Fig. 3

Extended elimination of microglia does not induce peripheral leukocyte or behavioral abnormalities. 1.5-month-old wild-type (WT) or 5xFAD mice were treated with control chow or PLX5622 for 10 or 24 weeks. a Experimental design. b Terminal PK of PLX5622 and 5xFAD + PLX5622 groups. Two-tailed independent t-test. n = 5–6 for PLX5622, n = 8–11 for 5xFAD + PLX5622. c Hemisphere stitches of microglia (IBA1 in green) in the 7-month-old cohort. Scale bar = 1000 μm. d Analysis of different subsets of leukocytes (CD11b, p = 0.059; CD19, p = 0.052; all others NS). Two-tailed independent t-test; n = 4–5 for Wild-type, n = 3–4 for PLX5622. e–f All analyses listed in respective order for retrosplenial (RS) cortex, somatosensory (SS) cortex, and thalamus. Microglial number in 4 e and 7 f month-old cohorts (4-month cohort: WT v 5xFAD, p = 0.001, NS, p < 0.001; WT v PLX5622, p < 0.001, p < 0.001, p = 0.044; 5xFAD v 5xFAD + PLX5622, p < 0.001, p < 0.001, p < 0.001. 7-month cohort: WT v 5xFAD, p = 0.015, p = 0.024, p < 0.001; WT v PLX5622, p < 0.001 for all regions; 5xFAD v 5xFAD + PLX5622, p < 0.001 for all regions). Two-way ANOVA with Tukey’s post hoc test; n = 4–6 for Wild-type, n = 4–6 for PLX5622, n = 5–8 for 5xFAD, n = 6–9 for 5xFAD + PLX5622. g–h Measurements for anxiety were performed by Elevated Plus Maze (EPM) and quantified as number of arm entries (g WT v 5xFAD, NS (open) and p = 0.050 (closed); PLX5622 v 5xFAD + PLX5622, p = 0.007 (open) and p < 0.001 (closed), 5xFAD v 5xFAD + PL5622, NS (open) and p = 0.056 (closed)) and time in arms (h WT v 5xFAD, p = 0.0157 (open) and p = 0.046 (closed); PLX5622 v 5xFAD + PLX5622, p < 0.001 (open) and p < 0.001 (closed); 5xFAD v 5xFAD + PLX5622, p < 0.001 (open), p < 0.020 (closed)). i Mean latencies to a hidden platform from the Morris water maze (MWM) acquisition trials. j–k Number of platform entries (j; PLX5622 v 5xFAD + PLX5622, p = 0.071) and time in platform zone (k; WT v PLX5622, p = 0.094; PLX5622 v 5xFAD + PLX5622, p = 0.073) in the MWM probe trial. For all behavioral analyses: Two-way ANOVA with Tukey’s post hoc test; n = 8–12 for Wild-type, n = 8–12 for PLX5622, n = 9–12 for 5xFAD, n = 8–9 for 5xFAD + PLX5622. Statistical significance is denoted by *p < 0.05, **p < 0.01, and ***p < 0.001. Statistical trends are denoted by ^#p < 0.10. NS, not significant (p > 0.10). Error bars indicate SEM