Pathology-related alterations in gene expression in Alzheimer's disease: Single cell resolution.

Abstract

Alzheimer's disease is pathologically defined by the presence of extracellular amyloid beta plaques and intracellular tau tangles, resulting in neurodegeneration and reactive gliosis in affected brain regions such as temporal cortex (TCX). The degree and distribution of AD pathology can be defined by Braak stage (tau) and Thal phase (amyloid). Brain tissue is comprised of multiple cell types that have different molecular profiles, the proportions of which can vary between individuals and in response to disease; creating challenges for the interpretation of bulk tissue transcriptome profiling study. We aimed to use single nucleus RNA sequencing (snRNAseq) to explore cell proportion and cell-type specific transcriptome changes associated with AD and neuropathology. We selected age and sex matched post mortem TCX from 12 AD and 12 control patients. We obtained snRNAseq profiles using 10x Genomics platform. After quality control 79,751 nuclei remained for analysis and were annotated for their type according to commonly used cellular markers. Nuclei proportions were tested for association with AD, Braak stage and Thal phase using Wilcoxon rank sum test. Differentially expressed genes (DEGs) and pathways were identified for each cell cluster using MAST and validated through in situ hybridization via RNAscope and qPCR. Annotation according to common cellular markers distinguished major and rare cell types of the brain in 35 clusters. We detected a unique excitatory neuron cluster, predominantly from control individuals that show less pathology and APOE ε4 genotype. DEG and pathway analysis revealed alterations in myelination pathways in oligodendrocytes, and synaptic signaling pathways in excitatory neurons. Analysis of the effects of sex, age, and APOE ε4 genotype on selected DEGs and pathways are conducted. In this study, we developed and optimized an snRNAseq approach for frozen human temporal cortex that captures the major brain cell types. We identified robust DEGs and pathways for these between AD vs control and demonstrated shared and unique cell-specific pathways and DEGs amongst different pathologies in AD.