Objective: A noncoding hexanucleotide repeat expansion in C9orf72 is the most common cause of amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration (FTLD). It has been reported that the repeat expansion causes a downregulation of C9orf72 transcripts, suggesting that haploinsufficiency may contribute to disease pathogenesis. Two protein isoforms are generated from three alternatively spliced transcripts of C9orf72; a long form (C9-L) and a short form (C9-S), and their function(s) are largely unknown owing to lack of specific antibodies. Methods: To investigate C9orf72 protein properties, we developed novel antibodies that recognize either C9-L or C9-S. Multiple techniques, including Western blot, immunohistochemistry, and coimmunoprecipitation, were used to determine the expression levels and subcellular localizations of C9-L and C9-S. Results: Investigation of expression of C9-L and C9-S demonstrated distinct biochemical profiles, region-specific changes, and distinct subcellular localizations in ALS tissues. In particular, C9-L antibody exhibited a diffuse cytoplasmic staining in neurons and labeled large speckles in cerebellar Purkinje cells. In contrast, C9-S antibody gave very specific labeling of the nuclear membrane in healthy neurons, with apparent relocalization to the plasma membrane of diseased motor neurons in ALS. Coimmunoprecipitation experiments revealed an interaction of the C9-isoforms with both Importin b1 and Ran-GTPase, components of the nuclear pore complex. Interpretation: Using these antibodies, we have shown that C9orf72 may be involved in nucleocytoplasmic shuttling and this may have relevance to pathophysiology of ALS/FTLD. Our antibodies have provided improved detection of C9orf72 protein isoforms, which will help elucidate its physiological function and role in ALS/FTLD. ANN NEUROL 2015;78:568-583 H exanucleotide (G 4 C 2 ) repeat expansions within a noncoding region of C9orf72 are the most common known genetic cause of amyotrophic lateral sclerosis (ALS) and frontotemporal lobar degeneration (FTLD), accounting for up to 50% of ALS, 29% of FTLD, and 88% of ALS/ FTLD patients, including familial and sporadic cases.

No abstract

Tar DNA-binding protein 43 (TDP-43) is an RNA-binding protein normally localized to the nucleus of cells, where it elicits functions related to RNA metabolism such as transcriptional regulation and alternative splicing. In amyotrophic lateral sclerosis, TDP-43 is mislocalized from the nucleus to the cytoplasm of diseased motor neurons, forming ubiquitinated inclusions. Although mutations in the gene encoding TDP-43, TARDBP, are found in amyotrophic lateral sclerosis, these are rare. However, TDP-43 pathology is common to over 95% of amyotrophic lateral sclerosis cases, suggesting that abnormalities of TDP-43 play an active role in disease pathogenesis. It is our hypothesis that a loss of TDP-43 from the nucleus of affected motor neurons in amyotrophic lateral sclerosis will lead to changes in RNA processing and expression. Identifying these changes could uncover molecular pathways that underpin motor neuron degeneration. Here we have used translating ribosome affinity purification coupled with microarray analysis to identify the mRNAs being actively translated in motor neurons of mutant TDP-43(A315T) mice compared to age-matched non-transgenic littermates. No significant changes were found at 5 months (presymptomatic) of age, but at 10 months (symptomatic) the translational profile revealed significant changes in genes involved in RNA metabolic process, immune response and cell cycle regulation. Of 28 differentially expressed genes, seven had a ≥ 2-fold change; four were validated by immunofluorescence labelling of motor neurons in TDP-43(A315T) mice, and two of these were confirmed by immunohistochemistry in amyotrophic lateral sclerosis cases. Both of these identified genes, DDX58 and MTHFSD, are RNA-binding proteins, and we show that TDP-43 binds to their respective mRNAs and we identify MTHFSD as a novel component of stress granules. This discovery-based approach has for the first time revealed translational changes in motor neurons of a TDP-43 mouse model, identifying DDX58 and MTHFSD as two TDP-43 targets that are misregulated in amyotrophic lateral sclerosis.