor cortex, brainstem motor nuclei of cranial nerves XII-X, VII, V, and spinal cord α-motoneurons (Stage 1). Increasing burdens of pathology showed involvement of the prefrontal neocortex (middle frontal gyrus), brainstem reticular formation, precerebellar nuclei, and the red nucleus (Stage 2). In Stage 3, pTDP-43 pathology involved the prefrontal (gyrus rectus and orbital gyri) and then postcentral neocortex and striatum. Cases with the greatest burden of pTDP-43 lesions showed pTDP-43 inclusions in anteromedial portions of the temporal lobe, including the hippocampus (Stage 4). At all stages, these lesions were accompanied by pTDP-43 oligodendroglial aggregates. Ten cases with C9orf72 repeat expansion displayed the same sequential spreading pattern as non-expansion cases, but a greater regional burden of lesions, indicating a more fulminant dissemination of pTDP-43 pathology.

Discussion: pTDP-43 pathology in ALS possibly disseminates in a sequential pattern that permits recognition of four neuropathological stages consistent with the hypothesis that pTDP-43 pathology is propagated along axonal pathways. Moreover, the fact that pTDP-43 pathology develops in the prefrontal cortex as part of an ongoing disease process could account for the development of executive cognitive deficits in ALS.

C95 C9ORF72 REGULATE PROTEIN DEGRADATION PATHWAYS

Farg M¹

Yang S²

Sundaramoorthy V¹

Halloran M³

Soo KY¹

Blair I²

Atkin J¹

^kcDepartment of Biochemistry, La Trobe University, Victoria, Australia

^kdAustralian School of Advanced Medicine, Macquarie University, NSW, Australia

^keDepartment of Neuroscience, School of Psychological Science, La Trobe University, Victoria, Australia

Email address for correspondence: [email protected]

Keywords: C9orf 72, endosomal markers, autophagy, heterogeneous nuclear ribonucleic acids

Background: Intronic expansion of a hexanucleotide GGGGCC repeat in the chromosome 9 open reading frame 72 (C9ORF72) gene was identified as the major cause of familial amyotrophic lateral sclerosis (ALS).

Objective: Elucidation of C9ORF72 normal cellular function in cellular trafficking and protein degradation.

Methods: We investigated the role of C9ORF72 in protein trafficking in neuronal cell lines, using reporter constructs, immunofluorescence microscopy, immunoprecipitation and mass spectrometry to identify the most abundant proteins that interact with C9ORF72. We also performed immunohistochemistry on C9ORF72 positive ALS patient tissues.

Results: We demonstrated that C9ORF72 was expressed in both the nucleus and cytoplasm as vesicular-like structures, and it was also secreted into the extracellular space. C9ORF72 co-localized strongly with Rab 1, 5, 7 and 11, consistent with previous studies that it bears Rab guanine exchange factor (GEF) activity. A physical interaction between C9orf72 and these Rab proteins was confirmed using co-immunoprecipitation studies. Depletion of C9ORF72 using siRNA inhibited transport of Shiga toxin from the plasma membrane to Golgi apparatus, revealing that C9ORF72 regulates endocytosis. Immunohistochemistry of C9ORF72 ALS patient motor neurons revealed increased co-localisation between C9ORF72 and Rab 7 and 11 compared to control patients, indicating dysregulation of Rab-mediated trafficking in patients bearing the C9ORF72 intronic mutation. C9ORF72 also co-localised with markers of the autophagosome and siRNA treatment altered the ratio of LC3I:LC3II, demonstrating that C9ORF72 regulates autophagy-related endosomal trafficking. Investigation of proteins interacting with C9ORF72 using mass spectrometry identified two other proteins linked to ALS, ubiquilin-2 and heterogeneous nuclear ribonucleoproteins, hnRNP A2/B1 and hnRNP A1. Treatment of cells over-expressing C9ORF72 with proteasome inhibitors induced the formation of intra-nuclear aggregates of C9ORF72 suggesting that proteasome activity alters C9ORF72 cellular distribution.

Conclusion: This study demonstrates that C9ORF72 regulates endolysosomal trafficking and protein degradation pathways in neuronal cell lines. Furthermore, these data demonstrated that C9ORF72 is associated with heterogeneous nuclear ribonucleic acids (hnRNPs), suggesting that C9ORF72 may have a role in RNA metabolism and transport.

C96 ASSESSMENT AND MANAGEMENT OF CALORIC NEEDS IN ALS

Tandan R¹

Bromberg M²

Kasarskis E³

Mitsumoto H⁴

Simmons Z⁵

Matthews D¹

^kfUniversity of Vermont, Burlington, VT, USA

^kgUniversity of Utah, Salt Lake City, UT, USA

^khUniversity of Kentucky, Lexington, KY, USA

^kiColumbia University, New York, NY, USA

^kjPenn State University, Hershey, PA, USA

Email address for correspondence: [email protected]

Keywords: malnutrition, caloric balance, enteral feeding

Caloric under nutrition or marasmus is common in ALS and although multifactorial, largely results from caloric imbalance due to decreased intake or increased expenditure, singly or in combination. The adverse effects of malnutrition include loss of weight and BMI, alterations in body composition, decreased survival, and possible effects on Quality of Life (QOL) and general motor and respiratory functions (1). Decreased caloric intake has been reported in up to 90% of patients during the disease course (2). Increased caloric expenditure is suggested by 12–20% and 27% higher measured than calculated resting metabolic rate in about 48–62% of sporadic and 100% of familial ALS, respectively, suggesting intrinsic hypermetabolism (3). The hypermetabolism correlated with age, gender and fat-free mass. Nevertheless, other causes of increased caloric use have not been explored, particularly physical activity caloric expenditure and non-exercise activity thermogenesis (NEATS) that are germane to ALS, and could result from weakness, spasticity, fasciculations, cramps, emotional lability and respiratory insufficiency. In the late stages of ALS, however, patients become hypometabolic especially after gastrostomy and mechanical ventilation support (4).

Routine enquiry about nutritional habits and caloric intake at regular visits, and recording of weight, ALSFRS-R scores, FVC, and employment of a ‘nutritional’ equation such as the Harris–Benedict equation, are advocated (5). Recommended surrogate indicators of the need for caloric supplementation include weight loss of 5–10% over healthy state, symptomatic dysphagia and FVC less than 50% of predicted; however, none of these criterion takes into account caloric balance, particularly caloric needs. Equations currently used to predict caloric needs are imprecise and inaccurate as they do not account for the variable topography and severity of the disease, and resultant variations in physical activity (6). We have developed a modified equation which can be used in clinics, validated against caloric needs measured directly by the gold standard doubly labeled water technique, to accurately predict caloric requirements in patients with different severity and phenotype of the disease (7).

Malnourished patients are encouraged to maximize oral nutritional rehabilitation, and when unsuccessful caloric supplementation is recommended through an enteral tube, such as a percutaneous endoscopic gastrostomy (PEG) or radiologically inserted gastrostomy (RIG) (8). Morbidity and mortality are slightly lower for RIG than PEG tubes, although further study is needed. Successful enteral feeding stems weight loss, and improves QOL and possibly survival in patients (9). Formal criteria for timing, rate and extent of caloric replacement are needed. Further, caution needs to be exercised to prevent the development of the refeeding syndrome, with its antecedent effects of enhanced morbidity and mortality (10).

References:

• Desport JC, Preux PM, Truong TC et al. Neurology 1999; 53:1059–63.
   PubMed Web of Science ®Google Scholar
• Daniel I, Greenwood BA. Nutr Clin Prac 2013; 28:392–9.
   Google Scholar
• Funalot B, Desport JC, Sturtz F et al. ALS 2009; 10:113–7.
   Google Scholar
• Ichihara N, Namba K, Ishikawa-Takata K et al. ALS 2012;13:544–9.
   Google Scholar
• Rio A, Cawadias E. J Hum Nutr Diet 2007;20:3–13.
   PubMed Web of Science ®Google Scholar
• Kasarskis EJ, Mendiondo MS, Wells S et al. ALS 2011;12:17–25.
   Google Scholar
• Kasarskis EJ, Mendiondo MS, Kryscio R et al. ALS FTD (Submitted).
   Google Scholar
• Stavroulakis T, Walsh T, Shaw PJ et al. ALS FTD 2013; 14:96–104.
   Google Scholar
• Spataro R, Ficano L, Piccolli F et al. J Neuro Sci 2011; 304:44–48.
   PubMed Web of Science ®Google Scholar
• Braun MM, Osecheck M, Joyce NC. Phys Med Rehabil Clin N Amer 2012; 23:751–71.
   PubMed Web of Science ®Google Scholar

C97 A MULTI-CENTRE EVALUATION OF SECRETION MANAGEMENT IN PATIENTS WITH MOTOR NEURONE DISEASE (MND)

McGeachan A

Hobson E

Stephenson J

Shaw PJ

McDermott C

Sheffield Institute of Translational Neuroscience; the University of Sheffield, Sheffield, UK

Email address for correspondence: [email protected]

Keywords: saliva, management, symptoms

Background: Excessive saliva is an unpleasant and challenging to treat symptom in MND. Consequential problems include drooling, embarrassment, higher risk of aspiration and the soiling of clothes. There is little evidence to guide clinicians when managing these problems and a recent Cochrane review highlighted the need for more studies.

Objectives: To carry out an evaluation of the management strategies that are being used to treat excessive saliva by clinicians treating MND at the Sheffield MND clinic.

Methods: Retrospective case note review of deceased patients who attended the MND clinic in Sheffield between 2002 and 2012.

Results: The case notes of 518 individuals with MND were reviewed. Problems with excessive saliva were identified in 266 patients (51%). Management for excessive saliva was described in 186 cases; seven anticholinergics were administered in 11 preparations and 91 doses. Hyoscine and amitriptyline were the most common therapies, used in 164 and 91 patients, respectively. Botulinum toxin was used in 45 patients.

Hyoscine patches were the anticholinergic that was most often effective. The response to hyoscine was documented in 79 patients; symptomatic relief was experienced in 67 (85%). Of the patients who were benefitting from hyoscine, 34 (51%) required additional management as their disease progressed, 13 (19%) of whom had to discontinue the hyoscine patches because of adverse effects. In 12 patients (18%) the adverse effect responsible for the discontinuation was a skin reaction.

Neurobloc botulinum toxin was given to 35 patients. The outcome of injecting both the submandibular and parotid glands was recorded in 21 patients. Eighteen (86%) had some symptomatic improvement. The outcome of injecting only the parotid glands was recorded in six patients, two patients (33%) symptoms improved. The comparative incidence of adverse effects in those who did and did not have their submandibular glands injected were nines patients (47%) and one patient (13%), respectively.

Discussion and conclusion: Hyoscine patches are the most effective anticholinergic for reducing excessive saliva and finding ways to manage the skin reactions to these patches may allow more patients to better control their symptoms. The benefits of hyoscine patches seem more limited as a patient’s symptoms progress, and more invasive therapies such as salivary gland botulinum toxin injection may be required.

Neurobloc injections solely into the parotid glands are less often successful than injections involving the submandibular glands. Patients with troublesome excessive saliva may require injections into their submandibular glands to experience a benefit, whilst patients with milder symptoms could benefit from injections solely into the parotids.

C98 ORAL SECRETION SCORE (OSS) PREDICTS BEST CARE INTERVENTIONS AND OUTCOMES OF PATIENTS WITH ALS/MND USING NONINVASIVE VENTILATION (NIV)

Cazzolli P¹

Brooks BR²

Lewarski J³

McKim D⁴

Nakayama Y⁵

Chatburn R⁶

^kkALS Care Project, Canton, Ohio, USA

^klCarolinas Medical Center, Neurology, Carolinas Health Care System-University of North Carolina School of Medicine-Charlotte Campus, Charlotte, North Carolina, USA

^kmInvacare Corporation, Elyria, Ohio, USA

^knUniversity of Ottawa-CANVent Respiratory Rehabilitation Services, The Ottawa Rehabilitation Centre, Ottawa, Ontario, Canada

^koTokyo Metropolitan Institute of Medical Science, Tokyo, Japan

^kpRespiratory Institute-Cleveland Clinic Foundation, Lerner College of Medicine-Case Western Reserve University, Cleveland, Ohio, USA

Email address for correspondence: [email protected]

Keywords: noninvasive ventilation, oral secretions, airway management

Background: Adherence to NIV in ALS/MND patients is limited by worsening oral secretions. The Oral Secretion Score (OSS) developed to assess oral secretions has been validated in ALS patients from USA, Japan, and France (1,2,3).

Objective: To relate OSS to compliance with treatment interventions in ALS patients requiring NIV.

Methods: This prospective observational study involved 135 consecutive ALS/MND patients (M: 51%: F: 49%; median age 62 years; range: 31–86 years) in Ohio, USA who were followed prospectively during home visits from start of NIV until they died or transitioned to tracheostomy.OSS, categorizing the severity of oral secretions in relationship to changes in swallowing and coughing (4 = normal automatic saliva swallow; no drooling; 3 = minimal drooling; automatic saliva swallow decreased; 2 = moderate drooling; conscious swallow required; 1 = severe drooling; difficult conscious swallow; ineffective cough force; 0 = most severe drooling; conscious swallow impossible;no cough force), was measured at beginning of NIV and throughout follow-up in clinic and home visits. Adherence to NIV, tolerance to NIV, and effective use of Mechanical In-Exsufflation (MIE) related to OSS were assessed longitudinally. Time to failure analysis of compliance with NIV as a function of OSS was evaluated by standard statistical evaluations (MedCalc Statistical Software www.medcalc.org).

Results: At start of NIV, 57% (77/135) patients had no bulbar impairment and (OSS = 4). Of the rest, 43% (58/135) had bulbar signs (13%(17) (OSS = 3), 18%(24) (OSS = 2), and 13%(17) (OSS(= 1)). Median time on NIV for the 135 was significantly longer in ALS patients (OSS = 4) (10.3 months; 95% CI = 8.0–14.0 months) compared with (OSS < 4) (3.1 months; 95% CI = 2.1–4.0 months); p < 0.001). NIV was better tolerated in 118 patients (OSS = 2–4) significantly compared with 17 patients (OSS = 1) (Chi Square = 135; p < 0.01). MIE was used for airway clearance by 16% (22/135). Of these 50% (11/22) (OSS = 0–1) were intolerant and discontinued its use, while 50% (11/22)-7 (OSS = 4)-4 (OSS = 2–3) used MIE effectively to remove phlegm/thick mucus Chi Square = 22; p < 0.01). Pharmacological agents were used to reduce sialorrhea in 44% (60/135). In 40% (54/135) who had (OSS = 2–3), patients reported drugs were effective in reducing saliva, but not if (OSS = 0–1). Oropharyngeal suctioning was used to clear upper airway in 38% (51/135). In 29% (39/135) who had (OSS = 2–3), patients indicated suctioning was effective in clearing airway, but ineffective in maintaining airway if (OSS = 0–1).

Conclusions: Higher OSS at initiation of NIV is significantly associated with improved tolerance/adherence to NIV and continued use of this intervention. Noninvasive airway clearance is not achieved in patients (OSS = 1). Saliva control (via medications) and expectoration control (via MIE) was significantly more effective when (OSS = 2–4). Replication of these findings in additional patient cohorts may establish the usefulness of the OSS in disease management of patients with ALS undergoing NIV intervention.

References:

• Cazzolli PA et al. ALS 2010;11(Suppl.1):140.
   Google Scholar
• Nakayama Y et al. ALS 2010;11(Suppl.1):141.
   Google Scholar
• Abdelnour-Mallet M et al. ALS and FD. 2013;14:302–307.
   Google Scholar

C99 NOCTURNAL TRANSCUTANEOUS CAPNOGRAPHY IN ALS IS A RELIABLE AND NON-INVASIVE PARAMETER FOR DECIDING NON-INVASIVE VENTILATION IN ALS PATIENTS

Pageot N

Juntas-Morales R

Alphandery S

Camu W

ALS center, Montpellier, France

Email address for correspondence: [email protected]

Keywords: non invasive ventilation, transcutaneous capnography, decision criteria

Background: Non-invasive ventilation (NIV) may improve quality of life as well as survival in ALS patients. There is an agreement about the importance of using NIV as early as possible. However, the best criteria for starting NIV remain in dispute.

Objectives: To study the usefulness of nocturnal transcutaneous capnography (NTC) as a parameter for deciding NIV in ALS.

Methods: We studied clinical characteristics (age and site of onset, ALS duration, and ALSFRS score) and respiratory data (SVC, FVC, PImax, SNIP, nocturnal oxymetry, PaO₂, and PaCO₂) of patients with definite ALS, that had NTC between March 2012 and March 2013, as well as their outcome after the NTC exam (delay for starting NIV).

Results: There were 21 patients, 8 women and 13 men, with a mean age of 62.6 yrs. Mean disease duration was 22.7 months, six patients had bulbar onset, 15 had a spinal onset and mean ALSFRS score was 35.4/48. At NTC, the median PCO₂ of the population was 48mm Hg. We compared the groups below and above this median, comprising 10 and 11 ALS patients, respectively. At the time of submission, 100% of the patients with high PCO₂ are now under NIV vs 10% in the other group. Mean delay for NIV in the first group was 6.55 months. Comparison between two groups for clinical characteristics and respiratory data allowed only one significant difference to be noted: SNIP was significantly lower in patients with hypercapnia (23% vs 52, p = 0.01).

Discussion: Hypercapnia is one criterion for NIV in ALS patients. This measure is an invasive one, while NTC is non-invasive and may be obtained at home (this is the case for all the patients described here). Our data suggest that this measure is an option for following respiratory involvement in ALS patients and that it may help for the decision of NIV. It appears to be as pertinent as the SNIP but SNIP measure may be difficult to obtain/interpret particularly in bulbar patients.

Conclusion: The measure of PCO₂ transcutaneously is a reliable, easy to obtain and non-invasive technique. We believe that it has to be considered as an additional criterion for NIV decision.

Acknowledgements:

We thank LVL medical for their technical support for NTC at patients’ home.

References:

• Kim SM, Park KS, Nam H et al. PLoS One 2011; 30:(3).
   Google Scholar
• Bauman KA, Kurili A, Schmidt SL et al. Arch Phys Med Rehabil 2013;94(1):46–52.
   Google Scholar

C100 CAN NIV PARAMETERS SETTINGS AND CHANGES OVERTIME PREDICT FUNCTIONAL AND SURVIVAL OUTCOME IN ALS PATIENTS?

Braga AC¹

Pinto A²

Almeida JP¹

de Carvalho M³

^kqTranslational and Clinical Physiology Unit - Molecular Medicine Institute/Faculty of Medicine of Lisbon, Lisbon, Portugal

^krDepartment of PM&R - Centro Hospitalar Lisboa Norte, Lisbon, Portugal

^ksDepartment of Neurosciences -Centro Hospitalar Lisboa Norte, Lisbon, Portugal

Email address for correspondence: [email protected]

Keywords: non-invasive ventilation, funtional decline, survival

Background: Early NIV may play a critical role in survival, and in its compliance. However, adherence have only been evaluated regarding the number of hours of use, and yet others variables recorded in the equipment's software could also be useful.

Objective: We looked for potential predictors among ALS patients fully compliant to NIV.

Methods: In a prospective trial design, we followed-up 60 ALS patients from January 2008 to May 2012 that at the end of study were ascribed to two groups according to whether they were dead (G1) or alive (G2). They were all early ventilated, based on abnormal oximetry, phrenic nerve response or ALSFRS-R score less than 12. At each 3 months, patients were extensively evaluated with respiratory function testing and all data from the NIV equipment were downloaded and registered. Primary outcomes: ALSFRS functional decline and disease duration to death or end of study; Secondary outcomes: time to NIV; NIV use and parameters settings at NIV adaptation.

Results: No clinical or demographic differences were observed between groups at admission. Disease duration from symptoms onset and time to NIV adaptation were also non-significant, but disease duration correlated positively with maximal inspiratory pressure, IPAP and backup breathing rate and nocturnal SpO₂. At the end of study, there were significant differences showing better functional status in G2 with lower heart and breath rate and reduced time spent with lower SpO₂ and have higher IPAP pressures. Multivariate Cox regression analysis showed that IPAP pressures more than 18 cm H₂O was a significant preditor of survival (−2 Log-L 82.56; sig. 0.0004; B−0,36; CI−0.53 to 0.77; sig. 0.02) and a trend regarding a lower respiratory decline (−2 Log-L 124.03; sig. 0.000; B−2.46; CI 0.53–0.77; sig. 0.05).

Conclusions: For the first time, determinants of functional decline and survival are significantly related to parameters settings of NIV equipment.

C101 DIAPHRAGM FUNCTIONAL ANALYSIS AT THE UPPER AND LOWER SPECTRUM OF FORCED VITAL CAPACITY (FVC) IN ALS/MND: FVC INADEQUATELY ASSESSES DIAPHRAGM FUNCTION OR UPPER MOTOR NEURON INVOLVEMENT FOR STIMULATABILITY

Onders R^1,2

Elmo M¹

Kaplan C¹

Katirji B^1,2

Schilz R^1,2

^ktUniversity Hospitals Case Medical Center, Cleveland, Ohio, USA

^kuCase Western Reserve University School of Medicine, Cleveland, Ohio, USA

Email address for correspondence: [email protected]

Keywords: FVC, diaphragm, diaphragm pacing

Background: Forced vital capacity (FVC) is widely used as an indicator of prognosis in ALS, need for non-invasive ventilation (NIV), diaphragm pacing (DP) and inclusion/assessment of efficacy in clinical trials. Clinical trials may only accept patients above a FVC of 65% predicted yet these patients may have significant diaphragm dysfunction that could affect study results. Also patients with FVC greater than 65% may need diaphragm dysfunction addressed with NIV or DP. Patients with a FVC below 45% may be excluded from DP or gastrostomy tubes yet have a stimulatable diaphragm.

Objective: Analyze patients with FVC’s greater than 65% or less than 45% to determine the relationship between FVC and other diaphragm functional tests including assessing operative diaphragm response with direct stimulation.

Methods: Retrospective analysis of a prospective database at a single site under an IRB protocol after FDA approval in November 2011.

Results: Ninty-six ALS subjects were evaluated with 86 patients undergoing DP (10 had non-stimulatable diaphragm at surgery or with pre-operative testing with FVC ranging from 28% to 65%). Twenty-nine patients had a FVC above 65% (average: 79, range: 65–110). This subgroup 26 (90%) had at least one additional test: 23 (79%) had a MIP less than 60; 18 (62%) had an elevated hemidiaphragm on CXR; 23 (79%) had reduced diaphragm excursion of 2 cm or less on fluoroscopy; 15 (52%) had a measured amplitudes of less than 0.2 on phrenic nerve conduction study; five (17%) had hypercarbia with pCO₂ above 45; and three patients had sniff with paradoxical diaphragm motion (FVC of 92%, 79% and 73%). Twenty-four patients had FVC 45% or less. All had MIP less than 60 (average: 23, range: 11–46). All had a stimulatable diaphragm at surgery with 80% having simultaneous feeding tube. There was no peri-operative mortality. Using regression analysis sniff fluoroscopy of diaphragm function had the greatest correlation to MIP (r = 0.47) then CO₂ levels (r = 0.33) then FVC with the weakest (r = 0.29). In operative assessment of diaphragm stimulatability; in those with FVC greater than 65%, 70% measured strong diaphragm response and in those below 45% FVC group, 50% had the strongest diaphragm rating. Intra-operatively, three patients with the weakest diaphragm score had FVC greater than 80%. Survival analysis based on each one of these functional tests is ongoing.

Discussion and conclusion: In ALS, a FVC above 65% is considered adequate yet many can have considerable diaphragm dysfunction. If this is identified and the diaphragm has primary upper motor neuron involvement, then therapy such as DP can be used to improve diaphragm function, respiration and in unilateral cases prevent paradoxical movement which leads to atrophy, muscle elongation and damage. Patients with ALS and FVC below 45% can still have significant intact lower motor neurons with stimulatable diaphragm muscle units where DP can be utilized. ALS patients need a thorough analysis of diaphragm function to assist in prognosis and management of their disease.

C102 AN AMBULATORY MODEL OF NON-INVASIVE VENTILATION IMPLEMENTATION IMPROVES SURVIVAL IN MOTOR NEURONE DISEASE

Sheers N^1,2

Berlowitz DJ^1,2

Smith C³

Rautela L^1,2

Batchelder I^1,2

Howard ME^1,2

^kvVictorian Respiratory Support Service, Austin Health, Melbourne, Australia

^kwInstitute for Breathing and Sleep, Melbourne, Australia

^kxPeninsula Health, Frankston, Australia

Email address for correspondence: [email protected]

Keywords: non invasive ventilation, ambulatory model of care, survival

Background and Objective: Non-invasive ventilation (NIV) increases survival and quality of life in Motor Neurone Disease (MND) (1). Historically at our institution, NIV was implemented during a multi-day inpatient admission, but increased demand lead to prolonged waiting times. We developed and implemented an ambulatory model and report the prospective evaluation of its impact.

Methods: All people with MND referred for implementation of NIV 6 months before and after the ambulatory model commenced were included. The ambulatory model involved a 4-hour stay to commence ventilation, including mask fitting, bedside titration of the spontaneous-timed mode bi-level pressure ventilator (VPAP III ST and ST-A, Resmed, San Diego USA) and education. Follow-up consisted of in-laboratory polysomnography and outpatient attendance. Waiting time, hospital length of stay, adverse events and polysomnography data were reviewed before and after the model change.

Results: Twenty-nine patients were included in the analysis, with similar baseline characteristics (mean (SD) FVC% pred 51.8 (15.2) pre- vs 54.2 (17.4) post-model change, p = 0.72). After changing to the ambulatory model the median waiting time to commence ventilation fell from 33 to 14 days (p < 0.04) and adverse events reduced (4 of 17 (3 deaths, 1 acute admission) pre vs 0 of 12 post). Survival was also prolonged (median (IQR) 278 (51–512) vs 580 (306–1355) days; hazard ratio 0.41, p = 0.04). Despite poorer sleep quality on follow-up polysomnography, daytime CO₂ was not different.

Discussion and conclusion: The introduction of an ambulatory model to commence NIV resulted in more rapid initiation of ventilation with an associated improved survival in our tertiary domiciliary ventilation service. This model of care provided an efficient option for implementing NIV, with waiting time reducing by 19 days and a 24% reduction in adverse events. Efficacy of ventilation, as measured by daytime PaCO₂ was similar. The flexibility of the ambulatory care model provided for prompt scheduling of NIV trials in patients who were deteriorating regardless of inpatient bed capacity.

Criteria for referring for ventilation did not change during the audit period; patients post-model change did not appear to have been referred earlier in their disease course; however, they were commenced on ventilation more quickly once the decision to ventilate was made. Our data suggest that once a decision to ventilate has been made, delays in commencing NIV are clinically important and that alternative models of implementation can be effective.

Reference:

• Bourke SC, Tomlinson M, Williams TL et al. Lancet neurol 2006;5:140–7.
   PubMed Web of Science ®Google Scholar

C103 THE FUTURE OF ALS THERAPEUTICS

Miller TM

Washington University, St. Louis, Missouri, USA

Email address for correspondence: [email protected]

Keywords: therapeutics, antisense oligonucleotides, familial ALS

Despite multiple trials and many promising leads, the development of new therapeutics for ALS has remained challenging. Yet, the future of drug development for ALS is encouraging. The vast amount of genetic data has defined and will continue to define new targets for therapeutics. Novel tools to modulate these targets will turn an increasing number of these discoveries into new drugs. Methods for targeted therapies include small molecules, passive immunization, siRNA, and antisense oligonucleotides. Each of these will be discussed, with a focus on recent experience with antisense oligonucleotides. Antisense oligonucleotides designed to lower SOD1 mRNA and protein were recently shown to be safe in a Phase I trial for SOD1-related ALS. The pharmacokinetic data in humans from this study will be helpful for conducting future antisense oligo trials. Antisense oligonucleotides strategies may be used to lower other mRNAs, change splicing isoforms, or to inhibit regulatory RNAs such as miRNAs. Inhibition of the regulatory miRNA, miR-155, for example, extends survival in the ALS model SOD1^G93A mice. Targeted therapies for distinct populations of patients with ALS have now become possible and are likely to become increasingly important. Some of these therapies developed for a small subset of ALS, may be worthy of testing in a broader group of ALS patients especially where shared pathophysiology may be demonstrated.