Single-cell variant landscape of the human mitochondrial genome reveals cell type-specific functions and disease mechanisms

Abstract

Mitochondrial DNA (mtDNA) heteroplasmy is a significant source of cellular genetic diversity and can cause disease yet its distribution and functional importance across different human cell types are poorly understood. Here we present scMOCHA (Single-Cell Mitochondrial Omics for Cellular Heteroplasmy Analysis), a scalable computational framework that resolves high-confidence, single-cell resolution mtDNA heteroplasmy directly from standard single-cell RNA sequencing (scRNA-seq) data. Using this tool to interrogate publicly available scRNA-seq datasets from human peripheral blood mononuclear cells (PBMC) and bone marrow (BM) uncovered the single-cell landscape of mtDNA single-nucleotide variants (SNVs) and heteroplasmy across ages in human hematopoietic cell types. PBMCs and BM share many inherited homoplasmic and highly heteroplasmic SNVs, whereas somatic SNVs are tissue- and cell type-specific and distributed across the mtDNA genome, suggesting that they arise largely stochastically. mtDNA replication error is a dominant force shaping homoplasmic variants, while oxidative stress potentially drives a significant portion of heteroplasmic and somatic mutations. Importantly, we discover previously unknown cell type-specific and human disease-associated mtDNA SNVs. By coupling mtDNA genotypes with matched single-cell nuclear transcriptomes and structural modeling, we uncover potential molecular mechanisms by which disease-associated mtDNA SNVs impact cellular processes such as inflammation and contribute to disease. Together, these findings define the single-cell variant landscape of the human mitochondrial genome and reveal previously unrecognized lineage- and disease-associated patterns of mtDNA variation in the human hematopoietic system.