Impact of Cerebral Amyloid Angiopathy in Two Transgenic Mouse Models of Cerebral β-Amyloidosis: A Neuropathological Study

Abstract

The pathological accumulation of parenchymal and vascular amyloid-beta (Aβ) are the main hallmarks of Alzheimer's disease (AD) and Cerebral Amyloid Angiopathy (CAA), respectively. Emerging evidence raises an important contribution of vascular dysfunction in AD pathology that could partially explain the failure of anti-Aβ therapies in this field. Transgenic mice models of cerebral β-amyloidosis are essential to a better understanding of the mechanisms underlying amyloid accumulation in the cerebrovasculature and its interactions with neuritic plaque deposition. Here, our main objective was to evaluate the progression of both parenchymal and vascular deposition in APP23 and 5xFAD transgenic mice in relation to age and sex. We first showed a significant age-dependent accumulation of extracellular Aβ deposits in both transgenic models, with a greater increase in APP23 females. We confirmed that CAA pathology was more prominent in the APP23 mice, demonstrating a higher progression of Aβ-positive vessels with age, but not linked to sex, and detecting a pronounced burden of cerebral microbleeds (cMBs) by magnetic resonance imaging (MRI). In contrast, 5xFAD mice did not present CAA, as shown by the negligible Aβ presence in cerebral vessels and the occurrence of occasional cMBs comparable to WT mice. In conclusion, the APP23 mouse model is an interesting tool to study the overlap between vascular and parenchymal Aβ deposition and to evaluate future disease-modifying therapy before its translation to the clinic.

Keywords: 5xFAD; APP23; cerebral microbleeds; cerebral β-amyloidosis; preclinical MRI.

Figure 1

Extracellular Aβ deposition in 5xFAD and APP23 mice. (A) Representative images of sagittal brain sections from female 5xFAD, APP23, and WT mice showing Aβ-positive fibrillar deposits stained with Thioflavin-S (ThS). Scale bar indicates 2 mm and 200 μm; m.o: months old. (B) Representation of the number of parenchymal Aβ deposits in male and female 5xFAD mice and APP23 mice at different ages. Data are expressed as the number of ThS-positive deposits (fibrillar Aβ deposits) per area. n = 4–7/group. Statistical differences were analyzed between APP23 and 5xFAD mice (females + males) and represented as: * p < 0.05, **** p < 0.0001.

Figure 2

Quantification of parenchymal Aβ burden in brains from male and female 5xFAD and APP23 mice at different ages. (A) Quantification of the number of ThS-positive deposits corrected by area. (B) Quantification of the average size of ThS-positive deposits. (C) Quantification of total area occupied by the deposits (%). n = 4–7/group. The effect of sex (S) and age (A) was evaluated in each subgroup of data. Statistical differences were analyzed among groups as indicated and represented as: # p < 0.1, * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001.

Figure 3

Vascular Aβ deposition in 5xFAD and APP23 mice. (A) Representative images of resorufin staining in 16-month-old WT, 5xFAD, and APP23 female mice. Scale bar indicates 200 μm and 20 μm, respectively. (B) Quantification of the number of Aβ-positive vessels identified by resorufin staining in APP23 mice. Data are expressed as the number of Aβ-positive vessels per area (pixels²). n = 4–7/group. The effect of sex (S) and age (A) was evaluated in each subgroup of data. Statistical differences were analyzed among groups as indicated and represented as: # p < 0.1, **** p < 0.0001.

Figure 4

Specific Aβ40 and Aβ42 immunodetection in the parenchyma and cerebral vasculature in APP23 and 5xFAD mice. Total Aβ (determined by anti-4G8), Aβ40, and Aβ42 immunodetection analyzed in brain sections of: (A) 16-month-old female APP23:and, (B) 5xFAD (mice. Zoom images from leptomeningeal vessels (green), cortical vessels (cyan), and amyloid plaques (dark blue) are shown in A2–A10 panels for the APP23 model and in B2-B10 panels for the 5xFAD model. A2–A4 and B2–B4 represent consecutive brain sections stained with anti-4G8 primary antibody; A5–A7 and B5–B7 represent consecutive brain sections stained with anti-Aβ40 primary antibody; and A8–A10 and B8–B10 represent consecutive brain sections stained with anti-Aβ42 primary antibody. Red arrows in A1 indicate amyloid-affected capillaries in APP23 brains. Black arrows in A4–A10 and B4–B10 indicate the same amyloid-plaques analyzed in consecutive sections using different primary antibodies. Scale bar in A1/B1 indicates 200 μm and in A2–A10 and B2–B10 indicates 40 μm.

Figure 5

MRI detection of CAA-related cerebral microbleeds in 5xFAD and APP23 mice. (A) Representative T2* magnetic resonance images (MRI) of cerebral microbleeds (cMBs) in 20-month-old male WT, 5xFAD, and APP23 mice. cMBs are indicated with red arrows. Scale bar indicates 2 mm. (B) Distribution of cMBs in WT, 5xFAD, and APP23 mice, according to the percentage of individuals affected. N = 4–7/group. (C) Comparison of cMBs in APP23 mice T2* sequences and Prussian blue staining showing iron hemosiderin deposits. Scale bar indicates 2 mm and 40 μm.