C74 MASS SPECTROMETRY BASED PROTEOMIC ANALYSIS OF CSF FOR BIOMARKERS OF ALS
BOWSER R, AN J, HOOD B, CONRADS T
University of Pittsburgh, Pittsburgh, PA, USA
Email address for correspondence: [email protected]
Keywords: proteomics, mass spectrometry, biomarkers
Background: There have been numerous studies reporting protein-based biomarkers in the cerebrospinal fluid (CSF) of ALS patients. Most studies have focused on candidate biomarkers based on biochemical pathways presumably involved in ALS. We previously performed an unbiased screen using mass spectrometry based proteomics to discover candidate biomarkers for ALS. To further our understanding of the pathogenic mechanisms of ALS and verify biomarker candidates, we performed a large unbiased proteomic analysis of CSF from ALS and control subjects.
Objectives: Perform an unbiased proteomic analysis of CSF from ALS, healthy control and disease controls using liquid chromatography tandem mass spectrometry (LC-MS/MS) to sequence the proteome in each subject group. CSF from 250 subjects was used for this study.
Methods: CSF was collected and immediately processed from 90 sporadic ALS, 20 familial ALS, 80 healthy control, 20 multiple sclerosis (MS), 20 Alzheimer's disease (AD), 10 upper motor neuron disease and 10 lower motor neuron disease subjects. We created 25-pooled samples for further analysis, each containing 10 subjects of the same gender ratio (6 male, 4 female) and were controlled for age. For ALS samples, each pool was also controlled for site of disease onset, sporadic or familial disease, and use of riluzole. Samples were carefully controlled for removal of abundant proteins, trypsin digestion, enrichment of peptides and analysis by LC-MS/MS using a Thermo LTQ orbitrap mass spectrometer. Proteins were identified by the presence of at least 2 peptides.
Results: Over 4,000 proteins were identified in the human CSF proteome, with significant alterations in 187 proteins between ALS and control groups. Unbiased cluster analysis easily distinguished ALS from the other subject groups. We identified alterations in CSF proteins for a number of signaling pathways and proteins expressed by neurons and/or glia. Specific pathways that exhibited a number of protein alterations in the CSF of ALS patients include the ubiquitin degradation system, RNA/DNA binding proteins, inflammatory proteins, extracellular matrix, cytoskeletal proteins and complement proteins. We also noted age-related changes in the proteome and differences based on site of disease onset.
Discussion: We performed an unbiased proteomic analysis of CSF from ALS and control subjects using LC-MS/MS to identify as many proteins as possible across all subject groups. While we verified findings for many proteins, we also identified additional proteins that are altered in the CSF of ALS patients and mapped these to specific biochemical pathways and expression by cell type. Further studies are required to validate these findings in another separate set of samples collected in a prospective manner across multiple clinics.
Conclusions: We have identified additional candidate biomarkers for ALS and have shown that LC-MS/MS based peptide analysis can distinguish ALS from neurologic disease controls, disease mimics and healthy control subjects.
C75 CHEMICAL GENETIC SCREENS FOR IN VIVO TDP-43 MODIFIERS AND ALS DRUG DISCOVERY
KABASHI E1,2, VACCARO A1, MAIOS C1, DURHAM H3, PARKER JA1, DRAPEAU P1
1University of Montreal, Montreal, Canada, 2Institute of Brain and Spinal Cord (ICM), Paris, France, 3Montreal Neurological Institute, Montreal, Canada
Email address for correspondence: [email protected]
Keywords: TDP-43, FUS, zebrafish, C.elegans
Background: Although our understanding of the genetics and molecular mechanisms leading to ALS and other motor neuron diseases has increased, there is no effective treatment for these disorders. Further, no high-throughput chemical screens have been thus far conducted to identify potential therapies using in vivo models for ALS and other motor neuron diseases.
Objectives: Our laboratories have developed stable transgenic lines expressing mutant and WT human TDP-43 and FUS in zebrafish (D. rerio) and worms (C. elegans). These models are invaluable to study genetic pathways of disease. Further, both these model organisms are amenable to chemical screening.
Methods: We established mutant and WT TDP-43 and FUS lines under an inducible ubiquitous heat shock promoter (zebrafish) or a constitutive motor neuron promoter (C. elegans). Mutant but not WT TDP-43 and FUS fish and worms had visible motor deficits accompanied with axonal abnormalities and protein aggregates in motor neurons. Zebrafish larvae and adult worms were placed in multiwall plates containing a variety of chemical compounds. Following overnight treatment, the motor phenotype was assessed. Further, axonal projections from motor neurons as well as protein aggregation using specific antibodies were quantified.
Results: In a small preliminary screen of compounds with neuroprotective properties, we identified Methylene Blue (MB) to consistently and potently rescue the motor phenotype in both our model organisms. Using the power of worm genetics, we further studied the chaperones in a variety of subcellular compartments upregulated by MB. We identified that treatment with MB specifically and solely upregulated one protein chaperone. Further, molecular analysis demonstrated that overexpression of mutant TDP-43 elicited cellular stress and specific inhibition of this cellular stress rescued the motor phenotype observed both in worms and zebrafish. Further, MB as well as reducing cellular stress were able to reduce motor neuron excitoxicity by glutamate treatment in mammalian primary motor neuron cultures.
Discussion: We have identified MB as a novel compound with therapeutic potential in ALS. MB rescued the motor phenotype induced by overexpression of both mutant TDP-43 and FUS. Since MB is a compound that is also under clinical trial in Alzheimer patients, this treatment might have significant effects in a number of neurological disorders. We have further elucidated the molecular mechanism through which MB exerts its potential beneficial effect to reduce protein aggregation by upregulating a specific chaperones and reducing cellular stress. Thus, our observations are consistent with other studies suggesting that reduced cellular stress is a crucial step in ALS pathogenesis.
Conclusions: This is the first chemical screen in multiple model organisms for TDP-43 and FUS. Its high-throughput potential and the confirmation of therapeutic compounds in different models give our assay unique capabilities in the quest to treat ALS and other motor neuron diseases.
C76 RESCUE OF MOTOR NEURON DEATH BY TARGETING THE BNIP3 PATHWAY
Kong J, Weng J, Chen X
University of Manitoba, Winnipeg, Canada
Email address for correspondence: [email protected]
Keywords: BNIP3, mitochondria, cell death pathway
Oxidative stress, mitochondrial dysfunction and morphologically necrotic-like motor neuron death are major features in ALS. Previously we showed that oxidative stress provided a redox signal to activate hypoxia-inducible factor 1a (HIF-1a), which is the primary, if not the only, transcriptional factor for the death-inducing gene BNIP3. Expression of BNIP3 caused a caspase-independent form of neuronal cell death in vitro and in vivo. Here we show that BNIP3 was induced to express at the onset of the disease in transgenic mice expressing the G93A and the G37R mutations of SOD1. BNIP3 was not detectable in the brain of control animals and in the G93A and the G37R mice before the onset of disease. Levels of BNIP3 expression increased with disease progression as evidenced by immunohistochemistry, Western blotting and RT-PCR analyses. The expressed BNIP3 was found to be primarily localized in motor neurons. BNIP3 was not detectable in the liver, kidney and lung tissues from the same groups of G93A and G37R animals that showed high levels of BNIP3 in the spinal cord.BNIP3 was detected in the mitochondrial membranes after alkaline extract, indicating that the expressed BNIP3 was active because inactive BNIP3 is known to be dissociated from mitochondria after alkaline treatment. To further determine the role of BNIP3 in mutant SOD1-induced neuronal death, a lentiviral shRNA vector targeting the nucleotides 167-188 of the BNIP3 mRNA, was injected into the lumbar spinal cord of the G93A mice at the age of 8 weeks. Animals injected with a scramble shRNA vector were used as controls. Inhibition of BNIP3 by RNAi significantly increased the number of axons in the L5 ventral roots (p = 0.015). Analysis of axon size distribution showed clearly the protection of middle to large (larger than 6 mm in inner diameter) axons by the lentiviral BNIP3 shRNA vector. We further analyzed the BNIP3 pathway and found that BNIP3 interacted with the ion channel VDAC to induce mitochondrial release of endonuclease G leading to a caspase-independent apoptosis. To look for an inhibitor for the BNIP3 pathway, we identified the small chemical necrostatin-1 that was able to inhibit BNIP3 cell death pathway by preventing integration of BNIP3 to the outer membrane of mitochondria. The results demonstrate that BNIP3 plays a role in mediating mutant SOD1-induced motor neuron death. The BNIP3-induced cell death pathway provides a molecular linkage for mitochondrial degeneration, oxidative stress and caspase-independent neuronal death. Necrostatin-1 appears to be a potent inhibitor for the BNIP3 pathway and may be a new therapy for ALS.
C77 THE BIS(THIOSEMICARBAZONATO)-COPPERII COMPOUND CUII(atsm) DELAYS SYMPTOM ONSET AND DISEASE END-STAGE IN SOD1G37R MICE
LIM N1,2, LIDDELL J1, PARKER S1, KANNINEN K1, LIM S1, PATTERSON B1, PRICE K1, LI Q-X1,2, DONNELLY P1, BARNHAM K1,2, WHITE A1,2, CROUCH P1,2
1University of Melbourne, Melbourne, Victoria, Australia, 2Mental Health Research Institute, Melbourne, Victoria, Australia
Email address for correspondence: [email protected]
Keywords: therapeutic, mouse models, SOD1
Background: In the absence of effective treatments for ALS the development of new therapeutic strategies is still needed. In preliminary pre-clinical studies we tested the bis (thiosemicarbazonato)-copperII compound CuII(atsm) as a novel therapeutic for ALS. Treating mice expressing a low copy number of human SOD1G93A slowed progression of disease symptoms and significantly delayed the age at which the mice reached disease end-stage.
Objectives: In the present study we tested therapeutic activity of CuII(atsm) in mice expressing human SOD1G37R.Our objective was to determine whether CuII(atsm) could prevent disease symptoms in a second, more aggressive mouse model for ALS.
Methods: Mice expressing human SOD1G37R and their non-transgenic littermates were treated with CuII(atsm) at 30 mg/kg body weight (or with the suspension vehicle as a sham control) by daily gavage. CuII(atsm) is not commercially available and was synthesised following published procedures. The onset and progression of locomotor deficits were monitored using rotarod and stride length assays. One cohort of mice was monitored until they reached disease end-stage and a second cohort of mice was culled at 24 weeks when disease symptoms became apparent in the sham treated mice. Tissues were collected from culled mice for immunohistochemistry and biochemical analyses.
Results: Treating with CuII(atsm) significantly delayed the onset of disease symptoms in SOD1G37R mice. Sham treated SOD1G37R mice developed an impairment on the rotarod at 22.5 weeks of age and in the stride length assay at 24 weeks. These deficits however were not detectable in CuII(atsm) treated mice until 26.5 weeks (rotarod) and 28 weeks (stride length assay). Irrespective of when disease symptoms developed CuII(atsm) did not alter the rate of their progression. Accordingly, disease end-stage was delayed in the CuII(atsm) treated mice by 4 weeks. Western blot analysis of whole blood samples collected from mice at 24 weeks revealed treating with CuII(atsm) did not alter expression of the endogenous mouse SOD1. By contrast, levels of the transgenic human SOD1G37R were increased 2.3-fold in the CuII(atsm) treated mice.
Discussion and conclusions: These data support our previous work using low copy SOD1G93A mice and demonstrate that CuII(atsm) delays the onset of disease symptoms in the more aggressive SOD1G37R mouse model for ALS. Increased levels of SOD1G37R in the blood of CuII(atsm) treated mice indicate that stabilisation of the mutant SOD1 may contribute to the therapeutic activity of CuII(atsm). Our on-going histological and biochemical analyses of brain, spinal cord and muscle samples should reveal more information on the mechanism of action for this potential ALS therapeutic.
C78 INTERMOLECULAR AND INTERCELLULAR PROPAGATION OF SUPEROXIDE DISMUTASE 1 MISFOLDING IN ALS
CASHMAN N, GRAD L, GUEST W
University of British Columbia, Vancouver, BC, Canada
Email address for correspondence: [email protected]
Keywords: superoxide dismutase 1, prion, template-directed misfolding
Background: Approximately 10% of ALS cases are familial, with ∼20% of these due to mutations in the gene encoding Cu/Zn superoxide dismutase 1 (SOD1), a widely-expressed free-radical defense enzyme. A consequence of SOD1 mutation and/or oxidation is a propensity of the protein to misfold and aggregate. Human wild-type (wt) SOD1 is known to co-aggregate with mutant SOD1 in familial ALS, in double transgenic mouse models, and in cell culture systems, but the structural determinants of this process and its functional consequences are unclear.
Objective: We sought to molecularly dissect the effects of intracellular obligately misfolded SOD1 mutant proteins on natively structured wild-type SOD1.
Methods and results: Expression of the enzymatically inactive, natural familial ALS SOD1 mutations G127X and G85R in human mesenchymal and neural cell lines induces misfolding of wild-type natively-structured SOD1, as indicated by: 1) acquisition of immunoreactivity with SOD1 misfolding-specific monoclonal antibodies; 2) markedly enhanced protease sensitivity suggestive of structural loosening; 3) non native disulfide-linked oligomer and multimer formation; and 4) generation of reactive oxygen species. Expression of G127X and G85R in mouse cell lines did not induce misfolding of murine wtSOD1, and a species restriction element for human wtSOD1 conversion was mapped to a region of sequence divergence in loop II and beta-strand 3 of the SOD1 beta-barrel (residues 24-36), then further refined surprisingly to a single tryptophan residue at codon 32 in human SOD1. Aggregated recombinant G127X is capable of inducing misfolding of recombinant human wtSOD1 in a cell-free system in buffered saline containing reducing and chelating agents. The presence of a tryptophan at codon 32 on recombinant G127X increases its effectiveness at inducing wtSOD1 misfolding, compared to a serine at position 32 as found in the murine SOD1 sequence. Furthermore, the culture medium from cells transiently transfected with wild-type or mutant SOD1 can induce misfolding of endogenous SOD1 when added to naive neuroblastoma cell cultures, and this process can be stably propagated in serial passage. The agent responsible for induction of misfolding was determined to be a misfolded SOD1 aggregate. Transmission of SOD1 misfolding in vitro is abrogated by extracellular pan- and misfolding-specific SOD1 antibodies.
Conclusions: SOD1 misfolding and toxicity can propagate within and between cells, which may prompt novel targeted therapies for ALS and other neurodegenerative diseases.
C79 NON-VIRAL GENE DELIVERY FOR MOTOR NEURONS
ROGERS M-L, MATUSICA D, SMITH K, RUSH R
Flinders Medical Science and Technology, Human Physiology and Centre For Neuroscience, Adelaide, South Australia, Australia
Email address for correspondence: [email protected]
Keywords: non-viral gene delivery, SOD1G93A mice, antibodies
Background: There are no effective treatments for motor neuron disease (MND) also known as Amyotrophic Lateral Sclerosis (ALS). We have developed a non-viral method of introducing exogenous genes into neurons using antibodies to specific receptors as targeting agents, which are termed ‘immunogenes’. The common neurotrophin receptor (p75NTR) is up-regulated on motor neurons of symptomatic SOD1G93A mice, the mouse model of MND (1) whereas the trkC receptor is present on motor neurons throughout the entire lifespan (2). Our group has shown functional outcomes are possible by specific delivery of siRNA to basal forebrain neurons, and GDNF encoding plasmid to injured motor neurons using immunogenes targeting the p75NTR (3,4).
Objectives: To demonstrate that immunogenes targeting either p75NTR or the TrkC receptor can deliver genes to motor neurons of SOD1G93A mice.
Methods: p75NTR and trkC targeting antibodies MLR2 and 2B7 were tested for ability to access motor neurons following systemic injections into mice in-vivo. Immunogenes were constructed by attaching polyethylene glycol (PEG) covalently to the DNA/ RNA binder polyethylenimine (PEI) that was then conjugated to monoclonal antibodies. The specificity and toxicity of the immunogenes were examined in primary motor neuron cultures. The ability of p75NTR immunogenes to target motor neurons in SOD1G93A mice was examined by intraperitoneal injections of immunogenes carrying plasmids that express green fluorescent protein (pGFP).
Results: Administration of p75NTR antibody by systemic injections into newborn mice (n = 4), showed that the antibody internalises into more 90% of motor neurons. TrkC receptor antibody also was able to access adult motor neurons after intraperitoneal injections (n = 3). PEG was attached covalently to PEI and then conjugated to the monoclonal antibodies. Immunogenes condensed and bound plasmid encoding GFP (n = 3). MLR2-PEG-PEI-pGFP was able to transfect pure motor neurons cultured in the presence of 10% serum (n = 4; 5) and motor neurons in mixed cultures containing primarily astrocytes (n = 3). In addition, MLR2-PEG-PEI-pGFP was not toxic to cultured neurons (n = 4). Finally, control mice given 2B7-PEG-PEI-pGFP or symptomatic SOD1G93A mice given intraperitoneal MLR2-PEI-PEG-pGFP resulted in GFP expression in spinal motor neurons (n = 4).
Discussion and conclusion: This study shows effective non-viral gene delivery to motor neurons in-vitro and in-vivo. Reduced toxicity and stability of the immunoporter may be important in developing gene therapies that target motor neurons in the SOD1G93A mice.
References
- Lowry KS, Murray SS, McLean CA, et al. ALS 2001; 2:127–314.
- Nishio T, Sunohara N, Furukawa S. Neuroreport. 1998; 9:1661–5.
- Barati S, Hurtado PR, Zhang SH, et al. Exp Neurol. 2006; 202:179–88.
- Berhanu D, R. Rush. Neuroscience 2008;153:1115–25.
- Wiese S, Herrmann T, Drepper C, et al. Nature Protocols. 2010;5:31–8.