Amyloid generation and dysfunctional immunoproteasome activation with disease progression in animal model of familial Alzheimer's disease

Abstract

Double-transgenic amyloid precursor protein/presenilin 1 (APP/PS1) mice express a chimeric mouse/human APP bearing the Swedish mutation (Mo/HuAPP695swe) and a mutant human PS1-dE9 both causative of familial Alzheimer's disease (FAD). Transgenic mice show impaired memory and learning performance from the age of 6 months onwards. Double-transgenic APP/PS1 mice express altered APP and PS1 mRNAs and proteins, reduced β-secretase 1 (BACE1) mRNA and normal BACE1 protein, all of which suggest a particular mechanism of amyloidogenesis when compared with sporadic AD. The first β-amyloid plaques in APP/PS1 mice appear at 3 months, and they increase in number and distribution with disease progression in parallel with increased levels of brain soluble β-amyloid 1-42 and 1-40, but also with reduced 1-42/1-40 ratio with age. Amyloid deposition in plaques is accompanied by altered mitochondria and increased oxidative damage, post-translational modifications and accumulation of altered proteins at the dystrophic neurites surrounding plaques. Degradation pathways are also modified with disease progression including activation of the immunoproteasome together with variable alterations of the different protease activities of the ubiquitin-proteasome system. Present observations show modifications in the production of β-amyloid and activation and malfunction of the subcellular degradation pathways that have general implications in the pathogenesis of AD and more particularly in specificities of FAD amyloidogenesis.

Figure 1

A. Memory performance in the V‐maze shows significant differences between APP/PS1 mice and corresponding WT from the age of 6 months onwards. Student's t‐test shows a reduction in the recognition index of APP/PS1 mice at 6, 9 and 12/14 months when compared with the corresponding WT littermates. B. Active avoidance test shows a decrease in the learning performance in APP/PS1 mice when compared with age‐matched WT littermates. Student's t‐test shows a reduction in the conditioned changes of APP/PS1 mice at 6, 9 and 12 months when compared with the corresponding age controls. *P < 0.05, **P < 0.01, ***P < 0.001 compared with age‐matched WT mice (Student's t‐test). #P < 0.05 compared with animals from the same genotype aged 3 months (Tukey's post hoc test).

Figure 2

A–C. APP, PS1 and BACE mRNA expression in the neocortex of male APP/PS1 and WT mice at 3 and 12 months of age. mRNA levels of APP, PS1 and BACE are constant in WT mice aged 3 and 12 months, in contrast to the reduced mRNA levels observed in APP/PS1 mice aged 12 months. Comparing genotypes, APP/PS1 mice aged 3 months exhibit increased expression levels of APP (A) and PS1 (B) and reduced levels of BACE (C) mRNAs when compared with corresponding controls. APP mRNA levels are still higher in APP/PS1 mice aged 12 months but PS1 levels shift below control values. D. Increased APP protein levels are observed in APP/PS1 mice at 3, 6 and 12 months when compared with WT animals. In both genotypes, the APP levels increase with age. E. Two bands (48 and 50 kDa) are detected in the PS1 immunoblots. A significant reduction in 48 kDa PS1 is observed in APP/P1 aged 6 months when compared with WT mice. F. In contrast to BACE mRNA expression, no significant differences related to age or genotype are seen in BACE protein expression levels. *P < 0.05, ***P < 0.001 compared with WT mice (Student's t‐test). ##P < 0.01, ###P < 0.001 compared with APP/PS1 animals aged 3 months (Student's t‐test).

Figure 3

Main neuropathological findings in APP/PS1 mice aged 6 months. β‐amyloid plaques in the neocortex (A), hippocampus (B) and entorhinal cortex (C). Amyloid plaques are surrounded by dystrophic neurites immunoreactive with mitochondrial porin/voltage‐dependent anion channel (VDAC), as revealed with Abcam (D) and Calbiochem (E), antibodies and ubiquitin (F). BACE 1 is also expressed in association with β‐amyloid plaques (G). Hypertrophic astrocytes, as revealed with GFAP antibodies (H) and microglia stained with Lycopericum esculentum lectin (I), are also observed in the vicinity of plaques. Paraffin sections slightly counterstained with hematoxylin. A–G, bar in C = 25 µm; H, I, bar in I = 40 µm.

Figure 4

β‐amyloid plaques increase in number with disease progression. A. Representative images of β‐amyloid plaques in the cortex of APP/PS1 mice aged 6, 12 and 20 months. B.β‐amyloid burden in three cortical regions from three different sections in APP/PS1 mice aged 6, 12 and 20 months. Data are expressed as the mean values ± standard error of the mean (SEM) of three to four animals per age. **P < 0.01, ***P < 0.001 compared with 6 months (Tukey's post hoc test). #P < 0.05 compared with 12 months (Tukey's post hoc test). Data are expressed as the mean ± SEM of three to seven animals per age. **P < 0.01, ***P < 0.001 compared with animals aged 3 months (Tukey's post hoc test). ##P < 0.01 compared with animals aged 6 months (Tukey's post hoc test). C. ELISA quantification of cortical homogenates reveals increased β‐amyloid 1–40 and 1–42 soluble levels because of age in APP/PS1 mice. D. The increase in the ratio between β‐amyloid 1–42 and 1–40 and soluble forms observed in APP/PS1 at 6‐month‐old with respect to 3‐month‐old mice is not found at 12 months (B). Data are expressed as the mean ± SEM of 3–7 animals per age. **P < 0.01, ***P < 0.001 compared with animals aged 3 months (Tukey's post hoc test). ##P < 0.01 compared with animals aged 6 months (Tukey's post hoc test).

Figure 5

A–C. Electron microscopy of β‐amyloid plaques showing a central, radiating core of amyloid (β) surrounded by aberrant huge neurites filled with altered mitochondria, polymorphous inclusions and vesicles. APP/PS1 mice aged 12 months. A–C, bar = 2 µm; D, bar = 0.5 µm. D,F. Immunoelectronmicroscopy showing immunogold particles (arrows) decorating β‐amyloid in the core of a plaque (D) and VDAC in the membrane of mitochondria and polymorphous inclusions in aberrant neurites surrounding β‐amyloid deposits (F). APP/PS1 mice aged 6 months. E, bar = 0.2 µm; F, bar = 0.5 µm.

Figure 6

(A) SAPK/JNK‐P, (B) p38‐P and (C) SOD1 immunoreactivity in cellular processes surrounding β‐amyloid plaques (unstained center) in APP/PS1 mice aged 6 months. Immunoreaction visualized with nickel. D. Confocal microscopy showing HNE immunoreactivity (red) in association with β‐amyloid (green) deposition in plaques. Nuclei are stained in blue with DRAQ5^TM. A–C, bar = 25 µm; D, bar = 10 µm.

Figure 7

Double‐labeling immunofluorescence and confocal microscopy showing altered phospho‐tau deposition (red, B, E, H, K) in association with increased neuroketal immunoreactivity in plaques (green, A, D, G, J). C, F, I, L: merge. Nuclei are stained in blue (DRAQ5). APP/PS1 mice aged 6 and 9 months. A–F, bar = 25 µm; G–L, bar = 25 µm.

Figure 8

Variable amounts of α‐synuclein (B) and α‐synuclein phosphorylated at Ser129 (E, red) accumulate at the periphery of β‐amyloid plaques (A, D, green). Rab3a (G, green) is found in dystrophic neurites surrounding β‐amyloid plaques (H, red). Similarly, APP (J, green) accumulates in the vicinity of amyloid deposits (K, red). Nuclei are stained in blue (DRAQ5). APP/PS1 mice aged 12/14 months. A–F, bar = 40 µm; G–L, bar = 20 µm.

Figure 9

Altered expression of constitutive and inducible proteasome components. A,B. No significant difference in the levels of 19S subunit S1 because age or genotype is observed in WT or APP/PS1 mice, in spite of a tendency to decrease in 12‐month‐old APP/PS1. C,D. Decreased expression of β5 subunit in APP/PS1 mice aged 6 and 12 months when compared with WT. In contrast, β5 subunit expression is higher at 12 months than at 3 months in WT animals. E,F. Protein levels of MECL‐1 are reduced in APP/PS1 animals aged 6 months when compared with younger animals. This reduction is not maintained at 12 months. G,H. Quantification for LMP2 indicates a significant increase in APP/PS1 mice at 12 months compared with WT and APP/PS1 animals aged 3 or 6 months. I,J. The pathology progression with age is associated with an upregulation of the proteasome activator PA28 regulatory complex in APP/PS1 mice. In contrast, PA28 expression is reduced in aged WT animals. K,L. The expression of UCHL‐1 decreases in APP/PS1 mice at 6 and 12 months with respect to control littermates. A,C,E,G,I,K. Representative immunoblots for β5, MECL‐1, LMP2, PA28 and corresponding β‐actin loading control indicating the molecular weight of each protein. B,D,F,H,J,L. Densitometric quantification of β5, MECL‐1, LMP2, PA28 and UCHL‐1 levels with respect to β‐actin. Data are represented as the mean ± SEM of three animals per group. *P < 0.05, **P < 0.01, ***P < 0.001 compared with WT animals (Student's t‐test). #P < 0.05, ##P < 0.01, ###P < 0.001 compared with animals aged 3 months (Tukey's post hoc test). $$P < 0.01 compared with animals aged 6 months (Tukey's post hoc test).

Figure 10

Altered proteasome 26S functionality in APP/PS1 mice. A. Chymotrypsin‐like activity is reduced in APP/PS1 at 6 and 12 months of age when compared with WT mice. However, this activity increases proportionally with age in both genotypes. B. Trypsin‐like activity is increased in APP/PS1 mice at 6 months of age although this increase is not maintained with age. C. Similarly, the increase in the PGPH activity exhibited by APP/PS1 mice at 6 months of age is not observed at 12 months. *P < 0.05, **P < 0.01, ***P < 0.001 compared with WT animals (Student's t‐test). #P < 0.05, ##P < 0.01 compared with animals aged 6 months (Student's t‐test).