Fig 2: Expansion of EBVST in 24-well plate (24w) and BECA-D. (A) Plot of cell population numbers counted for 24-well plate (24w) and BECA-D at the end of every activation for Donor 3, Donor 4 and Donor 5. (B) Percentage population of helper (CD4+) and cytotoxic (CD8+) T-cells under the subset of T-cells (CD3+) cells at the end of 5th activation for Donors 1–5.

Fig 3: VISTA+CD4+ T cells in mouse splenocytes and VISTA+CD3+ lymphocytes in patient blood increase following experimental or clinical sepsis. (A) Summary graph of VISTA+ CD4+ T cells in the wild-type mouse spleen. (B) Summary graph of CD4+ T cell frequency in the wild-type mouse spleen. (C) Summary graph of VISTA+ CD3+ T cell frequency in the peripheral blood lymphocytes. (D) Summary graph of CD3+ T cell frequency in the peripheral blood. (A–D) Summary graphs show mean ± SEM [WT-sham: n = 13, WT-CLP: n = 16]; significance **p < 0.01, ***p < 0.001. (E) Initial process (Strategy) for producing embryos deficient in the ~11.3-kb region containing exon (ex) 2 to exon 7 of the VISTA gene on mouse chromosome 10 with CRISPR/Cas9 followed by NHEJ-mediated repair. (F) Results of initial heterozygous cross of VISTA-/+ founder mice resulting from CRISPR/Cas9 technology that produced homozygous VISTA-/- mice for breeding (PCR genotyping strategy: 403/412: 11,745 bp from WT and ~0.4 kb from VISTA deletion alleles).

Fig 4: SARS-CoV-2-reactive, antigen-experienced CD4+ T lymphocytes are in the circulation of COVID-19 ICU patients.a Titration of SARS-Cov-2-S protein-specific IgG, IgA, and IgM antibodies in the serum of three COVID-19 ICU patients. ?MFI calculation as described in Fig. 2a. Dotted line indicates a ?MFI of 1500. b Workflow for the isolation and analysis of SARS-CoV-2-and measles-reactive CD4+ T lymphocytes. c Cells for single cell sequencing were sorted using FACS as PI-CD19-CD14-CD3+CD4+ CD154+ or CD137+ (gating strategy as used for the analysis shown in Supplementary Fig. 4). UMAP representation of 3727 CD4+ T cells from three COVID-19 ICU patients and two healthy controls (SARS-CoV-2-stimulated cells from one control and measles-stimulated cells from another control). Four clusters were identified based on transcriptional similarity using shared nearest-neighbor (SNN) modularity optimization (left). An additional CD14+ cluster (124 cells) and six outlying T cells with considerably less UMI-counts than the average were also identified but excluded from further analysis. Heatmap of signature genes of the four clusters. Depicted are genes with an absolute log2 fold change > log2(2) and a p-value < 0.01 after Bonferroni correction (two-sided Wilcoxon rank sum test). Colors corresponds to z-scores of the average expression. d Expression levels of selected signature genes and cytokines.

Fig 5: Thymic CD3+ and CD4+ lymphocytes demonstrated no change in HVEM+ BTLA+ co-expression after either Hem/CLP or S-CLP models, though CD3+ and CD4+ lymphocytes both demonstrated increased HVEM expression after S-CLP. (A) Thymic CD3+ lymphocyte abundance. (B) Thymic CD3+ lymphocyte HVEM+ expression. (C) Thymic CD3+ lymphocyte HVEM+ BTLA+ co-expression. (D) Thymic CD4+ lymphocyte abundance. (E) Thymic CD4+ HVEM+ expression. (F) Thymic CD4+ HVEM+ BTLA+ co-expression. (A–F) Summary graphs show mean ± SEM all after Hem/CLP and S-CLP treatments; [Sham Hem/CLP n = 11, Hem/CLP n = 11, Sham S-CLP n = 10, S-CLP n = 10]; significance *p < 0.05; **p < 0.01.