Systematic investigation of mitochondrial transfer between cancer cells and T cells at single-cell resolution

Mitochondrion (MT) as the energy-producing organelle participates in most metabolic activities of mammalian cells. Unidirectional mitochondrial transfer from T cells to cancer cells was recently observed “metabolically empowering” cancer cells while “depleting immune cells”, providing new insights into Tumor-T cell interaction and immune evasion. Here, we leveraged the increasingly adopted single-cell RNA-seq technology and introduced MERCI, a statistical deconvolution method for tracing and quantifying the process of mitochondrial trafficking between cancer and T cells. MERCI was benchmarked using data generated from a coculture system of MT-labeled cancer and T cells to accurately predict the recipient cells and their mitochondrial compositions. We independently demonstrated the cellular MT transfer signals and validated the MERCI performance by additionally generating a gold-standard mtscATAC-seq dataset. Application of MERCI to single cell human cancer samples identified a novel MT transfer phenotype and its signature genes involved in cytoskeleton remodeling, energy production and TNFα signaling pathways. Application of MERCI to ovarian cancer spatial transcriptomes revealed the prevalence of MT transfer in the tumor tissue of high T cell infiltration, which is elevated in the endothelium-rich regions. Finally, MT transfer is associated with high cell cycle activity and poor clinical outcome in our pan-cancer analysis through MERCI. In summary, MERCI enabled systematic investigation of a novel aspect of Tumor-T cell interaction and revealed MT-transfer related genes and pathways that may lead to new therapeutic opportunities.