Tau depletion diminishes vascular amyloid-related deficits in a mouse model of cerebral amyloid angiopathy

Abstract

Introduction: Tau is essential for amyloid beta (Aβ)-induced synaptic and cognitive deficits in Alzheimer's disease (AD), making its downregulation a therapeutic target. Cerebral amyloid angiopathy (CAA), a major vascular contributor to cognitive decline, affects over 90% of patients with AD. This study explores the impact of tau downregulation on CAA pathogenesis.

Methods: We crossed the Familial Danish Dementia mouse model (Tg-FDD), which develops vascular amyloid, with tau-null (mTau^-/-) mice to generate a CAA model lacking endogenous tau (Tg-FDD/mTau^-/-). Behavioral, electrophysiological, histological, and transcriptomic analyses were performed.

Results: Tau depletion ameliorated motor and synaptic impairments, reduced vascular amyloid deposition, and prevented vascular damage. Tau ablation also mitigated astrocytic reactivity and neuroinflammation associated with vascular amyloid accumulation.

Conclusion: These findings provide the first in vivo evidence of the beneficial effects of tau downregulation in a CAA mouse model, supporting tau reduction as a potential therapeutic strategy for patients with parenchymal and vascular amyloid deposition.

Highlights: Tau ablation improves motor function and synaptic impair, reduces cerebrovascular amyloid deposits, and prevents vascular damage in a mouse model of cerebral amyloid angiopathy (CAA). Tau reduction decreases astrocytic reactivity, alters neuroinflammatory gene expression, and enhances oligodendrocyte function, suggesting a protective role against neuroinflammation in CAA. These findings highlight tau reduction as a potential therapeutic strategy to mitigate CAA-induced pathogenesis, with implications for treating patients with both parenchymal and vascular amyloid deposition.

Keywords: cerebral amyloid angiopathy; neuroinflammation; tau; vascular amyloid.

FIGURE 1

Reducing endogenous tau ameliorates motor impairment in a mouse model of CAA. Tg‐FDD mice present lower grip strength compared to wild‐type mice, whereas Tg‐FDD/mTau^−/− mice present recovery of motor function in the (A) 2‐paw Grip Strength and in the (B) 4‐paw Grip Strength. Data are shown as mean ± SEM (females, inverted triangles; males, squares). Selected pairwise comparisons were assessed using one‐way ANOVA followed by Sidak's multiple comparison test. Wild‐type n = 29, Tg‐FDD n = 33, Tg‐FDD/mTau^−/− n = 15, mTau^−/− n = 17. ANOVA: analysis of variance; CAA, Cerebral amyloid angiopathy; SEM, Standard error of the mean.

FIGURE 2

Total ablation of tau decreases vascular amyloid in the Tg‐FDD mice. (A) Immunofluorescence against alpha smooth muscle actin (αSMA‐red) and thioflavin‐S staining (green) showing vascular amyloid deposition in the brain cortex of Tg‐FDD and Tg‐FDD/mTau^−/− mice. Scale bar: 20 µm. (B) Quantification of percentage of thioflavin‐S surrounding area in the perivasculature. Data are shown as mean ± SEM (females, inverted triangles; males, squares). Shapiro–Wilk Normality Test and Mann–Whitney test were performed. Tg‐FDD n = 15, Tg‐FDD/mTau^−/− n = 13.

FIGURE 3

Tau ablation reduces fibrinogen deposition in the Tg‐FDD mice. (A) Immunofluorescence against fibrinogen (cyan) and αSMA (red) in the brain cortex of wild‐type, Tg‐FDD, Tg‐FDD/mTau^−/−, and mTau^−/− mice. Scale bar: 20 µm. (B) Quantification of the percentage of positive area for fibrinogen in cortex. Data are shown as mean ± SEM (females, inverted triangles; males, squares), One‐way ANOVA Tukey's multiple comparisons test. Wild‐type n = 11, Tg‐FDD n = 12, Tg‐FDD/mTau^−/‐ n = 11, mTau^−/− n = 12).

FIGURE 4

Tau ablation affects astrocytic but not microglia's morphology. (A) Representative Imaris 3D reconstructions from astrocytes stained with GFAP. Individual astrocytes were selected, isolated, and skeletonized for analysis from brain cortex of wild‐type, Tg‐FDD, Tg‐FDD/mTau^−/−, and mTau^−/− mice. (B–C) Quantification of the average of astrocytic number of branches (B) and Scholl intersections (C). Data are shown as mean ± SEM, one‐way Kruskal–Wallis test followed by Dunn's multiple comparisons test. n = 11–13 animals per condition; an average of 40 astrocytes per animal were analyzed. (D) Representative Imaris 3D reconstructions from microglia stained with IBA1. Individual microglia were selected, isolated, and skeletonized for analysis from brain cortex of wild‐type, Tg‐FDD, Tg‐FDD/mTau^−/−, and mTau^−/− mice. (E–F) Quantification of the average of microglial number of branches (E) and Scholl intersections (F). Data are shown as mean ± SEM, One‐way Kruskal–Wallis test followed by Dunn's multiple comparisons test. n = 10–12 animals per condition; an average of 50 microglia per animal were analyzed (females, inverted triangles; males, squares).

FIGURE 5

Tau deletion affects the neuroinflammatory and oligodendrocyte response associated with vascular amyloid deposition in Tg‐FDD male mice. Total mRNA was isolated from the brain cortex of wild‐type, Tg‐FDD, Tg‐FDD/mTau^−/−, and mTau^−/− mice to perform the NanoString neuroinflammation panel. (A) Volcano plot of the DEGs between Tg‐FDD and wild‐type mice. (B) Volcano plot of the DEGs between Tg‐FDD and Tg‐FDD/mTau^−/− mice. Discontinued lines parallel to y indicates a fold change over 1.2 (log₂(FC) = ± 0.3) and discontinued lines parallel to x indicates p value of .05 (–log₁₀ (p‐value) = 1.3). Transcripts in green, red, and blue denote the genes up‐ or downregulated with the largest log₂ fold change belonging to the microglia function, oligodendrocyte function, and cytokine signaling, respectively. One gene in dark cyan was classified as both microglia function and cytokine signaling. (C) Heatmap showing changes between wild‐type, Tg‐FDD, Tg‐FDD/mTau^−/−, and mTau^−/− mice NanoString annotations. (D) Normalized counts of DEGs related to microglia function, cytokine signaling, and oligodendrocyte function that were consistently altered by CAA or tau depletion. Significance was determined by one‐way ANOVA, n.s. followed by Tukey's multiple comparisons. p‐value > .05. n = 5−6. (E) Network String analysis of the annotated genes highlighted in (D). FC: Fold Change.

FIGURE 6

Ablation of endogenous tau prevents the reduction of microglial CSF3R levels in the cortex of Tg‐FDD mice. (A) Immunofluorescence against CSF3R (red) and the nuclear staining NucBlue (blue) in the cortex of wild‐type, Tg‐FDD, Tg‐FDD/mTau^−/−, and mTau^−/− mice. Scale bar: 30 µm. (B) Quantification of CSF3R immunoreactivity. Data are shown as mean ± SEM, one‐way ANOVA analysis of variance followed by Sidak's multiple comparison test. n = 12, per condition. (C) Immunofluorescence against CSF3R (red) and IBA1 (green) and NucBlue (blue). Scale bar: 20 µm. (D) Quantification of CSF3R immunoreactivity within the IBA1‐positive area. Data are shown as mean + SEM, one‐way Kruskal−Wallis test followed by Dunn's multiple comparisons test. n = 12, per condition (females, inverted triangles; males, squares).