Figure 4.

Both C3H/He and CBA contribute to dentate granule cell loss of the original hybrid background. a, As expected from our initial analyses, the dentate granule cell layer was unaffected by TTA expression on the congenic B6 background. b–e, However, on several F1 hybrid backgrounds derived from the B6 congenic line, TTA expression significantly decreased the width of the granule cell layer between 14% (B6C3) and 30% (B6CBA). f, Dentate degeneration was also present in TTA-expressing offspring from the B6C3 F1 intercross, where 15 of 20 F2 TTA mice exhibited thinning of the dentate granule cell layer compared with NTG siblings. One-quarter of the cohort (5 of 20) appeared indistinguishable from NTG. Representative images from both groups are shown. Scale bar, 300 μm. g, The average width of the granule cell layer was measured for each background (n = 6–20 per genotype). One outlier, determined by Grubb's test, was removed from the B6C3 × CBA TTA group before analysis and appears near the mean for NTG animals of that background. Note that the y-axis starts at 30 μm. h, Cell counts performed in C3B6 × C3 mice confirm that the width of the granule cell layer is linearly related to the number of granule cells remaining. Data points from each microscopic field are shown in gray (6–9 fields per mouse) and average values from each animal are shown in black (n = 7–8 per genotype). Solid line indicates the best-fit linear regression line for the average values, dashed lines indicate the 95% confidence interval. Note the x-axis starts at 20 μm. i, Multipoint interval mapping was used to locate genetic modifiers of the neurodegenerative phenotype in the B6C3 F2 TTA cohort. LOD scores are plotted as a function of marker location in centimorgans, with the chromosome number designated. Black line represents the LOD significance threshold for α = 0.05, determined by Haley–Knott permutations. Odd-numbered chromosomes are shaded in light gray. *p < 0.05, ***p < 0.001, two-way ANOVA with Bonferroni post hoc test.