Fig. 3: Pyramidal neuron density in area CA2 becomes more CA1-like in MR KO.

a, b Immunostaining for ZnT3 to label the mossy fibers and WFS1 to label the CA1 pyramidal cells in WT (a) and MR KO (b) animals were used to localize CA1, CA2 and CA3. DAPI was used to label nuclei. In WT animals, CA2 is positioned at the distal end of the mossy fibers, and WFS1 does not overlap with ZnT3. CA2 was defined as the pyramidal cell nuclei in the pyramidal cell area overlapping with the distal-most 150 µm of the ZnT3 stain, and measurements were made from there. In MR KO animals, ZnT3 expression overlaps with the cells expressing WFS1, so CA2 was defined as the area of overlap between these two markers, and measurements corresponding to CA2 were made from that area. Inset squares represent the areas in each subregion that are shown enlarged on the right side. c Nuclear spacing, measured by nearest neighbor analysis, show that in WT animals, both CA2 and CA3 neurons have significantly greater nuclear spacing than CA1, and CA2 spacing does not differ from CA3 (F(2,10) = 81.24, p < 0.0001, one-way ANOVA; results of Tukey’s post-hoc test shown). In MR KOs animals, both CA2 and CA3 have significantly greater nuclear spacing than CA1, but CA2 has significantly less nuclear spacing than CA3. (F(2,16) = 122.3, p < 0.0001, one-way ANOVA; results of Tukey’s post-hoc test shown) Nuclear spacing in CA2 was significantly greater in MR KOs than in WTs (main effect of subfield: F(2,26) = 164.7, p < 0.0001; subfield x genotype interaction: F(2,26) = 44.71, p < 0.0001; two way-ANOVA, result of Sidak post-hoc test shown). d Nuclear density was measured across the depth of DAPI-stained section. In WT animals, CA2 had a significantly lower nuclear density than CA1 but not CA3 (F(2,10) = 97.75, p < 0.0001, one-way ANOVA; results of Tukey’s post-hoc test shown, but in MR KO animals CA2 nuclear density was significantly greater than in CA3 (F(2,16) = 146.7, p < 0.0001, one-way ANOVA; results of Tukey’s post-hoc test shown). Nuclear density was significantly greater in CA2 of MR KO animals than in WT (main effect of subfield: F(2,26) = 210.0, p < 0.0001; main effect of genotype: F(1,13) = 9.863, p < 0.001; subfield x genotype interaction: F(2,26) = 35.31, p < 0.0001; two way-ANOVA, result of Sidak post-hoc test shown), suggesting that CA2 moves toward an anatomical profile more closely resembling CA1 than CA3.