Fig 2: In the absence of B7-CD28 co-stimulation, the frequency of cells undergoing clonal deletion at DP and CD4SP stages is decreased.a–d WT and B7 DKO thymocytes were gated on developmental stages as defined by TCR and CD69 expression in Fig. 1. a Frequency of clonally deleting cells (Helios+ PD-1high) was decreased in Stage III (CD4+ CD8low) thymocytes in B7 DKO mice. Each group n = 3. Data are representative of three independent experiments. p = 0.0365. b Frequency of clonally deleting cells (Helios+) was decreased in Stage IV and V SP4 Tconv (CD25/Foxp3nega) in B7 DKO mice. Each group n = 3. Data are representative of three independent experiments. Stage IV p = 0.0004, Stage V p = 0.0003. c Frequency of clonally deleting cells (Active Caspase3+) was decreased in TCR-signaled DP and SP cells in B7 DKO mice. Each group n = 5. Data are pooled results of two independent experiments. Signaled DP p < 0.0001, Signaled SP4 p = 0.0038. d Frequency of clonally deleting cells (Helios+) was decreased in Stage IV and V SP4 Tconv (CD25/Foxp3nega) in Bcl2-Tg/B7 DKO mice. WT n = 9, B7 DKO n = 3, Bcl2-Tg n = 5, Bcl2-Tg/B7 DKO n = 3. Data are pooled results of three independent experiments. Stage IV (WT and B7DKO) p = 0.0003, Stage IV (Bcl2-Tg and Bcl2-Tg/B7DKO) p = 0.0037, Stage V (WT and B7DKO) p = 0.0008, Stage V (Bcl2-Tg and Bcl2-Tg/B7DKO) p = 0.020. e Nur77-GFPhigh SP4 cells were increased in B7 DKO mice. Left panel, representative FACS plot. Right panel, quantification of Nur77-GFPhigh and SP4 Nur77-GFPhigh populations. Each group n = 6. Data are pooled results of three independent experiments. Total p < 0.00001, SP4 p = 0.0010. f CD5 expression level (MFI) of SP4 Nur77-GFPhigh population. Left panel, representative FACS plot. Right panel, Quantification of CD5 MFI. Each group n = 4. Data are combined results of three independent experiments. p = 0.0005. a–f Data are mean ± SEM with dots representing individual values of biologically independent animals. Statistical differences between groups were calculated using (a–d) unpaired, two-tailed Student’s t-test, and (e, f) paired, two-tailed Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001. Source data are provided as a Source Data file.

Fig 3: Identification of a Distinct Population of ?-chain/Syk-Deficient NK Cells in rhCMV+ Rhesus Macaques(A) Representative gating strategy identifying ?g among bulk NK cells.(B) Representative histograms showing expression of NKp46, CD8a, Siglec-7, Helios, and Eomes in ?g and cNK cells.(C) Representative flow cytometry plots indicating distribution of CD56+ and CD16+ subpopulations in ?g and cNK cells.(D and E) Composite phenotypic t-SNE plot (D) and PCA analysis (E) of bulk NK cells down-selected from rhCMV-infected rhesus macaque cohorts showing independent multidimensional clustering of ?g NK cells; data points represent individual cells.Individual parameters are shown in Figure S3. All plots are representative of 6 to 70 animals depending on grouping and measured parameter.

Fig 4: ?g NK Cells Are Primed by CMV Status(A–D) Frequencies among bulk NK cells (A) and normalized circulating numbers of ?g NK cells (B) are shown for rhCMV-uninfected (n = 28) and rhCMV-infected (n = 14) rhesus macaques. Horizontal lines indicate medians. Density of intracellular Helios (MFI, median fluorescence intensity) (C) and surface CD16 on ?g NK cells (D) in peripheral blood nuclear cells (PBMCs) from rhCMV-uninfected and rhCMV-infected macaques. Bars indicate medians.(E) Composite phenotypic t-SNE plot of ?g NK cells only down-selected in both rhCMV-uninfected and rhCMV-infected rhesus macaque cohorts showing near complete independent multidimensional clustering by rhCMV-specific IgG levels; data points represent individual cells.(F) Correlation of circulating ?g NK cell numbers with rhCMV-binding antibody equivalents.Statistical evaluations were made with Mann Whitney U or Spearman’s correlation test; **p <0.01; ***p < 0.001; ****p < 0.0001.