In double-transgenic mouse models non-expanded ataxin-3 not suppresses polyglutamine neurodegeneration in the SCA3 disease – in vivo evidence

Central questions in polyQ-induced neurodegeneration are the influence of the function of the non-expanded and the disease protein, the formation of aggregates, and the modulation by quality control mechanisms (e.g. unfolded protein response, ERAD or ubiquitin proteasome pathway). To answer the question of the interaction of a non-expanded allele with an expanded polyQ stretch in vivo we generated double-transgenic (dt) mouse models for the neurodegenerative disease spinocerebellar ataxia type 3 (SCA3). SCA3, also known as Machado-Joseph disease (MJD), is the most common type of spinocerebellar ataxias in Europe, Japan and United States. SCA3 belongs to the group of polyglutamine repeat diseases (polyQ) as Huntington's disease, dentatorubro-pallidoluysian atrophy (DRPLA) and SCA 1, 2, 6, 7 and 17. Dt-mice were generated by mating mice which expressed the human SCA3 gene with a non-expanded allele (15Q) with mice from a SCA3 disease model which expressed an expanded polyQ stretch (70Q). Behaviour studies revealed that the dt-mice (expressed both 15 and 70Q) developed neurological symptoms with a premature death at the age of 6 months. This is comparable with the phenotype and the premature death which we detected in the disease model with 70 CAGs. Immunohistochemical staining revealed nuclear aggregates in the dt-mice and the model with 70 CAGs. Therefore, we can conclude that in this in vivo study the normal non-expanded polyQ stretch is not a striking suppressor of the polyglutamine-induced neurodegeneration in SCA3. This is in contrast to a published SCA3 Drosophila model were the non-expanded ataxin-3 allele is a suppressor in neurodegeneration (Warrick et al., 2005; Mol Cell 18:37–48). Another dt-mouse model which we generated harboured the human ataxin-3 gene with 70 CAGs (the same which we used for the described dt 15/70Q model) und a second human ataxin-3 gene with 148 CAGs and an external nuclear export signal (NES). The aim includes the question: is the external NES signal strong enough to rescue the cells for aggregation in the nucleus of the disease model with 70 CAGs. The answer is no – mice from the dt-mouse model (70/148 NES) developed a very early and severe neurological phenotype. These mice died at the age of 6 weeks and showed neurological symptoms from the first day of their life. The used single-transgenic mice (PrP-MJD 15.1; PrP-MJD 70.61 and PrP-MJD NES. 42) are already published (Bichelmeier et al., 2007; J o Neurosci 27:7418–28).