Fig 2: MAVS genetic variation does not affect intracellular cytokine levels of HIV-1-specific T cells. (a) PBMCs of individuals carrying a MAVS minor or MAVS major genotype obtained 2.5–3.5 years p.SC were left unstimulated (-), or were treated with HIV-1 consensus B Gag peptide pool (Gag) or staphylococcal enterotoxin B (SEB). After 6 h, the percentages of surface molecules expressing CD137+ and CD107a+ and intracellular expression of IFN-γ+, IL-2+, Mip-1β+, and TNF-α+ CD4+ T cells and (b) CD8+ T cells were analyzed using flow cytometry. Each square or dot represents a different study participant (median ± IQR). No significant differences between HIV-1-infected individuals with a MAVS minor or MAVS major genotype were observed.

Fig 3: CD46 activation has distinct outcomes in CD4+ and CD8+ T cells. a CD46 expression on resting CD4+ and CD8+ T cells from patient CD46-1 (red histograms) and a matched healthy donor (HD) (black histograms; gating strategy as in Supplementary Fig. 7a). b IFN-γ secretion by CD4+ and CD8+ T cells from patient CD46-1. Purified T cells were left non-activated (NA) or activated with the depicted immobilized antibodies and IFN-γ secretion assessed (60 h). Data shown are mean values of each condition performed in duplicate. c Degranulation (CD107a) and granzyme B expression of CTLs from patient CD46-1. Purified CTLs were activated with the indicated antibody combinations and CD107a and granzyme B expression measured at 60 h (gray histograms, isotype controls; black histograms, non-activated cells; green, blue, or red histograms, CD3, CD3 + CD28, or CD3 + CD46 activation, respectively; gating strategy shown in Supplementary Fig. 7b). d Cytotoxic activity of CTLs from patient CD46-1. CTLs from patient CD46-1 were CD3 + CD46-activated for 24 h and killing activity measured in comparison to CTLs from five HDs (gating in Supplementary Fig. 7c). e, f IFN-γ production and degranulation capacity of CTLs from CD46-deficient patients CD46-1 to CD46-4. Purified CD8+ T cells from patients CD46-1 to -4 and from four sex- and age-matched HDs were activated as in (b) and IFN-γ secretion (e) and CD107a and granzyme B+ cells (f, gating as in Supplementary Fig. 7b) measured (n = 4). g, h Effect of CD46 protein knockdown on HD CTL effector function. CD8+ T cells from three different healthy donors were transfected with either siRNA targeting CD46 mRNA or control siRNA for 48 h, then activated with immobilized antibodies to CD3 and CD46 for further 48 h and IFN-γ secretion (g) and CD107a and granzyme B expression measured (h, gating as in Supplementary Fig. 7b; representative cytometry plots of three similarly performed experiments; black histograms, control siRNA; purple histograms, CD46 siRNA) (n = 3). Error bars denote mean ± SEM. *p < 0.05; ***p < 0.005; ctrl.: control. Statistical analysis was performed using One-way ANOVA with Tukey Multiple Comparison test

Fig 4: HLA-G engages KIR2DL4 on NK cells to suppress trastuzumab-elicited ADCC. a FCM assay of the expression of the indicated receptors on human PBMC-derived primary NK cells. b NK cells in the presence of trastuzumab were cocultured with HER2-overexpressing breast cancer cells at the indicated E:T ratios supplemented with or without the indicated blocking antibodies. The toxicity of NK cells to malignant cells was measured. c, d NK cells and SK-BR-3 cells were cocultured as described in (b) (E:T = 30:1). IFN-? production was measured via ELISA (c), and the degranulation of NK cells was evaluated via FCM assay for CD107a expression (d). e Representative immunohistochemical staining of KIR2DL4 on infiltrating NK cells in clinical HER2-positive breast cancer tissues. f Representative immunofluorescent staining of HER2 and KIR2DL4 in breast cancer tissues of individual patients. All experiments were performed three times. Statistical significance was obtained by Student’s t test. *P < 0.05, **P < 0.01, and ***P < 0.001. n.s. Nonsignificant

Fig 5: CD46 co-stimulation augments nutrient influx and fatty-acid synthesis. a, b Differentially expressed genes in CD8+ T cells of four healthy donors (HD) and three CD46-deficient (CD46-def.) patients (CD46-2 to -4) upon CD3 + CD46 activation at 6 h (a) and Venn diagram indicating overlap between differentially expressed transcripts (b). c Heatmaps comparing non-activated (NA) and CD3 + CD46-activated (6 h) HD CTLs with focus on genes involved in lipid/lipoprotein metabolism (left), transmembrane solute transport pathways (right), and FABP5 (left below). d Effect of CD46 co-stimulation on LAT1 and FASN expression and on mTOR activation. CD8+ T cells (gated as in Supplementary Fig. 7b; gray histograms, isotype controls; black histograms, non-activated cells; blue, green, or red histograms, CD3, CD3 + CD28, or CD3 + CD46 activation, respectively) from HDs were left NA or stimulated as depicted and LAT1 and FASN expression and mTOR activity (70S6K phosphorylation) measured 48 h post activation (n = 4). e Effect of FASN inhibition on IFN-? production and degranulation of CTLs. CD8+ T cells were activated as in (d) in the presence or absence of the FASN inhibitor C75 and IFN-? secretion and CD107a expression measured at 48 h post activation (n = 3–4; gray bars, carrier addition; white bars, addition of 5 µM C75; black bars, addition of 10 µM C75). f Heatmap indicating FASN, FABP5 and SLC7A5 expression in HD (n = 4) vs. CD46-deficient (n = 3) CD8+ T cells in resting state and after CD3 + CD46 activation. g RT-PCR for FASN, FABP5, and SLC7A5 in CD3 or CD3 + CD46-activated CD8+ T cells from patients CD46-1, CD46-2, and CD46-4 (light red bars), and three matched HDs (light blue bars) at 6 h post activation (n = 3). Error bars denote mean ± SEM. *p < 0.05; **p < 0.01. Statistical analyses were performed using One-way ANOVA with Tukey Multiple Comparison test or the Paired Student’s t-test where appropriate