Lipid Membrane-Based Antigen Presentation to B Cells Using a Fully Synthetic Ex Vivo Germinal Center Model

Abstract

High-affinity antigen-specific B cells are generated within specialized structures, germinal centers (GCs), inside lymphoid organs. In GCs, follicular dendritic cells (FDCs) present antigens on their membrane surface to cognate B cells, inducing rapid proliferation and differentiation of the B cells toward antibody-secreting cells. The FDC's fluid membrane surface allows B cells to "pull" the antigens into clusters and internalize them, a process that frequently involves tearing off and internalizing FDC membrane fragments. To study this process ex vivo, liposomal membranes are used as the antigen-presenting FDC-like fluid lipid surface to activate B cells. In a fully synthetic in vitro GC model (sGC), which uses the microbead-based presentation of the CD40 Ligand and a cytokine cocktail to mimic T follicular helper cell signals to B cells, liposomes presenting a model antigen mimic effectively engage B cell receptors (BCRs) and induce greater BCR clustering compared to soluble antigens, resulting in rapid antigen internalization and proliferation of the B cells. B cells showed GC-like reactions and undergo efficient IgG1 class-switching. Taken together, the results suggest that fluid membrane-bound antigen induces a strong GC response and provides a novel synthetic in vitro system for studying GC biology in health and diseases, and for expanding therapeutic B cells ex vivo.

Keywords: B cell receptors; B cells; antigen; liposomes; vaccine.

Figure 1.

Liposome nanoparticles as FDC membrane mimics. A) Schematic of GC interactions where B cells engage with T_FH cells via CD40-CD40L and FDCs via antigen–BCR. B) On antigen engagement (1), BCRs are concentrated (2). The B cell uses mechanical forces to pull the antigen (3), leading to the capture of antigen and associated membrane fragments (4). C) Schematic of in vitro culture system consisting of murine splenic B cells, antigen-presenting liposomes, and CD40L-presenting beads. D) Size distribution profile of 100 nm liposomes. E) Fluorescence microscopy of liposomes conjugated to biotinylated R-phycoerythrin. Scale bar: 500 μm.

Figure 2.

Antigen-loaded liposomes induce uptake and activation of primary murine B cells. A) Loading of biotinylated anti-IgM per 250 μg liposomes. N = 3; mean ± SEM. B) Stability of particle diameter after 3 days of culture at 37 °C. N = 3; mean ± SEM; *p < 0.05. C) Stability of bound protein after 3 days of culture at 37 °C. N = 4; mean ± SEM. D) B cell uptake of FITC-liposomes either loaded or unloaded with anti-IgM. Confocal microscopy of B cells at 24 h. Zoomed region indicated by dotted box. E) B cell uptake of FITC-liposomes either loaded or unloaded with anti-IgM. Plots represent MFI of FITC-liposomes in B cells as a function of time. N = 4; mean ± SEM; A (p < 0.05 than no liposomes), B (p < 0.05 than unloaded); two-way ANOVA with Tukey’s post hoc comparison. F) Effect of liposomes on expression of B cell activation markers. Plots represent MFI of activation markers of B cells incubated for 24 h with liposomes either loaded or unloaded with anti-IgM as a function of liposome concentration. N = 3; mean ± SEM; A (p < 0.05 than unloaded), B (p < 0.05 than other doses of same formulation); two-way ANOVA with Tukey’s post hoc comparison.

Figure 3.

Proliferation and generation of GC B cells. A) Experiment timeline. B) Total cell counts per 200 000 starting cells at day 3 (D3) and day 5 (D5). C) Quantification of Cell Trace plots in (D). Normalized proliferation calculated by dividing the MFI of the samples from the MFI of the unloaded liposome controls. D) Retention of Cell Trace proliferation dye in CD19+ cells after 3 days of culture. The flow cytometry plots show one representative sample per group. Overlayed is the unproliferated control and unstained control. E) Count of GL7+ B cells per 200 000 starting cells at day 3 (left) and day 5 (right). F) Retention of Cell Trace proliferation dye in CD19+ GL7+ cells. The flow cytometry plots show one representative sample per group. Overlayed is the unproliferated control and unstained control. G) Left: Representative quadrant gating of GBCs (GL7+ Fas+ ). Right: Count of GL7 Fas B cells per 200 000 starting cells at day 3. In all studies, anti-IgM concentration was either 60 ng mL⁻¹ or 300 ng mL⁻¹ and presented either soluble (red circle) or on liposomes (blue squares). Negative control is unloaded liposomes. Experiments were done with n =4–6, presented as mean ± SEM; ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05; two-way ANOVA with Tukey’s post hoc comparison.

Figure 4.

Liposome-based antigen presentation promotes light zone-like GC B cells. A) Flow cytometry plots of one representative sample showing the gating strategy for centrocytes and centroblasts. B) Representative gating of centrocytes and centroblasts for 300 ng mL⁻¹ antigen conditions on days 3 and 5. C) Quantification of the flow cytometry in (B). Plots show the percentages of GC-like B cells that are centrocytes or centroblasts at days 3 and 5. Anti-IgM concentration was either 60 or 300 ng mL⁻¹ and presented either soluble (red circle) or on liposomes (blue squares). N = 4; mean ± SEM; ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05; two-way ANOVA with Tukey’s post hoc comparison.

Figure 5.

Liposome-based antigen presentation regulates BCR clustering. A) Representative confocal microscopy images of unstimulated naïve B cells or cells incubated with anti-IgM antigen for 30 min. Stained for IgM BCR in red. Scale bar: 10 μm. B) Quantification of total BCR cluster area per cell from the images in (A). C) Quantification of cell diameter from the images in (A). N = 28–45; mean ± SEM; ***p < 0.001, **p < 0.01, *p < 0.05; one-way ANOVA with Tukey’s post hoc comparison.

Figure 6.

Liposome-based antigen presentation regulates IgG1 class-switching. A) Representative quadrant gating of class-switched GBCs (GL7+IgG1+) shows the expansion of this class-switched population over 5 days. B) Count of GL7+IgG1+ B cells per 200 000 starting cells at day 0, day 3, and day 5. C) Ratio of IgG1+GL7+ cell count to IgM+GL7+ cell count. D) Ratio of IgG1+ cell count to IgM+ cell count. Anti-IgM concentration was either 60 or 300 ng mL⁻¹ and presented on liposomes. N = 4; mean ± SEM; ****p < 0.0001, ***p < 0.001, **p < 0.01, *p < 0.05; two-way ANOVA with Tukey’s post hoc comparison.