Figure 2.
Homogeneous distribution of microglial cells is lost in the aged brain. Histological sections (40 μm) were taken from young adult (3 months), adult (11–12 months), and aged (26–27 months) Iba-1-EGFP transgenic mice (7–8 animals/group). (A) Representative Iba-1 immunostained sections are shown for each age group. Scale bar is 50 μm and 20 μm for insert. (B) Stereological analysis showed a ~14% increase in Iba-1 stained microglia in the aged animals compared to the young adults (one-way ANOVA F2,18 = 1.395, P = 0.2733). Error bars indicate SEM. (C) Microglial distribution was analyzed by laser scanning confocal microscopy, performed on 40–μm-thick Iba-1 immunostained sections. Representative max. projections are shown for each age group. Scale bar is 50 μm. White circles show exemplary radii (20 μm, 40 μm, and 60 μm) used for each microglia to compute nearest neighbors and clusters. (D) Spatial distribution of microglial cells was analyzed by calculating local cell distribution patterns. A clustered cell was defined as having at least 3 microglial cells within the set radii. The comparison of clustered microglia across age groups revealed significant differences in their clustering pattern, with an increased number of microglia in close proximity to each other in the aged mice (*P < 0.05; 40 μm radius 95%CI young adult = 1.29–4.51, aged=1.87–81.32; 60 μm 95%CI young adult = 52.37–74.30, aged = 86.12–99.13; 80 μm radius 95%CI young adult = 90.20–96.41, aged = 98.29–100.14). (E) To further determine differences in distribution patterns, the total number of microglia was counted in set radii. This comparison showed a significant increase in the percentage of microglia in close proximity (0–20 μm) in aged animals vs. young adults and adults (Mann–Whitney U-test P < 0.05). Correspondingly, the number of cells present in a distance >60 μm decreased in aged animals (Mann–Whitney U-test P < 0.05). (F) Logistic regression analysis was used to model the probability of correctly predicting an animal’s respective age group by examining the distance to its nearest neighbor. The probability of an animal being aged significantly increases as the distance to the nearest neighbor decreases, when considering a distinction between young adult and aged groups (continuous line), and between adult and aged groups (dotted line) (Wald tests: young adults vs. aged P < 0.001 for up to 60 μm neighbors and P < 0.05 for up to 80 μm; adults vs. aged animals P < 0.001 for up to 60 μm neighbors and P < 0.01 for up to 80 μm; young adults vs. adult animals no significant difference). A clear trend is depicted, showing that the aged group can always be predicted in comparison with the other groups. Considering the probability of aging from young adult to adult group (dashed line), there is no effect of the distance to the nearest neighbor, suggesting a high spatial similarity in the microglia distribution of nonaged animals. Error bars are 95% confidence intervals.