THEME 9 IN VIVO EXPERIMENTAL MODELS

P210 ROLE OF ZPR1 IN MOTOR NEURON DEGENERATION AND SEVERITY OF SPINAL MUSCULAR ATROPHY

Gangwani L

Texas Tech Health Sciences University, El Paso, Texas, USA

Email address for correspondence: [email protected]

Keywords: SMA, SMN, neurodegeneration

Background: Spinal muscular atrophy (SMA) is caused by mutations of the survival motor neurons (SMN1) gene. SMA is characterized by degeneration of spinal motor neurons caused by low levels of SMN protein. The copy number of the SMN2 gene primarily influences the severity of SMA. Additional modifier genes that lie outside the SMA locus exist and one gene that could modify SMA is the zinc finger protein (ZPR1) gene (1,2). Currently, there is no treatment available to cure or reduce the burden of severity of SMA because of limited knowledge of modifier genes and the molecular mechanisms associated with SMA pathogenesis.

Objective: To examine the role of ZPR1 in motor neuron degeneration and severity of SMA.

Methods: In vivo studies using Zpr1 knockout (3) and SMA model mice (4). In vitro studies using cultured primary spinal cord neurons.

Results: To test the significance of ZPR1 down-regulation in SMA, we examined the effect of reduced ZPR1 expression in mice with mild and severe SMA. We report that the reduced ZPR1 expression causes increase in the loss of motor neurons, hypermyelination in phrenic nerves, increases in respiratory distress and disease severity that reduces lifespan of SMA mice. The deficiency of SMN-containing sub-nuclear bodies correlates with the severity of SMA. ZPR1 is required for accumulation of SMN in sub-nuclear bodies. We report that ZPR1 overexpression increases the levels of SMN and promotes accumulation of SMN in sub-nuclear bodies in SMA patient fibroblasts. ZPR1 stimulates neurite growth and rescues axonal growth defects in SMN-deficient spinal cord neurons from SMA mice. These data suggest that the severity of disease correlates negatively with ZPR1 levels and ZPR1 may be a protective modifier of SMA.

Discussion and conclusion: ZPR1 deficiency causes defects in phrenic nerve that may contribute to respiratory distress and increase the severity of SMA. Because SMA patients express low levels of ZPR1, our data suggest phrenic nerve as a potential therapeutic target to reduce the burden of respiratory distress in SMA. ZPR1 overexpression elevates SMN levels, corrects the defect in nuclear accumulation of SMN in SMA patient cells and rescues the axonal growth of SMN-deficient neurons from SMA mice. These findings suggest that ZPR1 may be a protective modifier of SMA and opens new avenues for SMA therapeutics.

Acknowledgements

Funding from the Muscular Dystrophy Association, Families of SMA and NIH (R01NS064224) to LG supported this study.

References

• Gangwani L, Mikrut M, Theroux S et al. Nat Cell Biol. 2001;3:376–83.
   PubMed Web of Science ®Google Scholar
• Doran B et al. Proc Natl Acad Sci USA 2006;103:7471–5.
   PubMed Web of Science ®Google Scholar
• Gangwani L, Flavell RA, Davis RJ. Mol Cell Biol 2005; 25: 274456 (2005).
   Google Scholar
• Le TT et al. Hum Mol Genet 2005;14:845–57.
   PubMed Web of Science ®Google Scholar

P211 POSTNATAL REQUIREMENTS FOR SURVIVAL MOTOR NEURON (SMN), A PROTEIN DEFICIENT IN SPINAL MUSCULAR ATROPHY, DURING MATURATION AND REMODELING OF THE NEUROMUSCULAR SYSTEM

Kariya S^1,2

Obis T^1,2

Garone C³

Akay T^1,4

Hirano M³

Monani U^1,2

^aMotor Neuron Center, Columbia University, New York, USA

^bDepartment of Pathology & Cell Biology

^cDepartment of Neurology

^dDepartment of Neurological Surgery, New York, USA

Email address for correspondence: [email protected]

Keywords: SMA, SMN, neuromuscular remodeling

Background: Spinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disease caused by survival motor neuron (SMN) protein deficiency due to mutations in the SMN1 gene. Growing evidence indicates that SMA is a developmental disorder primarily affecting the neuromuscular system; however, a consequence of SMN reduction in adults with fully matured organs remains unclear.

Results: Here, we demonstrate, through the analysis of tamoxifen-inducible SMN-deficient mice, that induction of low SMN levels ubiquitously to the extent equivalent to those in severe SMA model mice does not mediate any abnormal phenotype in the mice older than P21, but prevents them from remodeling matured neuromuscular synapses after nerve injury. The mice were also unable to reconstruct fully matured myofibers following the skeletal muscle damage, while epithelialization after the skin injury was not affected. In control mice, the expression of SMN protein in the L5-S2 spinal motor neurons had significantly increased during the neuromuscular junction (NMJ) re-maturation process after crushing the sciatic nerve, but the increase was not detectable in SMN-deficient mice. Collectively, these results indicate that high SMN levels are required during neuromuscular maturation. In fact, inducing low SMN levels in mice younger than P12 with immature neuromuscular system resulted in the muscle growth defects, NMJ degeneration, and progressive muscle weakness leading to death.

Discussion: Our findings suggest that patients with SMA receiving an SMN-increasing therapy can at least reduce their medication after reaching a certain age, although they need to re-augment their SMN levels under the conditions requiring neuromuscular remodeling, such as injury, diseases, and aging. This novel concept will contribute to designing further practical strategies for the treatment of individual patients with SMA.

P212 GENETIC BACKGROUND EFFECTS ON LIFESPAN OF DYNACTIN P150 GLUED MOUSE MODEL OF MOTOR NEURON DISEASE

Heiman-Patterson T¹

Blankenhorn E¹

Sher R³

Wong PC²

Jiang J¹

Alexander G¹

Cox G³

^eDrexel University College of Medicine, Philadelphia, Pennsylvania, USA

^fJohns Hopkins University, Baltimore, Maryland, USA

^gJackson Laboratories, Bar Harbour, Maine, USA

Email address for correspondence: [email protected]

Keywords: modifiers, genetic background, mouse

Background: There are background-dependent differences in disease phenotype in transgenic mice that carry the mutated G93ASOD1 transgene. Expression of G93A hSOD1Tg in ALR, NODRag1KO, SJL or C3H backgrounds show a more severe phenotype, whereas a milder phenotype is observed in B6, B10, BALB/c and DBA inbred strains compared to the original mixed (B6xSJL) hSOD1Tg mice (1). These background differences are due to disease modifying genes. We have identified a Chromosome 17 QTL associated with increased lifespan in animals carrying alleles derived from B6 or B10 at this locus. Once identified, these modifiers would be of great interest, especially if they were shown to affect other motor neuron disease phenotypes.

The Dynactin p150^glued mouse model is based on a slowly progressive autosomal dominant, lower motor neuron disease in humans that is linked to a mutation in the p150^Glued subunit of the dynactin complex, which demonstrates clinical and pathologic changes of motor neuron disease (2,3).

Objectives: To examine background effects on disease phenotype in the Dynactin p150^glued mouse model of motor neuron disease.

Methods: We developed two inbred strains (C57BL/6J and SJL/J) expressing the mutant human Dynactin p150^glued originally bred on a mixed background of C57BL/6J and SJL/J. Onset of tremor along with survival was examined and compared between the two congenic strains. Survival was determined by the ability of the mouse to right itself within 10 seconds when placed on its back.

Results: We observed an acceleration of tremor onset (127.8 ± 37.7 days; N = 22 vs. 158.3 ± 31.1 days; N = 23) and decreased survival (286.8 ± 54.7; days; N = 19 vs. 344.6 ± 49.7; N = 11) when Dynactin p150 ^glued was bred onto the SJL/J background. There was a milder phenotype with later onset (175.3 ± 30.5 days; N = 14 vs. 158.3 ± 31.1days; N = 23) when Dynactin p150^glued was bred on the C57BL/6J background compared to the original mixed B6/SJL background. Survival is not yet available for animals bred to the C57BL/6J background since these animals remain alive at 400 days old.

Discussion and conclusion: The genetic background influences phenotype in the Dynactin p150^glued model of motor neuron degeneration similar to hSOD1-G93A transgenic mice that also have an accelerated phenotype when bred on SJL/J and a milder phenotype on the C57BL/6J. This suggests that there are genetic modifiers in both these disorders and that they may be similar. Identification of modifier genes might highlight intracellular pathways involved in motor neuron degeneration and provide new therapeutic targets.

Acknowledgements

ALS Hope Foundation, Muscular Dystrophy Association.

References

• Heiman-Patterson TD, Sher RB, Blankenhorn EA et al. ALS 2011;12:79–86.
   Google Scholar
• Puls I, Oh SJ, Sumner CJ et al. Ann Neurol 2005;57:687–694.
   PubMed Web of Science ®Google Scholar
• Laird FM, Farah MH, Ackerly S et al. Jour Neurosci 2008;28:1997–2005.
   PubMed Web of Science ®Google Scholar

P213 ROLE OF ELP3 IN ALS

Bento-Abreu A

Timmers M

Robberecht W

V.I.B. - K.U.Leuven, Leuven, Belgium

Email address for correspondence: [email protected]

Keywords: elongator, modulator, methylation

Background: Elp3 is the catalytic subunit of the elongator complex, comprised of six subunits (Elp1–Elp6). Elongator complex is present mainly in the nucleus, where it associates with the hyperphosphorylated RNA polymerase II, but its presence in the cytoplasm has been associated with other functions. We have identified a polymorphism in the ELP3 gene that is associated with ALS. Lower expression levels of Elp3 were found in the brain of individuals with the ALS at-risk genotype. Moreover, two loss-of-function mutations in the drosophila ELP3 were identified to induce profound axonal and synaptic defects, and the knockdown of Elp3 in zebrafish induced motor axonal abnormalities.

Objectives: To determine whether ELP3 is a modulator gene in ALS.

Methods: Overexpression and knock-down of Elp3 in the SOD1^G93A mouse model of ALS and also in the SOD1A4V zebrafish model.

Results: We generated an ELP3^-/- mouse, embryonically lethal at E10.5. Nonetheless, ELP3^± are viable and ELP3^±/SOD1^G93A mice become symptomatic earlier than SOD1^G93A mice (100.4 ± 3.5 days vs. 114.6 ± 5.9 days). The survival of these animals is currently being monitored. We also generated an ELP3-overexpressing mouse that is now being crossed with SOD1^G93A mice. Finally, we established a procedure for the AAV9-mediated overexpression of Elp3 in the spinal cord of SOD1^G93A mice. Preliminary data show that Elp3 overexpression prolongs survival of SOD1^G93A mice by 13 days (145 days (AAV9:GFP) vs. 158.5 (AAV9:Elp3) days).

Discussion: Increasing the expression of ELP3 is beneficial in the SOD1G93A mouse whereas lowering ELP3 levels is deterimental, indicating that ELP3 may be a modulator of the disease. ELP3 has been shown to acetylate Histone H3, via this HAT domain. ELP3 also contains a SAM domain. The latest is involved in methylation/demethylation reactions. It is reasonable to speculate that ELP3 might regulate the transcription of certain genes by acetylation or methylation of histones H3. Further investigation is needed to clarify the role of ELP3 in ALS.

P214 LOSS OF FUNCTION C9ORF72 CAUSES MOTOR DEFICITS IN A ZEBRAFISH MODEL OF ALS

Kabashi E

Ciura S

Lattante S

Brain and Spinal Cord Institute, Paris, France

Email address for correspondence: [email protected]

Keywords: C9orf72, zebrafish, loss of function.

Objective: To define the role that repeat expansions of a GGGGCC hexanucleotide sequence of the C9orf72 gene play in the pathogenesis of amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration (FTLD). A genetic model for ALS was developed to determine whether loss of function of the zebrafish orthologue of C9orf72 (zC9orf72) leads to abnormalities in neuronal development.

Methods: C9orf72 mRNA levels were quantified in brain and lymphoblasts derived from FTLD and ALS/FTLD patients and in zebrafish. Knockdown of the zC9orf72 in zebrafish was performed using two specific antisense morpholino oligonucleotides to block transcription. Quantifications of spontaneous swimming and tactile escape response, as well as measurements of axonal projections from the spinal cord, were performed.

Results: Significantly decreased expression of C9orf72 transcripts in brain and lymphoblasts was found in sporadic FTLD and ALS/FTLD patients with normal-size or expanded hexanucleotide repeats. The zC9orf72 is selectively expressed in the developing nervous system at developmental stages. Loss of function of the zC9orf72 transcripts causes both behavioral and cellular deficits related to locomotion without major morphological abnormalities. These deficits were rescued upon overexpression of human C9orf72 mRNA transcripts.

Discussion: Our results indicate C9orf72 haploinsufficiency could be a contributing factor in the spectrum of ALS/FTLD neurodegenerative disorders. Loss of function of the zebrafish orthologue of zC9orf72 expression in zebrafish is associated with axonal degeneration of motor neurons that can be rescued by expressing human C9orf72 mRNA, highlighting the specificity of the induced phenotype. These results reveal a pathogenic consequence of decreased C9orf72 levels, supporting a loss of function mechanism of disease.

P215 CHARACTERISATION OF A UNIQUE SOD1 MOUSE MODEL FOR AMYOTROPHIC LATERAL SCLEROSIS

Saccon R¹

Joyce P²

Mcgoldrick P³

Fratta P¹

Greensmith L³

Acevedo A²

Fisher EM¹

^hDepartment of Neurodegenerative, UCL Institute of Neurology, London, UK

ⁱMRC National institute for Medical Research, Harwell, UK

^jSobell Department of Motor Neuroscience and Movement Disorders, London, UK

Email address for correspondence: [email protected]

Keywords: mouse model, SOD1, neurodegeneration

Background: Mutations in the superoxide dismutase 1 (SOD1) gene account for approximately 20% of familial amyotrophic lateral sclerosis (FALS) cases. Transgenic mouse models expressing mutant SOD1 have been crucial in furthering our understanding of amyotrophic lateral sclerosis (ALS). However, SOD1 transgenic mice overexpress the mutant protein and are therefore not direct genetic representations of human ALS. Thus, components of the resulting phenotype may arise from overexpression of the protein, rather than the effects of its mutation. Using MRC Harwell's N-ethyl-N-nitrosourea (ENU) bank of mutagenised mice, we identified a new mouse model that carries a point mutation in the endogenous mouse Sod1 gene and expresses endogenous levels of SOD1, making it biochemically more relevant to human condition of ALS. This mutation, Sod1 D83G is identical to a FALS causative mutation found in humans (1).

Objectives: Our aim is to characterize the Sod1^D83G model and gain insights into the pathological mechanisms of ALS, using immunohistochemical, biochemical, and molecular analyses.

Methods: SOD1 activity and protein levels from brain homogenate have been assessed using an in-gel assay and Western blotting technique. Muscle force and motor unit number have been measured in tibialis anterior and extensor digitorum longus muscles, stimulating the sciatic nerve and the distal tendons of anesthetized mice. Motor neuron survival and neuromuscular junctions (NMJs) were evaluated using immunostaining and confocal microscopy.

Results: Since the D83 residue of SOD1 coordinates zinc, we verified that the SOD1 D83G protein is dismutase inactive and found that SOD1 D83G protein is unstable in vivo. In particular, SOD1 activity in brain extracts from Sod1^+/D83G mice is reduced to ˜50% whereas Sod1^D83G/D83G are dismutase inactive. Sod1^D83G/D83G also phenocopy Sod1 null mice and develop hepatocellular carcinoma and progressive muscle denervation. Homozygote Sod1^D83G/D83G mutant mice develop progressive degeneration of lower and upper motor neurons, although they do not become paralysed, unlike transgenic models of ALS. Sod1^D83G/D83G mice do not develop overt inclusion pathology but show a progressive neuromuscular deterioration as demonstrated by the denervation of the NMJs.

Discussion and conclusions: The Sod1^D83G model demonstrates that a point mutation in the mouse Sod1 gene is sufficient to cause degeneration of upper and lower motor neurons, and presents a unique model in which to dissect the role of mutant SOD1 gain and loss of function on motor neuron function and survival. In particular Sod1^D83G mice will be important for the examination of early stage pathological mechanism of ALS and may provide an excellent new tool for testing disease therapeutics. Furthermore, by developing both motor neuron degeneration and hepatocellular carcinoma, this model offers the opportunity to study the link between cancer and neurodegeneration.

Acknowledgements

We thank MNDA for founding our research.

Reference

• Millecamps , Salachas , Cazeneuve et al. J Med Genet 2010.
   Google Scholar

P216 A NEW TRANSGENIC MOUSE MODEL BASED ON OVEREXPRESSION OF A CHMP2B MUTANT RECAPITULATES PARTS OF ALS AND FTD HALLMARKS

Vernay A¹

Therreau L¹

Blot B²

Risson V³

Grosch S¹

Schaeffer L³

Sadoul R²

Loeffler JP¹

Rene F¹

^kINSERM U 1118, UDS Faculté de Médecine, Strasbourg, France

^lInserm U836, Grenoble Institut des Neurosciences, Université Joseph Fourier, Grenoble, France

^mLaboratoire de Biologie Moléculaire de la Cellule, UMR5239 CNRS/ENS Lyon/UCBL/HCL Ecole normale supérieure de Lyon, Lyon, France

Email address for correspondence: [email protected]

Keywords: CHMP2B intron 5, mouse model, phenotyping

Background: Clinical, physiopathological and genetical cues converge to a pathological continuum between amyotrophic lateral sclerosis (ALS) and frontotemporal dementia (FTD). About 15% of FTD patients present with motoneuron disease and 30% of ALS patients experience FTD or other dementia symptoms (1). Both pathology can be either sporadic or of familial origin. Among the genes associated with these diseases, mutations in the CHMP2B (chromatin-modifying protein 2B) gene were found in patients with ALS, FTD and ALS-FTD. The CHMP2B^Intron5 mutant is a protein encoded by one of these mutations (2,3).

Objective: To study the clinical, pathological and genetic continuum between ALS and FTD and to obtain a model covering both diseases.

Methods: We have established a transgenic (Tg) mouse line expressing the human CHMP2B^intron5 mutant in neurons driven by the Thy1.2 mouse promoter. Mice were characterized using behavioural, biochemical, electrophysiological and immunohistochemical approaches.

Results: The CHMP2B^intron5 mice have decreased survival and show progressive neurodegenerative changes leading to motor and behavioural alterations. They show a strong expression of the mutant protein in neurons of the brain and the spinal cord, especially in the anterior and motor cortices and in motor neurons. This expression is associated with a gliosis and the presence of ubiquitin-positive inclusions.

The motor phenotype recapitulates several aspects of human ALS. Homozygous mutants exhibit severe and early locomotor impairments attested by a decrease in rotarod performance and grip strength, and gait abnormalities at 2 months of age. These impairments appear from 12 months in hemizygous mice and develop towards a final paralysis associated with muscle denervation, as assessed by electromyography. We further show behavioural defects relevant to FTD such as stereotypes (repetitive rearing, excessive grooming) and food intake abnormalities.

Discussion: Here we report the generation of a Tg line expressing a human ALS/FTD-causing mutation that reproduces part of the ALS-FTD symptoms. Our data provide robust in vivo evidences of neurodegenerative mechanisms driven by the expression of the CHMP2B^intron5 mutant. Indepth analysis of the mechanisms leading to neuronal dysfunction may help to better understand the common physiopathological mechanisms leading to these two pathologies. This Tg line is a new promising model of ALS/FTD that will be useful for mechanistic and preclinical studies.

References

• Lomen-Hoerth C Clinical phenomenology and neuroimaging correlates in ALS-FTD. Journal of molecular neuroscience 2011;45:656–62.
   Google Scholar
• Cox LE et al. Mutations in CHMP2B in lower motor neuron predominant amyotrophic lateral sclerosis (ALS). PloS one 2010;5:e9872.
   PubMed Web of Science ®Google Scholar
• Isaacs AM, Johannsen P, Holm I, Nielsen JE. Frontotemporal dementia caused by CHMP2B mutations. Current Alzheimer research 2011;8:246–51.
   PubMed Web of Science ®Google Scholar

P217 LOSS OF P62/SQSTM1 EXACERBATES MOTOR DYSFUNCTION IN A MUTANT SOD1-EXPRESSING MOUSE ALS MODEL

Pan L¹

Otomo A¹

Abe K¹

Ogawa H¹

Chiba T²

Koike M³

Uchiyama Y³

Aoki M⁴

Yoshii F⁵

Ishii T⁶

Yanagawa T⁶

Hadano S¹

ⁿDepartment of Molecular Life Sciences, Tokai University School of Medicine, Isehara, Kanagawa, Japan

^oBiomedical Gerontology Laboratory, School of Human Sciences, Waseda University, Tokorozawa, Saitama, Japan

^pDepartment of Cell Biology and Neuroscience, Juntendo University Graduate School of Medicine, Bunkyo-ku, Tokyo, Japan

^qDepartment of Neurology, Tohoku University Graduate School of Medicine, Sendai, Miyagi, Japan

^rDepartment of Neurology, Tokai University School of Medicine, Isehara, Kanagawa, Japan

^sFaculty of Medicine, University of Tsukuba, Tsukuba, Ibaraki, Japan

Email address for correspondence: [email protected]

Keywords: p62/SQSTM1, SOD1, autophagy

Background: Recent studies have revealed mis-sense variants and/or mutations in SQSTM1 in familial as well as sporadic ALS. SQSTM1 encodes p62/SQSTM1 that regulates the selective-autophagy via association with ubiquitinated misfolded proteins. It is notable that abundant p62/SQSTM1-positive inclusions in the brain are a typical pathological feature of ALS or ALS-FTD. Thus, p62/SQSTM1 might play a crucial role in maintenance and/or survival of motor neurons. However, it remains to be determined as to whether loss-of-function or gain-of-toxic-function of the p62/SQSTM1 mutants is associated with ALS. Furthermore, it has been shown that genetic inactivation of Sqstm1 in mice results in the accumulation of hyperphosphorylated tau and neurodegeneration, and also in an accelerated presentation of aging phenotypes. However, no clear experimental evidence showing p62/SQSTM1 linking to motor neuron degeneration has been demonstrated.

Objectives: To examine whether p62/SQSTM1 plays a role in the onset and/or progression of ALS/MND that is associated with mutant SOD1 in vivo.

Methods: We generated SOD1^H46R transgenic mice on a Sqstm1-null background by crossing Sqstm1-KO mice with the SOD1^H46R mouse line expressing familial ALS-linked SOD1^H46R under the control of inherent human SOD1 promoter. We first generated congenic lines of both SOD1^H46R transgenic and Sqstm1^-/- mice by each backcrossing more than 10 generations with C57BL/6N mice. Next, we produced Sqstm1 ^± SOD1^H46R mice by crossing male SOD1^H46R and female Sqstm1^± mice, and then generated six different genotype mice: wild-type, Sqstm1^±, Sqstm1^-/-, SOD1^H46R, Sqstm1 ^± SOD1^H46R, and Sqstm1^-/-SOD1^H46R, by crossing male Sqstm1 ^± SOD1^H46R and female Sqstm1^± mice. Body weight and survival of each animal were monitored. Motor coordination and balance were also measured using a balance-beam test to evaluate the motor dysfunction in the mice.

Results: Sqstm1^-/- mice did not show any gross abnormal phenotypes during the experimental period (˜28 weeks). By contrast, SOD1^H46R and Sqstm1 ^± SOD1^H46R mice both exhibited progressive motor dysfunction and paralysis with average life spans of 174.0 ± 16.4 (n = 82) and 175.3 ± 11.4 days (n = 41), respectively. Remarkably, Sqstm1^-/-SOD1^H46R mice showed a shorter life span of 153.5 ± 8.0 days (n = 43). Survival in Sqstm1^-/-SOD1^H46R mice was significantly shorter than that in wild-type or Sqstm1±; SOD1^H46R mice (p < 0.0001, Kaplan–Meier analysis with log-rank test). Furthermore, a balance-beam test revealed that motor dysfunction in Sqstm1^-/-SOD1^H46R mice (˜17 weeks of age) occurs at an approximately 3 weeks earlier than that in SOD1^H46R mice (˜20 weeks of age).

Discussion and conclusion: These results indicate that lack of p62/SQSTM1 exacerbates motor dysfunction in SOD1^H46R mice. Thus, loss of function of p62/SQSTM1 is, at least in part, associated with motor neuron diseases. Further characterization of these mice will clarify the implication of the p62/SQSTM1-mediated functions in SOD1 and/or SQSTM1-linked ALS.

Acknowledgements

This work was supported by Grant- in-Aid for Scientific Research from the Japanese Society for Promotion of Science (JSPS).

P218 FATTY ACID PROFILE REVEALS PROFOUND ALTERATIONS OF LIPID METABOLISM IN SOD1 MICE

Henriques A^1,2

Schmitt F^1,2

Lequeu T^1,2

Hussain G^1,2

Metz-Boutigues M-H³

Bindler F⁴

Marchioni E⁴

Gonzalez de Aguilar J-L^1,2

Loeffler JP^1,2

^tINSERM, U1118, Mécanismes Centraux et Périphériques de la Neurodégénerescence, Strasbourg, France

^uUniversité de Strasbourg, UMRS1118, Strasbourg, France

^vINSERM, U1121, Biomatériaux et Ingénierie Tissulaire, Strasbourg, France

^wIPHC-DSA, UMR 7178, Illkrich-Graffenstaden, France

Email address for correspondence: [email protected]

Keywords: lipids, SCD-1, metabolism

Background: Lipid species are various molecules, ranging from pure energetic supplies to key structural player, and all are formed with at least one fatty acid. Reports point the lipid metabolism to the pathophysiology of ALS. Patients with dyslipidemia present extended life expectancy (1). An animal model of ALS, the SOD1 mice, reproduces the hypermetabolism trait of ALS, and high-fat diets extent their life expenctancy (2). The length and the number of saturations present in the carbon chain of fatty acids relate to the management of energetic metabolism. A low activity of Stearoyl-Coa Desaturase-1 (SCD-1), an enzyme converting saturated fatty acid (SFA) to mono-unsaturated fatty acids (MUFAs) promotes peripheral axonal regeneration in mice (3).

Objectives: This study was designed to determine the tuning of energetic metabolism in SOD1 mice, at different disease stages, and the role of SCD-1 activity in ALS metabolic disorders.

Methods: SOD1 mice and mice treated with an inhibitor of SCD-1 were used in this study. After sacrifice, lipids from serum and liver were collected and extracted (Bligh and Dyer method). Lipids were transmethylated, and fatty acid methyl esters were analyzed by gas chromatography. Fatty acid profiles were compared to muscular denervation assessed by electromyography. Levels of circulating lipids were measured using enzymatic assays. Gene expressions were performed by quantitative PCR. Animal experiments followed European Union regulations and were approved by the ethical committee of the University of Strasbourg (No. AL/01/20/09/12).

Results: We found that relative SCD-1 activity is decreased in the serum and in the liver of SOD-1 mice, already at disease onset. The reduction of SCD-1 activity was accompanied by low levels of MUFA and increased susceptibility for lipid peroxidation. We noticed a reduction of triglycerides and total cholesterol in the serum and a downregulation of genes involved in the hepatic de novo lipogenesis. Pharmacological inhibition of SCD-1 reproduces the same metabolic traits found in SOD-1 mice.

Discussion and conclusion: Energetic lipids suffer changes in their composition, at the time that muscular denervation occurs in an animal model of ALS. These alterations reveal (i) a higher susceptibility for lipid peroxidation and (ii) a silencing of de novo lipogenesis in liver. Pharmacological inhibition of SCD1 reproduces these observations in a non-transgenic animal. Our study suggests a general role for SCD1 in the pathophysiology of amyotrophic lateral sclerosis.

Acknowledgements

This work was supported by the Thierry Latran Foundation, the European Community’s Health Seventh Framework Programme under grant agreement n° 259867 (FP7/2007–2013), by ARSLA, by AFM and by AREMANE.

References

• Dupuis et al. Neurology, 2008;70:1004–9.
   PubMed Web of Science ®Google Scholar
• Dupuis et al. PNAS, 2004;101:11159–11164.
   PubMed Web of Science ®Google Scholar
• Hussain et al. PloS One, 2013; In press.
   Google Scholar

P219 DISEASE STAGE-SPECIFIC ALTERATIONS IN FAT METABOLISM IN THE HSOD1G93A MOUSE MODEL OF AMYOTROPHIC LATERAL SCLEROSIS (ALS)

Steyn F¹

Lee K¹

Chen C¹

Mccombe P^2,3

Borges K¹

Ngo S^1,2

^xThe University of Queensland, Queensland, Australia

^yUniversity of Queensland Centre for Clinical Research, Queensland, Australia

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

Email address for correspondence: [email protected]

Keywords: body mass index, fat metabolism, fatty acids

Background: While motor neurone loss and muscle atrophy are characteristic of ALS, increased whole body fat mass (adiposity) is correlated with slower disease progression. Increased metabolism in ALS results in reduced fat mass and lower body mass index (BMI), which is negatively associated with survival (1,2). In line with this, nutritional supplementation with a diet rich in protein and calories promotes weight gain and slows disease symptoms (3). The mechanisms by which increased metabolism and reduced BMI contribute to ALS pathology are unknown. Identification of aberrations to the key factors that maintain fat mass will provide greater understanding of ALS disease mechanisms.

Objective: To characterize the mechanisms underlying increased fat catabolism that occurs in ALS.

Methods: Male wild-type and hSOD1G93A transgenic mice were studied at various stages of disease progression. We assessed epididymal fat weight, levels of circulating non-esterified free fatty acids (NEFAs) and ketones, and levels of NEFAs in skeletal muscle. PCR arrays were used to determine the expression of genes encoding proteins and enzymes involved in the catabolism and mobilization of long-chain fatty acids (LCFAs), and genes that oxidize ketones. Data were analyzed using unpaired t-test.

Results: When compared to wild-type age-matched controls, hSOD1G93A mice failed to accumulate fat mass after the onset of disease symptoms (n ≥ 6/group). This was followed by subsequent increases in the expression of circulating ketones and skeletal muscle NEFAs (n ≥ 5/group). Gene expression analysis of white adipose tissue collected from hSOD1G93A mice at the onset of disease symptoms and wild-type age-matched controls indicated an increase in the catabolism and mobilization of LCFAs, and oxidation of ketones in ALS (n = 4/group).

Discussion and conclusion: We report evidence of increased LCFA catabolism and mobilization in the hSOD1G93A mouse. The inability to gain fat mass, coupled with increased breakdown and movement of fat from storage, and fat accumulation in skeletal muscle suggests that endogenous mechanisms promote the use of fat as an energy source in skeletal muscle during periods of increased energy need.

Acknowledgements

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

References

• Chio A et al. Amyotroph Lateral Scler, 2009;10(5–6):310–23.
   PubMed Web of Science ®Google Scholar
• Jawaid A et al. Amyotroph Lateral Scler, 2010;11(6):542–8.
   Google Scholar
• Silva LB et al. Arq Neuropsiquiatr, 2010; 68(2):263–8.
   Google Scholar

P220 ANALYSIS OF HISTONE POST-TRANSLATIONAL MODIFICATIONS ASSOCIATED WITH AMYOTROPHIC LATERAL SCLEROSIS (ALS) ONSET AND PROGRESSION

Masala A

Esposito S

Dedola S

Galioto M

Iaccarino C

Crosio C

University of Sassari, Department Biomedical Science, Sassari, Italy

Email address for correspondence: [email protected]

Keywords: epigenetic, histone post-translational modifications, DNA methylation

Background: Epigenetic changes encompass an array of molecular modifications to both DNA and chromatin, including post-translational covalent modification of histones at distinct amino acid residues on their amino (N)-terminal tails, such as acetylation, phosphorylation, and methylation. As for other neurodegenerative and neurological disorders, recent lines of evidence associate epigenetic gene regulation with ALS pathogenesis suggesting that transcriptional dysregulation may play a central role in the pathogenesis of ALS. Although the HDAC inhibitor (HDACi) phenylbutyrate is used in a phase II clinical trial for ALS, the efficacy of HDACi for ALS therapy awaits clinical confirmation. In fact although HDACi reasonably prolongs elevated transcriptional gene activity, many aspects of epigenetic gene regulation in ALS remain elusive. At present, different aspects of epigenetic gene regulation, including a global picture of histone modifications, have not been investigated. Filling this gap might have major implications for ALS disease progression and upcoming pharmacological interventions for epigenetic ALS therapy.

Objectives: The major objective of our work is to investigate whether in cellular and animal models for ALS, it is possible to highlight chromatin modifications associated with disease onset and progression and to study the epigenetic status of key genes known to be involved in ALS development.

Methods: In order to study the chromatin modifications induced by the expression of the ALS causative-genes SOD1, TDP43 and FUS, we use the following cellular and animal models to mimic the genetic alterations that cause ALS: (i) transgenic mice SOD1-G93Aand (ii) adenoviral delivery of ALS causative-gene in neuronal cells. Western blot analysis and immunohistochemistry have been performed on infected SHSY5Y and NSC34 cells, and on Tg spinal cord to draw a map as complete as possible as histone modifications associated with ALS.

Discussion: Preliminary results indicate that expression of wild-type or mutant ALS-causative genes correlates with specific alterations in histone modifications associated with transcriptional activation.

P221 NEW INSIGHTS INTO THE MECHANISMS UNDERLYING THE PATHOGENESIS OF ALS USING FDG-PET AND 1H-MRS STUDIES IN VIVO AND ADENOVIRAL-MEDIATED GENE TRANSFER OF DNA/RNA-BINDING PROTEINS IN VITRO

Oksman J¹

Ahtoniemi T¹

Huhtala T¹

Cerrada-Gimenez M¹

Lehtimaki K¹

Woloszynek J²

Etzel K²

Mcintosh D⁵

Moore C³

Haq M⁴

Lopez H²

Haq S⁵

^aaCharles River Discovery Research Services, Kuopio, Finland

^abMurigenics Inc., Vallejo, CA, USA

^acPorstmouth Hospitals NHS Trust, Portsmouth, Hampshire, UK

^adMaidstone and Tunbridge Hospital NHS Trust, Tunbridge, Kent, UK

^aeDaval International Ltd., Eastbourne, East Sussex, UK

Email address for correspondence: [email protected]

Keywords: neuropeptides, FDG-PET, MRS

Objectives: The objective of the study was to determine whether a specific novel stabilized neuropeptide, the principle active component in hyperimmune caprine sera (HICS), could elicit measurable efficacy in multiple models of ALS in vivo and in vitro, and to understand better the features of the mechanisms behind ALS.

Results: Age-matched SOD1^G93A male and female mice(n = 20/treatment group naive/wild type [WT], naive/superoxide dismutase 1[SOD1] and hyperimmune caprine sera [HICS/SOD1]) were injected once daily (100 μg s.c.) using a double-blind experimental protocol starting from 60 days of age and continuing to the end-stage of the disease. Significant maintenance was observed in rotarod latency and grip strength, and concomitant changes were observed in several key cellular brainstem metabolites using 1H-MRS at 110 days. Delayed onset of disease and prolonged survival were also observed. In FDG-PET, SOD1 mice showed a significant reduction in brain glucose metabolism compared to normal naive/WT mice controls. HICS treatment was able to completely normalize and rescue the SOD1 glucose hypometabolism to normal levels (p < 0.048) using 100 μg s.c. b.i.d.

Discussion: In a separate study arm that constitutes work in progress, we are currently investigating the efficacy of HICS in primary cortical neurons and induced pluripotent stem cells transduced using adenoviral-mediated gene transfer (Ad) of several DNA/RNA-binding proteins that include the following: Ad-FUS, Ad-TDP-43, Ad-SOD1, and Ad-C9ORF72. Cell survival, protein mis-folding, and mechanisms implicated in neurotoxicity are being determined.

HICS may not only serve as a potential therapeutic agent but also help decipher aspects of the mechanisms that underlie the complex pathogenesis of ALS.

P222 CHARACTERIZATION OF INNATE AND ADAPTIVE IMMUNE RESPONSES IN THE HSOD1G93A-MCP1-CCR2 TRIPLE TRANSGENIC ALS MOUSE

Jara J¹

Farris C²

Trimarchi J²

Miller R¹

Ozdinler PH^1,2

^afNorthwestern University, Chicago, IL, USA

^agIowa State University, Ames, IA, USA

Email address for correspondence: [email protected]

Keywords: neuroinflammation, cytokines, SOD1

Background: Building evidence revealed the involvement of innate and adaptive immune responses in both the spinal cord and motor cortex of ALS patients, and in mouse models of ALS at different stages of disease pathology (1). Secretion of cytokines including MCP1 (monocyte chemoattractant protein-1) has been detected in both cerebrospinal fluid and spinal cord of ALS patients and mouse models of ALS (2). Furthermore, MCP1-mediated recruitment of monocytes that express CCR2 (CC chemokine receptor 2) is supported by decreased levels of CCR2 + monocytes in the blood of ALS patients (3).

Objectives: To elucidate the cellular components and the molecular basis of innate and adaptive immune response in ALS using the hSOD1^G93A-MCP1-CCR2 triple transgenic ALS mouse model, in which cells that express MCP1 and CCR2 are visualized and isolated based on their expression of mRFP (monomeric red fluorescent protein) and eGFP (enhanced green fluorescent protein), respectively. This approach uses MCP1 and CCR2 expression as a bait to genetically label cells of interest and to isolate them from the complex and heterogeneous structure of the brain as a pure group.

Methods: We generated hSOD1^G93A-MCP1-CCR2 triple transgenic ALS mice, in which cells that MCP1 and CCR2 are genetically labeled in an ALS mouse model. Fluorescence-activated cell sorting (FACS) and microarray analysis are performed to evaluate the cellular identity and transcription profile of MCP1 + and CCR2 + cells located in the motor cortex and spinal cord at different disease stage.

Results: Our results demonstrate the presence of MCP1 + and CCR2 + cells in both the motor cortex and the spinal cord at different stages of disease initiation and progression. MCP1 + cells express macrophage/microglia lineage-specific markers at P30 and P60, suggesting their early contribution to pathology. Interestingly, a subset of CCR2 + cells expressed Ly6C, a maker for infiltrating monocytes. Microarray analysis of FACS-purified MCP1 + and CCR2 + cells at P30 started to reveal genes that are upregulated and downregulated with respect to increased immunologic response. Preliminary results suggest RNA post-translation modifications, B- and T-cell receptor signaling, and chemokine signaling pathways to be activated.

Discussion and conclusion: Using pure populations of MCP1 + and CCR2 + cells will reveal details of the molecular controls over initiation and progression of immunity in ALS. Cellular mechanisms involved in the immunologic response to vulnerable motor neurons will help identify novel therapeutic targets for building effective treatment strategies.

Acknowledgments

We thank Les Turner and Wenske Foundation, ALS Association for his contribution.

References

• Ilieva H, Polymenidou M, Clevel DW. J Cell Biol 2009; 187:761–772.
   PubMed Web of Science ®Google Scholar
• Henkel JS, Beers DR, Siklos L et al. Mol Cell Neurosci. 2006;31:427–437.
   PubMed Web of Science ®Google Scholar
• Zhang R, Gascon R, Miller RG et al. J Neuroimmunol. 2006;179:87–93.
   PubMed Web of Science ®Google Scholar

P223 DEVELOPMENT OF A NEW CONFORMATION-SPECIFIC ANTI-SOD1 ANTIBODY (AJ10) USING A P2X4-LIKE HUMAN SOD1 PEPTIDE AS IMMUNIZING AGENT: IMPLICATIONS FOR IMMUNOTHERAPY IN ALS MICE

Sábado J

Casanovas A

Hernández S

Piedrafita L

Hereu M

Arqué G

Rodrigo H

Esquerda JE

Universitat de Lleida, Facultat de Medicina-IRBLLEIDA, Dept. Medicina Experimental, Lleida, Catalonia, Spain

Email address for correspondence: [email protected]

Keywords: anti-SOD1 antibody, SOD1 misfolding, immunotherapy

Background: Previous reports from our laboratory showed that some anti-purinergic receptor P2X4 antibodies were able to cross-react with misfolded forms of amyotrophic lateral sclerosis (ALS)-linked mutant Cu/Zn superoxide dismutase (SOD1). Cross-reactivity could have been caused by the abnormal exposure of an epitope in the inner hydrophobic region of SOD1 that shared structural homology with the P2X4-immunizing peptide (1,2).

Objectives: To develop antibodies against human SOD1 epitope mimicked by the P2X4 immunizing peptide and evaluate its immunotherapeutic capabilities.

Methods: The human SOD1 sequence used to raise the antibody was VKVWGSIKGLTEGLHGFHVHEFGDNTAGC. The specificity of the antibody was tested using ELISA, immunoprecipitation, and immunohistochemistry in tissues from transgenic SOD1^G93A and SOD1^G85R mice. NSC34 cells overexpressing mutant SOD1 were also used. The possible therapeutic effects of this peptide used as a vaccine was investigated in low-copy SOD1^G93A mice (life span, 7–8 months) and evaluated using motor behavioral testing (rotarod, catwalk XT, open field, and ladder task) and histopathological analysis.

Results: AJ10 antibody recognizes mutant or misfolded forms of ALS-linked mutant but not SOD1 WT in native conformation. This was demonstrated in the cell line NSC34 expressing G93A or A4V mutant SOD1. AJ10 immunoreactivity was selectively associated with degenerating neurons, but not with glial cells from ALS mice overexpressing either mutant SOD1^G93A or SOD1^G85R. Neurons with strongly positive AJ10 immunostaining are often associated with activated microglial cells displaying neuronophagic activity. AJ10-immunopositive SOD1 aggregates were also found in human spinal cord from a patient with a SOD1-linked familial ALS. AJ10-immunoreactive mutant SOD1 conformers were localized in large intracellular protein aggregates with a filamentous amyloid-like organization by ultrastructural immunolabeling and also detected in neuronal organelles. AJ10 peptide in adjuvant was administered at distinct points. Although AJ10 vaccination induced specific antibody response, preliminary results did not show any significant improvement of disease as a result.

Discussion and conclusion: All these data are congruent with the ability of the AJ10 antibody to recognize misfolded conformations of SOD1 shared by different ALS-linked SOD1 mutations but not with the native protein. Although positive effects of immunotherapy have been obtained with another SOD1 peptide (3), vaccination with AJ10 peptide does not improve ALS in our models. We conclude that neuronal mutant SOD1 conformers detected with AJ10 antibody may have pathogenetic relevance in the promotion of neuroinflammation and may define a new epitope in SOD1 for ALS research and therapy.

Acknowledgements

Supported by Spanish Ministerio de Ciencia e Innovación (Plan Nacional I+ D+ i 2008–2011; SAF 2011–22908).

References

• Casanovas A, Hernández S, Tarabal O et al. J Comp Neurol. 2008;506:75–92.
   Google Scholar
• Hernández S, Casanovas A, Piedrafita L et al. J Neuropathol Exp Neurol. 2010;69:176–187.
   PubMedGoogle Scholar
• Liu H, Tjostheim S, DaSilva K et al. J Neurosci 2012;32:8791–8799.
   Web of Science ®Google Scholar

P224 EFFECT OF THYMIC STIMULATION OF CD4 + T-CELLS ON DISEASE ONSET AND PROGRESSION IN MUTANT SOD1 MICE

Sheean R¹

Weston R¹

Perera N¹

D’Amico A²

Nutt S²

Turner B¹

^ahFlorey Institute of Neuroscience and Mental Health, Melbourne, VIC, Australia

^aiWalter and Eliza Hall Institute, Melbourne, VIC, Australia

Email address for correspondence: [email protected]

Keywords: thymus, regulatory T lymphocytes, SOD1

Background: There is increasing evidence for a role of inflammation mediated by astrocytes, microglia and peripheral T cells in the progression of ALS and motor neuron death. Specifically, ablation of CD4 + T cells accelerates disease course in mutant SOD1 mice, while passive transfer of CD4 + T cells, particularly T regulatory cells, ameliorates progression. T cells must first migrate to the thymus where they mature into CD4 + and/or CD8 + cells. Here, we use castration to naturally activate the thymus through circulating androgen deprivation to stimulate CD4 + cells in mutant SOD1 mice.

Objectives: To investigate the effects of thymic stimulation on the disease course and neuropathology in male transgenic SOD1^G93A mice.

Methods: Male pre-pubescent SOD1^G93A mice were castrated or sham operated, and disease onset and progression were measured using body weight loss, locomotor ability and survival. Thymic stimulation was confirmed by measuring thymus weight and total, CD4 + and CD8 + T-cell numbers were determined using FACS analysis of blood. Motor neuron, astrocyte and microglia were counted from spinal cords using immunohistochemistry. Androgen receptor (AR) expression was also examined in spinal cords of mice using Western blotting and immunohistochemistry.

Results: Castration significantly reversed thymus atrophy and enhanced peripheral CD4 + T-cell numbers in SOD1^G93A mice from pre-symptomatic disease. Despite stimulating thymic activity and CD4 + T-cell production, disease onset and progression were not significantly affected in SOD1^G93A mice. Thymic stimulation did not prevent motor neuron loss or reactive astrocytosis in spinal cords of SOD1^G93A mice; however, microgliosis was suppressed. We also show that AR is principally expressed on motor neurons, but not astrocytes and microglia, in spinal cord, and that AR is depleted from spinal cords of presymptomatic SOD1^G93A mice and is exacerbated by castration.

Discussion and conclusion: These results demonstrate that augmenting thymic function and related CD4 + T-cell production confer no benefit on the disease evolution or course in mutant SOD1 mice, in contrast to passive transfer of CD4 + T-cells. We show for the first time that AR expression is diminished in spinal motor neurons mutant SOD1 mice. Reduced AR expression on spinal motor neurons which links to another motor neuron disease, spinal bulbar muscular atrophy (SBMA), may therefore be a player in ALS pathogenesis and suggests an overlap in pathogenesis of ALS and SBMA.

P225 GENDER-SPECIFIC CONTRIBUTIONS TO ALTERED MOTONEURON SIZE IN SOD1G93A MOUSE MODEL OF ALS

Quinlan KA¹

Jiang M¹

Elbasiouny SM²

Lamano JB¹

Eissa T¹

Samuels J¹

Heckman CJ¹

^ajNorthwestern University, Feinberg School of Medicine, Chicago, IL, USA

^akWright State University, Dayton, OH, USA

Email address for correspondence: [email protected]

Keywords: gender, anatomy, SOD1 mouse

Background: Previous studies have shown a gender bias in both ALS patients and ALS animal models, with males showing more vulnerability. Within those patients and in SOD1 mutant mice, the first motor units lost are the largest, fast twitch motor units starting with the largest fast fatigable (IIB) units and progressing to the intermediately sized fast fatigue-resistant units (IID/X), while small, slow motor units (IIA) show most resistance. In our previous electrophysiological studies on neonatal, juvenile and adult SOD1^G93A mice, we noted a consistent alteration in input conductance in spinal motoneurons. This indicates (indirectly) that SOD1G93A motoneurons may be physically larger in size, and that size could contribute to vulnerability along with sex.

Objective: We are directly comparing spinal motoneuron morphology in male and female adult mice during the time of fast twitch motor unit loss, well in advance of symptom onset.

Methods: Two-photon imaging was performed on motoneurons in transverse slices of sacral spinal cord and neuronal reconstructions compiled with Neurolucida software. Transgenic, overexpressed SOD1^G93A mutant, transgenic overexpressed SOD1 wild-type, and non-transgenic wild-type mice are compared at postnatal days 30 and 50. The parameters under examination include soma surface area; soma volume; and largest soma cross sectional area; number of stem dendrites; number of dendritic nodes (branching points); dendritic surface area; and dendritic volume.

Results: We find an increase in size in SOD1^G93A motoneurons, particularly evident in the significantly increased soma surface area, increased number of stem dendrites, and increased total dendritic length. These anatomical differences between wild-type and G93A mice are most pronounced in males. Male SOD1^G93A motoneurons have significantly greater soma surface area and significantly more stem dendrites than SOD1WT motoneurons, while female SOD1^G93A motoneurons have significantly greater dendritic length than their SOD1WT counterparts.

Discussion and conclusion: Motoneuron size is increased in SOD1^G93A mice, and more prominently in males than in females. This more pronounced phenotype in males could correspond to the delayed disease onset and longer lifespan of female vs. male SOD1^G93A mice.

Acknowledgements

Support was provided by NIH NINDS NS 077863 (CJH), with additional funding from ALS Association 1626; NIH NINDS F32 NS063535 (KAQ); Tim E Noël fellowship from ALS Society of Canada (SME); Undergraduate Research Grant from NU Office of the Provost (JBL). Authors made use of equipment supported by NS054850.

P226 VISUAL ANALYSIS AND INVESTIGATION OF TIMING AND EXTENT OF SENSORY NERVOUS SYSTEM DEGENERATION IN ALS USING A NOVEL REPORTER LINE

Genc B¹

Menichella D²

Yasvoina M¹

Tu M¹

Miller R²

Ozdinler PH^1,3

^alDavee Department of Neurology & Clinical Neurology Science

^amDepartment Molecular Pharm & Biol Chem

^anRobert H Lurie Comp Cancer Centre & Cog. Neurol & Alzheimer’s Dis Ctr Northwestern Univ, Chicago, IL, USA

Email address for correspondence: [email protected]

Keywords: sensory nervous system, dorsal root ganglia, uchl1

Background: Amyotrophic lateral sclerosis (ALS) is mainly characterized by motor neuron degeneration; however, potential involvement of the peripheral nervous system (PNS) to disease pathology has been suggested (1,2). The spatial and temporal pattern of SNS degeneration has not been studied with respect to disease progression. We recently generated a novel ALS mouse model reporter line, the hSOD1^G93A-UeGFP mice, in which corticospinal and spinal motor neurons, together with sensory neurons, are genetically labeled with enhanced green fluorescent protein (eGFP) expression that lasts into adulthood (3). This mouse model offers, for the first time, a comparative study of peripheral neurons together with motor neurons and allows investigation of the timing and extent of potential SNS degeneration with respect to motor neuron degeneration and disease pathology.

Objectives: To investigate the timing and extent of PNS degeneration at different stages of disease, and to assess the potential importance and relevance of sensory neuropathology in ALS, by measuring both anatomical and behavioral differences.

Methods: Sensory neurodegeneration was studied by quantification of epidermal nerve density using eGFP⁺ distal peripheral nerves in the footpads of the hSOD1^G93A-UeGFP mice at postnatal days (P) 30, 60, 90, and 120. Based on our initial results, von Frey test was used to determine their behavioral responses at P30 and P50.

Results: Preliminary data show a progressive and significant decrease in epidermal nerve density in the hSOD1^G93A-UeGFP mice by P90. Von frey test also reveal a significant difference in 50% withdrawal threshold at P50 with the hSOD1^G93A mice displaying allodynia, further confirming early sensory defect.

Discussion and conclusion: Decrease in epidermal nerve density has been reported in skin biopsies from late-stage ALS patients (4). Our results suggest that sensory neuropathy is an early defect in ALS, and that sensory neuron degeneration parallels that of motor neurons in ALS, adding complexity to the disease.

Acknowledgements

This work has been supported by grants from Les Turner ALS Foundation, Wenske Foundation and Brain Research Foundation (to P.H.O.), NIH M.A.D. Training Grants 5T32AG020506-09 (to B.G) and 5T32AG020506-10 (to M.V.Y).

References

• Fischer LR, Culver DG, Davis AA et al. Neurobiol Dis. 2005;19:293–300.
   PubMedGoogle Scholar
• Guo YS, Wu DX, Wu HR et al. Exp Mol Med. 2009;41:140–50.
   PubMed Web of Science ®Google Scholar
• Yasvoina MV, Genç B, Jara JH et al. J Neurosci. 2013;33:7890–7904.
   PubMed Web of Science ®Google Scholar
• Weis J, Katona I, Müller-Newen G et al. Neurology. 2011; 76:2024–9.
   PubMed Web of Science ®Google Scholar

P227 ALSIN KO-UEGFP MICE REVEAL LACK OF CSMN LOSS, BUT AXONAL DEFECT IN THE ABSENCE OF ALSIN FUNCTION AT P500

Yasvoina M

Jara J

Yang N

Ozdinler PH

Northwestern University, Chicago, USA

Email address for correspondence: [email protected]

Keywords: alsin, upper motor neurons, CST

Background: Mutations in ALS2 geneis found to be one of the underlying causes for juvenile primary lateral sclerosis (1), infantile onset ascending hereditary spastic paraplegia (2), and is the most common cause for autosomal recessive juvenile ALS (1). In addition, upper motor neuron signs and bulbar symptoms are often prevalent in patients with juvenile ALS. Even though Alsin KO mice show no apparent motor function defect, it is still not clear whether upper motor neurons are affected at a cellular level in this disease model.

Objectives: To visualize and assess corticospinal motor neuron (CSMN) cellular degeneration in Alsin Ko mice (3), using a novel reporter line, the Alsin KO-UeGFP, which is generated by crossbreeding of Alsin KO and UCHL1^-eGFP mouse models.

Methods: We generated UCHL1^-eGFP mice, which express eGFP selectively in CSMN in the motor cortex. Expression of eGFP lasts up to P800 in the mouse, allowing detailed cellular analysis of CSMN in disease models with longer life-spans (4). Using quantitative measures of genetically labeled CSMN in WT-^UeGFP and Alsin KO-^eGFP mice, we investigated the timing and the extent of CSMN cellular degeneration and axonal defects at P300 and P500.

Results: Our studies suggest the absence of prominent CSMN cellular degeneration and cell loss even at 500. We determined that CSMN are preserved and lack signs of degeneration in the motor cortex of the aged Alsin KO-^UeGFPmice. However, there were signs of axonal degeneration at the level of pons at P500.

Discussion: It is important to visualize CSMN to assess the extent and timing of their degeneration with respect to disease pathology in mouse. Generation of Alsin KO-^UeGFP mouse allowed visualization of CSMN at P300 and P500, but revealed the absence of CSMN loss event at P500. However, there was axonal defect, and CST degeneration was detected at the level of pons by P500.

Acknowledgements

We thank Dr. Siddique for providing Alsin^-/- breeding stock. This work was supported by the grants from Les Turner ALS Foundation, Wenske Foundation and NIH M.A.D; and by training Grants 5T32AG020506-10 (to M.V.Y)

References

• Yang Y, Hentati A, Deng HX et al. Nat Genet. 2001;29:160–165.
   PubMed Web of Science ®Google Scholar
• Eymard-Pierre E, Lesca G, Dollet S et al. Am J Hum Genet. 2002;71:518–527.
   PubMed Web of Science ®Google Scholar
• Deng HX, Zhai H, Fu R et al. Hum Mol Genet. 2007; 16:2911–2920.
   PubMed Web of Science ®Google Scholar
• Yasvoina MV, Genc B, Jara JH et al. J Neurosci. 2013;33:7890–7904.
   PubMed Web of Science ®Google Scholar

P228 NEUREGULIN-1 IS ASSOCIATED WITH POSTSYNAPTIC SITES CONTACTING AFFERENT C-TYPE CHOLINERGIC TERMINALS ON LOWER MOTONEURONS: CHANGES IN MURINE MODELS OF MOTONEURON DISEASES

Gallart-Palau X

Tarabal O

Casanovas A

Sábado J

Correa F

Cerveró C

Hereu M

Piedrafita L

Calderó J

Esquerda JE

Universitat de Lleida, Facultat de Medicina-IRBLLEIDA, Dept. Medicina Experimental, Lleida, Catalonia, Spain

Email address for correspondence: [email protected]

Keywords: neuregulin, C-terminals, spinal muscular atrophy

Background: Pioneering ultrastructural studies have defined different types of synapses (F-, S- and C-type boutons) on lower motoneurons (MNs). C-Type boutons are large nerve terminals characteristic of somatic α-MNs and display a subsynaptic cistern adjacent to postsynaptic membrane (1). The cholinergic nature of C-terminals has been established (2), and their origin from local interneurons has been deciphered (3). In a recent study (4) concerning neuregulin-1 (NRG-1) on phrenic MNs, it has been reported that this trophic factor is expressed in cholinergic terminals synapsing these MNs and a presynaptic localization of this protein was suggested.

Objectives: 1) To study the distribution of NRG-1 immunoreactivity (IR) amongst different MN groups; 2) to examine the localization of NRG-1 on spinal cord MNs by immunoelectron microscopy; 3) to perform a developmental analysis of NRG-1 expression; and 4) to explore NRG-1 changes in MNs of transgenic mouse models of spinal muscular atrophy (SMA) and ALS.

Methods: Immunofluorescence was performed on spinal cord from WT, SMNΔ7 and SOD1^G93A mice and chick embryos. Analysis was done using confocal microscopy and ultrastructural immunolabelling.

Results: In normal adult WT animals, NRG-1-IR was concentrated close to VAChT-positive cholinergic terminals around MN somata and proximal dendrites. However, a displacement of NRG-1 signal respect to presynaptic VAChT-positive spots was evidenced, suggesting a different compartmentalisation. Ultrastructural examination revealed that NRG-1 was not associated with presynaptic terminals but with postsynaptic subsynaptic cisterns corresponding to C-terminals. A similar pattern of NRG-1-IR was found in brainstem MNs with the exception of those in oculomotor nuclei, in which NRG-1 expression was much lower. In chick embryo MNs, NRG-1 is early expressed (before E6); only at advanced developmental stages, NRG-1-IR appeared concentrated in relation to synaptic sites being downregulated after axotomy. In SMNΔ7 and SOD1^G93A, mice synaptic NRG-1 spots on MNs increased early during disease progression with a subsequent fall at the end stages.

Discussion and conclusion: 1) NRG-1 is concentrated at the subsynaptic cysterns of α-MN afferent cholinergic C-terminals; 2) ALS-resistant MNs at the oculomotor nuclei show low NRG-1 expression; 3) NRG-1 is developmentally regulated and depends on the maintenance of nerve muscle interactions; 4) the transient increase in NRG-1-positive spots on MN somata during SMA or ALS progression is in concordance with the described sprouting of C-terminals in ALS (4).

Acknowledgements

Supported by the Ministerio de Economía y Competitividad and FEDER (SAF2011–22908; SAF2012–31831).

References

• Conradi S. Acta Physiol Scand 1969;332(Suppl):5–48.
   Google Scholar
• Hellström J et al. J Comp Neurol. 1999;411:578–590.
   Google Scholar
• Zagoraiou L et al. Neuron 2009;64:645–662.
   PubMed Web of Science ®Google Scholar
• Issa AN et al. J Comp Neurol. 2010;518:4213–4225.
   Google Scholar
• Pullen AH, Athanasiou D. Eur J Neurosci. 2009;29:551–561.
   Google Scholar

P229 NEUROINFLAMMATION IN ALS: THE COMPLEX ROLE OF P2X7 RECEPTOR

Apolloni S^1,2

Amadio S²

Parisi C^1,2

Montilli C²

Carrì MT^2,3

Cozzolino M^2,4

Volonté C^1,2

D’Ambrosi N¹

^aoCellular Biology and Neurobiology Institute, CNR, Rome, Italy

^apSanta Lucia Foundation, IRCCS, Rome, Italy

^aqDepartment of Biology, University of Rome “Tor Vergata”, Rome, Italy

^arInstitute of Translational Pharmacology, CNR, Rome, Italy

Email address for correspondence: [email protected]

Keywords: P2X7 receptor, neuroinflammation, NADPH oxidase

Background: Inflammation and oxidative stress play determinant roles in the pathogenesis of ALS (1). Degenerating motor neurons produce signals that activate microglia to release ROS and proinflammatory cytokines determining a vicious cycle of neurodegeneration. Extracellular ATP through P2X7 receptor constitutes a neuron-to-microglia alarm signal, and P2X7 has been shown to be implicated in ALS (2). The comprehension of the mechanisms underlying the action of P2X7 might thus be important in counteracting the progression of the disease.

Objectives: Aims of our study were to investigate: a) the effects of pharmacological modulation and genetic ablation of P2X7 activity in microglia on the main ROS-producing enzyme NADPH oxidase 2 (NOX2), a well-known player in the pathogenesis of ALS; b) the disease outcome determined by thelack of P2X7 in SOD1^G93A mice.

Methods: For the in vitro study, primary microglia derived from SOD1^G93A and P2X7^-/-SOD1^G93A mice were challenged with P2X7 agonist 2’-3’-O-(benzoyl-benzoyl) ATP and specific antagonists in order to analyze NOX2 activity and related pathways. For the in vivo approach, hetero- and homozygous P2X7 receptor knock-out SOD1^G93A mice were generated and analyzed for body weight, disease onset and progression, and survival.

Results: We observed that stimulation of P2X7 enhanced NOX2 activity and ROS production in SOD1^G93A microglia. We also found that phosphorylation of ERK1/2 was augmented in ALS-microglia and there was a mutual dependency between the NOX2 and ERK1/2 pathways. All these damaging mechanisms were prevented by blocking P2X7. Despite these in vitro results, in P2X7^-/-/SOD1^G93A mice, the clinical onset was significantly anticipated and the disease progression worsened with consistent increased astrogliosis, microgliosis, and motoneuron loss in the lumbar spinal cord of mice at end stage.

Discussion and conclusion: Altogether, these results show that although a noxious mechanism leads in isolated ALS-microglia to enhanced oxidative stress through P2X7, its constitutive deletion in SOD1^G93A mice aggravates the pathogenesis of ALS. These findings suggest that P2X7 might have a complex role with beneficial inflammatory effects exerted in at least definite stages of the disease. This supports the emerging dual role of neuroinflammation in ALS (3) and highlights the importance of a successful time window of therapeutic intervention against it.

Acknowledgements

This work was supported by AriSLA-Agenzia Italiana per la Ricerca sulla Sclerosi Laterale Amiotrofica (grant PRALS 2009, co-financed with the support of “5 × 1000”– Healthcare Research of the Ministry of Health).

References

• Philips T, Robberecht W. Lancet Neurol. 2011;10:253–263.
   PubMed Web of Science ®Google Scholar
• Volonté C, Apolloni S, Carrì MT et al. Pharmacol Ther. 2011;132:111–122.
   PubMedGoogle Scholar
• Beers DR, Zhao W, Liao B et al. Brain Behav Immun. 2011;25:1025–1035.
   PubMed Web of Science ®Google Scholar

P230 KNOCKING-DOWN MGLUR1 AND MGLUR5IN SOD1G93A MICE AMELIORATES SURVIVAL AND DISEASE PROGRESSION

Bonifacino T¹

Milanese M¹

Melone M²

Giribaldi F¹

Musante I³

Vergani L⁴

Voci A⁴

Puliti A³

Conti F²

Bonanno G^1,5

^asDepartment of Pharmacy, Pharmacology and Toxicology Unit

^atDepartment of Earth, Environment and Life Sciences

^auCenter of Excellence for Biomedical Research University of Genoa, Genova, Italy

^avDepartment of Experimental and Clinical Medicine, Unit of Neuroscience and Cell Biology, Università Politecnica delle Marche and Center for Neurobiology of Aging, INRCA IRCCS, Ancona, Italy

^awDepartment of Neuroscience, Rehabilitation, Ophthalmology, Genetics and Maternal and Child Health, Medical Genetics Unit, University of Genoa and Gaslini Institute, Genova, Italy

Email address for correspondence: [email protected]

Keywords: G = group I metabotrophic glutamate receptors, glutamate-mediated excitotoxicity, genetic mouse model

Background: Glutamate (Glu)-mediated excitotoxicity plays a major role in the degeneration of motor neurons (MNs) in amyotrophic lateral sclerosis (ALS), and reduced astrocytic uptake was suggested as a cause for the increased synaptic availability of Glu (1). On the basis of our studies, we have proposed that abnormal release may represent another source for excessive extracellular Glu levels. Acting at the altered Glu release mechanisms may represent a possible strategy for new therapeutic approaches to ALS (2,3). Our previous results indicated the existence of excessive Glu release in the spinal cord of SOD1G93A mice following activation of Group I metabotropic Glureceptors (mGluR1 and mGluR5) expressed at the pre-synaptic level (4).

Objectives: To prove the positive impact of mGluR1 or mGluR5 blockade in experimental ALS.

Methods: To provide a genetic tool to evaluate the role of mGluR1 and mGluR5 in ALS, we generated mice carrying half expression of mGluR1 in the SOD1^G93A background, by crossing SOD1G93A mice with heterozygous Grm1^± mice. In the same line, we also generated mice carrying half expression of mGluR5, by crossing SOD1^G93A mice with Grm5^± mice. Life span, motor abilities, MN preservation, mitochondrial damage, oxidative stress markers, astrogliosis and microglia activation, receptor expression and Glu release were investigated to characterize double-mutant mice compared to the SOD1^G93A ALS model.

Results: SOD1^G93A_Grm1^± double-mutant mice showed prolonged survival probability respect to SOD1^G93A mice. Accordingly, slower disease progression and improved motor performances were observed. Interestingly, knocking down mGluR1 also reduced mGluR5 expression in the spinal cord of double-mutant mice. Histological studies performed at the late symptomatic phase of the disease showed a significant reduction on MNs death in spinal cordventro-lateral horns accompanied by lower astrocyte and microglia activation and normalization of the up-regulation of metallothionein mRNA expression in SOD1^G93A_Grm1^± double mutants respect to SOD1^G93A mice. Reduced mitochondrial damage was also observed in soma, dendrites, axons and axons terminal in double-mutant mice. Abnormal Glu release induced by the activation of Group I mGluRs was reduced in SOD1^G93A_Grm1^± compared to that in SOD1^G93A mice. Also SOD1^G93A_Grm5^± mice, lacking of mGluR5, showed remarkable prolonged survival and phenotype amelioration.

Discussion and conclusion: mGluR1 or mGluR5 downregulation has a significant impact in vivo on experimental ALS. These results would provide the rationale for pharmacological approaches to ALS by selectively blocking Group I mGluRs.

References

• Rothstein JD, Van Kammen M, Levey AI et al. Ann. Neurol. 1995;38:73–84.
   Web of Science ®Google Scholar
• Milanese M, Zappettini S, Onori F et al. J. Neurochem. 2011;116:1028–1042.
   PubMed Web of Science ®Google Scholar
• Uccelli A, Milanese M, Principato MC et al. Mol. Med. 2012;18:794–804.
   Google Scholar
• Giribaldi F, Milanese M, Bonifacino T et al. Neuropharmacol. 2013;66:253–263.
   PubMed Web of Science ®Google Scholar

P231 CALCITONIN GENE-RELATED PEPTIDE SIGNALING INFLUENCES MOTOR SYMPTOM ONSET AND DISEASE PROGRESSION IN THE SOD1-G93A MOUSE MODEL OF ALS

Ringer C¹

Tsujikawa K²

Weihe E¹

Schütz B¹

^axInstitute of Anatomy and Cell Biology, Philipps-University, Marburg, Germany

^ayLaboratory of Molecular and Cellular Physiology, Graduate School of Pharmaceutical Sciences, Osaka University Yamadaoka, Osaka, Japan

Email address for correspondence: [email protected]

Keywords: neuropeptide, neuroinflammation, cgrp

Background: In ALS, neuroinflammation contributes to disease initiation and progression. Recently, we have shown that alterations in the subcellular distribution of the β-isoform of the neuropeptide calcitonin gene-related peptide (CGRP) in motor neurons precede astrogliosis (1), and that CGRP expression levels predict motor neuron vulnerability (2) in the SOD1^G93A mouse model of ALS.

Objectives: To further elucidate the possible pathogenic role of CGRP on ALS disease progression, we crossbred SOD1 mice with mice depleted of the CGRP-specific receptor component, receptor activity-modifying protein 1 (RAMP1), and monitored clinical and histological symptom development and progression under the presence and absence of functional CGRP signaling.

Methods: Survival rate, body weight, and motor functions were monitored for wild-type; RAMP1^-/-; SOD1; and SOD1:RAMP1^-/- mice with at least 10 animals per group. In addition, motor neuron numbers, activations of astro- and microglia, and lymphocyte infiltrations were investigated in all genotypes by immunohistochemistry at different disease stages.

Results: SOD1:RAMP1^-/- mice showed an earlier onset of hind limb motor deficits compared to SOD1:RAMP1^+/+ mice (56 ± 30 days vs. 92 ± 20 days), while overall survival was similar (median, 142 days vs. 137 days). At the histological level, advanced onset of motor neuron degeneration in the lumbar aspect of the spinal cord was present in SOD1:RAMP1^-/- mice compared to that in SOD1:RAMP1^+/+ mice, but followed by decelerated motor neuron loss throughout disease progression. In addition, morphological activations of astrocytes and microglia, and lymphocyte infiltrations were attenuated in SOD1:RAMP1^-/- mice, in pre-symptomatic, early symptomatic, and end-stage as compared to SOD1:RAMP1^+/+ mice.

Discussion and conclusion: The observed effects under suppression of CGRP signaling in SOD1 mice suggest that the secretion of CGRP by motor neurons is neuroprotective predominantly at disease-onset, possibly through stimulation of a protective neuro-inflammatory milieu. In contrast, continuous CGRP signaling on glia at later stages furthers disease progression by promoting chronic neurodestructive neuroinflammation.

Acknowledgements

This study was supported by a research fund from the University Medical Center Giessen and Marburg (UKGM), the Foundation P.E. Kempkes Mar-burg, and the German Society for the Muscular Diseased (DGM).

References

• Ringer C, Weihe E, Schütz B Neurobiol . Dis 2009;35:286–295.
   Google Scholar
• Ringer C, Weihe E, Schütz B Neurobiol . Dis 2012;45:547–554.
   Google Scholar

P232 ERYTHROPOIETIN MODULATES IMMUNE-INFLAMMATORY RESPONSE IN A SOD1 (G93A) MOUSE MODEL OF AMYOTROPHIC LATERAL SCLEROSIS

Noh M-Y

Cho KA

Kim H

Choi H

Kim SH

Department of Neurology, Hanyang University College of Medicine, Seoul, Republic of Korea

Email address for correspondence: [email protected]

Keywords: amyotrophic lateral sclerosis, erythropoietin, inflammation

Background: Amyotrophic lateral sclerosis (ALS) is characterized by the appearance of reactive microglia and astroglia cells, a process referred to as neuro-inflammation. The inflammatory reaction has recently received attention as an unexpected potential target for the treatment of ALS. Erythropoietin (EPO) has shown various potent neuroprotective effects, including reduction of inflammation, enhancement of survival signals, and prevention of neuronal cell death. Recently, it has been shown that recombinant human EPO (rhEPO) is able to cross the BBB, making it a good candidate to target the CNS.

Objectives: This study was undertaken to evaluate the temporal pattern of inflammation cytokine by rhEPO treatment on SOD1^G93A mouse model of ALS.

Methods: We treated ALS model mice with vehicle only or 5 iμ of rhEPO/g of mouse, twice every other week after they were 60 days old. Mice were analysed for inflammation cytokines involvement in the ALS pathogenic process at asymptomatic stage (30and 60 days), at early symptomatic stage (90 days) and at 120 days.

Results: In the real-time quantitative PCR (qPCR) analysis, spinal cord tissue from SOD1^G93A mice at 90 days expressed higher levels of anti-inflammatory cytokines including TGF-β, interleukin (IL)-4 and IL-10 mRNA. In addition, late-symptomatic stage at 120 days expressed higher levels of pro-inflammatory cytokines including interferon gamma (IFN)-γ, MCP-1, IL-1β, CCL5, CXCL10 and tumour necrosis factor (TNF)-α mRNA. More importantly, when injected with rhEPO at 60 days, these anti-inflammatory cytokines (TGF-β and IL-10) were continued to the mRNA level of symptomatic stage and IL-10 was up-regulated. However, pro-inflammatory cytokines were markedly reduced from 90 days.

Discussion and conclusion: The treatment of rhEPO significantly changed temporal inflammatory cytokine in SOD1 ^G93A mouse model of ALS. These findings suggest that rhEPO was a potential therapeutic strategy through the modulation of neuro-inflammation in ALS.

Acknowledgements

This study was supported by a grant of the Korean Health Technology R&D Project, Ministry for Health, Welfare & Family Affairs, Republic of Korea. (A120182).

P233 NEUROINFLAMMATION AND MUSCLE DENERVATION IN WOBBLER MICE

Kano O¹

Kawabe K¹

Yoshii Y¹

Ishikawa Y²

Ishii T²

Ikeda K¹

Iwasaki Y¹

^azDepartment of Neurology, Toho University Omori Medical Center, Tokyo, Japan

^baDepartment of Pathology, School of Medicine, Toho University, Tokyo, Japan

Email address for correspondence: [email protected]

Keywords: inflammation, denervation, microglia, astrocyte

Objective: We evaluated the temporal relationship among diaphragm denervation, phrenic nerve and corresponding cervical cord inflammation. In addition, we examined how an inflammatory immune response contributes to the balance between neuroprotection and neurotoxicity.

Background: Neuroinflammation plays a role in the pathogenesis of amyotrophic lateral sclerosis. Neuropathological hallmarks of Wobbler mice revealed cervical motor neuron degeneration and motor axonopathy. However, little is known about the role of inflammation in Wobbler mice.

Methods: Cervical cord (C4)-phrenic nerve-diaphragm unit from 20 (early symptomatic stage)-, 40 (rapid progressive stage)-, 60 (chronic stage)-day-old Wobbler mice and age-matched wild-type littermates (C57BL/6, n = 3 each) were assessed for an inflammatory immune response using quantitative RT-PCR (CD68, GFAP, BDNF, GDNF, IL-4, TNF-α, IL-1β and NOX2) and immunofluorescent histochemistry. Denervation was evaluated in diaphragm using qRT-PCR for the temporal changes in the mRNA levels of γ(fetal) and ϵ(adult) acethylcholine receptor (AChR) subunits. Data were analysed using two-tailed Student's t–test, and group means were plotted ± SEM; p < 0.05 was considered statistically significant. Differences between groups were analysed using a two-way ANOVA.

Results: Compared with WT control mice, expression of GFAP (marker of astrocytes) in cervical cord, and AchR γ subunit in diaphragm increased at 20 days. Whereas CD68 (marker of monocytes/macrophages (microglia)) was elevated in cervical cord after 40 days. The expression of CD68 was not seen in phrenic nerve. Anti-inflammatory factors, including BDNF, GDNF and IL-4, were not increased over an entire period. In contrast, the neurotoxic markers TNF-α, IL-1β and NOX2 were increased after 20 days. Immunohistochemical study showed that astrocytes secrete TNF-α in cervical cord in Wobbler mice.

Conclusions: We suggest that activated astrocytes play an important role at the early symptomatic stage in Wobbler mice. Muscle denervation precedes microglial activation in cervical cord.

P234 TESTOSTERONE AND SYNTHETIC ANABOLIC STEROIDS ARE MODIFIERS FOR MUTANT SOD1-RELATED ALS PATHOGENESIS

Aggarwal T¹

Galbiati M²

Rizzuto E³

Musaro` A³

Poletti A²

Pennuto M^1,4

^bbIstituto Italiano di Tecnologia, Genoa, Italy

^bcUniversità degli Studi di Milano, Milan, Italy

^bdUniversity La Sapienza, Rome, Italy

^beCIBIO, University of Trento, Trento, Italy

Email address for correspondence: [email protected]

Keywords: androgens; androgen receptor; anabolic steroids

Background: Evidence indicates a higher incidence of sporadic ALS (sALS) in males than in females. This difference is very marked in juvenile forms of ALS, but it declines with age, when androgen levels in the serum decrease. Varsity athletes, such as Italian soccer and American football players, and military veterans show a higher risk of developing ALS. Endogenous androgens act on the androgen receptor (AR), and anabolic synthetic steroids, such as nandrolone decanoate, are androgenic compounds illegally used as performance-enhancing agents. This suggests a possible correlation between the risk of developing SALS and dysregulation of androgen signalling.

Objective: To verify this hypothesis, we tested the impact of androgen signalling in the pathogenesis of SOD1-related ALS in transgenic mice expressing mutant SOD1^G93A. Mice were either castrated to eliminate serum androgens or treated with nandrolone decanoate to mimic the condition occurring in doped athletes.

Results: Castration resulted in a significant decrease in body weight of SOD1^G93A mice, but it enhanced muscle strength and motor coordination in ALS-affected mice. Castration also resulted in delayed disease onset and increased survival, although there was no significant effect on disease progression. Castration was associated with decreased levels of expression of AR. On the other hand, treatment of intact mice with nandrolone did not affect the body weight of SOD1^G93A mice, but it resulted in significant improvement in muscle strength and motor coordination. Nandrolone treatment did not affect disease onset and progression, though it increased survival compared to that of control mice. Nandrolone treatment resulted in increased EDL weight, strength and specific force, which was associated with an increase in speed contraction. However, nandrolone effects were also associated with increased fatigue. Surprisingly, treatment of castrated mice with nandrolone deteriorated phenotype, and this was associated with increased AR aggregation. Altogether, these results indicate that androgen signalling plays a role in SOD1-related ALS pathogenesis.

P235 H63D HFE SHORTENS SURVIVAL AND ACCELERATES DISEASE PROGRESSION IN AN ALS MOUSE MODEL

Nandar W

Neely E

Simmons Z

Connor J

The Pennsylvania State University, Hershey, Pennsylvania, USA

Email address for correspondence: [email protected]

Keywords: SOD1/H67D, iron, oxidative stress

Background: There is increasing evidence suggesting an association between H63D HFE and amyotrophic lateral sclerosis (ALS). The presence of H63D HFE is associated with disease processes implicated in ALS such as iron accumulation, oxidative stress, increased glutamate release, and endoplasmic reticulum stress. Therefore, we hypothesize that H63D HFE increases the risk of ALS by promoting the convergence of disease processes implicated in ALS.

Objectives: To create a double transgenic mouse line that carries H67D HFE (homologous to H63D in humans) and the SOD1^G93A mutation to examine how H63D HFE influences ALS pathogenesis.

Methods: We crossed an ALS mouse model SOD1^G93A with H67D mice to generate double transgenic mice (SOD1/H67D). Disease onset was determined by monitoring the motor performance on a rotarod. End-stage was defined as the inability of the animal to right itself within 30 s after being placed on its side. Disease duration was the mean time from onset to end-stage. A grip strength meter was used to measure forelimb and hindlimb strength as markers for disease progression. We measured expression of proteins involved in iron homeostasis and oxidative stress to determine mechanisms by which H63D HFE contributes to ALS. In all experiments, SOD1/H67D mice were compared with SOD1^G93A and wild-type littermates.

Results: The SOD1/H67D mice have a shorter survival and more rapid rate of disease progression than SOD1^G93A mice, although age of disease onset is not different between two groups. Expression of the transferrin receptor (TfR) involved in cellular iron uptake is decreased in both SOD1/H67D and SOD1^G93A mice starting at 90 days. However, in SOD1/H67D mice, decreased TfR expression is accompanied by increased L-ferritin, an iron storage protein. Increased hemeoxygenase-1 and caspase-3 expressions are present at an earlier age in SOD1/H67D compared to SOD1^G93A mice. The elevated oxidative stress in SOD1/H67D mice may be the result of impaired cellular stress response by nuclear factor E2-related factor 2 (Nrf2), which is lower in SOD1/H67D than in SOD1^G93A mice.

Discussion and conclusion: It is noteworthy that no dietary or environmental challenges were presented to the double transgenic mice; therefore, our findings strongly indicate that H63D HFE is a genetic modulator of ALS. Mechanisms underlying more rapid disease progression observed in SOD1/H67D mice include disrupted iron homeostasis and oxidative stress. Because as many as 30% of ALS patients carry the H63D HFE polymorphism, our animal model has meaningful clinical implications and can serve as a preclinical model when evaluating the impact of H63D HFE on treatment strategies for ALS patients.

Acknowledgements

This work is supported by Judith and Jean Pape Adams Charitable Foundation, the Paul and Harriett Campbell Fund for ALS research, Zimmerman Family Love Fund and the Robert Luongo ALS Fund.

P236 CHARACTERISING DISTAL DYSFUNCTION AND DEGENERATION IN ALS: THE POTENTIAL FOR AXON PROTECTION?

Clark JA¹

Blizzard CA¹

Southam K¹

Chuckowree J¹

King A²

Dickson TC¹

^bfMenzies Research Institute

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

Email address for correspondence: [email protected]

Keywords: distal degeneration, neuromuscular junction, axon protection

Background: ALS is characterised by degeneration of the motor axons in the corticospinal tract and lower motor neurons. This die-back also involves the loss of acetylcholine receptor (AChR's) at the neuromuscular junction (NMJ). The precise time course of distal pathology development and NMJ dysfunction in ALS is yet to be fully elucidated. This information is critical for revealing novel drug targets and for specifically investigating the potential therapeutic strategy of ‘distal protection’ for improving ALS outcomes.

Objectives: To characterise the development of distal dysfunction and pathology development at both the axon and NMJ in mSOD1^G93A and YFPxmSOD1^G93A cross transgenic models, and characterise the recently developed novel model of site-specific excitotoxicity prior to determining the effect of treatment with the microtubule stabilizing agent Epothilone D on the development of disease phenotype.

Methods: YFP (Thy1) effects on motor phenotype were undertaken to account for transgene effects (n = 5) using rotarod performance test (TSE and USA), grip strength analysis (Chatillion, USA), and weight change over 11 weeks. Comprehensive immunohistochemical and confocal investigation of mSOD1^G93A (4, 8, 12, 16, and 20week) and YFP x mSOD1^G93Across (8 and 20weeks) transgenic gastrocnemus muscles (n = 4/group, PFA-fixed, 40- or 80-μm frozen sections) was subsequently performed.

Results: Analysis of motor phenotype in YFP mice identified no significant change compared to that in wild type due to transgene expression. Immunohistochemical analysis showed that at 20-week-old YFP x mSOD1^G93A mice had a ratio of 1:1 degenerating axons to healthy axons (2.252 ± 0.92, 2.67 ± 1.01, respectively), with a global loss of axons also apparent. AchR staining in both the YFP x mSOD1^G93A cross, and the mSOD1^G93A tissue resulted in NMJs with fragmented morphology. Non-co-locolized NMJs (αBt-YFP) increased from 8 to 20 weeks (2.15 ± 0.84, p < 0.001). Fully co-localized NMJs showed degenerative morphology at 8 weeks (1.31 ± 0.42, p < 0.05). A significant (p < 0.001) increase in the size of degenerating NMJs occurred over disease progression from 8 to 20 weeks. Between 8 and 20 weeks, NMJs positive for S100 decline (52.4 ± 3.4, 20.5 ± 12.8 SOD; 65.4 ± 4.8, 45.8 ± 6.7 WT); conversely, nestin immunoreactivity significantly (p < 0.05) increases at 8 weeks (73.3 ± 6.7 SOD; 5.0 ± 2.6 WT).

Discussion and conclusion: The changes in axonal pathology typical of ALS offer a key site for potential intervention for future therapeutics. Complete characterization of the functional and pathological changes occurring in the distal axon and NMJ in the mSOD1^G93A and our novel site specific excitotoxicity model will allow for subsequent testing of novel therapeutic targets directed at distal protection. Specifically, drugs such as the microtubule stabilizer Epothilone D may allow improved microtubule function, resulting in sustained axonal transport and subsequently less degeneration and pathology.

Funding

This work was supported by MNDRIA- Mick Rodger Research Grant, and s'ship top-up to JC.

P237 TARGETED ABLATION OF MYELINATING SCHWANN CELLS ENHANCES DISEASE SEVERITY IN SOD1G93A MICE

Sheean R

Stratton J

Merson T

Turner B

Florey Institute of Neuroscience and Mental Health, Melbourne, VIC, Australia

Email address for correspondence: [email protected]

Keywords: SOD1, Schwann cells, diptheria toxin

Background: Mutant SOD1 action within non-neuronal cells is implicated in damage to motor neurons in genetic forms of ALS. Astrocytes, microglia and oligodendrocytes drive progression in mutant SOD1 mice, while the role of Schwann cells is less clear. Schwann cells, the myelinating glial cells of the peripheral nervous system, are intimately associated with motor neurons and are vital for nerve conduction, axonal development, transport and support. The role of Schwann cells in ALS pathogenesis is unequivocal. Selective removal of mutant SOD1 from Schwann cells accelerates disease progression, while restricted expression of mutant SOD1 in Schwann cells is not harmful to mice. To resolve the contribution of Schwann cells to ALS, we generated double transgenic mutant SOD1; MBP-DTR mice that allows selective elimination of myelinating Schwann cells.

Objectives: To investigate the effect of myelinating Schwann cell depletion on disease onset, progression and spinal cord and peripheral nerve pathology in SOD1^G93A mice.

Methods: SOD1^G93A mice were crossed with novel transgenic MBP-DTR mice which express diphtheria toxin receptor (DTR) driven by the myelin basic protein (MBP) promoter. Exogenous administration of diphtheria toxin (DT) to MBP-DTR mice results in selective ablation of 25% of myelinating Schwann cells that is sublethal. Double transgenic SOD1^G93AMBP-DTR mice and control genotypes SOD1^G93A; MBP-DTR; and wild-type (WT) were injected with DT (10μg/kg, ip) at presymptomatic age (P60). Disease onset and progression was determined using rotarod and grid test performance, and survival was assessed. Spinal cords and sciatic nerves were analysed by immunohistochemistry and electron microscopy for motor neuron and axonal counts, Schwann cell apoptosis and myelination.

Results: Administration of DT provoked hindlimb weakness and muscle wasting in MBP-DTR mice, which peaked at 22 days post injection (P 82), followed by rapid recovery by 28 days post injection (P88). This resulted from 25% depletion of Schwann cells and demyelination in sciatic nerves. Double transgenic SOD1^G93AMBP-DTR mice showed increased severity of muscle weakness and wasting at peak symptoms and Schwann cell loss and demyelination. Administration of DT to SOD1^G93A or WT mice did not elicit symptoms or Schwann cell death as mice are naturally resistant to diphtheria.

Discussion and conclusion: Our data demonstrate that DT-induced myelinating Schwann cell ablation and resulting motor dysfunction is enhanced by mutant SOD1 expression in Schwann cells. This suggests that mutant SOD1 damage to Schwann cells sensitises them to death in this toxin-induced model.

P238 BLOCKING DEATH RECEPTOR 6 (DR6) PROMOTES NEUROMUSCULAR JUNCTION INTEGRITY AND FUNCTIONAL RECOVERY IN MOUSE MODEL FOR ALS

Mi S

Huang G

Lee X

Shao Z

Bian Y

Pepinsky B

Biogenidec, Cambridge, USA

Email address for correspondence: [email protected]

Keywords: motor neuron protection, axon integrity, functional recovery

Background: Death receptor 6 (DR6) is a TNFR superfamily member with death domain selectively expressed in CNS. DR6 levels are elevated in spinal cords from postmortem samples of human ALS and from SOD1G93A transgenic mice.

Objective: To determine whether blocking DR6 by its antagonist antibody promotes motor neuron survival in vitro and in SOD1G93A mice.

Methods: Human and rat primary motor neuron cultures were used to determine whether blocking DR6 by anti-DR6 antagonist monoclonal antibody, 5D10 (1,3), protected motor neurons from death using three methods: growth factor removal; sodium arsenite to induce mitochondrial oxidative stress; and astrocyte (SOD1^G93A)-induced cytotoxicity in motor neuron and astrocyte co-culture. We also investigated the neuroprotective effects of blocking DR6 in SOD1^G93A mice by using immunohistochemical staining to quantify the effects of anti-DR6 on neuromuscular junctions (NMJ), motor neurons in spinal cord, gliosis and plasma levels of phosphorylated neurofilament heavy chain (pNFH).

Results: Blocking DR6 with an antagonist antibody (5D10) promotes motor neuron survival in vitro resulting from growth factor withdrawal, sodium arsenite treatment, or co-culturing with SOD1^G93A astrocytes. When compared to control treated littermates, 5D10 treatment protects NMJs from denervation, increased numbers of motor neurons, decreased gliosis in the spinal cords, decreased plasma pNfH levels, and promotes functional recovery in the SOD1G93A mice.

Discussion and conclusion: DR6 has emerged as an important regulator of oligodendrocyte and neuronal cell death (1–3). DR6 antagonism can protect motor neuron survival in vitro and in SOD1^G93A mice. Blocking DR6 function may represent a new approach for the treatment of neurodegenerative disorders involving motor neuron death and axon degeneration such as ALS.

References

• Mi S et al. Death receptor 6 negatively regulates oligodendrocyte survival, maturation and myelination. Nat Med 2011; 17:816–821.
   Google Scholar
• Nikolaev A, McLaughlin T, O’Leary DD, Tessier-Lavigne M. APP binds DR6 to trigger axon pruning and neuron death via distinct caspases. Nature 2009;457:981–989.
   PubMed Web of Science ®Google Scholar
• Hu Y et al. A DR6/p75(NTR) complex is responsible for beta-amyloid-induced cortical neuron death. Cell death & disease. 2013;4:e579.
   Google Scholar

P239 EXCITOTOXICITY AND NEUROMUSCULAR JUNCTION DEGENERATION FOLLOWING SITE-SPECIFIC EXCITOTOXIN EXPOSURE IN VIVO

Blizzard CA

Clark JA

Dickson TC

Menzies Research Institute Tasmania, Hobart, Tasmania, Australia

Email address for correspondence: [email protected]

Keywords: excitotoxicity in vivo mouse

Background: Amyotrophic lateral sclerosis (ALS) is likely to be a multifactorial disease of neuronal dysfunction and loss; however, recent investigations indicate that axonal dysfunction, prior to cell loss, may be the causative factor of the initial symptoms of ALS and that distal axonal degeneration may occur before the onset of disease symptoms.

Purpose: Our investigations are focused on determining the degenerative changes underlying ALS-like axonopathy using site-specific excitotoxic insults, via osmotic mini pumps, to the spinal cord and muscle.

Methods: To achieve site-specific excitotoxicity osmotic mini pumps (Alzet, model 1004), a constant chronic infusion to either the L3–4 lumbar region of the spinal cord or the gastrocnemius muscle in the hind limb was delivered. A constant and chronic infusion of Kanic acid (KA, 5–10mM, in cortex buffer with 1μM Fluro Ruby) was delivered to the subarachnoid space of the lumbar region (L3–5) of C57/Bl6 mice and transgenic mice which express yellow fluorescent protein (YFP) in a subset of motor neurons on a C57/BL6 background. At the gastrocnemius muscle, glutamate (5–10mM in saline with 1μM Fluro Ruby) was chronically infused in YFP mice. Animals were perfused at a range of time points, and the degree of axonal degeneration was investigated through immunohistochemistry and confocal microscopy.

Results: Fluoro ruby labelling was present throughout cells within the subarachnoid space in L3–L5 and muscle fibres of the gastrocnemius muscle, indicating that a targeted delivery can be achieved with the osmotic pumps. Quantitation of the number of neurons (cell body > 20 μm) stained with toluidine blue within the anterior ventral horn at 7, 14 and 28 days post surgery (DPS) demonstrated a significant (p < 0.05) decrease in number of motor neurons at 28 DPS in the 5 mM- and 10 mM KA-treated mice in comparison with vehicle control. Gastrocnemius muscles of the KA and vehicle control spinal cord treated mice were double labelled with synaptophysin and alpha-bunglarotoxin to determine the amount of neuromuscular junction (NMJ) degeneration. Synapses were graded as either intact or degenerating. There was a significant (p < 0.05) increase in the percentage of degenerating synapses in the KA mice in comparison with that of control. Excitotoxin application to the gastrocnemius muscle (5–10 mM glutamate) resulted in a localised synapse loss. However, the cell bodies in the spinal cord did not undergo degeneration over the 28 DPS time course.

Conclusion: Our results indicate that motor neuron degeneration is dependent upon the site of exposure to excitotoxin. Identifying the site of the initial effects of excitotoxicity will identify mechanisms of distal axon degeneration that may provide novel therapeutic targets directed at axon protection.

P240 IDENTIFICATION OF NOVEL THERAPEUTICS TO TREAT NEURODEGENERATION USING THE SOD1 ZEBRAFISH MODEL OF AMYOTROPHIC LATERAL SCLEROSIS (ALS)

Mcgown A

Shaw PJ

Ramesh T

University of Sheffield, Sheffield, UK

Email address for correspondence: [email protected]

Keywords: zebrafish, therapeutic, screening

Background: We have utilised a mutant SOD1 zebrafish model containing the mutant sod1 gene in tandem with stress readout (HSP70-DsRed) reporter gene as a measure of neuronal stress. Previously it has been shown that our model exhibits neuronal stress in the spinal inhibitory interneurons that eventually transfers to stress and results in functional changes in the motor neurons and neuromuscular synapse formation. We have also previously shown that using riluzole, currently the only approved treatment for ALS, we can reduce neuronal stress in our model.

Objectives: To utilise this zebrafish model and to develop and perform a high-throughput compound screen to identify compounds that reduce SOD1-mediated neuronal toxicity. The secondary aim is to design and optimise secondary screens to further refine our key hit compounds and to identify potential future clinical therapeutics.

Methods: Until 6 days post fertilisation (dpf) by immersion, 48hpf sod1 transgenic zebrafish were treated with compounds obtained through the Microsource spectrum library . At 6dpf, the larvae were imaged, sonicated and total fluorescence measured using a fluorescence plate reader. Strictly standardised mean difference (SSMD) was used to select hit compounds (β-value below −0.5 were considered as hits in this assay). Dose responses, adult compound dosing, seahorse mitochondrial analysis, and imaging were used as secondary screens.

Results: A high-throughput drug screen has been developed with the capability to screen over 100's of compounds per day with an assay specificity and sensitivity of approaching 100%. We have screened 2000 compounds from the spectrum library in triplicate and have identified 61 different modulators of neuronal stress. From this screen, we have identified 42/2000 compounds which show reduced neuronal stress and 19/2000 which induced a strong hsp70 activation, a cellular repair pathway. Using secondary screens, we have identified a selection of hit compounds that have a positive role in reducing neuronal stress in mutant SOD1 zebrafish. In conclusion, we have designed and validated a drug screen capable of screening 1000's of compounds rapidly and accurately. Using this screen we have identified compounds with a positive role in reducing neuronal stress which are being taken further developed as potential new treatments for human disease.

P241 CHEMICAL GENETIC SCREENS OF TARDBP AND FUS MODIFIERS IN C. ELEGANS AND ZEBRAFISH

Drapeau P¹

Kabashi E²

Parker JA¹

^bhUniv. Montréal, Montréal, Québec, Canada

^biICM, Paris, France

Email address for correspondence: [email protected]

Keywords: drug discovery, TDP-43, FUS

Background: Two recently discovered causative genes for ALS are the TARDBP gene (coding for TDP-43 and also mutated in rare cases of FTLD) and FUS/TLS. Since TDP-43 and FUS are evolutionarily conserved, we turned to the model organisms Caenorhabditis elegans and zebrafish to learn more about their biological functions and screen for potential therapeutic modifiers.

Methods: Transgenic expression of wild-type and mutant human TARDBP (1) and FUS (2) was obtained upon mRNA injection into zebrafish blastulae. We obtained deletion mutants for the worm orthologues tdp-1 and fus-1 and engineered strains to express human TARDBP or FUS in worm motor neurons (3). We then screened 3,700 FDA-approved compounds in the Microsource Discovery Spectrum, Biomol, Prestwick and Sigma libraries using mutant TARDBP worms and validated hits in zebrafish and in mutant FUS lines.

Results: We performed motility assays in zebrafish and lifespan and stress response assays in worms. We observed that TARDBP and FUS have roles in the response to oxidative and osmotic stress. The expression of mutant TDP-43 or FUS in worm motor neurons produces robust, adult onset motility defects, and in both models this was caused by motor neuron deficits. We isolated a number of chemical suppressors of mutant TARDBP toxicity.

Discussion: Under normal conditions, TDP-43 and FUS regulate specific aspects of the cellular stress response. The transgenic models allowed us to isolate chemical suppressors of motor defects. In particular, several neuroleptics protected against development of the motor phenotype and one is currently in clinical trial.

Conclusion: Together these data provide clues to help unravel the mechanism for TDP-43 and FUS toxicity that should also provide leads for early drug discovery.

Acknowledgements

Funded by the Frick Foundation, Genome Quebec and US Dept of Defense.

References

• Kabashi E et al. Human Mol. Gen 2010;19:671–83.
   PubMed Web of Science ®Google Scholar
• Kabashi E et al. PLoS Genet 2011;7: e1002214.
   Google Scholar
• Vaccaro A et al. PLoS One 2012;7:e31321.
   Google Scholar

P242 IDENTIFICATION OF RNA BOUND TO TDP-43 SUPPORTS ITS ROLE IN SYNAPTIC FUNCTION

Mangelsdorf M¹

Narayanan R¹

Chaplin J¹

Noakes PG^1,2

Hilliard MA¹

Wallace RH¹

^bjQueensland Brain Institute

^bkSchool of Biomedical Science, The University of Queensland, Australia

Email address for correspondence: [email protected]

Keywords: TDP-43, RNA-binding protein, C. elegans

Background: Transactive response DNA-binding (TDP-43) protein is an RNA binding molecule that is known to regulate activities including transcription and pre-mRNA splicing. The importance of TDP-43 in ALS pathogenesis has been highlighted by its presence in cytoplasmic inclusions in the central nervous system (CNS) of ALS patients, as well as by the discovery of mutations in TARDBP (the gene encoding TDP-43). The normal role of TDP-43 in the CNS is yet to be fully determined, and how mutations lead to neurodegeneration remains unknown.

Objectives: The aim of the present study was to identify RNA targets of TDP-43 in the mammalian CNS, and to generate a Caenorhabditis elegans model to study molecular pathways that are involved in TDP-43-mediated neurodegeneration.

Methods and results: RNA immunoprecipitation followed by microarray analysis (RIP-chip) was used to identify RNAs bound to TDP-43 in the mouse brain. The RIP-chip analysis produced a list of 1839 potential target genes many of which overlap with previous studies, and whose functions include RNA processing and synaptic function. We have also shown by immunohistochemistry that TDP-43 is localized at the mouse neuromuscular junction. We next generated a C. elegans transgenic strain in which human TDP-43 (normal TDP-43 as well as the A315T and M337V ALS causing variants) was selectively expressed in the GABAergic motor neurons. Worms expressing mutant TDP-43 show disrupted presynaptic loci, prior to the onset of changes to neuronal morphology. Using a GFP-tagged version of TDP-43, we investigated the intracellular localization and found that TDP-43 was predominantly nuclear localized, but was also seen in axons and puncta consistent with synaptic loci.

Discussion and conclusion: We are characterizing both mouse and C. elegans models to study the function of TDP-43 in the CNS. Our results support a role of TDP-43 in the transport of synaptic mRNAs down the axon to synapses for local translation. Disrupted axonal mRNA transport to distal processes may be a mechanism that leads to motor neuron death in ALS.

P243 TRANSLATIONAL PROFILING IN TDP-43 TRANSGENIC MOUSE MODEL OF ALS

Macnair L^1,2

Zhao B^1,2

Miletic D²

Ghani M²

Rogaeva E^2,3

Keith J⁴

Zinman L⁴

Julein JP⁵

Robertson J^1,2

^blDepartment of Laboratory Medicine and Pathobiology

^bmCentre for Research in Neurodegenerative Diseases

^bnDepartment of Medicine

^boDepartment of Internal Medicine, Sunnybrook Health Science Centre, University of Toronto, Toronto, Ontario, Canada

^bpCentre de Recherche du Centre Hospitalier Universitaire de Quebec, Department of Psychiatry and Neuroscience, Laval University, Laval, Quebec, Canada

Email address for correspondence: [email protected]

Keywords: translational profiling, microarray, immunofluorescence

Background: A major feature of degenerating motor neurons in ALS is the mislocalization of the transactive response DNA-binding protein of 43kDa (TDP-43) from the nucleus to the cytoplasm, forming ubiquitinated inclusions. Mutations in TDP-43 account for a small portion of ALS cases; however, TDP-43 pathology is observed in over 90% of cases, indicating that abnormalities in TDP-43 are an important contributor to disease pathogenesis.

Objective: Since TDP-43 is a nuclear DNA- and RNA-binding protein that has known functions in regulating RNA metabolism, it is likely that abnormalities in TDP-43 will be reflected in changes in RNA processing and expression. Our objective is to identify these changes as a means to understanding how abnormal TDP-43 contributes to ALS pathogenesis.

Methods: Typical approaches to identifying changes in RNA expression (transcriptional profiles) rely on analyzing total mRNA pools from a tissue region or specific cell type. We used a novel technique entitled translating ribosome affinity purification (TRAP) to identify mRNAs actively being translated from spinal cord motor neurons of TDP-43^A315T mice. Bacterial artificial chromosome transgenic mice expressing an EGFP-tagged ribosomal protein, L10a, under the control of the choline acetyltransferase promoter were crossed with TDP-43^A315T mice to facilitate affinity purification of translating mRNAs from motor neuron polysomes. Translational profiles were obtained using microarray analysis in symptomatic (10 months old) TDP-43^A315T mice and were compared to wild-type (WT) littermates. Unpaired two-tailed t-test with Benjamini–Hochberg correction was used to determine significantly varying probes between TDP-43^A315T and WT. The Biological Networks Gene Ontology Tool (BiNGO) plugin in Cytoscape was used to identify overrepresented GO terms. Genes with a ≥ 2-fold change between TDP-43^A315T and WT were validated using immunofluorescence.

Results: Symptomatic translational profiles showed that 28 genes were significantly misregulated. BiNGO analysis demonstrated that there was overrepresentation of genes involved in RNA metabolic process (GO-ID 0016070, p = 3.27E-02), immune response (GO-ID 0006955, p = 3.39E-02), and regulation of mitotic recombination (GO-ID 0000019, p = 3.27E-02) in the TDP-43A315T mice. Of the 28 misregulated genes, 20 are mapped and 7 had a fold change of ≥ 2. Validation with immunofluorescence was representative of microarray results.

Discussion and conclusion: Symptomatic TDP-43^A315T mice showed misregulation of genes enriched for RNA processing, immune response, and cell cycle regulation, all of which are highly implicated in ALS pathogenesis and other neurodegenerative diseases. Immunofluorescence in TDP-43^A315T motor neurons was indicative of microarray and will be validated in patient samples. This discovery-based approach has, for the first time revealed translational changes in motor neurons of a TDP-43 mouse model and will provide a greater understanding of the mechanistic basis of motor neuron degeneration.

Acknowledgements

CIHR, University of Toronto Department of Laboratory Medicine and Pathobiology

Reference

• Heiman M, Shaefer A, Gong S et al. Cell 2008;135:738–748.
   PubMed Web of Science ®Google Scholar

P244 SENATAXIN MOUSE MODELS OF ALS4 RECAPITULATE HUMAN ALS TDP-43 PATHOLOGY, DEVELOP NEUROMUSCULAR PHENOTYPES, AND EXHIBIT TRANSCRIPTIONAL ALTERATIONS RESULTING IN DEMYELINATION

van Es M^1,2

Bennett C.¹

Ling S¹

Lagier-Tourenne C¹

Liu P¹

Crain B³

Shelton D¹

Cleveland DW¹

Yeo GW¹

La Spada A¹

^bqUniversity of California, San Diego (UCSD), CA, USA

^brUMC Utrecht, Utrecht, The Netherlands

^bsJohn’s Hopkins, Baltimore, USA

Email address for correspondence: [email protected]

Keywords: SETX, TDP-43 mislocalization, myelin

Background: Autosomal dominant, gain-of-function mutations in the senataxin (SETX) gene cause a juvenile onset form of amyotrophic lateral sclerosis, known as ALS4. To determine the mechanistic basis of ALS4 motor neuron degeneration, we derived two different mouse models carrying human ALS4 mutations.

Methods: In this study transgenic lines expressing wild-type senataxin and the R2136H mutation in murine prion protein promoter expression constructs (PrP-SETX^wt and PrP-SETX^R2136H), and a knock-in line containing the L389S substitution mutation at the mouse senataxin locus (SETX-^L389S-KI) were generated. Characterization of these mice was performed using composite phenotype scoring, and by evaluating motor function through rotarod testing and stride length measurements.

Results: This analysis revealed that both mutant mouse lines develop a slowly progressive motor phenotype, with impaired rotarod performance, presence of hind limb clasping, and ledge test abnormalities. When we immunostained lumbar spinal cord sections from SETX mutant mice, we observed nuclear clearing of TDP-43, accompanied by TDP-43 cytosolic aggregation, which was indistinguishable from the hallmark pathology observed in human ALS patients.

To elucidate the molecular basis of senataxin gain-of-function pathology in the spinal cord, we isolated RNA from ventral horn samples obtained from sets of early symptomatic PrP-SETX-^R2136H and SETX-^L389S-KI mice, and performed RNA-Seq to catalogue transcriptome changes associated with the ALS4 disease state. We found that PrP-SETX-^R2136H mice exhibited significant expression alterations in 188 spinal cord-expressed genes, and SETX-^L389S-KI mice displayed significant expression changes in 304 spinal cord-expressed genes, with coordinate overlapping gene expression alterations detected for 71 genes. Analysis of the shared gene set subject to alteration revealed a preponderance of genes implicated in the myelination pathway: Egr2; Pmp22; Prx; and Mpz mutated in Charcot-Marie-Tooth disease. When we re-examined autopsy material from human ALS4 cases, we noted loss of myelin in peripheral nerves, ventral horn, and dorsal horn regions.

Discussion: These findings are consistent with evidence from sporadic ALS for myelin loss, suggesting that these ALS4 mouse models have uncovered a previously unappreciated aspect of ALS disease pathogenesis in human patients.

P245 A BAC-BASED MOUSE MODEL OF TDP-43-ASSOCIATED ALS AS A TOOL TO EXPLORE EARLY-PHASE PATHOGENESIS

Gordon D¹

Mutihac R²

Alegre-Abarrategui J²

Davies B³

Ansorge O⁴

Wade-Martins R²

Talbot K¹

^btNuffield Department of Clinical Neurosciences

^buDepartment of Physiology, Anatomy and Genetics

^bvWellcome Trust Centre for Human Genetics

^bwDepartment of Neuropathology, University of Oxford, Oxford, UK

Email address for correspondence: [email protected]

Keywords: TDP-43, bacterial artificial chromosome, mouse model

Background: Mutations in TAR DNA-binding protein 43 (TDP-43) are associated with ALS, though the molecular mechanisms leading to pathogenesis are still unclear. TDP-43 becomes mis-localised to the cytoplasm of affected motor neurons in most sporadic ALS patients, where it forms characteristic insoluble, ubiquitinated inclusions. TDP-43 cDNA overexpression models driven by ubiquitous or neuronally expressed promoters display highly variable disease phenotypes.

Objective: To more accurately model ALS, we have generated transgenic mice using bacterial artificial chromosome (BAC) constructs expressing human wild-type (WT) or ALS-associated mutant (M337V) TDP-43 under the control of the endogenous human promoter.

Methods: BAC vectors containing the full-length human genomic locus of the WT or M337V mutation, with an Ypet tag, were targeted to the ROSA26 locus in embryonic stem cells (ESCs) by PhiC31 integrase-mediated cassette exchange. Chimeric mice were generated by blastocyst injection of recombinant ESCs which were subsequently crossed with C57BL/6J female mice to generate two isogenic human TDP-43 transgenic lines, differing only by the presence or absence of the M337V mutation.

Results: We have confirmed that a single BAC copy of the TDP-43-WT-Ypet and TDP-43-^M337V-Ypet is sufficient to drive physiologically relevant levels of transgene expression by immunoblotting. By 2 months of age, elevated levels of human TDP-43 and total TDP-43 are observed in insoluble brain and spinal cord protein fractions from homozygous TDP-43^M337V-Ypet mice compared to TDP-43-WT-Ypet and non-transgenic littermate controls. Accompanying this is a decrease in human TDP-43 mRNA levels in the spinal cord of TDP-43^M337V-Ypet mice compared to TDP-43-WT-Ypet mice, which is not observed in the brain. Expression of endogenous mouse TDP-43 mRNA is unchanged in brain and spinal cord of both transgenic lines compared to non-transgenic controls. Immunostaining of brain and spinal cord sections with antibodies directed against total TDP-43, human-specific TDP-43, and GFP confirm the ubiquitous expression of TDP-43 throughout the CNS, and specifically the expression of human TDP-43 in brain and spinal cord of the transgenic lines.

Discussion and conclusion: BAC transgenic mice represent the state-of-the-art in transgenic mouse models, since transgenes are expressed at physiological levels from native promoter regions. Parallel cohorts of non-transgenic, heterozygous and homozygous mice from each line are being monitored with an array of behavioural, biochemical and neuropathological testing to study the emergence of phenotypic changes. In combination with longitudinal RNA splicing analysis timed to compare pre-symptomatic with various stages of symptomatic mice, we will address the fundamental role of TDP-43 mutation in ALS. Elucidation of early pre-symptomatic changes in cellular homeostasis has the potential to allow the identification of biomarkers for human ALS and to identify the most therapeutically tractable targets for drug development.

P246 ABERRANT PERIPHERIN EXPRESSION AND STABILITY AND SPLICING OF GLT-1 IN TDP-43 (A315T) TRANSGENIC MOUSE MODEL OF ALS

Barri M

Hafezparast M

University of Sussex, Brighton, East Sussex, UK

Email address for correspondence: [email protected]

Keywords: TDP-43, peripherin, GLT-1B

Background: TDP-43 is an RNA/DNA-binding protein implicated in ALS. Cytoplasmic TDP-43-positive inclusions are found in motor neurons of patients with ALS suggesting a potential loss of function (1). The function of TDP-43 in the nucleus is to regulate transcription and RNA metabolism, including RNA splicing and editing. Aberrant RNA splicing of peripherin and EAAT2 (excitatory amino acid glutamate transporter 2) have been implicated in ALS (2,3).

Objective: We investigated whether TDP-43 regulates RNA splicing of peripherin and GLT-1, the murine homologue of EAAT2, in the TDP-43^A315T transgenic mouse model of ALS. We also examined peripherin expression and stability in TDP-43^A315T and SOD1^G93A transgenic mice.

Methods: To identify novel splice variants, we designed primers that amplify the entire coding sequence of both peripherin and GLT-1 and also Per-61-specific primers. We performed sequential extraction of neurofilaments from lumbar spinal cords (LSCs) using low- and TX-100-containing high salt buffer to determine peripherin expression and stability in both TDP-43^A315T and SOD1^G93A. We also performed Western blot to check the protein levels of both the glial-specific GLT-1 and the neuronal isoform GLT-1B in TDP-43^A315T.

Results: We found that Per-61 was expressed in all genotypes including the wild-type mice at both RNA and protein levels. No novel alternative splice variants were found in peripherin nor GLT-1. In symptomatic mice, Per-45 translation was selectively up-regulated in TDP-43^A315T. Moreover, treating neurofilaments with Triton X-100 resulted in significant increase in soluble Per-56 and Per-45 in samples isolated from TDP-43^A315T mice (P-values: 0.02 and 0.003, respectively). Conversely, soluble Per-45 and Per-56 levels were reduced significantly (P-values: 0.009 and 0.03, respectively) in SOD1^G93A mice. In contrast, no change was detected in the Triton-insoluble fractions of both genotypes. In addition, the relative expression of GLT-1B was significantly downregulated in both symptomatic and non-symptomatic TDP-43^A315T mice (P-values: 0.03 and 0.004, respectively).

Discussion and conclusion: Our data suggest that overexpression of TDP-43^A315T induces upregulation of Per-45 translation and leads to the instability of the filament network. Conversely, overexpression of SOD1^G93A enhances neurofilament stability. Moreover, the reduction of GLT-1B expression in TDP-43^A315T mice suggests impaired splicing and therefore might exacerbate disease progression.

Acknowledgements

This study has been supported by a scholarship from Saudi Arabia’s King Abdullah Scholarship Program.

References

• Neumann M, Sampathu DM, Kwong LK et al. Science 2006;314:130–133.
   PubMed Web of Science ®Google Scholar
• Robertson J, Doroudchi MM, Nguyen MD et al. J Cell Biol 2003;160:939–949.
   PubMed Web of Science ®Google Scholar
• Lin CL, Bristol LA, Jin L et al. Neuron 1998;20:589–602.
   PubMed Web of Science ®Google Scholar

P247 MUTANT TDP-43 DEREGULATES THE AMPK SIGNALLING CASCADE THROUGH NOVEL ACTIVATION OF PROTEIN PHOSPHATASE 2A (PP2A)

Perera N

Sheean R

Horne M

Turner B

Florey Institute of Neuroscience and Mental Health, Melbourne, Australia

Email address for correspondence: [email protected]

Keywords: AMPK, PP2A, TDP-43

Background: Hypermetabolism and defective energy homeostasis are implicated in motor neuron degeneration in ALS patients and mutant SOD1 mouse models. However, whether energy hypermetabolism occurs in mutant TDP-43-mediated ALS remains unclear. Here, we investigated activation of the AMP-activated protein kinase (AMPK) signalling cascade and its regulation by upstream kinases and downstream phosphatases in models expressing mutant TDP-43 linked to ALS.

Objectives: To examine AMPK signalling and its regulation by kinases and phosphatases in cell culture and mouse models expressing mutant TDP-43.

Methods: Activation of the AMPK signalling pathway was examined using Western blotting and immunocytochemistry for phosphorylated AMPK and downstream target acetyl-CoA carboxylase (ACC) in NSC-34 cells stably transfected with wild-type (WT) or mutant TDP-43. AMPK signalling was also examined in brains and spinal cords of presymptomatic (P30, 60) and symptomatic (P90) transgenic TDP-43^A315T mice and compared to age-matched WT controls. AMPK regulation was examined in stable cells and mice using Western blotting for kinases (CAMKII and LKB1) and phosphatases (PP2A, PP2Cand PP1-R6).

Results: AMPK activity was drastically diminished by 80% and 60% in spinal cords and brains of transgenic TDP-43^A315T mice at pre-symptomatic and symptomatic disease, respectively. AMPK activation was also reduced ˜50% in NSC-34 cells expressing different TDP-43 mutants. To determine how mutant TDP-43 deregulates AMPK signalling, we screened AMPK kinases and phosphatases, revealing that PP2A level was sharply increased in tissues of mice and NSC-34 cells expressing mutant TDP-43. Furthermore, treatment of NSC-34 cells with the AMPK agonist AICAR did not normalise AMPK activity level, consistent with PP2A-dependent regulation.

Discussion and conclusion: In contrast to mutant SOD1 models, we show that energy homeostasis mediated by the AMPK signalling cascade is severely deregulated in both pre-symptomatic mice and cell cultures expressing mutant TDP-43. We also uncover a novel regulation of PP2A activity by mutant TDP-43. Thus, overstimulation of PP2A signalling in motor neurons by mutant TDP-43 presents a new potential player in ALS pathogenesis.

P248 ADAR2 PLAYS A KEY ROLE FOR DEATH AND TDP-43 MISLOCALIZATION IN ALS MOTOR NEURONS

Yamashita T¹

Teramoto S¹

Chai H-L¹

Muramatsu S-I³

Kwak S^1,2

^bxThe University of Tokyo, Tokyo, Japan

^byInternational University of Health and Welfare, Chiba, Japan

^bzJichi Medical University, Tochigi, Japan

Email address for correspondence: [email protected]

Keywords: TDP-43, AMPA receptor, gene therapy

Background: Failure of GluA2 RNA editing resulting from downregulation of the RNA-editing enzyme adenosine deaminase acting on RNA 2 (ADAR2) occurs in the majority of ALS cases and causes death of motor neurons via a Ca²⁺-permeable AMPA receptor-mediated mechanism. Notably, downregulation of ADAR2 is associated with TDP-43 mislocalization in the motor neurons of conditional ADAR2 knockout (AR2) mice that mimic molecular changes in sporadic ALS motor neurons, but the underlying molecular mechanism remained to be elucidated.

Methods: We found that the expression of Q/R site-unedited GluA2 and resultant upregulation of Ca²⁺ permeable AMPA receptors upregulated calpain, a Ca²⁺-dependent serine protease, via increasing cytoplasmic Ca²⁺. Calpain cleaved TDP-43 in the C-terminal region and generated aggregation-prone N-terminal fragments, which served as seeds for cytoplasmic inclusions. It is likely that full-length TDP-43 that shuttles between the nucleus and the cytoplasm sequestrates into the inclusion, which results in the elimination of TDP-43 from the nucleus, thereby forming TDP-43 pathology. Calpain sequentially cleaved TDP-43 into soluble smaller fragments, indicating that only an optimum level of calpain activation allowed formation of aggregates. Both TDP-43 mislocalization and neuronal death were inhibited when the Ca²⁺ influx though the AMPA receptors were normalized.

Results: These lines of evidence prompted us to develop gene therapy for sporadic ALS by means of global delivery of the ADAR2 gene to motor neurons. AAV9-ADAR2 injected in the tail vein of AR2 mice caused effective expression of ADAR2 in the motor neurons, and motor dysfunction and neuronal death were prevented in pre- and post-symptomatic AR2 mice. Notably, TDP-43 mislocalization in the motor neurons was normalized in the remaining motor neurons. Therefore, delivery of ADAR2 using AAV9 as a vector enabled restoration of RNA-editing activity at the GluA2 Q/R site, which provides mechanistic therapeutic strategy for patients with sporadic ALS.

Discussion: In conclusion, ADAR2 downregulation induces death of motor neurons via Ca²⁺-permeable AMPA receptor-mediated mechanism and causes mislocalization of TDP-43 via activating calpain-mediated cleavage. Delivery of the ADAR2 gene globally to motor neurons effectively rescued motor neurons with normalization of TDP-43 localization in sporadic ALS model mice.

Funding

Supported by CREST JST, grants in aid from MEXT Japan.

P249 ULTRASTRUCTURAL CHANGES IN THE BLOOD–SPINAL CORD BARRIER IN TDP-43 CONDITIONAL KNOCKOUT MICE

Sasaki S¹

Iguchi Y²

Katsuno M²

Sobue G²

^caTokyo Women’s Medical University, Tokyo, Japan

^cbNagoya University Graduate School of Medicine, Nagoya, Japan

Email address for correspondence: [email protected]

Keywords: TDP-43 knockout mice, blood-spinal cord barrier, ultrastructure

Background: TDP-43 conditional knockout (TDP CKO) mice develop progressive weight loss and motor impairment around the age of 60 weeks, and degeneration of large anterior horn neurons at the age of 100 weeks. Although the loss-of-function of TDP-43 protein seems to affect the pathogenesis of ALS, the interaction between the loss of TDP-43 protein and the changes in the blood–spinal cord barrier (BSCB) remains unclear.

Objectives: To clarify whether the loss of TDP-43 protein causes changes in the BSCB in TDP CKO mice.

Methods: We studied lumbar spinal cords in eight TDP CKO mice and eight control littermates using electron-microscopy. TDP CKO mice were divided into four groups: early presymptomatic (aged 20 weeks); late presymptomatic (36 weeks); early symptomatic (50 weeks); and late symptomatic (100 weeks) stages (n = 2, respectively). TDP-43^flox/flox mice served as age-matched control littermates in each group (n = 2).

Results: In controls, endothelial cells of the capillaries were characterized by tight junctions and some cytoplasmic organelles such as mitochondria, Golgi apparatus, rough endoplasmic reticulum, free ribosomes, vesicles and multi-vesicular bodies. The perivascular space of the capillaries was bordered on one side by the basal lamina of the endothelial cell and on the other by the basal lamina of the astrocytic foot processes. Sometimes, the perivascular space was virtually obliterated and the two laminae formed a single homogenous structure. In the anterior horn, the majority of the capillaries were surrounded by a narrow perivascular space without connective tissue. Pericytes were found outside the endothelial cells, and completely surrounded by a basal lamina.

In TDP CKO mice, the capillaries at the early presymptomatic, late presymptomatic, and late symptomatic stages were well preserved as in age-matched controls. At the early symptomatic stage, tremor appeared as the earliest symptom of motor deficit and the capillaries showed various alterations. In the anterior horn, the cytoplasm of most endothelial cells was severely vacuolated. The endothelium occasionally exhibited marked edema, detachment of the cytoplasm, and protrusion of the cytoplasm into the lumen of the blood vessel, although the tight junction appeared relatively intact. Perivascular spaces of the capillaries were frequently edematous. Mitochondria in the cytoplasm of endothelial cells and pericytes, and in perivascular spaces were frequently swollen and vacuolated. In the posterior horn, the structure of the capillaries was almost always normal.

Conclusions: The temporary and reversible breakdown of the BSCB with leakage or increased permeability at the early symptomatic stage could be a direct consequence of the loss of TDP-43 protein in the endothelial cells. The damage by the loss-of-function of TDP-43 protein to the vasculature contributes to initiate non-cell autonomous pathogenesis of degeneration of motor neurons in the mouse model observed.

P250 THE EFFECT OF CLUSTERIN ON THE TOXICITY OF GLIAL TDP-43 IN A DROSOPHILA MODEL OF ALS

Brown R¹

Luheshi LM²

Wilson M¹

Dobson CM²

^ccSchool of Biological Sciences, University of Wollongong, Wollongong, Australia

^cdDepartment of Chemistry, University of Cambridge, Cambridge, UK

Email address for correspondence: [email protected]

Keywords: TDP-43, clusterin, Drosophila

Background: Amyotrophic lateral sclerosis (ALS) is characterised by the presence of intracellular protein inclusions within degenerating neurons, and astro and microgliosis. Protein aggregation is implied as an underlying cause of disease pathology and recent studies suggest that non-neuronal cells play important roles in neurodegeneration through non-cell autonomous mechanisms (1,2). Clusterin is a conserved glycoprotein which under normal conditions is secreted from the endoplasmic reticulum (ER) into the extracellular space, where it acts as an extracellular chaperone. Recently, it has been shown that under conditions of ER stress, clusterin can be retro-translocated to the cytosol where it may act as an intracellular chaperone (3).

Objectives: To determine whether expression of TAR DNA-binding protein 43 (TDP-43; a major component of inclusions in ALS) in the glia of Drosophila is pathogenic, and whether co-expression of clusterin reduces this toxicity.

Methods: TDP-43 was expressed in the glial cells of Drosophila by crossing female TDP-43 flies with the male driver stocks eaat-1-GAL4 or eaat-1-GFP. The brains of larvae were dissected and stained for TDP-43 and neuronal markers.

Survival assays were performed on adult Drosophila expressing TDP-43, TDP-43 + Clusterin or clusterin in glial cells, as well as on 51D (non-transgenic) flies. Females of these stocks were crossed with the male driver line eeat-GFP-Gal80. A Kaplan–Meier survival curve was generated and differences between genotypes analysed using Kaplan Meir statistics.

Results: Expression of TDP-43 in larval glial cells resulted in a reduction of larvae size, locomotor defects and decreased hatching. Immunohistochemistry revealed motor neuron degeneration and suggested that TDP-43 was able to spread from glia to other cell types, including motor neurons.

Expression of TDP-43 in the glia of adult Drosophila was also toxic, with flies having a reduced lifespan (˜12 days) when compared to that of the control flies (˜30 + days). Flies that co-expressed clusterin had a significantly (p value < 0.0001) extended lifespan (˜14 days).

Discussion and conclusion: These results support the theory that non-neuronal cells play important roles in neurodegeneration. The ability of clusterin to rescue TDP-43 toxicity suggests that it may play a role in the pathogenesis of neurodegenerative diseases in which protein aggregation underlies disease pathology. The pathogenic mechanisms of ALS are multifactorial, and it is likely that effective treatments can only be developed once these mechanisms are better understood. Potential therapeutic targets may eventually include drugs targeting glial cells and the use of chemical or peptide chaperones.

Acknowledgements

The authors thank Dr Jenna Gregory and Dr Justin Yerbury for their valuable contribution.

References

• Lino M, Schneider C, Caroni P. Journal of Neuroscience 2002;22:4825–4832.
   PubMed Web of Science ®Google Scholar
• Clement A, Nguyen M, Roberts E et al. Science 2003;302:113–117.
   PubMed Web of Science ®Google Scholar
• Nizard P, Tetley S, Le Drean Y et al. Traffic 2007;8:554–65.
   PubMed Web of Science ®Google Scholar

P251 INVESTIGATING PROPAGATION OF TDP-43 AGGREGATION IN AMYOTROPHIC LATERAL SCLEROSIS USING A DROSOPHILA MELANOGASTER MODEL

Hanspal M¹

Luheshi LM¹

Yerbury J²

Dobson CM¹

^ceUniversity of Cambridge, Cambridge, Cambridgeshire, UK

^cfUniversity of Wollongong, Wollongong, New South Wales, Australia

Email address for correspondence: [email protected]

Keywords: TDP-43, Drosophila, propagation

Background: Symptoms of amyotrophic lateral sclerosis (ALS) characteristically have a focal onset and spread directionally and progressively throughout the central nervous system (CNS) (1). While the directional advancement of ALS pathology is well characterised, the molecular mechanisms underlying this process remain unknown. TDP-43, one of the major components of ubiquitinated inclusions associated with this disease, forms mislocalised aggregates capable of seeding aggregation of natively folded protein (2). Cell-to-cell transmission of misfolding proteins has been demonstrated for a number of neurodegenerative diseases, but little work has been carried out to investigate whether TDP-43 aggregation, initiated at a focal site, propagates throughout the CNS (3). Finding evidence for this proposed molecular model of ALS pathogenesis would explain how the disease progresses over time, and reveal new therapeutic targets to treat this debilitating condition.

Objectives: The objective was to develop a Drosophila melanogaster (fruit fly) model to investigate propagation of TDP-43 aggregates within the CNS to determine whether cell-to-cell transmission of the protein takes place.

Methods: TDP-43 expression was directed to a subset of photoreceptor neurons in the adult fly visual system using the Gal4-UAS system. Brains of 3-day-old adult flies were dissected and the distribution of TDP-43 analysed using immunohistochemistry and confocal microscopy.

Results: Transgenic TDP-43 driven in a subset of photoreceptors was localised in both the nuclei and the cytoplasm of these neurons, including the axonal projections targeting the optic lobes. TDP-43 also formed predominantly cytoplasmic puncta in cell populations beyond the Gal4 driver expression pattern when compared to controls.

Discussion and conclusion: Spreading of TDP-43 in the CNS was successfully modelled using the fly visual system. The results of this project show that TDP-43 can spread from a subset of neurons to other cell populations in the CNS, demonstrating that TDP-43 is transmissible between cells. Furthermore, TDP-43 forms cytoplasmic aggregates in recipient cells, recapitulating characteristic features of TDP-43-associated ALS pathology: nuclear mislocalisation and formation of aggregates. These findings support the proposal that TDP-43 aggregation can be propagated within the CNS from a focal site of onset, thus providing an explanation for how spatiotemporal spreading of ALS pathology occurs.

Acknowledgements

Dr T Pereira de Barros, Dr J Gregory, R Brown, J Ng.

References

• Ravits J, Paul P, Jorg C. Neurology 2007;68:1571–1575.
   PubMed Web of Science ®Google Scholar
• Furukawa Y, Kaneko K, Watanabe S et al. Journal of Biological Chemistry 2011;286:18664–18672.
   PubMed Web of Science ®Google Scholar
• Münch C, Bertolotti A. Journal of Molecular Biology. 2012;421:491–498.
   PubMed Web of Science ®Google Scholar

P252 IDENTIFYING THE ROLE OF TDP-43 IN AMYOTROPHIC LATERAL SCLEROSIS (ALS) THROUGH INTERACTOME ANALYSIS OF PATHOGENIC TDP-43 IN A TRANSGENIC MOUSE MODEL

Chiang H^1,2

Xiao S²

Zhao B^1,2

Miletic D²

Ho K^1,2

Mount H^1,2

Schmitt-Ulms G^1,2

Robertson J^1,2

^cgUniversity of Toronto, Toronto, ON, Canada

^chTanz Centre for Research in Neurodegenerative Diseases, Toronto, ON, Canada

Email address for correspondence: [email protected]

Keywords: interactome, alternative splicing, TDP-43

Background: TAR DNA-binding protein 43 (TDP-43) has been identified as a major protein in pathological inclusions of ALS and frontotemporal lobar degeneration (FTLD). A biochemical signature of TDP-43 proteinopathy is the presence of lower molecular weight (LMW) TDP-43 fragments. In understanding the origin of these LMW TDP-43 species, we have identified an abnormal splice variant of TDP-43 migrating at 35kD, herein referred to as TDP-35. TDP-35 expression is elevated in ALS tissues, and overexpression of TDP-35 in cell culture induces aggregate formation and cellular toxicity.

Objectives: To identify the role of TDP-35 in the pathogenesis of ALS, we have characterized transgenic mice overexpressing TDP-35 and performed interactome analysis of TDP-35 in order to elucidate disease-associated cellular pathways.

Method: Transgenic mice overexpressing human TDP-35 under the hamster prion promoter were characterized using protein biochemistry, immunohistochemistry, and motor and cognitive function tests. Interactome analysis was performed by perfusion crosslinking of animals, co-immunoprecipitation of TDP-35 complexes from brain homogenate, and use of mass spectrometry with iTRAQ labelling to identify proteins that co-purify with TDP-35. Candidates were validated using reciprocal IP and immunohistochemistry.

Results: Human TDP-35 is overexpressed in the brain and to a lesser extent in the spinal cord of transgenic animals. Younger transgenic mice exhibit a predominantly nuclear localization of TDP-35 while older mice show rare cytoplasmic inclusions and increased gliosis. No axonal loss, weight change, or overt motor phenotype is observed. Novel object recognition test revealed that at 11 months of age, but not at 6 months of age, transgenic animals exhibit significantly lower memory score than non-transgenic animals. Interactome analysis of TDP-35 in 12-month-old mice reveals several potential candidates. One of which, a protein involved in the transport of an excitatory neurotransmitter, has been validated by reciprocal IP, while the remaining are undergoing validation.

Discussion and conclusion: Mice overexpressing TDP-35 exhibit progressive cognitive dysfunction accompanied by increased gliosis in the brain. The lack of motor dysfunction and presence of cognitive phenotype may be attributed to the preferred expression of the hamster prion promoter in the brain versus the spinal cord. Given the presence of TDP-43 proteinopathy in FTLD and cognitive symptoms in a subset of ALS patients, TDP-35 may play a role in the cognitive aspect of these neurodegenerative diseases.

In conclusion, overexpression of TDP-35, an abnormal splice variant of TDP-43, is associated with cognitive dysfunction in mice and may underlie cognitive phenotypes in ALS and FTLD. The disease mechanism may involve abnormal transport of an excitatory neurotransmitter. Current investigation into TDP-35 interactome will shed light on the cellular pathways involved in neurodegeneration and provide additional insight into the role of excitotoxicity in ALS pathogenesis.

Acknowledgements

CIHR Doctoral Research Award, ALS Canada.