Single-cell transcriptomics reveal a unique memory-like NK cell subset that accumulates with aging and correlates with disease severity in COVID-19

For the scRNA-seq analysis of human blood lymphocytes and NK cells, lymphocytes were sorted as CD45+ cells, and NK cells were sorted as CD3/CD19/CD20/CD14−CD7+ cells using FACS (SONY SH800S); the purity was above 95%. The cells from 4 elderly individuals were pooled in one tube and the cells from 4 young individuals were pooled in another tube. We then counted and resuspended the pooled cells at a concentration of 1000 cells/μL, aiming for an estimated 6000 cells per library, following the instructions of the single-cell 3ʹ solution v3 reagent kit (10X Genomics). Single-cell libraries were constructed strictly according to the manufacturer’s standard protocols. Each sequencing library was generated with a unique sample index. Libraries were sequenced on the Illumina NovaSeq 6000 system. Lymphocytes and sorted NK cells in peripheral blood from young (<=30 years) and elderly (>=65 years) healthy individuals were collected and subjected to single-cell sequencing.