Characterization of T stem cell memory heterogeneity in humans

TSCM cells are a rare, antigen-experienced T cell population, generated directly from TN upon priming and endowed with both stemness and memory features. After the identification and characterization in mice, non-human primates and humans, TSCM cells have been placed at the apex of the memory T cell differentiation tree. Their longevity together with their capacity to reconstitute the entire T cell memory compartment makes TSCM highly exploitable for clinical interventions in the ACT field. Recently, it has been suggested that the TSCM compartment is actually a heterogeneous pool comprising two distinct subpopulations, differing for their longevity and proliferation capacity. scRNA-Seq and bulk RNA sequencing have helped in the discrimination of (at least) two subsets, by identifying CD29 and Plexin B2 as putative markers. A preliminary functional characterization of these two subsets, termed LT-TSCM (CD29lo Plexin B2hi) and ST-TSCM (CD29hi Plexin B2lo), as well as an initial analysis of their kinetics in the course of an immune response, have been addressed in this work. CFSE dilution in sorted, TCR-stimulated subsets revealed substantial heterogeneity between independent experiments, precluding definitive conclusions on differences between the TSCM subsets. Plated ST-TSCM had a greater fold expansion than LT-TSCM, hinting to a higher proliferation capability, as detected in vivo by TREC analysis (unpublished findings). ST-TSCM also showed lower multipotency, as the phenotypic analysis of their progeny highlighted their ability to generate ST-TSCM, TCM, and TEM, but not LT-TSCM. This evidence seems to confirm the idea that ST-TSCM are a more proliferative and more differentiated TSCM subset than LT-TSCM. To address the polyfunctionality, ST-TSCM and LT-TSCM were compared regarding the percentages of IL-2-, IFNγ-,TNFα-, GzmB-, Prf1-producing, CD107a-exposing and Ki-67-positive cells. Both subsets presented a higher functionality than TN coherently with their memory nature. ST-TSCM were more polyfunctional than LT-TSCM and contained more producers of every cytokine and cytotoxicity marker, confirming their more advanced differentiation level. Interestingly, frequencies of Ki-67 positive LT-TSCM were the highest among all subsets, requiring further investigation in this direction. Finally, to identify Ag-specific LT-TSCM and ST-TSCM and analyze their kinetics during the course of an immune response, a tetramer staining was performed in the setting of YF-17D vaccination. The high-quality reagent we validated allowed to efficiently identify in human blood samples the A2/LLW-specific CD8 T cell response, that peaked at W2 after vaccination and steadily decreased in the next 6 months. From our preliminary findings in three vaccinees, LT-TSCM were less abundant than ST-TSCM at any timepoint, and the LT-TSCM:ST-TSCM ratio was decreased in comparison to the one observed both in the tetramer negative population and among CD8+ T cells at steady state. If this has to be imputed to an underdetection of the memory marker CD95 or to the LT-TSCM and ST-TSCM biology needs to be assessed in future experiments and might help to give further insight into the development of memory in the context of infectious diseases and vaccination. Altogether, our preliminary findings on the analysis of the heterogeneity of the TSCM population support the hypothesis that it comprises two phenotypically and functionally distinct subsets, of which LT-TSCM are of particular interest, as they feature the least differentiated phenotype and greatest multipotency. A deeper understanding of memory generation and stemness regulation in these subsets is an enticing prospect for both basic and translational research, as it could eventually lead to the generation of more effective T cell immunotherapies.