Machado-Joseph disease or spinocerebellar ataxia type 3 (MJD/SCA3) is a fatal, autosomal dominant disorder caused by a cytosine-adenine-guanine expansion in the coding region of the MJD1 gene. RNA interference has potential as a therapeutic approach but raises the issue of the role of wild-type ataxin-3 (WT ATX3) in MJD and of whether the expression of the wild-type protein must be maintained. To address this issue, we both overexpressed and silenced WT ATX3 in a rat model of MJD. We showed that (i) overexpression of WT ATX3 did not protect against MJD pathology, (ii) knockdown of WT ATX3 did not aggravate MJD pathology and that (iii) non-allele-specific silencing of ataxin-3 strongly reduced neuropathology in a rat model of MJD. Our findings indicate that therapeutic strategies involving non-allele-specific silencing to treat MJD patients may be safe and effective.