Dual-specificity phosphatases (DUSPs), which constitute type I cysteine-based protein tyrosine phosphatases, are a subfamily of protein phosphatases that can dephosphorylate both tyrosine and serine/threonine residues within the same substrate.1 To date, at least 16 mammalian DUSPs showing dephosphorylation activity towards MAP kinases (MAPKs) have been identified. 2 Although the actual physiological roles of many DUSP proteins are still largely unknown, growing information have revealed their correlation with numbers of human diseases e.g., multiple types of cancer, cerebral ischemia and neurodegenerative diseases. [3][4][5][6] It is generally assumed that DUSP proteins have distinct roles in diverse cellular processes; however, the understanding of specific functions of individual DUSP protein has been somewhat difficult partly due to the overlapping substrate specificity.7 A member of DUSP subfamily, DUSP8 (also known as M3/6 or hVH5) is reported to selectively regulate JNK pathway and, to a lesser extent, p38 pathway. 7,8 An interesting characteristic of DUSP8 is that protein damaging conditions can inactivate this phosphatase by increasing insolubility or decreasing stability. 9,10 In a group of neurodegenerative diseases collectively known as polyglutamine (polyQ) diseases, the expansion of CAG (glutamine) repeats leads to misfolding and aggregation of causative proteins.11 Similar to protein damaging conditions (e.g., heat shock and oxidative stress), polyQ-induced toxicity involves the aggregation of aberrant proteins and the concomitant activation of MAP kinase through the inhibition of MAPK phosphatases. It was shown that the solubility of DUSP8 is impaired by the expansion of polyQ-repeats. 6 In the previous study, we demonstrated that polyglutaminerepeat protein ataxin-1 (ATXN1; the causative protein of spinocerebellar ataxia type 1) interacts with a number of DUSP proteins including DUSP8 and the cellular state of oxidative stress influences the association between ataxin-1 and DUSP8.12 During the course of studies, we repeatedly observed that the cellular levels of ataxin-1 were diminished by DUSP8 over-expression, implicating that it may facilitate the down-regulation of ataxin-1. Since it was plausible that DUSP8 decreases the solubility of ataxin-1 rather than directly promoting protein degradation and thereby causes its accumulation in insoluble fraction, we first measured the effects of DUSP8 on the relative amounts of ataxin-1 in soluble and insoluble fractions. To this end, we fractionated the lysate of HEK293T cells co-expressing ataxin-1 and DUSP8 and then examined whether DUSP8 causes increased degradation of ataxin-1 or enhanced partitioning of polyQ proteins into insoluble fraction.Over-expression of DUSP8 significantly reduced the steadystate level of ataxin-1 both in the soluble and insoluble fractions supporting the notion that the down-regulation of this polyQ protein is indeed caused by the increased degradation ( Fig. 1(a)). In fact, DUSP8 over-expression almost completely abolish...