Fig. 7

MUT ATXN7 specifically traps MBNL1 but not MBNL2 in the mouse cerebellum. a–c Laser confocal microscopy in cerebellum shows co-aggregation of MUT ATXN7 (green) and MBNL1 (red) in aggregates (~80 %) in the the GCL (B); no MBNL1 aggregation was observed in WT ATXN7-expressing PCs (A). The MBNL2 protein was not trapped in nuclear MUT-ATXN7 aggregates (C). d Representative western-blot of cerebellar lysates shows increased levels of the MBNL1 protein (55 kDa, 35 kDa and 25 kDa) in a mouse overexpressing MUT ATXN7 compared to a mouse overexpressing the control WT ATXN7 (n = 3). H) Western-blot: no differences in MBNL2 protein levels were observed in the cerebella of mice injected with LV-WT-ATXN7 or LV-MUT-ATXN7 (n = 3). Tubulin was used as a loading control. e, f, g and i Optical densitometry was normalized according to the amount of tubulin loaded in the corresponding lane. A partition ratio was calculated and expressed as optical densitometry (arbitrary units) relative to the sample with highest value for the normalization control set at 1. Values are expressed as mean ± standard error (SEM) of the mean. *p ≤ 0.05 (Student’s T test). All data are from 3 mice/group. Bars: A–C: 10 μm