Untangling the diversity in the genetic architecture of late-onset Alzheimer's disease using single cell multi-omics

点击翻译按钮获取中文摘要

ABSTRACT Late Onset Alzheimer's Disease (LOAD) genome wide association studies (GWAS) discovered numerous loci. But there remains an unmet need to translate the GWAS findings to disease mechanisms through the identification of the specific genes involved, the causal variants, and the molecular mechanisms by which they exert their pathogenic effects. Most LOAD-associated SNPs are in noncoding regions pointing to gene regulation as an important disease mechanism. Another challenge in LOAD genetics is diversity, as most studies were conducted in subjects from European ancestry, while other populations are largely understudied. Our central hypothesis is that LOAD-specific epigenomic signatures, as well as noncoding functional genetic variants result in dysregulation of genes with key roles in LOAD pathogenic biological pathways. While omics studies using bulk brain tissue from European ancestry donors have produced informative data for a few genes at LOAD loci, single-cell omics data from brains of patients and controls from diverse populations will provide new knowledge in unprecedented brain cell-subtype precision across multiple racial and ethnical groups. We will investigate the relationships between LOAD-specific gene expression, chromatin accessibility and genetic variability in European and African ancestries by single-nuclei multi-omics approaches following three specific aims. Aim 1 will generate matched single-nuclei (sn)RNA-seq and ATAC-seq datasets using the 10X Genomics platform (Single Cell Multiome) to characterize cell-subtype specific changes in transcriptomic and chromatin accessibility landscape, respectively, in LOAD compared to control, that are shared and distinct across European and African ancestries. Aim 2 will integrate these datasets to identify open/closed chromatin sites that function as regulatory elements to impact gene expression in LOAD state, which will be then validated in the relevant cell-subtype using isogenic hiPSC-derived models by CRISPR/Cas9 genome editing. Aim 3 will identify LOAD specific gene regulatory variants within specific brain cell-subtypes through integrative single-cell genomics. We will perform expression(e)QTL and chromatin(c)QTL analyses by cell- subtype focusing specifically on the QTLs that fall within previously published GWAS regions to determine whether GWAS signals can be explained by the identified regulatory interactions. We will then catalogue the SNPs that identified as both strong and significant eQTL and cQTL and prioritize those that predicted to affect transcription factor binding sites. Last, we will validate the top prioritized variants in genome edited isogenic hiPSC-derived models. Successful accomplishment of these aims is expected to be high impact as it will advance the understanding of the genetic complexity underpinning LOAD in diverse populations and will decipher the regulatory elements and the corresponding genes mediating LOAD risk. This knowledge will be translational by promoting the refinement of Polygenic Risk Scores, and the development of novel therapeutic targets for LOAD based on manipulation of dysregulated genes.