Characterization of human immune cell dynamics based on methylation patterns in circulating cell-free DNA

Counting and expression profiling of immune cells in peripheral blood provide rich information about immune states. However such snapshot measurements are blind to cell birth and death dynamics, and to immune processes taking place in remote locations. We explored the utility of circulating cell-free DNA (cfDNA) fragments released from dying cells as a window into immune cell turnover dynamics. Using comparative methylome analysis we established a panel of cell type-specific DNA methylation markers that can accurately quantify DNA derived from specific immune cell types including T cell subtypes, B cells, monocytes, neutrophils and eosinophils. cfDNA levels of these markers reveal the integration of cell number and cell turnover, such that DNA from short-living cell types is over represented in cfDNA. We document the baseline levels of immune cell cfDNA in the plasma of 190 individuals aged 1 to 85, as well as intra-individual variation during the time scale of weeks. cfDNA levels vary more than cell counts, suggesting that homeostatic maintenance of specific immune cell numbers is achieved by modulation of cell birth and death rates. Longitudinal analysis of immune cell cfDNA following influenza vaccination revealed an early and transient elevation of B cell cfDNA, indicative of altered B cell turnover in response to antigen stimulation. Analysis of immune cell cfDNA allows for sensitive detection of immune processes not reflected in the cellular fraction of peripheral blood.