Single cell transcriptomes and multiscale networks from persons with and without Alzheimer's disease

Abstract

The emergence of single nucleus RNA sequencing (snRNA-seq) offers to revolutionize the study of Alzheimer's disease (AD). Integration with complementary multiomics data such as genetics, proteomics and clinical data provides powerful opportunities to link cell subpopulations and molecular networks with a broader disease-relevant context. We report snRNA-seq profiles from superior frontal gyrus samples from 101 well characterized subjects from the Banner Brain and Body Donation Program in combination with whole genome sequences. We report findings that link common AD risk variants with CR1 expression in oligodendrocytes as well as alterations in hematological parameters. We observed an AD-associated CD83(+) microglial subtype with unique molecular networks and which is associated with immunoglobulin IgG4 production in the transverse colon. Our major observations were replicated in two additional, independent snRNA-seq data sets. These findings illustrate the power of multi-tissue molecular profiling to contextualize snRNA-seq brain transcriptomics and reveal disease biology.

Fig. 1. Single nucleus RNA-seq from superior frontal gyrus samples from 101 human decedents.

A Experimental workflow for generation of transcriptomes from 481,840 nuclei. B, C Major cell type identification followed by detection of cell populations that are differentially abundant between AD and Control subjects. D With varying associations with clinicopathological AD traits. DA differentially abundant, SFG superior frontal gyrus, snRNA-seq single nucleus RNA-sequencing, Ex excitatory neurons, In inhibitory neurons, Mic microglia, Oli oligodendrocytes, Opc oligodendrocyte progenitor cells, Ast astrocytes. Panel A created withBioRender.com released under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 Internationallicense.

Fig. 2. Cell-type expression QTL detection in superior frontal gyrus.

A Integration of whole genome sequences with cell type expression to detect cell type eQTL associations B with shared and cell type-specific distribution C revealing multiple cell type eQTL linking with DNA loci that have been implicated in diverse neurological and neurodegenerative traits via GWAS data. eQTL expression quantitative trait loci, Ex excitatory neurons, In inhibitory neurons, Mic microglia, Oli oligodendrocytes, Opc oligodendrocyte progenitor cells, Ast astrocytes.

Fig. 3. GWAS loci rs679515 impacts CR1 in oligodendrocytes, known erythrocyte regulators and peripheral blood hematocrit.

A AD-specific eQTL association between AD GWAS loci rs679515 genotype and CR1 expression within oligodendrocytes. B Colocalization of genetic signal underpinning AD GWAS risk and CR1 expression within oligodendrocytes driven by rs679515. C Causal inference network illustrating conditional relationships between AD GWAS locus rs679515, CR1 expression, and downstream molecular networks, including impacts on EPO receptor expression. D Molecular functional enrichments of the rs679515/CR1 network revealed themes of erythrocyte biology and hematological traits. E Antemortem hematocrit levels from 49 Banner AD subjects, stratified according to rs679515 dosage. Enrichments based on Fisher’s Exact Test. AD Alzheimer’s disease, GWAS Genome-Wide Association Study, HCT hematocrit, FDR false discovery rate.

Fig. 4. Exploration of CD83(+) microglia.

A Multiscale gene coexpression network constructed from differentially abundant CD83(+) microglial subpopulation reveals novel antisense transcripts functioning as network hubs. B Photomicrograph of CD83(+) microglia within Layer II of superior frontal gyrus in a Braak IV AD subject. CD83 reactive microglia appear (see insert) compared to unreactive microglia (arrow). C APOE gained as a network hub and D at increased expression in CD83(+) microglia compared with CD83(−) microglia. The mean APOE expression for each subject (stratified by microglial type) is shown. E Representative molecular functional enrichments among CD83(+) microglia hub genes. F Increased abundance of Immunoglobulin Heavy Chain protein IGHG4 protein in the transverse colon as a function of the presence of CD83(+) microglia in superior frontal gyrus, while controlling for AD, Age, Sex, and PMI (Samples N = 26).