HDAC2 dysregulation in the nucleus basalis of Meynert during the progression of Alzheimer's disease

Abstract

Aims: Alzheimer's disease (AD) is characterized by degeneration of cholinergic basal forebrain (CBF) neurons in the nucleus basalis of Meynert (nbM), which provides the major cholinergic input to the cortical mantle and is related to cognitive decline in patients with AD. Cortical histone deacetylase (HDAC) dysregulation has been associated with neuronal degeneration during AD progression. However, whether HDAC alterations play a role in CBF degeneration during AD onset is unknown. We investigated global HDAC protein levels and nuclear HDAC2 immunoreactivity in tissue containing the nbM, changes and their association with neurofibrillary tangles (NFTs) during the progression of AD.

Methods: We used semi-quantitative western blotting and immunohistochemistry to evaluate HDAC and sirtuin (SIRT) levels in individuals that died with a premortem clinical diagnosis of no cognitive impairment (NCI), mild cognitive impairment (MCI), mild/moderate AD (mAD) or severe AD (sAD). Quantitative immunohistochemistry was used to identify HDAC2 protein levels in individual cholinergic nbM nuclei and their colocalization with the early phosphorylated tau marker AT8, the late-stage apoptotic tau marker TauC3 and Thioflavin-S, a marker of β-pleated sheet structures in NFTs.

Results: In AD patients, HDAC2 protein levels were dysregulated in the basal forebrain region containing cholinergic neurons of the nbM. HDAC2 nuclear immunoreactivity was reduced in individual cholinergic nbM neurons across disease stages. HDAC2 nuclear reactivity correlated with multiple cognitive domains and with NFT formation.

Conclusions: These findings suggest that HDAC2 dysregulation contributes to cholinergic nbM neuronal dysfunction, NFT pathology, and cognitive decline during clinical progression of AD.

Keywords: Alzheimer's disease; basal forebrain; epigenetics; mild cognitive impairment; neurofibrillary tangles; nucleus basalis of Meynert.

Figure 1.

Representative immunoblots and box plots of levels of (A) HDAC1, (B) HDAC2, (C) HDAC3, and (D) HDAC4 in the basal forebrains of non-cognitively impaired (NCI), mild cognitively impaired (MCI), mild/moderate AD (mAD), and severe AD (sAD). Immunoreactive signals obtained by densitometry were normalized to levels of β-tubulin. (A) HDAC1 levels were stable across clinical groups. (B) HDAC2 levels were significantly decreased in the sAD group compared with those in the NCI group (p = 0.03). (C) HDAC3 and (D) HDAC4 levels did not change significantly across groups. Circles in box plots indicate outliers. Asterisk indicates p < 0.05.

Figure 2.

Representative immunoblots and box plots of (A) HDAC6, (B) SIRT1, and (C) ChAT basal forebrain levels in the non-cognitively impaired (NCI), mild cognitive impairment (MCI), mild/moderate AD (mAD), and severe AD (sAD) groups. Immunoreactive signals obtained by densitometry were normalized to levels of β-tubulin. HDAC6 (A, p = 0.46) and SIRT1 (B, p = 0.20) levels were stable across clinical groups. (C) ChAT levels were significantly decreased in the sAD group compared with those in the NCI (p < 0.001) and MCI (p = 0.04) groups. Circles in box plots indicate outliers. One asterisk indicates p < 0.05. Two asterisks indicate p < 0.001.

Figure 3.

(A-D) Photomicrographs of sections stained for HDAC2 and cresyl violet, and (E and F) box plots of HDAC2 intensity and nuclear area in non-cognitively impaired (NCI), mild cognitive impairment (MCI), mild/moderate AD (mAD), and severe AD (sAD) groups. Nuclei positive for HDAC2 lost their rounded shape (A) and were displaced to the periphery of the soma in the MCI, mAD, and sAD groups (B, C, and D). (E) HDAC2-ir was significantly decreased in mAD compared to NCI and MCI (p < 0.001), and sAD compared to NCI, MCI, and mAD (p < 0.001). (F) The area of the nuclei positive for HDAC2-ir was significantly decreased in mAD compared to NCI (p = 0.03) and in sAD compared to NCI (p < 0.001), MCI (p < 0.001), and mAD (p = 0.04) groups. Circles in box plots indicate outliers. One asterisk indicates p < 0.05; two asterisks indicate a p < 0.001. Scale bars: 10μm.

Figure 4.

Photomicrographs of nbM tissue dual immunostained for p75^NTR (brown) and AT8 (dark blue/black) in (A) NCI, (B) MCI, (C) mAD, and (D) sAD groups. Note the presence of more p75^NTR/AT8-positive (closed arrow) and AT8-positive perikarya (open arrow) in (B) MCI, (C) mAD, and (D) sAD. The percentage of p75^NTR-positive neurons (red circles, E) was significantly greater in the NCI and MCI groups than in the mAD and sAD groups (p < 0.001). The percentage of AT8-positive perikarya (green circles, E) increased significantly in the mAD and sAD groups (p < 0.01) compared to the NCI and MCI groups. No significant differences in the percentage of p75^NTR/AT8-positive neurons (blue triangle, E) were detected across the groups (p = 0.05). One asterisk indicates p < 0.05; two asterisks indicate a p < 0.001. Scale bars: 50 μm, 10 μm for inset.

Figure 5.

(A-T) Fluorescent photomicrographs of p75^NTR-ir nbM tissue stained for HDAC2 (cyan), p75^NTR (green), AT8 (red), TauC3 (red), and Thioflavin-S (blue). HDAC2-ir decreases in nbM neurons labelled for p75^NTR and AT8 from (A) NCI, (F) MCI, to (K) mAD, and (P) sAD (arrow identifies HDAC2-ir nucleus shown in colour inset). Nuclear HDAC2-ir declines in p75^NTR-positive neurons labelled for AT8/HDAC2 and HDAC2 alone (arrow identifies same nucleus in black and white images) in NCI (B, C), MCI (G, H), mAD (L, M), and sAD (Q, R) and in nbM neurons triple-labelled for p75^NTR/HDAC2/TauC3, and HDAC2 alone in NCI (D, E), MCI (I, J), mAD (N, O), and sAD (S, T). Note displacement and shrinkage of nuclei positive for HDAC2-ir in tangle-bearing cholinergic nbM neurons. Scale bars: 50 μm (A, F, K, P); 10 μm (B-E, G-J, L-O, Q-T).