Single-cell analyses identify tobacco smoke exposure-associated, dysfunctional CD16+ CD8 T cells with high cytolytic potential in peripheral blood

Abstract

Tobacco smoke exposure has been found to impact immune response, leukocyte subtypes, DNA methylation, and gene expression in human whole blood. Analysis with single cell technologies will resolve smoking associated (sub)population compositions, gene expression differences, and identification of rare subtypes masked by bulk fraction data. To characterize smoking-related gene expression changes in primary immune cells, we performed single-cell RNA sequencing (scRNAseq) on >45,000 human peripheral blood mononuclear cells (PBMCs) from smokers (n=4) and nonsmokers (n=4). Major cell type population frequencies showed strong correlation between scRNAseq and mass cytometry. Transcriptomes revealed an altered subpopulation of Natural Killer (NK)-like T lymphocytes in smokers, which expressed elevated levels of FCGR3A (gene encoding CD16) compared to other CD8 T cell subpopulations. Relatively rare in nonsmokers (median: 1.8%), the transcriptionally unique subset of CD8 T cells comprised 7.3% of PBMCs in smokers. Mass cytometry confirmed a significant increase (p = 0.03) in the frequency of CD16+ CD8 T cells in smokers. The majority of CD16+ CD8 T cells were CD45RA positive, indicating an effector memory re-expressing CD45RA T cell (TEMRA) phenotype. We expect that cigarette smoke alters CD8 T cell composition by shifting CD8 T cells toward differentiated functional states. Pseudotemporal ordering of CD8 T cell clusters revealed that smokers cells were biased toward later pseudotimes, and characterization of established markers in CD8 T cell subsets indicates a higher frequency of terminally differentiated cells in smokers than in nonsmokers, which corresponded with a lower frequency in naive CD8 T cells. Consistent with an end-stage TEMRA phenotype, FCGR3A-expressing CD8 T cells were inferred as the most differentiated cluster by pseudotime analysis and expressed markers linked to senescence. Examination of differentially expressed genes in other PBMCs uncovered additional senescence-associated genes in CD4 T cells, NKT cells, NK cells, and monocytes. We also observed elevated Tregs, inducers of T cell senescence, in smokers. Taken together, our results suggest smoking-induced, senescence-associated immune cell dysregulation contributes to smoking-mediated pathologies.