Abstract 968: Distinct pathways of DC-induced CD8+T cell differentiation revealed by single-cell mRNA sequencing analysis

Abstract

Abstract Purpose of Study: Dendritic Cells (DC) maturing in the conditions of acute/early versus late/chronic inflammation (such as alpha-type-1-polarized DCs [αDC1s] and “standard” nonpolarized DCs; sDCs) deliver differential “signal 3” to resting T cells, resulting in their different induction of effector functions in tumor-specific CD8+ T cells. Here, using single-cell mRNAseq (scRNAseq), a unique technology allowing for unbiased analysis of gene expression (GEX), we evaluated the GEX patterns and cellular heterogeneity of CD8+ T cells expanded by cancer-cell loaded αDC1s or sDCs. Procedure: Freshly isolated autologous (compared to DCs) CD8+ T cells were in vitro sensitized for 1 week, using either αDC1s loaded with UV/gamma-radiation-killed ovarian cancer (OvCa) cell lines OVCAR or SKOV3, or sDCs loaded with the same antigens. αDC1s and sDCs not loaded with exogenous antigen were used as controls. The expanded CD8+ T cells were harvested, captured at the single-cell level and immobilized in a vertical flow array chip (VFAC), which contain 100-microchambers packed with 105 beads immobilizing 1010 oligo(dT) probes with unique cell-ID, UMI, and PCR-tags. Each cell was lysed and its mRNA was individually tagged with unique cell ID sequences prior to library preparation and targeted scRNAseq using a custom 94-gene inflammation RNA-seq panel was performed. Post-clustering (Scanpy), unsupervised cell-type identification (louvain), and downstream analyses were performed on GEX data to determine impact of αDC1s versus sDCs (loaded with OVCAR or SKOV3 versus not loaded with antigens) on the effector and memory pathways of differentiation of the expanded T cells and their unique functional status. Results: PCA and unsupervised clustering revealed sample groups with distinct phenotypic signatures associated with loaded versus unloaded aDC1s, with differential GEX subclusters identified for the αDC1- activated T cells (independently on DC loading with either OVCAR or SKOV3 cells) to include at least two populations with differential expression of genes typical of central memory, effector memory and effector cells (killer molecules and chemokine receptors typical of both lymphoid and peripheral homing potential). In sharp contrast, sDC1 activated T cell showed a more uniform gene distribution patterns lacking the effector T cell component. These differences correlated with the differential ability of αDC1- and sDC-activated T cells to respond to migrate to different chemokines and to recognize and kill cancer cells. Conclusion: scRNAseq was able to identify distinct single-cell T-cell phenotypes caused by induction of αDC1s or sDCs (loaded with OVCAR or SKOV3), consistent with the differential ability of these two DC types to induce effector, effector memory and central memory CD8+ T cells in response to different tumor targets and with their differential ability to modulate peripheral tissue homing patterns. Citation Format: Bowen Dong, Sarabjot Pabla, Vincent Giamo, Weijian Jiang, Kiyomi Taniguchi, Sean T. Glenn, Masataka Shirai, Pawel Kalinski. Distinct pathways of DC-induced CD8+T cell differentiation revealed by single-cell mRNA sequencing analysis [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 968.