Deletion of Msh2 in medium-spiny neurons delays nuclear huntingtin phenotypes. A, B.

<p>Nuclear mutant huntingtin immunostaining is decreased in the striata of five-month old <i>HdhQ111</i>/+ mice with deletion of <i>Msh2</i> in MSNs. <b>A.</b> Fluorescent micrographs of striata double-stained with anti-huntingtin mAb5374 and anti-histone H3 antibodies for three CAG repeat length-matched mice (<i>Msh2</i>+/+ CAG 113, <i>Msh2Δ</i>/<i>Δ</i> CAG 112, <i>Msh2−/−</i> CAG 113). <b>B.</b> Box plot showing upper and lower quartiles, median and range for the normalized mAb5374 immunostaining intensity (total mAb5374 staining intensity normalized to the number of H3-positive nuclei). Outlier (circle) is defined by a standard interquartile method and is included in the analysis. Multiple regression analysis was used to determine the effect of <i>Msh2</i> genotype on mAb5374 staining using normalized mAb5374 intensity (continuous variable) as a dependent variable and <i>Msh2</i> genotype (discrete variable), constitutive CAG length (continuous variable) and position (medial versus lateral, discrete variable) as independent variables. Both constitutive CAG length (P<0.05) and medial versus lateral position (P<0.001) were significantly associated with normalized mAb5374 intensity. Asterisks above the bars indicate a significant difference from <i>Msh2</i>+/+ at a p-value cut-off of p<0.05(*), p<0.01 (**), p<0.001 (***) in the regression analysis. <i>Msh2Δ</i>/− was not significantly different from <i>Msh2</i>+/− (p = 0.18). The five-month mice used in the quantitative analysis are as follows: <i>Msh2</i>+/+ (n = 6, CAG 113, 118, 119, 121, 123, 125), <i>Msh2</i>+/− (n = 4, CAG 114, 114, 120, 123), <i>Msh2Δ</i>/<i>Δ</i> (n = 5, CAG 113, 121, 121, 126, 129), <i>Msh2Δ</i>/− (n = 7, CAG 113, 121, 121, 122, 125, 125, 133) and <i>Msh2</i>−/− (n = 3, CAG 112, 120, 123). Note that the relatively “weak” effect of the <i>Msh2</i>−/− genotype likely reflects the small number of mice of this genotype and hence the least accurate estimate of the relationship of mAb5374 intensity to CAG length in the regression analysis. <b>C, D.</b> Intranuclear inclusions are decreased in the striata of ten-month old <i>HdhQ111</i>/+ mice with deletion of <i>Msh2</i> in MSNs. <b>C.</b> Fluorescent micrographs of striata stained with mAb5374 from mice with <i>Msh2</i>+/+ (CAG 121), <i>Msh2</i>+/− (CAG 123), <i>Msh2Δ</i>/<i>Δ</i> (CAG 133), <i>Msh2Δ</i>/− (CAG 123) and <i>Msh2</i>−/− (CAG 132) genotypes. <b>D.</b> Quantification of the percentage of cells containing an inclusion (more than one inclusion per cell was rarely observed). The total number of cells was determined by co-staining with histone H3 (not shown). The ten-month mice used in the quantitative analysis are as follows: <i>Msh2</i>+/+ (n = 6, CAG 118, 121, 121, 123, 126, 134), <i>Msh2</i>+/− (n = 4, CAG 116, 118, 123, 131), <i>Msh2Δ</i>/<i>Δ</i> (n = 1, CAG 133), <i>Msh2Δ</i>/− (n = 7, CAG 115, 115, 117, 120, 121, 122, 123) and <i>Msh2</i>−/− (n = 1, CAG 132). Bars represent mean ±S.D.