Abstract 14781: Simultaneous Profiling of Genotype and Transcriptome at Single Cell Resolution Identifies Cellular Programs Underlying Clonal Hematopoiesis Pathology

Abstract

Introduction: Clonal hematopoiesis of indeterminate potential (CHIP) results from a clonal expansion of hematopoietic stem cells due to somatic mutations in genes such as DNMT3A or TET2. Patients with CHIP have poor cardiovascular outcomes. TET2 CHIP clones that make up >20% of blood increases risk. Whether CHIP mutated cells confer risk directly or through polarizing other cells is presently unknown as prior work has been unable to simultaneously resolve mutational status and transcriptome of an individual cell. Methods: We performed single cell RNA-seq on blood from patients with DNMT3A CHIP (N=9), TET2 CHIP (N=8), and controls (N=6). We identified individual cell CHIP mutational status with a novel method that leverages mitochondrial mutations to identify cell lineage. Results: Across 104,556 single cells, CHIP patients had a higher proportion of CD14 monocytes (32% vs. 18%). Within individuals, CHIP mutated cells exhibited a strong myeloid bias. At the extreme, 93% of TET2 mutated cells were monocytes in one patient. Comparing TET2 CHIP monocytes to controls identified a hyper-inflammatory phenotype with increased expression of inflammatory cytokines and their downstream targets, and elevated expression of adhesion molecules. Comparing TET2-mutated CD14 monocytes to unmutated monocytes within the same individual localized the inflammatory phenotype to the mutated cells. Conversely, this phenotype was not observed in DNMT3A CHIP monocytes either in aggregate or within the same individual. Conclusions: Simultaneous, high-throughput, single-cell resolution of DNA mutation status and RNA transcriptomes identifies aberrant cellular programs in TET2-mutated CD14 monocytes. Our data support a simple mechanistic model for CHIP cardiovascular disease risk that is proportional to the number of mutated cells present. Differences observed between TET2 and DNMT3A provide a potential explanation for the differential pathogenicity of these two mutations.