Molecular characterization of extracellular vesicles for the spread of misfolded tau protein

Neurofibrillary tangles, composed of intracellular aggregates of hyperphosphorylated microtubule-associated protein tau (tau), are by far the most correlated pathology for clinical symptoms of Alzheimer disease (AD). Emerging evidence suggests that extracellular vesicles (EVs, such as exosomes and microvesicles), transfer pathological tau between cells as vehicles, propagating tau pathology. It is urgently important to find the molecular basis of brain-derived EVs, which may critically regulate EV uptake by neurons and aggregation of tau protein in EVs and/or recipient neurons. We have recently established the method for isolating EVs from human brain samples and successfully performed their proteomic profiling. We found that selective molecules from the EV proteomics datasets were able to differentiate human AD-EV from healthy control (CTRL)-EV with 88% accuracy by machine learning analysis, confirming pathogenic character of AD-EV molecules. Furthermore, our exciting preliminary data have shown that AD-EV have significantly higher tau seeding activity compared to CTRL-EV by FRET sensor tau seeding assay with subsets of EV molecules showing significant association with tau seeding activity. This proposed project will fortify these preliminary results and find the converging or specific mechanisms among tauopathies for mediating tau aggregation and its seeding via EV uptake through proteomics and biological examination of brain-derived EV samples. To meet this challenge, we assembled a multi-disciplinary team of investigators who have a strong record of accomplishments in biologic (Ikezu), proteomic (Emili) and bioengineering and bioinformatic analysis (Issador). In Aim1, we will examine EV samples from 240 new brain specimens (40 AD, 40 CTE, 40 LBD, 40 PSP, 40 CBD and 40 CTRL) for precision mass-spectrometry-based proteomics and tau-interactomes, and analyze those datasets by the machine learning approach. Aim 2 will examine the efficiency of tau propagation using tau fibrils, oligomers and EVs isolated from the same donors of the 5 different tauopathies and control cases in vitro and in vivo using FRET-based tau seeding assay and EV uptake by primary cultured mouse cortical neurons in vitro. EV-associated tau will be further characterized by the biochemical and microscopy-based analysis for their conformational and posttranslational changes. We will evaluate the difference in tau propagation after the intracranial injection of the tau seeds from different tauopathy brains using our recently established mouse models. Aim3 will identify candidate molecules most likely involved in EV uptake and tau seeding activity by bioinformatic analysis of the proteome dataset (Aim 1) and biological datasets (Aim 2). We will then test the functional roles of the identified molecules on EV uptake, tau seeding activities and neuronal firing activities in vitro. The candidate molecules will be specifically targeted by gene silencing or antagonists for their therapeutic potential to halt tau propagation in vitro and in vivo. Successful identification of responsible molecules for tau propagation will serve as a foundation for understanding EV-mediated disease progression.