Mitochondrial oxidative stress contributes to the pathological aggregation and accumulation of tau oligomers in Alzheimer's disease

Abstract

Tau oligomers (oTau) are thought to precede neurofibrillary tangle formation and likely represent one of the toxic species in disease. This study addresses whether mitochondrial reactive oxygen species (ROS) contribute to tau oligomer accumulation. First, we determined whether elevated oxidative stress correlates with aggregation of tau oligomers in the brain and platelets of human Alzheimer's disease (AD) patient, tauopathy mice, primary cortical neurons from tau mice and human trans-mitochondrial 'cybrid' (cytoplasmic hybrid) neuronal cells, whose mitochondria are derived from platelets of patients with sporadic AD- or mild cognitive impairment (MCI)-derived mitochondria. Increased formation of tau oligomers correlates with elevated ROS levels in the hippocampi of AD patients and tauopathy mice, AD- and MCI-derived mitochondria and AD and MCI cybrid cells. Furthermore, scavenging ROS by application of mito-TEMPO/2-(2,2,6,6-Tetramethylpiperidin-1-oxyl-4-ylamino)-2-oxoethyl)triphenylphosphonium chloride, a mitochondria-targeted antioxidant, not only inhibits the generation of mitochondrial ROS and rescues mitochondrial respiratory function but also robustly suppresses tau oligomer accumulation in MCI and AD cybrids as well as cortical neurons from tau mice. These studies provide substantial evidence that mitochondria-mediated oxidative stress contributes to tau oligomer formation and accumulation.

Figure 1

Tau oligomers pathology in human Alzheimer’s disease (AD) hippocampus. (A, B) Tau oligomeric complex 1 (TOC1) immunohistochemistry on AD hippocampus. (A) Representative images showed TOC1 immunohistochemistry on hippocampus (I: ND and II: AD; TOC1: green, MAP2: red). (B) Quantification of TOC1 immunohistochemistry was performed with the hippocampus from the indicated hippocampal sections. n = 4 per group. (C–F) The graph presents quantification of immunodot blotting (D and F) for TOC1 normalized to β-actin on hippocampus (C, D) and cerebellum (E, F) of ND and AD brains. n = 7 per group. The representative immunodot blotting for TOC1 from indicated hippocampal (C and D) and cerebellum (E and F) homogenates, and β-actin served as a loading control.

Figure 2

Tau oligomers associate with ROS levels in human AD hippocampus. Quantification of EPR (A) and representative spectra of EPR spectra (B) in the indicated hippocampal homogenous. The peak height in the spectrum indicates levels of ROS. n = 7 per group. (C) Correlation analysis of the relationship between tau oligomer accumulation based on the intensity of immunoreactive oTau dot bots and ROS levels quantified by EPR on hippocampus. n = 14 for the correlation analysis.

Figure 3

Age-dependent Tau oligomers pathology in P301S mutant human tau transgenic mice. (A, B) TOC1 immunohistochemistry on entorhinal cortex and hippocampus from 9-month-old P301S mutant human tau transgenic mice. (A) Representative images showed TOC1 immunohistochemistry on entorhinal cortex (I and II) and hippocampus (III and IV). I, III: nonTg, and II, IV: P301S tau transgenic mice (TOC1: green, MAP2: red). Scale bar = 50 μm. (B) Quantifications of TOC1 immunohistochemistry were performed in the entorhinal cortex (open bar) and hippocampus (black bar) from the indicated mice. n = 5 mice (3 males and 2 females) per group. (C, D) The bar graph presents quantification of immunodot blotting for TOC1 normalized to β-actin on entorhinal cortex (C) and hippocampus (D) of the indicated mice. n = 5 mice (3 males and 2 females) per group. The representative immunodot blotting for TOC1 from indicated entorhinal cortex (C) and hippocampus (D) homogenates, and β-actin served as a loading control.

Figure 4

Tau oligomers associate with ROS levels in P301S mutant human tau transgenic mice. Quantitative data of EPR spectra (A) (up: entorhinal cortex and down: hippocampus) and representative spectra of EPR (B) (left: entorhinal cortex and right: hippocampus) in the indicated mice. The peak height in the spectrum indicates levels of ROS. n = 5 mice (3 males and 2 females) per group. (C, D) Correlation analysis of the relationship between tau oligomer based on the quantification of the intensity of immunoreactive oTau dot bots and ROS levels on entorhinal cortex (C) and hippocampus (D). N = 5 mice (3 males and 2 females) per group.

Figure 5

Effect of Mito-TEMPO on tau oligomer pathology and ROS levels in hippocampal neurons cultured from tau mice in vitro. (A, B) TOC1 immunocytochemistry 21 days in vitro (DIV) hippocampal neurons. (A) Representative images showed TOC1 immunocytochemistry on hippocampal neurons (TOC1: green, MAP2: red). I: nonTg, II: TAU, and III: TAU + TEMPO. The right panels are the enlarged views of TOC1 in the left panels. (B) Quantification of TOC1 immunocytochemistry in (A). n = 8 neurons per group. Scale bars = 25 μm. (C) The representative immunodot blottings for TOC1 were shown from indicated neurons, and β-actin served as a loading control. (D) The bar graph presents quantifications of immunodot blottings for TOC1 normalized to β-actin of the indicated neurons. n = 3 per group. Complex I (E) activities and ATP levels (F) were determined in indicated groups. Data are expressed as fold change relative to the vehicle group (N = 3 independent experiments). Quantification of EPR (G) and representative spectra of EPR spectra (H) in the indicated hippocampal neurons. The peak height in the spectrum indicates levels of ROS. n = 4 per group. (I) Correlation analysis of the relationship between tau oligomer based on the quantification of immunoreactive oTau immunodot blots for TOC1 and ROS levels on hippocampal neurons. n = 9 for the correlation analysis.

Figure 6

Tau oligomer pathology in differentiated AD and MCI trans-mitochondrial cybrid cells and human platelets. (A, B) The bar graph presents quantification of immunodot blotting for TOC1 normalized to β-actin on differentiated MCI and AD trans-mitochondrial cybrid cells (A) and human platelets (B). n = 5 for cybrids and 7 for platelets per group. The representative immunodot blottings were shown for TOC1 from indicated cybrids (A) and platelets (B), and β-actin served as a loading control. (C) The bar graph presents quantification of immunodot blotting for TOC1 normalized to β-actin on differentiated MCI and AD trans-mitochondrial cybrid cells with or without mito-TEMPO treatment. The representative immunodot blottings were shown for TOC1 from indicated cybrids and β-actin served as a loading control.

Figure 7

Tau oligomers associate with ROS levels in differentiated AD and MCI trans-mitochondrial cybrid cells and human platelets. Quantifications of EPR spectra and representative spectra of EPR in the indicated cybrids (A) and platelets (B). The peak height in the spectrum indicates levels of ROS. n = 5 for cybrids and 7 for platelets per group. (C, D) Correlation analysis of the relationship between Tau oligomers based on the quantification of immunodot blots for TOC1 and ROS levels on cybrids (C) and platelets (D). n = 15 for cybrids and 21 for platelets for the correlation analysis. (E) The bar graph presents quantification of immunodot blotting for TOC1 normalized to β-actin on differentiated MCI and AD trans-mitochondrial cybrid cells with or without mito-TEMPO treatment. The representative immunodot blottings were shown below for TOC1 from indicated cybrids and β-actin served as a loading control. n = 5 for cybrids per group.