Neuronal, glial and microglial maps of epigenetic and transcriptomics dysregulations in Alzheimer’s disease

Abstract Background Regions of open chromatin house regulatory elements required to mediate cell‐type and tissue‐specific gene expression. Human brain studies have shown that dysregulation of these regulatory mechanisms is associated with Alzheimer’s Disease (AD). Here, we present the largest cell type and brain region‐specific differential chromatin accessibility analysis and fine‐mapping analysis in AD. Method Using both fresh and frozen postmortem tissue from over 200 cases with AD and controls, we performed ATAC‐seq to profile chromatin accessibility in three distinct populations of cells (neurons, glia, and microglia), isolated by FACS from three brain regions. We characterized epigenetics changes associated with multiple AD phenotype ratings. We further studied the correlation structures in chromatin accessibility to define high‐resolution maps of cis‐regulatory domains. Lastly, we examined population‐level variation of gene expression and chromatin accessibility to pinpoint genetically driven regulation of transcription. Result We observed widespread differences in chromatin accessibility associated with AD. The differential cis‐regulatory domains were highly concordant with gene expression perturbations suggesting coordinated changes of the 3D genome regulation of the transcriptome in AD. For microglia cells, the fine‐mapping analysis identified putative regulatory mechanisms for 21 AD risk loci, of which 18 were refined to a single gene, including 3 novel genes (KCNN4, FIBP and LRRC25). Transcription factor regulatory networks captured AD risk variation and identified SPI1, IRF1, and PURA as the key regulators of microglia expression changes in AD. The same analysis in neurons pinpointed upstream transcription factor 2 (USF2), which regulates the genes participating in lysosome function, which we validated by overexpression and knockdown experiments. Lastly, we determined enhancer‐promoter interactions, by generating additional omics in brain tissue (Hi‐C, H3K27ac ChIP‐seq) and jointly analyzing with ATAC‐seq based on the “activity‐by‐contact” approach. To prove the relevance of these putative enhancers, we performed validations in neuronal progenitor cells. Conclusion Overall, our human neuronal, glia, and microglia multi‐scale omics datasets uncovered disease‐associated perturbations impacting chromatin accessibility, transcription factor regulatory networks, and the 3D genome, and implicated transcriptional dysregulation in AD. Furthermore, we were able to fine‐map multiple AD loci, identifying not only the relevant genes but, in some cases, proposing the regulatory mechanisms contributing to disease.