Dissecting the role of ataxin-3 isoforms for the pathogenesis of Machado Joseph disease

Abstract

The autosomal-dominantly inherited, neurodegenerative disorder Machado-Joseph Disease (MJD) is caused by an expanded CAG repeat in the ataxin-3 encoding gene ATXN3. This sequence is translated into a polyglutamine repeat in the protein. ATXN3 is spliced alternatively leading to protein isoforms which differ in their their C-terminus and number of ubiquitin interacting motifs. Additionally, the isoforms are modified by single nucleotide polymorphisms of which one is leading to a premature stop codon. The objective of this study was to assess the influence of both, alternative splicing and the nonsense polymorphism on major aspects of ataxin-3's physiological function and main disease mechanisms. It was found that the premature stop and alternative splicing of the ATXN3 gene affect the physiological characteristics on multiple levels and modulate the proteins' stability enzymatic activity and subcellular localization. Interestingly, also an isoform-specific interaction of ataxin-3 with other proteins could be observed. This demonstrates that ataxin-3 isoforms are individual proteins with distinct functions and interaction networks within the cell. The modulatory effects, however, were not limited to the physiological properties of non-expanded ataxin-3. An expansion of the polyglutamine repeat entailed a stabilization of the respective isoform and caused aggregation. Surprisingly, alternative splicing and the nonsense mutation comprehensively modified the aggregation profperties. Thus ataxin-3 isoforms supposedly show a differential contribution to the pathogenesis of MJD. Further, a functional interaction between non-expanded and expanded ataxin-3 isoforms could be demonstrated for the first time. Interestingly, this cross-talk likewise modified physiological and pathophysiological properties of the protein. In order to convert the present data into a treatment strategy, the possibility of correcting the premature termination codon by translational read-through was assessed. Utilizing aminoglycoside antibiotics allowed for a partial compensation of the nonsense polymorphism and proved that it could be an interesting therapeutic intervention in order to ameliorate the disease progression. Lastly, in order to additionally verify the obtained results in vivo, a new isoform-specific zebrafish model was generated and initially characterized. Altogether, alternative splicing, the nonsense polymorphism and the mutual interaction of ataxin-3 isoforms modify both, major physiological characteristics of ataxin-3 and MJD pathogenesis. However, new therapeutic strategies like stop codon read-through may allow for a compensation of the negative effects of the nonsense polymorphism. The data provided in this study stresses the importance of considering isoforms of disease causing proteins, their modification by single nucleotide polymorphisms and their interplay with the normal protein allele as disease modifiers, not only in MJD but in autosomal-dominantly inherited diseases in general.