Mitochondrial RNA in Alzheimer's Disease Circulating Extracellular Vesicles

Abstract

Alzheimer's disease (AD) is the most common type of dementia. Amyloid β (Aβ) plaques, tau-containing neurofibrillary tangles, and neuronal loss leading to brain atrophy are pathologic hallmarks of AD. Given the importance of early diagnosis, extensive efforts have been undertaken to identify diagnostic and prognostic biomarkers for AD. Circulating extracellular vesicles (EVs) provide a platform for "liquid biopsy" biomarkers for AD. Here, we characterized the RNA contents of plasma EVs of age-matched individuals who were cognitively normal (healthy controls (HC)) or had mild cognitive impairment (MCI) due to AD or had mild AD dementia (AD). Using RNA sequencing analysis, we found that mitochondrial (mt)-RNAs, including MT-ND1-6 mRNAs and other protein-coding and non-coding mt-RNAs, were strikingly elevated in plasma EVs of MCI and AD individuals compared with HC. EVs secreted from cultured astrocytes, microglia, and neurons after exposure to toxic conditions relevant to AD pathogenesis (Aβ aggregates and H₂O₂), contained mitochondrial structures (detected by electron microscopy) and mitochondrial RNA and protein. We propose that in the AD brain, toxicity-causing mitochondrial damage results in the packaging of mitochondrial components for export in EVs and further propose that mt-RNAs in plasma EVs can be diagnostic and prognostic biomarkers for MCI and AD.

Keywords: Alzheimer’s disease; biomarker; extracellular vesicles; mitochondrial RNA; mitochondrial protein.

FIGURE 1

Preparation of extracellular vesicles (EVs) for RNA sequencing (RNA-seq) analysis. (A) Subject characteristics of individual human blood donors for RNA-seq analysis. Five samples were collected in each group (HC, healthy donors; MCI, patients with mild cognitive impairment due to AD; AD, patients with mild dementia due to Alzheimer’s disease). Mean Age is 76.2 for healthy donors, 77.6 for MCI, and 76 for AD. (B) Workflow of EV isolation, RNA extraction, and RNA-seq analysis. (C,D) Bioanalyzer chronographs of cDNA amplified by ovation kit (C) and cDNA libraries (D). Data in (C,D) are representative of individuals in each of the 3 groups.

FIGURE 2

RNA-sequencing (RNA-seq) analysis. (A) Pie chart of total transcripts in EVs (for breakdown of noncoding RNA biotypes, see Supplementary Figure S1). (B) Volcano plot of significantly changed transcripts, MCI vs. HC and AD vs. HC. (C) List of mitochondrial RNAs that are found in EVs using RNA-seq analysis: 13 mitochondrial mRNAs, 22 mitochondrial tRNAs (transfer RNAs), and 2 mitochondrial rRNAs (ribosomal RNAs). Changes in abundance were quantified as log₂ fold change (FC) differences in MCI vs. HC and AD vs. HC. (D) UCSC genome browser of mitochondrial chromosome (Chr M) from 5 HC, 5 MCI, and 5 AD persons. Mitochondrial genes were indicated above the Chr M schematic.

FIGURE 3

Validation of mt-RNAs in EVs by RT-qPCR analysis. (A) Subject characteristics of additional individual human blood donors for RNA-sequencing validation. Five samples were collected in each group as in Figure 1A. Mean age is 75.8 years for HC, 66.4 years for MCI, and 74.6 years for AD. (B) RT-qPCR analysis of human mt-RNAs (MT-ND1, MT-ND2, MT-ND3, MT-ND4, MT-ND4L, MT-ND5, MT-ND6, MT-ATP6, MT-ATP8, MT-CYTB, MT-CO1, MT-CO2, MT-CO3 mRNAs, and MT-RNR1 rRNA) and GAPDH mRNA. Data represent all individuals of each group, 5 HC, 5 MCI, and 5 AD.

FIGURE 4

Treatment with Aβ42 and H₂O₂ increases mt-RNAs in EVs. (A) RT-qPCR analysis of mt-RNAs in EVs obtained from human astrocytoma, microglia, and neuroblastoma cells after treatments with Aβ42 oligomers, as indicated (section “Materials and Methods”). (B) RT-qPCR analysis of mt-RNAs in EVs obtained from astrocytoma, microglia, or primary mouse neurons after treatments with H₂O₂ as indicated (section “Materials and Methods”). Data are normalized to GAPDH mRNA values and represent the means ± SEM from three independent experiments. *P < 0.05, **P < 0.01.

FIGURE 5

Treatment of astrocytoma with Aβ42 and H₂O₂ increases the presence of inclusions and mitochondrial proteins in EVs. (A) EVs were isolated from astrocytoma cells treated with 1 μM Aβ42 for 3 days. Representative transmission electron microscopy (TEM) images showing inclusions of structures (arrows) consistent with mitochondria fragments. Black bars, size scale. (B) Western blot analysis of COXIV, OPA1, TOMM20, GAPDH, CD81, and CD63 using whole-cell lysates (‘Cell’) and EVs obtained from astrocytes after the indicated treatments.

FIGURE 6

Model. We propose that in the vicinity of lesions (plaques, tangles) found in persons with mild cognitive impairment (MCI) or Alzheimer’s disease (AD), local damage to brain tissues can lead to the release of EVs containing dysfunctional or damaged mitochondria from cells including microglia, astrocytes, and neurons. As these EVs cross the blood-brain barrier (BBB), the mitochondrial contents (including mitochondrial RNA) present as EV cargo can be measured in peripheral blood. Created in BioRender.