CD70 recruitment to the immunological synapse is dependent on CD20 in B cells

Abstract

CD20 is a four-transmembrane protein expressed at the surface of B cells from late pro-B cells to memory B cells, with the exception of plasma cells. Its expression pattern makes it an attractive therapeutic target for different B cell malignancies and autoimmune diseases. Despite the clinical success of CD20-targeting antibodies, the biology of the CD20 protein is still not well understood. We investigated CD20 binding partners in the membrane of human B cells using immunoprecipitation followed by mass spectrometry analysis. We identified a molecular interaction between CD70 and CD20, and confirmed this using proximity ligation assays. CD20-CD70 spatiotemporal colocalization was validated at the plasma membrane of B cells using high-resolution microscopy. Cell surface expression of CD70 was found to be enhanced upon CD20 overexpression, suggesting a role for CD20 in stabilizing CD70 at the B cell membrane. Moreover, we observed impaired B-T cell synapse formation and defective recruitment of CD70 to the immunological synapse in the absence of CD20. Impaired synapse formation was confirmed by deleting CD20 in primary B cells, and analysis of B cells from a CD20-deficient patient. Finally, CD20-deletion resulted in diminished T cell activation and cytokine secretion. Together, this study demonstrates that CD20 interacts with CD70 at the B cell membrane, and that CD20 is required for immune synapse formation between B and T cells and consequent T cell activation.

Keywords: B lymphocytes; CD20; CD70; immune synapse.

Fig. 1.

CD20 is highly clustered at the cell surface of B cells. (A–D) Representative Airyscan confocal images of CD20 staining (Top), expression intensity shown by color gradient (bottom bar), and defined CD20 clusters (Bottom), for BJAB (A), Oci-Ly1 (B), PB B cells (C) and tonsil B cells (D). (Scale bar: 5 μm.) (E and F) Quantification of CD20 cluster size in μm² (E) and the number of CD20 clusters (F) defined on the bottom of the plasma membrane for BJAB, Oci-Ly1, PB, and tonsil B cells. Datapoints are 46 (BJAB), 38 (Oci-Ly1), 47 (PB B cell), and 25 (Tonsillar B cells) cells derived from 4 (BJAB), 2 (Oci-Ly1), 3 (PB), and 1 (tonsil) independent experiments or donors. Data represent mean ± SEM.

Fig. 2.

CD20 is mobile at the B cell surface. (A) Representative confocal images of CD20 FRAP time course in BJAB cells. Images show CD20 signal and distribution prebleach (Left), and postbleach imaging (Middle) and final post bleach image (Right). The yellow circle indicates the bleached area and timing (seconds) as indicated in figure. (Scale bar: 5 μm.) (B) Mean recovery curves of CD20 and HLA-DR (positive control) signal normalized to prebleach intensity in BJAB cells. (C) Quantification of CD20 and HLA-DR mobile fractions in BJAB cells. (D) CD20 recovery speed indicated by T-half in seconds in BJAB cells. (E) PB B cells were analyzed as described in A. (Scale bar: 5 μm.) (F) Mean recovery curves of CD20 and HLA-DR (positive control) signal normalized to prebleach intensity in PB B cells. (G) Quantification of CD20 and HLA-DR mobile fractions in PB B cells. (H) CD20 recovery speed indicated by T-half in seconds in PB B cells. Datapoints are 30 (CD20, BJAB), 47 (HLA-DR, BJAB), or 33 (PB B cells) cells derived from three (CD20, BJAB), four (HLA-DR, BJAB), or three (PB B cells) independent experiments or donors. Statistical significance was assessed by the Mann–Whitney U test, mean ± SEM is shown (*P < 0.05, ****P < 0.0001).

Fig. 3.

CD20 interacts with CD70 at the cell surface of B cell lines. (A) IP of CD20 using detergents Digitonin (Left) and Brij97 (Right). (B) Heatmap of iBAQ values showing the highest enriched proteins hits detected by MS upon CD20 Co-IP using Brij97 (B) or Digitonin (D). Enrichment intensity shown by gradient scale (right bar). (C) Representative confocal microscopy images of CD20 (Left column, magenta) and CD70 (middle column, green) and the merged image (Right column) in BJAB (Top row), Oci-Ly1 (Middle row), and PB B cells (Bottom row). (Scale bar: 5 μm.) (D) Manders’ overlap coefficient (M1) of CD70 overlap over CD20 for BJAB, Oci-Ly1, and PB B cells. Datapoints derived from four (BJAB), two (Oci-Ly1), and three (PB B cells) independent experiments or donors. Mean ± SEM shown. (E) Copurification of CD20 (Left) and CD70 (Right) in HEK293 cells. (F) Representative confocal microscopy images of in situ PLA on BJAB cells stained for CD20 and CD70 (Left), CD20 and HLA-ABC (Middle panel, positive control), and CD20 and CD55 (Right panel, negative control). (Scale bar: 5 μm.) (G) Dots per cell were quantified from three independent experiments (CD70 n = 74, HLA-ABC n = 53, and CD55 n = 56 cells). Data represent mean ± SEM. Statistical significance was assessed by one-way ANOVA with Tukey’s post hoc test. (****P < 0.0001).

Fig. 4.

Effect of CD20 on CD70 cell surface expression in B cell lines and primary B cells. (A) CD20 and CD70 surface expression in WT and CD20KO BJAB cells. Representative flow cytometry histogram showing CD20 expression in WT (blue) and CD20KO (green) BJAB cells (Left). Isotype control is shown in black. Quantification of normalized CD70 MFI values in WT and CD20KO BJAB cells (Right). Datapoints derived from three independent experiments. (B) CD20 and CD70 surface expression in WT and CD20KO Oci-Ly1 cells. Representative flow cytometry histogram showing CD20 expression in WT (blue) and CD20KO1 (green), CD20KO2 (orange), and CD20KO3 (pink) Oci-Ly1 cells (Left). Isotype control is shown in black. Quantification of normalized CD70 MFI values in WT and CD20KO Oci-Ly1 cells (Right). Datapoints derived from four independent experiments. (C) Schematic representation of the generation of CD20KO primary B cells. (D) CD20 (Left) and CD70 (Right) surface expression in WT and CD20KO activated primary B cells by flow cytometry (seven donors). (E) Representative confocal image of BJAB WT cells expressing CD20-GFP (green, Bottom Left) and stained for CD20 (endogenous and exogenous) (magenta, Top Left). Merged image (Right) shows overlap. (Scale bar: 5 μm.) (F) Quantification of normalized CD70 MFI values in BJAB cells transfected with CD20-GFP, gated for GFP⁺ or GFP⁻. Datapoints derived from four independent experiments. (G) Correlation between CD20 and CD70 surface expression in CD20-GFP-positive BJAB cells. Statistical significance was assessed by the Mann–Whitney U test (A), Friedmann test followed by Dunn’s multiple comparison test (B), or paired t test (D and F). (*P < 0.05, ****P < 0.0001). All graphs depict mean values ± SEM.

Fig. 5.

CD20 deletion in BJAB cells results in defective IS formation and less CD70 recruitment toward the synapse. (A and B) Representative epifluorescence images of cocultures with Jurkat and WT (A) or CD20KO (B) BJAB cells pulsed with SEE. Cells were stained against CD3 (blue, Top Left), MHC II (magenta, Top Right), and CD70 (green, Bottom Left). Bottom Right shows merged panel and yellow arrows indicate immune synapses. (Scale bar: 10 μm.) (C) Quantification of mean number of synapses per Jurkat T cell in cocultures with WT and CD20KO BJAB cells. (D) Quantification of synapses with 1.5 times enriched CD70 intensity in CD3-enriched areas as percentage of total synapses in cocultures between Jurkat T cells and WT or CD20KO BJAB cells. (E) Quantification of CD70 intensity in CD3-enriched areas as a percentage of mean CD70 intensity in cocultures of Jurkat T cells and WT or CD20KO BJAB cells. Graphs show mean ± SEM values from three independent experiments. 30 images per condition (WT or CD20KO) were analyzed, ranging between 300 to 400 synapses analyzed per condition. Statistical significance was assessed by the Mann–Whitney U test (C and D) or unpaired t test (E), (**P < 0.01, ***P < 0.001, ****P < 0.0001).

Fig. 6.

CD20 deficiency in primary B cells impairs synapse formation, affecting T cell activation and cytokine secretion. (A) Quantification of number of synapses per B cell in cocultures of primary T cells with activated primary B cells from healthy donors (WT) or a CD20-deficient patient (CD20KO), with/without SEE. Datapoints derived from five donors and two independent experiments. 85 (-SEE, WT), 90 (-SEE, CD20KO), 93 (+SEE, WT), and 90 (+SEE, CD20KO) images were analyzed. (B) Representative epifluorescence zoomed images of these cocultures stimulated with SEE. Shown are merged images (Left), CD20 (magenta, Middle Left), CD3 (blue, Middle Right), and HLA-DR (green, Right). Yellow arrows indicate immune synapses. (Scale bar: 10 μm.) (C) Quantification of number of synapses per B cell in cocultures of CMV-specific CD4+ T cells with WT or CD20KO activated primary B cells with/without CMV peptide. Datapoints derived from three donors. 67 (-CMV) and 72 (+CMV) images were analyzed. (D) Representative epifluorescence zoomed images of these cocultures stimulated with CMV. Shown are merged images (Left), CD20 (magenta, Middle Left), CD3 (blue, Middle Right), and HLA-DR (green, Right). Yellow arrows indicate immune synapses. (Scale bar: 10 μm.) (E) Quantification of CD25 expression in CMV-specific CD4+ T cells after 24 h of coculture with WT or CD20KO primary B cells, with/without CMV. (F and G) Quantification of IFNγ secretion by these CMV-specific CD4+ T cells in time (F) and quantification at day 7 (G). Datapoints derived from three donors. Statistical significance was assessed by two-way ANOVA with Sidak’s multiple comparisons test (A, C, F, and G), or one-tailed unpaired t test with Welch’s correlation (E). (*P < 0.05). Mean ± SEM is shown.