Fig. 1.

RNAi-mediated suppression of expanded CAG repeat-containing genes. Expanded CAG repeats are not direct targets for preferential inactivation (a), but a linked SNP can be exploited to generate siRNA that selectively silences mutant ataxin-3 expression (b–f). (a) Schematic of cDNA encoding generalized polyQ-fluorescent protein fusions. The bars indicate regions targeted by siRNAs. HeLa cells cotransfected with Q80-GFP, Q19-RFP, and the indicated siRNA. Nuclei are visualized by 4′,6-diamidino-2-phenylindole staining (blue) in merged images. (b) Schematic of human ataxin-3 cDNA, with bars indicating regions targeted by siRNAs. The targeted SNP (G987C) is shown in color. In the displayed siRNAs, red or blue bars denote C or G, respectively. (c) Quantitation of fluorescence in Cos-7 cells transfected with WT or mutant ataxin-3 (Atx)-GFP expression plasmids and the indicated siRNA. Fluorescence from cells cotransfected with siMiss was set at 1. The bars depict mean total fluorescence from three independent experiments ± SEM. Rel. Fl., relative fluorescence. (d) Western blot analysis of cells cotransfected with the indicated ataxin-3 expression plasmids (Upper) and siRNAs (Lower). Appearance of aggregated, mutant ataxin-3 in the stacking gel (seen with siMiss and siG10) is prevented by siRNA inhibition of the mutant allele. (e) Allele specificity is retained in the simulated heterozygous state. Western blot analysis of Cos-7 cells cotransfected with WT (Atx-3-Q28-GFP) and mutant (Atx-Q166) expression plasmids along with the indicated siRNAs (mutant ataxin-3 detected with 1C2, an antibody specific for expanded polyQ, and WT ataxin-3 detected with anti-ataxin-3 antibody). (f) Western blot of Cos-7 cells transfected with ataxin-3-GFP expression plasmids and plasmids encoding the indicated shRNA. The negative control plasmid, phU6-LacZi, encodes siRNA specific for LacZ. Both normal and mutant proteins were detected with anti-ataxin-3 antibody. Tubulin immunostaining is shown as a loading control in d–f.