Age, but Not Amyloidosis, Induced Changes in Global Levels of Histone Modifications in Susceptible and Disease-Resistant Neurons in Alzheimer's Disease Model Mice

Abstract

There is increasing interest in the role of epigenetic alterations in Alzheimer's disease (AD). The epigenome of every cell type is distinct, yet data regarding epigenetic change in specific cell types in aging and AD is limited. We investigated histone tail modifications in neuronal subtypes in wild-type and APP/PS1 mice at 3 (pre-pathology), 6 (pathology-onset) and 12 (pathology-rich) months of age. In neurofilament (NF)-positive pyramidal neurons (vulnerable to AD pathology), and in calretinin-labeled interneurons (resistant to AD pathology) there were no global alterations in histone 3 lysine 4 trimethylation (H3K4me3), histone 3 lysine 27 acetylation (H3K27ac) or histone 3 lysine 27 trimethylation (H3K27me3) in APP/PS1 compared to wild-type mice at any age. Interestingly, age-related changes in the presence of H3K27ac and H3K27me3 were detected in NF-labeled pyramidal neurons and calretinin-positive interneurons, respectively. These data suggest that the global levels of histone modifications change with age, whilst amyloid plaque deposition and its sequelae do not result in global alterations of H3K4me3, H3K27ac and H3K27me3 in NF-positive pyramidal neurons or calretinin-labeled interneurons.

Keywords: H3K27me3; H3K4me3; H4K27ac; calretinin; epigenetics; neurofilament triplet proteins.

Figure 1

No difference in the percentage of nuclei immunolabeled with H3K4me3, H3K27me3 or H3K27ac in layers 1-6 of the cortex in APP/PS1 mice compared to wild-type mice. Representative images of coronal sections stained with DAPI (blue, A,C,D,F,G,I) and immunolabeled for H3K4me3 (red, B,C), H3K27ac (red, E,F) and H3K27me3 (red, H,I) in a 3 month old APP/PS1 mouse (A–C), a 12 month old APP/PS1 mouse (D–F) and a 3 month old wild type (WT) mouse (G–I). (J) Bar graph displaying the density of DAPI stained nuclei in layers 1-6 of 3-month (3M), 6-month (6M) and 12-month (12M) old WT and APP/PS1 transgenic (TG) mice. Bar graphs showing the percentage of nuclei co-localized with H3K4me3 (K), H3K27ac (L) and H3K27me3 (M) at 3, 6 and 12 months of age in APP/PS1 and wild-type mice. All data presented as mean ± standard error of the mean (SEM). Scale bar = 400 μm.

Figure 2

No change in the proportion of nuclei in layer 2/3 of the cortex co-localized with H3K4me3, H3K27me3 or H3K27ac in APP/PS1 mice vs. wild-type mice. Example images of areas layer 2/3 stained for DAPI (blue, A,C,D,F,G,I) and immunolabeled for H3K4me3 (red, B,C), H3K27ac (red, E,F) and H3K27me3 (red, H,I) in a 6 month old WT mouse (A–C), a 6 month old APP/PS1 mouse (D–F) and a 12 month old WT mouse (G–I). Note nuclei positive (arrows) and negative (arrowheads) for histone mark immunoreactivity. (J) Bar graph displaying the density of nuclei in layer 2/3 of 3-month (3M), 6-month (6M) and 12-month (12M) WT mice and TG APP/PS1 mice. Bar graphs showing the percentage of nuclei immunoreactive for H3K4me3 (K), H3K27ac (L) and H3K27ac (M). All data presented as mean ± SEM. Scale bar =50 μm.

Figure 3

Nuclear H3K4me3, H3K27me3 and H3K27ac immunolabeling was observed in cortical layer 2/3 calretinin-positive interneurons and neurofilament (NF)-labeled pyramidal neurons. Representative images of layer 2/3 in a 12 month old WT mouse (A–D) showing NF-labeled pyramidal neurons (A,D green) co-localized (arrows) with DAPI staining (B,D blue) and H3K27ac immunoreactivity (C,D red). An example image of layer 2/3 in a 3 month old APP/PS1 mouse (E–H) showing NF-labeled pyramidal neurons (E,H, green) co-localized (arrows) with DAPI staining (F,H, blue) and H3K27me3 labeling (G,H, red). A representative image layer 2/3 in a 3 month old wild-type mouse (I–L) showing NF-labeled pyramidal neurons (I,L, green) co-localized (arrows) with DAPI (J,L, blue) and H3K4me3 immunoreactivity (K,L, red). An example image of layer 2/3 of a 3 month old APP/PS1 cortex (M–P) with calretinin-labeled interneurons (M,P green) co-localized (arrows) with DAPI staining (N,P blue) and H3K27ac immunoreactivity (O,P red). Representative image of layer 2/3 of a 3 month old APP/PS1 mouse (Q–T) illustrating calretinin-positive interneurons (Q,T green) containing (arrows) DAPI staining (R,T blue) and H3K27me3 labeling (S,T red). An example region of layer 2/3 of the cortex of a 6 month old wild-type mouse (U–X) showing calretinin-labeled interneurons (U,X green) co-localized (arrows) with DAPI staining (V,X blue) and H3K4me3 immunopositive nuclei (W,X red). Scale bar = 18 μm.

Figure 4

No difference in the percentage of layer 2/3 NF-labeled pyramidal neurons co-localized with H3K4me3, H3K27me3 or H3K27ac in APP/PS1 vs. wild-type mice. Representative images of layer 2/3 NF-positive pyramidal neurons (SMI-32, green) stained with DAPI (blue) and labeled for H3K27ac (red, arrows) in 3- (A,D), 6- (B,E) and 12- (C,F) month-old wild-type (A–C) and APP/PS1 (D–F) mice. Bar graphs showing the percentage of NF-positive neuronal nuclei co-localized with H3K27ac (G), H3K27me3 (H) and H3K4me3 (I) immunolabeling. All data are presented as mean ± SEM. ^*p < 0.05. Scale bars: 25 μm.

Figure 5

No difference in the percentage of layer 2/3 calretinin-positive interneurons co-localized with H3K27me3, H3K4me3 or H3K27ac in APP/PS1 vs. wild-type mice. Representative images of layer 2/3 calretinin-positive interneurons (green), stained with DAPI (blue) and labeled with H3K27me3 (red, arrows) in 3- (A,D), 6- (B,E) and 12- (C,F) month-old wild-type (A–C) and APP/PS1 (D–F) mice. Bar graphs showing the percentage of calretinin-labeled interneurons co-localized with H3K27me3 (G), H3K4me3 (H) and H3K27ac (I). All data are presented as mean ± SEM. ^*p < 0.05. Scale bars 25 μm.