1508 Single-cell epigenetic profiling highlights genetic impact on chromatin accessibility in SLE

Abstract

<h3>Background</h3> Chromatin accessibility (CA) is a critical epigenetic feature identifying genomic loci that actively participate in gene-regulating functions, such as transcription and DNA repair. In turn, genetic polymorphisms within these loci can affect the magnitude of CA, i.e. a chromatin accessibility quantitative trait locus (caQTL). Changes in CA have been implicated in inflammatory disease, and previous genetic research has identified several risk haplotypes for systemic lupus erythematosus (SLE). However, it remains unclear how CA may interact with genetic risk factors in SLE pathogenesis. To better understand how CA in SLE may be driven by genetic variation, we performed single-cell assay for transposase accessible chromatin (sciATAC-seq) on peripheral blood mononuclear cells (PBMCs). SciATAC-seq is an efficient, scalable and sensitive assay that allows the epigenetic profiling of thousands of cells from an individual. By combining these profiles with genotypic data, we can search for caQTLs as evidence of genetic-epigenetic interaction specific to SLE. <h3>Methods</h3> PBMCs were isolated from 45 SLE patients and 50 healthy controls. Each PBMC sample underwent both genotyping and sci-ATAC sequencing. DNA was then sequenced by Illumina Next-seq. Cell-specific ATAC reads were demultiplexed and quantified by custom software developed by the BROAD Institute (Cambridge, MA). Genotyping data was phased and imputed using the IMPUTE2 tool suite. Cell type-specific caQTL analysis was performed by RASQUAL. <h3>Results</h3> An average of 980 cells were sequenced per sample, with a total of 745,697 CA sites measured. We identified a total of 153,716 caQTL relationships across 17 distinct immune cell types, involving 59,715 unique variants; 59% of which are also reported as expression QTLs in whole blood. The majority of caQTLs already implicated in autoimmune disease risk haplotypes occurred predominantly in B cells and plasmacytoid dendritic cells. CA profiles exhibit cell type-specific cluster orientation highly correlated with caQTL genotype. Genotypes at variant rs1131665, previously associated with an SLE risk haplotype in IRF7 and here as a caQTL, distinguished subpopulations of B cells and monocytes on the basis of global CA profiles. Increases of CA at the variant were present in SLE individuals compared to controls, suggesting caQTL variants contribute to a genome-wide epigenetic phenotype for SLE risk. <h3>Conclusion</h3> Using advances in single-cell epigenetic profiling, we were able to identify thousands of genetic variants which influence epigenetic functions, in a cell type-specific way, through their association to CA. Understanding the molecular mechanisms for how caQTLs alter cell type-specific chromatin accessibility will provide new insights into the role of epigenetic regulation in SLE pathogenesis. <h3>Acknowledgments</h3> The research reported in this abstract was supported by Institutional Development Awards (IDeA) from the National Institute of General Medical Sciences (U54GM104938 and P30GM110766), the National Institute of Arthritis and Musculoskeletal and Skin Diseases (P30AR073606 and R01AR073750); the National Institute of Allergy and Infectious Diseases (UM1AI144292 and R01AI156724), and by the Presbyterian Health Foundation (OKC, OK).