Towards a better understanding of genetic architecture of Alzheimer's disease with human iPSC models

点击翻译按钮获取中文摘要

PROJECT SUMMARY Alzheimer's disease (AD) is a progressive neurodegenerative disease and the leading cause of dementia with high heritability (~70%). It is increasingly clear that AD is highly polygenic, and for most of AD cases it is the polygenicity of the risk variants across the genome that predisposes the disease risk. In contrast to the rapid identification of risk loci associated with AD by recent genome-wide association studies (GWAS), identifying the potential causal variants/genes at the reported risk loci and decoding these variants/genes into molecular and cellular pathology have lagged far behind. Since disease variants, mostly locating in noncoding regions of the human genome, have been shown to affect cellular function through multi-level regulations such as DNA accessibility and histone modifications, DNA methylation and RNA expression in a cell type-specific manner, comprehensive and unbiased investigating the cell type-specific influence of generic risk variants on AD risk at multiple levels, including epigenomic, transcriptomic, and cellular levels, in an isogenic background is crucial to understand the genetic basis of AD pathogenesis. In the current application, by combining human induced pluripotent stem cells (hiPSCs) with gene editing and comprehensive multi-omics and cellular analyses, we will dissect the AD genetic risk variants into cell type-specific molecular and cellular pathology. Given the polygenic nature of AD, and the heterogeneity of AD risk genes on the cellular level, we hypothesize that multiple genetic risk variants act synergistically among different compartments (e.g. cell types) to contribute to pathogenesis of AD. First, we will identify AD risk variants and genes with comprehensive analyses of AD genetic architecture using machine learning approaches including DVAR, eVAR and iRIGS (Aim 1). Second, we will delineate the cell type-specific epigenetic and transcriptomic signatures associated with AD candidate risk variants using human iPSC-derived neurons/microglia/astrocytes (Aim 2). Last, we will determine the functional impact of AD candidate risk variants on AD-like cellular pathology in neurons, microglia, astrocytes, and their co-cultures (Aim 3). Our proposal may advance our understanding of the complex genetic architecture of AD, leading to a better understanding of AD pathogenesis and facilitating the development of novel therapeutic strategies.