Gene exression in single T cells across division states.

Purpose: To compare diversity of primary human CD8+ T cells that have divided 0, 1, or 2 times on day 3 of ex vivo expansion from naïve resting state. Methods: Naïve T cells were enriched from human peripheral blood monoluclear cells (PBMCs), labeled with CFSE dye, and expanded for 3 days using rapid expansion protocol (Li, Y. & Kurlander, R.J. Journal of Translational Medicine, 2010). On day 3, 10,000 single live CFSE+ CD8+ T cells from each of divisions 0, 1, and 2 were sorted and immediately processed using 10X Genomics single-cell RNA-sequencing platform. Results: We found that undivided cells display the highest gene expression diversity. Using 1,000 most variably expressed genes, we created a force-directed layout, representing a phenotypic map of cellular differentiation across division states. To understand the basis of T-cell diversity, we defined and quantified regions of interest on this map based on diffusion pseudo-time (DPT), a metric of cell differentiation state. Finally, we examined gene expression in cells from each region and found that undivided cells acquire gene expression associated with effector cell function, while remaining cells go on to grow and differentiate. Conclusions: Our study provides insights into T-cell differentiation within an ex vivo expansion system for cancer immunotherapy applications. Overall design: A total of 4,060 cells (division 0: n = 552 cells, division 1: n = 1,777 cells, division23: n = 1,731 cells) were sequenced to an average of 52,040 post-normalization reads per cell capturing a median of 18,770 unique molecular identifier (UMI) counts per cell mapping to 3,544 unique genes per cell.

3

Good Z, Borges L, Vivanco Gonzalez N, Sahaf B, Samusik N, Tibshirani R, Nolan GP, Bendall SC