Spinocerebellar ataxia type-3 (SCA3) is a neurodegenerative disorder caused by a polyglutamine repeat expansion in the ataxin-3 protein. Cleavage of mutant ataxin-3 by proteolytic enzymes yields ataxin-3 fragments containing the polyglutamine stretch. These shorter ataxin-3 fragments are thought to be involved in SCA3 pathogenesis due to their increased cellular toxicity and their involvement in formation of the characteristic neuronal aggregates. As a strategy to prevent formation of toxic cleavage fragments， we investigated an antisense oligonucleotide-mediated modification of the ataxin-3 pre-mRNA through exon skipping of exon 8 and 9， resulting in the removal of a central 88 amino acid region of the ataxin-3 protein. This removed protein region contains several predicted cleavage sites and two ubiquitin-interacting motifs. In contrast to unmodified mutant ataxin-3， the internally truncated ataxin-3 protein did not give rise to potentially toxic cleavage fragments when incubated with caspases. In vitro experiments did not show cellular toxicity of the modified ataxin-3 protein. However， the modified protein was incapable of binding poly-ubiquitin chains， which may interfere with its normal deubiquitinating function. Low exon skipping efficiencies combined with reduction in important ataxin-3 protein functions suggest that skipping of exon 8 and 9 is not a viable therapeutic option for SCA3.