Fig 2: Clustering of CNS cells following WNV infection by scRNA-seq. a Experimental design for scRNA-sequencing. Mice were infected (i.c.) with 1×104 p.f.u. WNV-NS5-E218A and harvested 25 DPI. N=2 per group, each N includes 4 mice pooled. b tSNE plot of all immune cells analyzed in both mock- and WNV-infected animals, showing 9 clusters, colored by density clustering, and annotated by cell-type identity and number of cells in parentheses. c tSNE plots of all cells analyzed in mock (left panel)- and WNV (right panel)-infected mice, separated by treatment group, colored by density clustering. d Heatmap of single cells representing the mRNA levels of the top three well-known genes used for cellular identification of each cluster. e tSNE plot of all immune cells with mRNA of Cxcr6 (left panel) and mRNA of Cxcl16 (right panel). f Violin plots showing immune cells with Cxcr6 mRNA (left panel) or Cxcl16 mRNA (right panel) in clusters 0–5. Each dot represents a cell

Fig 3: CXCR6 signaling potentiates gliosis in the recovered hippocampi of WNV-infected animals, contributing to synaptic elimination. a Representative immunostaining and quantification of IL-1ß (green), GFAP (red), and DAPI (blue) in the CA3 region of the hippocampus of mock or WNV-infected WT or Cxcr6 -/- mice at 52 DPI. GFAP quantified by percent positive area, IL-1ß quantified by percent positive area, followed by quantification of GFAP+IL-1ß+ area, normalized to the total GFAP+ area. b Representative immunostaining and quantification of IBA1 (green), CD68 (red), and DAPI (blue) in the hippocampus of mock- or WNV-infected WT or Cxcr6 -/- animals at 52 DPI. IBA1+ quantified as percent positive area. c Representative immunostaining and quantification of C1qa in the hippocampus of mock- or WNV-infected WT or Cxcr6 -/- animals at 52 DPI showing staining for C1qa (green) and DAPI (blue). C1qa quantified as percent positive area. d Representative immunostaining and quantification of synapses in the CA3 region of the hippocampus in mock- or WNV-infected WT or Cxcr6 -/- animals at 52 DPI showing staining for synaptophysin (red) and DAPI (blue). Synaptophysin quantified by percent positive area. Scale bars, 50 µm (a–c) or 20 µm (d). Data represent the mean±s.e.m. and were analyzed by two-way ANOVA and corrected for multiple comparisons. *P<0.05, **P<0.005, ***P < 0.001

Fig 4: CXCR6 is expressed by CD8+ T cells that persist in the CNS after viral recovery. a, b Immunohistological analysis for CXCR6 and CD3 in the cortex (a) and hippocampus (b) of mock- or WNV-infected mice at 25 DPI, presented as microscopy. The left graph shows the percenter CD3+ area, the middle left graph shows the percent CXCR6+ area, the right middle graph shows the CD3+CXCR6+ area normalized to the total CD3+ area, indicative of co-localization, and the right graph shows the CD3+CXCR6+ area normalized to the total CXCR6+ area. Scale bars, 50 µm. c Gating strategy and quantification of the percentage of CD8+ T cells that are CXCR6+ at 7, 25, and 52 DPI in the cortex (left graph) and hippocampus (right graph). Cells were gated on CD45high (Additional file 1: Fig. S8a) before gating on CD4+ and CD8+ T cells (left panel). CD8+ T cells were then gated on CXCR6 (middle panels). d Gating strategy and quantification of the percentage of CD103+ cells that are CXCR6+ at 7, 25, and 52 DPI in the cortex and hippocampus. e Gating strategy and quantification of the percentage of NS4B+ cells that are CXCR6+ at 7, 25, and 52 DPI in the cortex and hippocampus. f Gating strategy and quantification of the percentage of CD8+ T cells that are CXCR6+ at 7, 25, and 52 DPI in the blood. Cells were gated on CD45high before gating on CD4+ and CD8+ T cells. CD8+ T cells were then gated on CXCR6. g Quantification of the percentage of CD8+ T cells that are CXCR6+ at 7, 25, and 52 DPI in the cervical lymph nodes. h Quantification of the percentage of CD8+ T cells that are CXCR6+ at 25 and 52 DPI in the meninges. Gating for CXCR6 based off Cxcr6 -/- mice (Additional file 1: Fig. S8b, c). Data represent the mean±s.e.m. and were analyzed by unpaired Student’s t-test. *P<0.05, **P<0.005, ***P < 0.001, ****P < 0.0001

Fig 5: High dimensionality analysis reveals KFE8-Ag85B-dependent shift in Trm surface marker profile of Ag85B-specific CD4+ T cells in BCG-primed and boosted mice.a We gated for tetramer+ CD4+ T cells in each group. Data shown as mean ± SE. Significance was determined by one-way ANOVA followed by a Benjamini and Hochberg test. b The tetramer+ cells from each group were analyzed as a whole and visualized in a tSNE plot. The cells isolated from each treatment group are highlighted c and heatmaps were generated to describe intensity of specific markers CD44, CD62L, CD69, CD45, and CXCR3. Area highlighted in light blue represents the cell populations most augmented by KFE8-Ag85B and KFE8-Ag85B-pulsed DCs. d Heat maps are representative of light blue population and include markers CD127, CD69, CXCR6, CD49a, and CD103. Dark blue regions of heat maps correspond with lowest the lowest intensity of fluorescent antibody staining, whereas red corresponds with the highest intensity and ranges from -1622 up to 262856.