C25 ALTERED RNA FUNCTION IN ALS: LESSONS FROM GENETICS
BROWN JR. R
Day Neuromuscular Research Lab, University of Massachusetts Medical School, Worcester, MA, USA
Email address for correspondence: [email protected]
Keywords: RNA, genes, mutations
Amyotrophic lateral sclerosis (ALS) is a degenerative disorder that selectively targets motor neurons in the brain, brainstem and spinal cord. At least 10% of cases are caused by gene defects that are transmitted as dominant traits. Multiple gene mutations have now been identified as causes of either ALS or ALS and frontotemporal dementia. These mutations highlight several themes in ALS pathogenesis. Studies of mutations in the SOD1 gene and protein have highlighted protein instability as an upstream factor in this disease. By contrast, other recently defined ALS genes, prominently including TDP-43 and FUS/TLS, identify perturbations of RNA function as fundamental components of motor neuron disease. Multiple aspects of RNA biology are potentially implicated including transcription, splicing, shuttling between the nucleus and cytoplasm, protein translation and activity-dependent local control of protein synthesis in dendritic spines. These defects, which may be particularly critical after cellular stress, entrain a complex series of downstream events (e.g. mitochondrial dysfunction, excitotoxicity, disturbances of axonal transport) that ultimately lead to the demise of the motor neuron, with relevance both to familial and sporadic ALS. This presentation will review these emerging concepts as well as implications for further investigations of the role of gene variations in ALS pathogenesis.
C26 IDENTIFICATION OF FUS/TLS-MEDIATED RNA-PROCESSING ALTERATIONS IN AMYOTROPHIC LATERAL SCLEROSIS
POLYMENIDOU M1,3, LAGIER-TOURENNE C1,3, HUTT KR2,3, HUELGA SC2,3, LIANG TY2,3, LING S-C1,3, KORDASIEWICZ H1,3, WANCEWICZ E4, MAZUR C4, DONOHUE JP5, SHIUE L5, BENNETT CF4, CLEVELAND DW1 ,3, YEO GW2,3
1Ludwig Institute for Cancer Research, 2Stem Cell Program and Institute for Genomic Medicine, 3Department of Cellular and Molecular Medicine; UCSD, La Jolla, CA, USA, 4Isis Pharmaceuticals, Carlsbad, CA, USA, 5RNA Center, Department of Molecular, Cell and Developmental Biology, Sinsheimer Labs, UCSC, Santa Cruz, CA, USA
Email address for correspondence: [email protected]
Keywords: FUS/TLS, RNA targets, CLIP-seq
Background: TDP-43 and FUS/TLS, two RNA/DNA binding proteins, are central in the pathogenesis of Amyotrophic Lateral Sclerosis (ALS), yet their physiological roles in the central nervous system are poorly understood. Using cross-linking immunoprecipitation coupled with high-throughput sequencing (CLIP-seq) we recently identified binding sites in 6,304 genes as the brain RNA targets for TDP-43. Following depletion of TDP-43 from mouse adult brain with antisense oligonucleotides, we found that levels of 601 mRNAs were changed and 965 splicing events were altered. RNA-targets whose levels were most depleted by reduction in TDP-43 were derived from genes that are essential for the maintenance of neuronal integrity. A fundamental remaining issue is the precise role(s) of FUS/TLS in RNA metabolism regulation and how alterations in its properties may underlie neurodegeneration. Like TDP-43, FUS/TLS has been proposed to participate in several steps of RNA processing, including alternative splicing and transcription regulation. Few RNA targets of FUS/TLS have been identified and a comprehensive protein-RNA interaction map still needs to be defined.
Objectives: To identify in vivo RNA targets of FUS/TLS, to validate the roles of FUS/TLS in the processing of these targets and to determine TDP-43-FUS/TLS overlapping mRNA targets and/or RNA-processing alterations.
Methods: We have used cross-linking immunoprecipitation CLIP-seq to identify RNAs bound by FUS/TLS in mouse brain. We have subsequently determined the effects of FUS/TLS loss of function on RNA expression and splicing patterns by using high-throughput sequencing of cDNA (RNA-seq) and splicing-sensitive arrays.
Results: Greater than 8 million uniquely mapped reads enabled the accurate generation of clusters using gene-specific thresholds to define FUS/TLS binding sites. To validate the role of FUS/TLS in the regulation of transcription and alternative splicing via these sites, downregulation of FUS/TLS in vivo was achieved in an otherwise normal adult mouse brain, using direct injection of antisense oligonucleotides against FUS/TLS. Transcriptome profiling from brains with FUS/TLS reduction to 90% of endogenous levels confirmed its roles in alternative splicing and gene expression regulation.
Discussion and conclusions: Genome-wide identification of validated RNA targets is a first step in the elucidation of the molecular mechanisms underlying death of motor neurons in ALS. Since mutations in either TDP-43 or FUS/TLS cause a similar disease phenotype, and both proteins are involved in the same RNA-processing steps, we anticipate that the RNA-targets affected by both TDP-43 and FUS/TLS may be the most relevant for disease. This study reinforces the crucial role of RNA-processing regulation for neuronal integrity and potentially identifies candidate genes whose altered processing is central to ALS pathogenesis.
C27 PATHOLOGICAL FRAGMENT OF TDP-43 IN ALS IS GENERATED BY ALTERNATE TRANSLATION INITIATION
XIAO S, SANELLI T, CHIANG H, SO K, KEITH J, ROGAEVA E, BILBAO J, ZINMAN L, ROBERTSON J
University of Toronto, Toronto, Canada
Email address for correspondence: [email protected]
Keywords: TDP-43, RNA, splicing
Background: TAR DNA binding Protein-43 (TDP-43), is a major component of the cytoplasmic inclusions characteristic of amyotrophic lateral sclerosis (ALS). In addition to the full-length protein of 43 kDa, biochemical profiles of TDP-43 from diseased tissues have shown the presence of lower molecular weight species, most predominantly of 35 kDa (TDP-35) and 25 kDa (TDP-25). These species have been considered caspase-3 (C-3) degradation products since cleavage at the C-3 consensus sequences DETD89A and DVMD219V can generate species of 35 kDa and 25 kDa in vitro, respectively. However, there is no direct proof that these species are generated by C-3 cleavage in disease.
Objectives: Since TDP-43 is a splicing regulator and is itself alternatively spliced, we explored the possibility that the lower molecular weight species of TDP-43 found in diseased tissues may not be C-3 degradation products, but instead, alternatively spliced (AS) variants of TDP-43.
Methods: Genome bioinformatics databases were mined for AS transcripts of TDP-43. We focused on one AS transcript that showed elevated expression in ALS spinal cord compared to controls, as assessed by RT-PCR. This transcript was cloned and expression studies undertaken in cultured SHSY5Y human neuroblastoma cells and primary motor neurons. Antibody specific to the splice variant was generated and used to assess its expression in ALS spinal cord tissues.
Results: We identified an AS variant of TDP-43 with 91 bp skipped in exon 2 that was upregulated in ALS spinal cord compared to controls. The 91bp deletion caused a frameshift and use of downstream alternate translation initiation codon, ATG Met85. Expression of this transcript in cells generated a species of 35 kDa, named AS-TDP-35, that partitioned to the urea soluble fraction, and formed cytoplasmic aggregates both in SHSY5Y cells and primary neurons, causing neurotoxicity. To determine if TDP-35 is generated by expression from Met85 (AS-TDP-35) or C-3 cleavage at DEVD89A (C3-TDP-35), we made neoepitope antibodies corresponding to the different N-terminal sequences that would be generated, respectively. We show that the pathological TDP-35 species observed on immunoblots of ALS lumbar spinal cord tissues is labeled with antibody specific to AS-TDP-35, but not C3-TDP-35 antibody. We also show that AS-TDP-35 antibody labels both skein-like and round inclusions, whereas C3-TDP-35 antibody labeling was negative.
Discussion: These results show that the lower molecular weight TDP-43 species of 35kDa present in ALS spinal cord tissues is generated by expression from Met85. We propose that this occurs through expression of an AS variant of TDP-43, in which there is a 91 bp deletion in exon 2, causing a frameshift and alternate translation initiation. This identifies abnormal splicing of TDP-43 to generate TDP-35 as well as the downstream neurotoxic effects of its expression, as potential therapeutic targets.
C28 INTRACELLULAR INCLUSIONS OF THE RNA BINDING PROTEIN RBM45 IN ALS AND FTLD PATIENTS
BOWSER R, RIASCOS D, KOVALIK T, KRUPA K, KRUPA K, COLLINS M
University of Pittsburgh, Pittsburgh, PA, USA
Email address for correspondence: [email protected]
Keywords: RNA binding protein, protein aggregation, intracellular inclusion
Background: The emergence of defects associated with RNA binding proteins, predominantly TDP-43 and FUS, as pathogenic mechanisms in ALS and frontotemporal lobar dementia (FTLD) has fundamentally changed our view of these diseases. These proteins form cytotoxic inclusions in motor neurons and often exhibit an aberrant cytoplasmic localization. Moreover, genomic studies have identified sequence alterations in the TDP-43 and FUS genes in familial forms of ALS. We recently used unbiased proteomics methodologies to identify proteins that exhibit abnormal levels in the CSF of ALS patients. We detected an increase in the novel RNA binding protein RBM45 in the CSF of ALS patients.
Objectives: We used a combination of immunohistochemistry, immunofluorescence microscopy, and immunoblot to characterize RBM45 protein expression and distribution in control, ALS and FTLD patients.
Methods: Paraffin tissue and snap frozen tissue from the spinal cord, motor cortex and hippocampus of ALS, FTLD, and age-matched control patients were obtained from the University of Pittsburgh ALS Tissue Bank. RBM45 expression and subcellular distribution were examined by immunohistochemistry, immunofluorescence, and immunoblot. We also used anti-TDP43 to examine co-localization with RBM45.
Results: Immunoblot analysis verified our mass spectrometry results and demonstrated increased RBM45 levels in the CSF and spinal cord of ALS patients. We observed weak RBM45 immunoreactivity in the nucleus of motor neurons and glial cells in the spinal cord and brain in control subjects. However in ALS patients we detected robust RBM45 nuclear immunostaining and intracytoplasmic inclusions of RBM45 in spinal cord motor neurons and dendate granule cells of the hippocampus. We also observed cytoplasmic inclusions of RBM45 in the dendate gyrus of FTLD patients. RBM45 inclusions bear a striking resemblance to those containing TDP-43 or FUS in ALS/FTLD. In some cases, RBM45 inclusions co-localized within TDP-43 inclusions. However many RBM45 inclusions did not contain TDP-43. No inclusions were observed in control subjects. We also identified prominent glial RBM45 immunoreactivity in all brain areas of interest.
Discussion: We have identified a new RNA binding protein that exhibits intracellular inclusions of affected neurons in ALS and FTLD. Our results support the role for RNA binding proteins in the pathogenesis of ALS and FTLD. The overlapping distribution of RBM45 and TDP-43 suggests a commonality between the mechanisms that result in inclusion formation for each protein. Our findings raise the possibility that RBM45 expression can be induced in response to pathologic stressors. Moreover, we found prominent glial expression of RBM45.
Conclusions: We identified intracellular inclusions of the novel RNA binding protein RBM45 in affected neurons of ALS and FTLD patients. Ongoing studies will determine the functional role of RBM45 in regulating mRNA metabolism and its role in ALS and other neurodegenerative diseases.
C29 DOWNREGULATION OF RNA EDITING ENZYME ADAR2 AND SPORADIC ALS
HIDEYAMA T1, YAMASHITA T1, AIZAWA H2, TSUJI S1, KAKITA A3, TAKAHASHI H3, KWAK S1
1Graduate School of Medicine, The University of Tokyo, Tokyo, Japan, 2Tokyo National Hospital, Tokyo, Japan, 3Brain Research Institute, Niigata University, Niigata, Japan
Email address for correspondence: [email protected]
Keywords: glutamate receptor, neuronal cell death, RNA editing
Introduction: Ca2+ -permeable AMPA receptors play a pivotal role in neuronal death in ALS. GluA2 is a subunit of the AMPA receptor, playing a key role in the regulation of Ca2+ permeability after adenosine to inosine conversion (RNA editing) at the glutamine/arginine (Q/R) site, where the Q codon (CAG) is substituted by the R codon (CIG = CGG). A subset of motor neurons of patients with sporadic ALS express unedited (Q at the Q/R site) GluA2, hence abundant Ca2+ -permeable AMPA receptors. Because GluA2 Q/R site-editing is specifically mediated by adenosine deaminase acting on RNA 2 (ADAR2), it is likely that ADAR2 activity is reduced in ALS motor neurons.
Methods: Using a laser-microdissector, single motor neurons were dissected from autopsy-obtained frozen spinal cords of 29 ALS patients. After analyzing the extent of GluA2 Q/R site-editing in an individual motor neuron, expression levels of three members of the ADAR family (ADAR1, ADAR2 and ADAR3) were analyzed on pooled cDNAs from ALS motor neurons expressing unedited GluA2 and those expressing only edited GluR2. In addition, enzymatic activities of ADAR1 and ADAR2 in ALS motor neurons were analyzed by the measurement of the extent of RNA editing positions specifically mediated by either one of ADARs. The results were compared with those on control subjects and patients with other neurological diseases.
Results: We demonstrated that a considerable proportion of motor neurons express unedited GluA2 in all the ALS cases examined, while all the motor neurons of control cases expressed only edited GluA2. ADAR2, but not ADAR1 or ADAR3, was significantly downregulated in all the motor neurons of ALS patients, more extensively in those expressing unedited GluA2 than those expressing only edited GluA2. Extents of RNA editing at ADAR2-specific RNA editing positions, but not those at ADAR1-specific RNA editing positions were significantly lower in ALS motor neurons than in control motor neurons.
Discussion: The present results demonstrate that ADAR2 is universally downregulated in motor neurons of sporadic ALS patients irrespective of the variety of clinical manifestations. This indicates that loss of ADAR2-immunoreactivity in about half of motor neurons of ALS patients (1) is not likely a consequence of accelerated degradation of ADAR2 proteins but reflects the severity of ADAR2 downregulation. Because loss of ADAR2 induces death of motor neurons specifically by failure to edit the GluA2 Q/R site in conditional ADAR2 knockout mice (2), our results suggest that progressive downregulation of ADAR2 is involved in the ALS pathogenesis.
References
- Aizawa, et al. Acta Neuropathol, 2010;120:75–84.
- Hideyama, et al. J Neurosci 2010;30:11917–25.
Support: Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology; Health and Labour Sciences Research Grant.