Plasmodium infection induces phenotypic, clonal, and spatial diversity among differentiating CD4+ T cells

Naive CD4+ T cells must differentiate in order to orchestrate immunity to Plasmodium, yet understanding of their emerging phenotypes, clonality, spatial distributions, and cellular interactions remains incomplete. Here, we observe that splenic polyclonal CD4+ T cells differentiate toward T helper 1 (Th1) and T follicular helper (Tfh)-like states and exhibit rarer phenotypes not elicited among T cell receptor (TCR) transgenic counterparts. TCR clones present at higher frequencies exhibit Th1 skewing, suggesting that variation in major histocompatibility complex class II (MHC-II) interaction influences proliferation and Th1 differentiation. To characterize CD4+ T cell interactions, we map splenic microarchitecture, cellular locations, and molecular interactions using spatial transcriptomics at near single-cell resolution. Tfh-like cells co-locate with stromal cells in B cell follicles, while Th1 cells in red pulp co-locate with activated monocytes expressing multiple chemokines and MHC-II. Spatial mapping of individual transcriptomes suggests that proximity to chemokine-expressing monocytes correlates with stronger effector phenotypes in Th1 cells. Finally, CRISPR-Cas9 gene disruption reveals a role for CCR5 in promoting clonal expansion and Th1 differentiation. A database of cellular locations and interactions is presented: https://haquelab.mdhs.unimelb.edu.au/spatial_gui/. [Display omitted] •T helper cells exhibit expansion-linked Th1 skewing and rare iTreg/TCMP emergence in malaria•Spatial transcriptomics maps splenic microanatomy and cellular co-locations during infection•Spatial transcriptomics maps molecular interactions of Th1 and Tfh cells•CCR5 promotes CD4+ T cell differentiation in experimental malaria Williams et al. employ spatial transcriptomics at single-cell resolution and scRNA-seq during experimental malaria to map differentiation and cellular interactions of CD4+ T cells in the spleen, presenting this as a graphical user interface. They validate their predictions in vivo using confocal microscopy and CRISPR-Cas9-mediated gene disruption.