Functional Analysis of Transposon Activation in Alzheimer's Disease

Project: Research

View graph of relations

Description

There are an increasing number of data suggesting a large scale of transcriptional activation of transposable elements (TE) in the progression of Alzheimer’s disease (AD), likely followed by changes in the epigenetic landscape induced by the accumulation of the pathogenic Tau proteins. The TE activation can lead to genomic insertion mutations and promote genotoxicity such as DNA damages or genome instability, or double stranded RNAs that can trigger inflammatory responses. It is therefore highly plausible that the TE activation plays a critical role in the AD pathogenesis. Here we plan to investigate the cellular perturbations and the molecular mechanisms followed by the TE activation. Specifically, we will test three hypotheses, including (1) the transcription of TE generates double-stranded RNAs (dsRNAs), alerts the antiviral surveillance machinery and triggers inflammatory responses, (2) the transcription of active retrotransposons leads to genomic mobile element insertions (MEIs), precipitates DNA rearrangement and promotes further genotoxicity, and (3) TE-encoded proteins induces the genomic insertion of messenger RNAs, and leads to the overexpression and mutations of genes such as the amyloid precursor protein (APP). My lab will exploit a brain organoid model, where induced pluripotent stem cell (iPSC)-derived neural stem cells divide, differentiate and mature in a 3-D environment, generating Tau pathology that is highly relevant to the TE activation. We will study three different developmental stages (day 20, day 110 and day 220) of three pairs of AD and control organoids, by measuring (1) transcriptome changes of TE subfamilies and the inflammatory pathways via long-read RNAseq, (2) TE-induced dsRNAs via a targeted approach, dsRNA-seq, and (3) the somatic mutations via single-cell whole genome sequencing. Notably, TE sequences are highly repetitive and whole genome amplification can create false positive insertions, and therefore we expect the detection of somatic MEIs to have a low signalto- noise ratio. We have previously established a computational pipeline that can correct these technical artifacts, and plan to further improve the machine learning method, by adding 60 times more training data and new machine learning models. With the new computational and experimental investigations, our work will provide direct answers to the mechanisms and functional consequences of the TE activation in AD, which can open new possible therapeutic interventions through alleviating the TE-related stress, and shed light to the general pathological impacts of transposons in human diseases. 

Detail(s)

Project number9048238
Grant typeECS
StatusNot started
Effective start/end date1/01/23 → …