SESSION 9A IN VIVO MODELS

C70 SMN DEFICIENCY ACCELERATES PROGRESSION IN A MOUSE MODEL OF AMYOTROPHIC LATERAL SCLEROSIS

TURNER B, PARKINSON N, DAVIES K, TALBOT K

MRC Functional Genetics Unit, Department of Physiology, Anatomy and Genetics, University of Oxford, United Kingdom

e-mail address for correspondence: [email protected]

Keywords: survival motor neuron, spinal muscular atrophy, SOD1

Background: SMA and familial ALS are fatal motor neuron diseases resulting from mutations in SMN and SOD1 respectively. SMN genotypes predicting lower SMN protein levels are associated with increased risk and severity of sporadic ALS, suggesting a potential modulating role. Furthermore, a major genetic modifying locus encompassing the SMN region is implicated in the delayed phenotype of transgenic familial ALS mice on different background strains. Thus, several lines of evidence suggest possible genetic or molecular interactions between SOD1 and SMN in ALS.

Objectives: To investigate SMN expression in familial ALS models and assess the impact of SMN genetic deficiency on the clinical phenotype and neuropathology of transgenic SOD1^G93A mice.

Methods: SMN expression was analysed in stably transfected NSC34 cell lines and spinal cords of transgenic mice expressing SOD1 mutations using quantitative PCR and immunoblotting. Transgenic SOD1^G93A mice were crossed with SMN heterozygous knockout mice and examined for motor function, survival and neuropathology. In addition, SMA type I mice were bred onto a background of wild-type SOD1 overexpression to determine rescue.

Results: ALS-linked mutant SOD1 expression depleted SMN protein levels in cellular and mouse models with significant downregulation from presymptomatic disease (60 days). Accordingly, genetic disruption of SMN significantly worsened rotarod performance and survival in transgenic SOD1^G93A mice. In contrast, transgenic elevation of normal SOD1 in SMA mice failed to modify severity, suggesting an interaction specifically between mutant SOD1 and SMN.

Conclusions: These results establish that genetic reduction of SMN enhances phenotypic severity in transgenic familial ALS mice, supporting association studies where reduced SMN gene copy number predisposed to sporadic ALS. Thus, SMN reduction may be an enhancing genetic modifier of ALS. Secondly, SMN protein was progressively diminished in spinal cords of transgenic ALS mice, suggesting a relationship to lower motor neuron degeneration. We therefore propose that SMN replacement and upregulation strategies being developed as therapies for SMA may have potential benefit in patients with ALS.

C71 A NOVEL MUTATION IN GLYCINE TRNA SYNTHETASE AMELIORATES SOD1G93A MOTOR NEURON DEGENERATION.

BROS-FACER V, BANKS G, WILLIAMS H, GREENSMITH L, FISHER E

Institute of Neurology, London, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: glycine tRNA synthetase, Gars, SOD1G93A

Background: We are studying several novel mouse mutants potentially with motor neuron degeneration in order to characterise their phenotypes and further our understanding of human motor neuron disease.

Objectives: We have positionally cloned and characterised (with a PhD studentship from the UK Motor Neurone Disease Association and other funding) a new mouse mutant that has a defect in the glycine tRNA synthetase gene, Gars. Mutations in the human Gars gene results in a range of clinical manifestations from Charcot-Marie-Tooth disease to severe infantile muscular atrophy. We have characterised this mouse and have gone on to cross it to SOD1G93A transgenics.

Methods: We have undertaken a range of studies (paper submitted) of our Gars mice and proceeded to produce Gars,SOD1G93A double mutant; these have been characterised for lifespan and physiological and histological status.

Results: Our mouse has a dominant mutation resulting in loss of grip strength and sensory deficits. When crossed to SOD1G93A animals, the Gars,SOD1G93A double mutants have a striking and significant extension of lifespan and show delay in the histological characteristics of motor neuron disease compared to their SOD1G93A littermates.

Discussion and Conclusions: Suprisingly, the SOD1G93A, Gars double mutants have a significantly increased lifespan compared to their SOD1G93A littermates. This extension of lifespan is, from current data, even more than the ∼20–28% shown in another cross between SOD1G93A and Loa (Legs at odd angles) mice that we and Chen, Popko and colleagues have previously published. Both Gars and dynein are involved either directly or indirectly in local translation in axons and in neurite outgrowth, suggesting an interesting connection between them that may relate to the protection they offer from SOD1G93A induced motor neuron death.

C72 LOSS OF THE HSJ1 MOLECULAR CHAPERONE EXACERBATES DISEASE PHENOTYPE IN SOD1^G93A MICE

MUSTILL W, NOVOSELOV S, DICK J, GREENSMITH L, CHEETHAM M

University College London, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: SOD1, HSJ1, chaperone

Background: The SOD1 ^{G93A + /-} mouse and SOD1 ^{G93A + /-} transfected cell cultures are commonly used models of ALS. Although the precise pathogenesis of ALS remains unclear, several lines of evidence now indicate that molecular chaperones play a role in this disease. For example, SOD1 mutants have been shown to alter the activity of molecular chaperones. Molecular chaperones facilitate refolding or proteolysis of mutant or damaged proteins. Therefore, a reduction in the activity of molecular chaperones will affect the ability of motor neurons to defend themselves against the toxic properties of mutant SOD1.

Human HSJ1 proteins are neuronal members of the DnaJ family of molecular chaperones. Recent studies have indicated that over-expression of HSJ1 proteins, both in vitro and in vivo, provide beneficial effects on various pathological states associated with protein misfolding and protein aggregation. For example HSJ1a has been shown to reduce protein aggregation associated with polyglutamine expansions in Huntington's disease and SMBA. [1], [2]. In this study we examined the role of HSJ1 proteins in the SOD1^G93A mouse model of ALS.

Objective: To investigate the effect of manipulating the HSJ1 gene on disease progression in SOD1^G93A mice crossed to HSJ1 knockout mice.

Methods: HSJ1 knock-out (KO) mice were produced by homologous recombination. SOD1^{G93A + /-} male mice were mated to HSJ1^+/- females, and the offspring were back crossed in order to produce homozygous knockouts, both with and without the SOD1^{G93A + /-} mutation.

This cross resulted in progeny of 4 different genotypes: SOD1; WT; HSJ1^-/- and HSJ1^-/-/SOD1. The effects of HSJ1 KO on disease progression was monitored by observation of behaviour and body weight. At 120 days of age, mice of each genotype were anaesthetised and prepared for in vivo physiological analysis of hindlimb muscle force and motor unit survival, as previously described [3].

Results: HSJ1^-/- mice were healthy, viable and fertile. At 120 days of age, we observed no significant difference in weight or force of hind limb muscles between HSJ1 KO and WT mice. Although there was a reduction in muscle weight and force in SOD1 and HSJ1^-/-/SOD1 mice, there was no significant difference between these 2 groups. By 120 days of age, as previously reported, there is a significant reduction in the number of motor units in EDL muscles of SOD1 mice (ANOVA; p < 0.001). Importantly, in HSJ1^-/-/SOD1 mice, there was a significant reduction in motor unit survival compared to SOD1 mice (p < 0.05).

Discussion and Conclusions: These results suggest that HSJ1 may play an important role in motor neuron survival and indicate that the absence of HSJ1 chaperones in HSJ1^-/-/SOD1 mice increases the extent of motor neuron degeneration observed in SOD1 mice at 120 days. We are currently examining the mechanisms underlying these deficits and are investigating the effect of HSJ1 over-expression in SOD1 mice.

C73 A COPPER-BIS (THIOSEMICARBAZONATO) COMPLEX DELAYS DISEASE PROGRESSION AND INCREASES SURVIVAL IN A TRANSGENIC SOD1^G93A MOUSE MODEL OF ALS

SOON CPW¹, BARNHAM KJ¹, CROUCH PJ¹, LI SA¹, LAUGHTON KM¹, MASTERS CL¹, WHITE AR¹, DONNELLY PS², LI QX¹

¹Department of Pathology, Centre for Neuroscience, The University of Melbourne and The Mental Health Research Institute, Parkville ²School of Chemistry, The University of Melbourne, and Bio21 Molecular Science and Biotechnology Institute, The University of Melbourne, Parkville, Victoria, Australia

E-mail address for correspondence: [email protected]

Keywords: transgenic SOD1G93A, matrix metalloprotease-9, metal complex

Background: Amyotrophic Lateral Sclerosis (ALS) is an adult-onset fatal neurodegenerative disorder that involves progressive deterioration of motor neurons. It is clinically manifested by weight loss, muscle wasting, and spasticity leading to paralysis and eventually death through respiratory failure. Although the aetiology of this debilitating disease remains unclear, more than 100 different mutations in the copper-zinc superoxide dismutase (SOD1) gene can cause familial ALS, implicating a role for SOD1 in ALS pathogenesis. Transgenic mice overexpressing human mutant SOD1^G93A (TgSOD1^G93A) produce a phenotype that closely replicates both clinical and pathological hallmarks of human ALS. Mutations in SOD1 are believed to induce a gain in cytotoxic function causing degeneration of the motor neurons, possibly via oxidative stress. Diacetylbis(N⁴-methyl-3-thiosemicarbazonato) copper(II) (Cu(ATSM)) is a metal complex that crosses the BBB and has the potential to modify ALS through inhibition of oxidative stress.

Objectives: This study was carried out to evaluate the physical and biochemical effects of Cu(ATSM) treatment on the TgSOD1^G93A murine model of ALS.

Methods: TgSOD1^G93A mice in C57B6 background with delayed phenotype due to low copy of transgene were treated with Cu(ATSM) (n = 14) or vehicle (n = 18). Treatment was orally administered 5 days per week and commenced at the pre-symptom age of 140 days. Clinical assessment and motor function tests including rotarod and stride length were performed for all mice, and zymography was also performed to investigate matrix metalloprotease-9 (MMP-9) activity in spinal cord and serum samples.

Results: Disease onset was significantly delayed in TgSOD1^G93A mice treated with Cu(ATSM) (mean onset age (±SEM) of 261±5.4 days, compared to 241±1.7 days for vehicle treated control mice, p < 0.001). Cu(ATSM) also extended the life span of TgSOD1^G93A mice by 37 days (14%, p<0.0001). MMP-9 activity, which decreases as disease symptoms develop, was significantly restored to pre-symptomatic levels in Cu(ATSM) treated TgSOD1^G93A mice.

Discussion: Delayed development of ALS-like symptoms in Cu(ATSM) treated TgSOD1^G93A mice indicates that Cu(ATSM) may prevent motor neuron deterioration caused by the SOD1^G93A mutation. The mechanism of action may involve the compound's potential to restore MMP-9 activity. Further studies will be carried out to investigate the effects of Cu(ATSM) on oxidative stress and motor neuron survival in these mice. This will help define the mechanisms of motor neuron degeneration in ALS, and will facilitate the development of new therapeutic strategies.

C74 DROSOPHILA AS A MODEL SYSTEM TO ELUCIDATE THE MOLECULAR MECHANISMS UNDERLYING MOTOR NEURON DISEASES

CHAI A, WITHERS J, PARRY K, PATTERSON H, YU BIN, PENNETTA G

University of Edinburgh, United Kingdom

E-mail address for correspondence: [email protected]

Keywords: Drosophila, VAPB, Neurodegeneration

Background: Motor Neuron Diseases (MNDs) encompass a group of inherited neurodegenerative disorders characterized by selective dysfunction and death of motor neurons leading to spasticity, muscle atrophy and paralysis. In 2004, hVAPB (human VAMP-associated protein B) was shown to be the causative gene of a clinically diverse group of MNDs in humans including Amyotrophic Lateral Sclerosis (ALS), atypical ALS and late-onset spinal muscular atrophy. In recent years, Drosophila has proven to be a very powerful and flexible model system for studying human neurodegenerative diseases such as Alzheimer's, Parkinson's and Huntington's diseases. Despite the progress made in the previous pathologies, research on MNDs has been stalled somewhat due to the lack of a versatile genetic system.

Objectives: To understand the patho-physiology underlying VAP-induced MNDs in humans, we undertook a detailed functional characterization of VAP proteins in flies [1].

Methods: By using transgenic and loss-of-function approaches we generated a model of VAP-induced MNDs in Drosophila. Extensive phenotypic analysis was performed to dissect the patho-mechanism underlying these devastating diseases.

Results: We found that Drosophila VAP-33 (DVAP-33A), the structural homologue of hVAPB in flies, regulates synaptic remodelling by affecting the size and number of boutons at neuromuscular junctions (NMJs). Associated with these structural alterations, are compensatory changes in the physiology and ultrastructure of NMJs which maintain synaptic transmission within functional boundaries. Loss of DVAP-33A also induces axonal path-finding defects in Mushroom Bodies (MBs), the brain centres controlling learning and memory in flies. This phenotype can be rescued by targeting the expression of DVAP-33A in MBs. We also found that hVAPB and DVAP-33A are functionally interchangeable since the expression of hVAPB in neurons rescues the lethality, the morphological and the electrophysiological phenotypes associated with DVAP-33A loss-of-function mutations. Moreover, transgenic expression of hVAPB in neurons induces phenotypes similar to the overexpression of DVAP-33A. These data clearly indicate that the human and the Drosophila proteins perform homologous functions at the synapse. We also found that transgenic expression of DVAPP58S (the Drosophila protein carrying the pathogenic mutation) in neurons recapitulates several hallmarks of the human diseases, including locomotion defects, neuronal apoptosis and aggregate deposition. We are currently performing a modifier screen aimed at identifying VAP-interacting proteins and the relative data will be presented.

Conclusions: These findings point to a possible role of hVAPB in synaptic homeostasis and emphasize the relevance of our model in elucidating the patho-physiology underlying motor neuron degeneration in humans.

C75 AN SOD1 MISSENSE MUTATION IN DOGS WITH DEGENERATIVE MYELOPATHY: A SPONTANEOUS ANIMAL MODEL FOR AMYOTROPHIC LATERAL SCLEROSIS

COATES J¹, JOHNSON G¹, AWANO T¹, KATZ M¹, JOHNSON G¹, TAYLOR J¹, PERLOSKI M², BIAGI T², KHAN S¹, O'BRIEN D¹, WADE C², LINDBLAD-TOH K²

¹University of Missouri, Columbia, MO, United States, ²Broad Institute of Harvard and MIT, Cambride, MA, United States, ³Uppsala University, Uppsala, Sweden, Sweden

E-mail address for correspondence: [email protected]

Keywords: Canine, Degenerative Myelopathy, SOD1 mutation

Background: Canine degenerative myelopathy (DM) is an adult-onset neurodegenerative disease, characterized by progressive pelvic limb paresis and ataxia.[1] If euthanasia is delayed, the clinical signs will ascend causing flaccid tetraparesis/plegia. Axonal and myelin degeneration of the spinal cord is consistently most severe in the dorsal portion of the lateral funiculus within the mid-thoracic to lumbar region.[1], [2] Although DM is most commonly diagnosed in German Shepherd Dogs, it occurs in many other breeds and has been reported most recently in the Pembroke Welsh Corgi.[2] Amyotrophic lateral sclerosis (ALS) is an adult-onset, progressive paralysis of humans characterized by loss of motor neurons and sclerosis of the lateral funiculus. Mutations in the superoxide dismutase 1 (SOD1) gene cause some forms of familial ALS.

Objectives: The purpose of this study was to identify the gene and mutation responsible for canine DM.

Methods: Genome-wide association mapping of DM was performed with 38 cases and 17 controls from the Pembroke Welsh Corgi breed using the Affymetrix Canine Genome 2.0 Array^TM.[3] The strongest association was detected on CFA31 (p_raw=1x10⁻⁵,p_genome=0.18) where all affected dogs were homozygous for a common haplotype from 28.91 Mb to 29.67 Mb (CanFam2.0) containing 3 genes: SOD1, TIAM1 and SFRS15. Clinical similarities between DM and familial ALS made SOD1 a viable candidate gene. Exons 2 to 5 of canine SOD1 were resequenced from DM-affected and normal dogs.

Results: Resequencing of SOD1 in normal and affected dogs revealed a G to A transition, resulting in an E40K missense mutation. Homozygosity for the A allele was associated with DM in five dog breeds (Boxer, Pembroke Welsh Corgi, German Shepherd Dog, Chesapeake Bay Retriever, and Rhodesian Ridgeback). To verify our localization of the DM mutation, we fine mapped 63 SNPs across a 1.9 Mb region in five breeds which segregate for DM. Affected dogs from all five breeds share a five SNP haplotype surrounding the E40K mutation and no other haplotype in the region was concordant with the recessive mode of inheritance and disease phenotype, providing strong evidence that the E40K mutation underlies the disease phenotype. Microscopic examination of spinal cords from affected dogs revealed myelin and axon loss affecting the lateral white matter and neuronal cytoplasmic inclusions that bind anti-SOD1 antibodies. Such inclusions are also a feature of some forms of ALS caused by SOD1 mutations

Discussion and Conclusions: We identify DM as the first spontaneously occurring animal model for ALS. These dogs could be used to investigate the processes that underlie motor neuron degeneration in DM and ALS and to evaluate therapeutic interventions.