Fig 1.
Interaction of NK cells with B cell-bound EBV VP in the presence of EBV S+ reduces transformation without inducing B cell death.
(A, B) B cells were stained with fluorescent dye, coated with B95.8 VP for 1h at 4°C, washed, and cultured with autologous NK cells in the presence of EBV S-, EBV S+ serum or the indicated doses of Rituximab. After 4h, the percentages of CD107α+ NK cells (A) and of DAPI+ B cells (B) were determined by flow cytometry. Background levels without B cells (for CD107α) or without NK cells (for DAPI) were substracted. Plots of a representative experiment (left) and results of three (right) are shown. Statistical analysis was performed with Friedman test with Dunn’s post-test. (C, D) B cells were incubated with B95.8 VP for 1h at 4°C, washed and cultured alone or with autologous NK cells in the presence of EBV S- or S+ serum for 4h. Subsequently, B cells separated from NK cells were plated at 50,000 cells /well, cultured for 2 weeks and stained for CD19, CD21, CD23 and DAPI. (C) Representative gating strategy. (D) Data are expressed as Ln of the number of live CD23+ CD21+ CD19+ cells detected in 20–26 wells per condition from 3 experiments. Statistical analysis was performed with one-way ANOVA test with Bonferroni’s multiple comparison test.
Fig 2.
NK cells in the presence of EBV+ serum remove VP attached to autologous B cells.
(A) B cells were pre-incubated with B95.8 VP for 1h at 4°C, washed, stained with anti-gp350 and cultured alone or with autologous NK cells in the presence of EBV S- or EBV S+ for 4h. The proportions of B cells (CD19+) displaying gp350 were analyzed; representative plots (left) and results of four different experiments (right) are shown. (B) B cells were pre-incubated with AKBM SN, washed and stained with 0.3μM CFSE and for gp350. Samples were cultured alone or with NK cells in the presence of EBV S- or EBV S+ for 4h. The proportions of gp350+ B cells (CFSE+) were analyzed; representative plots (left) and results of four experiments (right) are shown. (C) B cells were pre-incubated with AKBM SN for 1h at 4°C, washed, cultured with NK cells in the presence of EBV S- or EBV S+ for 4h and the proportions of GFP+ B cells (CD19+) were analyzed; representative plots (left) and results of three experiments are shown. (A-C) Statistical analysis was performed with Friedman test with Dunn’s post-test.
Fig 3.
EBV-specific Ab-dependent removal of B cell-bound VP by NK cells is an active process independent of perforin and proteases.
(A, C-G) B cells were pretreated with AKBM VP (A, C, E) or with B95.8 VP (D, F, G), stained after washing with anti-gp350 and cultured with NK cells together with EBV S- or EBV S+ sera for 4h. (A) NK cells were pre-incubated for 2h with 50nM Dasatinib or vehicle (DMSO), or for 30 min with EDTA (20mM) or EGTA (10mM) and the same concentrations were maintained during the co-culture. The proportions of gp350+ B cells detected in three (Dasatinib, EDTA) or two experiments (EGTA) are shown. (B-D) NK cells were pre-treated or not for 2h with 1μM of Concanamycin A, 50μM of Chloroquine or vehicle (DMSO) and co-cultured with 721.221 with or without 100ng/ml of Rituximab (B), or with AKBM VP- (C) or B95.8 VP-coated (D) B cells; the same concentrations of inhibitors were maintained during co-cultures. Proportions of DAPI+ 721.221 cells (B) or of gp350+ B cells (C, D) from four (B) or three experiments (C, D) are shown. (E) Proportions of CD16low-neg cells detected among CD56dim NK cells from six experiments are shown. Statistical analysis was performed with Wilcoxon test. (F-G) NK cells were pre-treated or not for 2h with 25μM of GM6001, protease inhibitor cocktail of 200μM Leupeptin, 8μM Aprotinin, 15μM Pepstatin, 25μM GM6001 and 10μM Amastatin or the equivalent volume of vehicle (DMSO or H2O plus DMSO). B and NK cells were co-cultured for 4h in the presence of EBV S- or S+ sera, with or without fresh inhibitor or vehicle at the same concentration used for pre-treatment. Proportions of CD16low-neg cells within CD56dim NK cells (F) and of gp350+ B cells (G) are shown. Representative plots of the indicated conditions (left) and results of four experiments (right) are shown. (H) B cells were incubated with B95.8 VP, stained for gp350 and cultured alone or with NK cells in the presence of EBV S-, S+ or Rituximab (25ng/ml) for 4h. The proportions of gp350+ B cells were analyzed. Representative plots (left) and results of four experiments are shown. For statistical analysis Friedman test with Dunn’s post-test was applied to the last 3 columns.
Fig 4.
EBV S+ induces VP transfer from B to NK cells.
(A) B cells were coated with B95.8 VP, stained for gp350 and incubated for 4h with NK cells in the presence of EBV S-, S+ serum or 25ng/ml of Rituximab. Proportions of gp350+ NK cells detected in representative plots (left) or four experiments (right) are shown. Statistical analysis with Friedman test and Dunn’s post-test was applied. (B) B cells were incubated with AKBM SN, stained for gp350, co-cultured for 4h with NK cells in the presence of sera from various EBV S- (n = 5) or S+ (n = 6) donors, and analyzed for the proportions of gp350+ NK cells. (C-D) B cells were treated and co-cultured with NK cells as in (A), in the presence of EBV S- or S+ sera. (C) A gp350 fluorescence index was calculated for B and NK cells multiplying the percentage of the population by the corresponding gp350 MFI; the absolute variation (|Δ|) of the index comparing conditions with EBV S- and S+ was calculated. For each experiment, the absolute variation of the index in NK cells was expressed as a percentage of that calculated for B cells. (D) For each experiment, MFI of all live cells in the co-culture with EBV S+ was expressed as a percentage of that calculated for EBV S- co-cultures. In C-D results of eight experiments are shown. (E-G) B cells were incubated with B95.8 VP (E, F) or AKBM VP (G), stained for gp350 and co-cultured with NK cells, in the presence of EBV S- or S+ for 4h (E) or during the indicated times (F, G). The proportions of gp350+ B (F, G; left) and NK cells (E-G, right) corresponding respectively to four (E) and three (F, G) experiments are shown. Statistical analysis was performed with Friedman test with Dunn’s post-test. (H-I) B cells coated with B95.8 VP were cultured alone or with NK cells in the presence of EBV S- or S+ serum for 4h, followed by separation of B and NK cells. DNA was extracted from both cell fractions and analyzed by qPCR for the EBV gene EBNA1 and the endogenous gene 36B4. Relative quantification of EBNA1 in relation to 36B4 for B (H) and NK cells (I) was calculated and normalized to the first condition of each graph. Data correspond to three experiments.
Fig 5.
Transfer of B cell bound-EBV particles to NK cells involves contact.
(A) B cells were coated with B95.8 VP, stained for gp350 and co-cultured with NK cells pre-treated for 2h with 20μg/ml of Cytochalasin B or the equivalent volume of vehicle (DMSO). Co-cultures were performed in the presence of EBV S- or S+ serum, with or without 20μg/ml of Cytochalasin B or vehicle for 4h. The proportions of gp350+ B (left) and NK cells (right) corresponding to four experiments are shown. Statistical analysis was performed with Friedman test with Dunn’s multiple comparison test. (B-C) B cells were coated with PKH26-stained B95.8 VP, cultured with CFSE-stained NK cells and EBV S- or S+ for 15min, 1h or 4h, adhered onto coverslips and analyzed by confocal microscopy. (A) Images of Maximum Intensity of the stack of the EBV S- (top) or S+ (bottom) conditions at 15 min. Scale bar: 5μm. Data are representative of three experiments. (B) Quantification of total PKH26 fluorescence per B (left, 31–158 cells per condition) or NK cell (right, 34–116 cells per condition) from two different experiments. Statistical analysis was performed with Mann-Whitney U test.
Fig 6.
Transfer of B cell-bound EBV particles to NK cells is encompassed by trogocytosis.
(A-E) B cells were coated with B95.8 VP (A-C) or AKBM VP (D, E), co-cultured with NK cells and EBV S- or S+ serum for the indicated times, and stained for CD21, CD19, CD20 and CD56. The proportions of CD21+ (A, D), CD19+ (B, E) and CD20+ (C) NK cells (gated as in Fig 4A) are shown. Representative plots of EBV S- at 4h and EBV S+ at 15min (left) and of four (A-B) and three (C-E) experiments (right) are displayed. Statistical analysis was performed with Friedman test with Dunn’s multiple comparison test. (F) B cells were stained with 2μM PKH26, coated with B95.8 VP and co-cultured for the indicated times with NK cells and EBV S- or S+. Proportions of PKH26+ NK cells in representative plots of EBV S- at 4h and EBV S+ at 15min (left) and of three experiments are shown. (G, H) B cells were treated and co-cultured as in (F) but for 4h in the presence of EBV S-, EBV S+ or 12.5 ng/ml of Rituximab. The proportions of CD19+ (E) and PKH26+ (F) NK cells corresponding to two experiments are shown.
Fig 7.
EBV particles are internalized by NK cells and co-localize with endosome and lysosome markers.
(A-C) B cells were coated with B95.8 VP and stained by indirect immunofluorescence with anti-gp350 Ab, followed by the PE-conjugated secondary Ab either before (left) or after (right) being incubated alone (A) or with NK cells (B, C) in the presence of EBV S- or S+ for the indicated times. PE MFI was normalized to that detected on B cells alone (A, B) or NK cells (C) at time 0; data corresponding to three experiments is shown. Statistical analysis was performed with repeated measures mixed model ANOVA with Bonferroni post-test. (D) B cells were coated with B95.8 VP, stained for gp350, mixed with NK cells in the presence of EBV S- or S+ sera, incubated at 4°C or at 37°C for 4h, and treated or not with trypsin for 30 min. After trypsin neutralization, cells were washed and stained for CD19 detection. Since CD56 was eliminated by trypsin, NK cells were gated as CD19- cells. The percentage of maximum gp350+ B (left) or NK cells (right) of each experiment is shown. Data correspond to three experiments. Statistical analysis was performed with Friedman test with Dunn’s post-test. (E-F) B cells were coated with PKH26-stained B95.8 VP, cultured with CFSE-stained NK cells and EBV S+ for 15min, 1h or 4h, adhered onto coverslips, stained for Rab5 (E) or CD107α (F), and analyzed by confocal microscopy. Left: Representative slices at 1h (E) and 4h (F). Right: Corrected Manders coefficient and Pearsons coefficient (R) of co-localization of VP with Rab5 at 15 min and 1h (E) or with CD107α at 1h and 4h (F) within VP+ NK cells are shown. Data correspond to 51–78 cells per condition from two experiments. Scale bar: 5μm.