Forebrain microglia from wild-type but not adult 5xFAD mice prevent amyloid-β plaque formation in organotypic hippocampal slice cultures

Abstract

The role of microglia in amyloid-β (Aβ) deposition is controversial. In the present study, an organotypic hippocampal slice culture (OHSC) system with an in vivo-like microglial-neuronal environment was used to investigate the potential contribution of microglia to Aβ plaque formation. We found that microglia ingested Aβ, thereby preventing plaque formation in OHSCs. Conversely, Aβ deposits formed rapidly in microglia-free wild-type slices. The capacity to prevent Aβ plaque formation was absent in forebrain microglia from young adult but not juvenile 5xFamilial Alzheimer's disease (FAD) mice. Since no loss of Aβ clearance capacity was observed in both wild-type and cerebellar microglia from 5xFAD animals, the high Aβ1-42 burden in the forebrain of 5xFAD animals likely underlies the exhaustion of microglial Aβ clearance capacity. These data may therefore explain why Aβ plaque formation has never been described in wild-type mice, and point to a beneficial role of microglia in AD pathology. We also describe a new method to study Aβ plaque formation in a cell culture setting.

Figure 1. Microglia incorporate synthetic amyloid-β and prevent plaque formation.

(A,B,D,E) Iba1/DAPI (red/blue) staining on wild-type OHSC (dg = dentate gyrus; ca1/ca3 = cornu ammonis 1/3) following treatment with FAM-labeled Aβ₁₋₄₂ indicates a lack of plaque formation in microglia-containing slices (A). In contrast, microglia depletion resulted in an increase of Aβ plaques (B). Higher magnification reveals plaque-like structures (insert in (B)). Thioflavin S staining (C) and Aβ immunohistochemistry (D) confirmed the formation of Aβ plaques. (E) Confocal analyses of OHSC demonstrate the microglial uptake of Aβ (arrowhead). (F) Replenishment of microglia-depleted OHSC with primary forebrain microglia significantly reduces the Aβ plaque-load. Similar results were obtained in at least 4 independent experiments (G) Quantification of Aβ positive material using Image J revealed reduced Aβ-positive area fraction (0.903 ± 0.138%) in replenished slices compared to microglia-free slices (1.756 ± 0.318%) (Data are expressed as mean ± SEM from n = 2 experiments with 11 slices/group; Mann-Whitney U; *p = 0.038). (H) Quantitative Western immunoblot analysis of Aβ fragments using 6E10 antibody and densitometric analyses revealed significantly higher Aβ levels in microglia-free slices (189.31 ± 7.39%) compared to microglia-containing tissue (100.00 ± 14.11%; Student’s t test; p = 0.001). Scale bars: A, B, C, D, F: 100 μm; E: 10 μm.

Figure 2. Ultrastructural analyses reveals ingestion of amyloid-β to the lysosomal compartment of microglia.

(A,B) Immuno-electron microcroscopy using Iba1-DAB-staining as a microglial marker and immunogold-labeling for 6E10 in wild-type OHSC following treatment with synthetic amyloid-β (Aβ). (A) A high magnification electron micrograph revealed incorporation of Aβ into the lysosomal compartment (ly) of microglia (mg; white line) (B). In the absence of microglia Aβ is detected in pyramidal cells (pc, white line) showing signs of nuclear degeneration (n, nucleus) and in forms of extracellular plaques (asterisks). Scale bars: A: 2,500 nm B: 500 nm.

Figure 3. Forebrain microglia from adult FAD mice do not prevent Aβ-plaque formation.

Upper Panel Low magnification images of microglia-depleted wild-type OHSC following replenishment with either 5-weeks-old (A,C) or 6-months-old forebrain microglia (B,D) from wild-type (A,B) or 5xFAD mice (C,D). Replenished microglia (Iba1/red) were evenly distributed throughout the slice (neuronal staining by NeuN/blue). Thioflavin S staining (green) revealed plaque-like structures ((D) insert) only in OHSCs replenished with microglia from adult 5xFAD mice. Scale bar: 100 μm Lower Panel High magnification images of replenished cerebellar microglia (Iba1/red) from either 5-weeks-old (E,G) or 6-months-old (F,H) wild-type (E,F) or 5xFAD mice (G,H). Replenished microglia (Iba1/red) were evenly distributed throughout the slice and acquired a ramified morphology (E–H). Scale bar: 10 μm. Densitometric analysis of ThioflavinS-positive material in OHSCs was performed by Image J. ANOVA followed by Scheffé post-hoc test revealed that wild-type juvenile (0.331 ± 0.036%, n = 35, p < 0.001) or adult forebrain microglia (0.393 ± 0.043%, n = 33, p < 0.001) impeded plaque formation compared to endogenous wild-type microglia (0.178 ± 0.017%, n = 33, p < 0.001). No significant difference was found between microglia-free OHSCs (0.892 ± 0.106%, n = 30) and slices replenished with forebrain microglia from adult 5xFAD mice (0.750 ± 0.089%, n = 20), while forebrain microglia from juvenile 5xFAD mice prevented plaque formation (0.318 ± 0.033%, n = 31, p < 0.001) compared to endogenous wild-type microglia. Data are expressed as mean ± SEM from three independent experiments (I). In contrast one way ANOVA revealed no differences in densitometric analysis of ThioflavinS-positive material between microglia cells isolated from the cerebellum of juvenile or young adult wild type or 5xFAD mice. Number of OHSC analysed: juv.wt n = 18, ad.wt n = 11, juv.FAD n = 14, ad.FAD n = 16, data are expressed as mean ± SEM from two independent experiments (J).