BCL6 controls contact-dependent help delivery during follicular T-B cell interactions

Abstract

BCL6 is required for development of follicular T helper (Tfh) cells to support germinal center (GC) formation. However, it is not clear what unique functions programmed by BCL6 can explain its absolute essentiality in T cells for GC formation. We found that ablation of one Bcl6 allele did not appreciably alter early T cell activation and follicular localization but inhibited GC formation and Tfh cell maintenance. BCL6 impinged on Tfh calcium signaling and also controlled Tfh entanglement with and CD40L delivery to B cells. Amounts of BCL6 protein and nominal frequencies of Tfh cells markedly changed within hours after strengths of T-B cell interactions were altered in vivo, while CD40L overexpression rectified both defective GC formation and Tfh cell maintenance because of the BCL6 haploinsufficiency. Our results reveal BCL6 functions in Tfh cells that are essential for GC formation and suggest that BCL6 helps maintain Tfh cell phenotypes in a T cell non-autonomous manner.

Keywords: BCL6; CD40L; T-B interactions; Tfh cells; calcium signaling; follicle; germinal center; intravital imaging; lymph node; stim1.

Figure 1

Haploinsufficiency of BCL6 impairs the follicular response (A) Representative histograms and MFIs of surface CXCR5 expression on OT-II T cells of indicated genotypes in draining lymph nodes 4 days after NP-OVA immunization. Each symbol represents one experiment in which cells from 2–3 mice of each group were pooled. n.s., not significant. (B and C) Representative distribution patterns (B) and FRIs (C) of GFP-transduced Bcl6^+/+ and Bcl6^+/− OT-II T cells in draining lymph nodes as in (A). Each symbol represents one follicle and its adjacent T cell zone, and lines denote means. One of 2 experiments with similar results is shown, and each group had 3 mice per experiment. n.s., not significant. Scale bar, 100 μm. (D and E) Representative flow-cytometry profiles and frequencies of FAS^hi GL7^hi GC B cells in total CD19⁺ B cells (D) or flow-cytometry profiles and frequencies of CXCR5^hi PD1^hi Tfh cells in OT-II T cells (E) 8 days after NP-OVA immunization in Sap^−/− mice that received OT-II T cells of indicated genotypes. Each symbol represents one mouse, and lines denote the means. Data are pooled from 3 independent experiments, with at least three mice per group per experiment. ^∗∗p < 0.01. See also Figure S1.

Figure 2

BCL6 is required for efficient T-B cell entanglement inside the follicle (A–C) GFP-expressing MD4 B cells, control CFP-expressing Bcl6^+/+ and dsRed-expressing test Bcl6^+/−, or test Bcl6^+/+ OT-II T cells were visualized 4 days after immunization. (A) Bcl6^+/+ and Bcl6^+/− OT-II T cells interacting with MD4 B cells, with contacts highlighted by arrowheads. Scale bar, 20 μm. See corresponding Video S4. Duration (B) and SEI (C) of contacts of indicated types. See corresponding Videos S3 and S4. Each symbol is one contact, and lines denote the means. Data are pooled from three independent experiments. ^∗∗∗p < 0.001. See also Figure S3.

Figure 3

BCL6 is required for optimal calcium signaling during T-B cell entanglement in vivo (A–C) dsRed-expressing MD4 B cells and YC-Nano-50^CD-transduced Bcl6^+/+ or Bcl6^+/− OT-II T cells were visualized in B6 recipients 4 days after immunization. (A) Image sequences showing a Bcl6^+/+ and a Bcl6^+/− OT-II T cell interacting with MD4 B cells. CFP fluorescence from the YC-Nano-50^CD reporter identifies T cells in the fluorescence overlays (columns 1 and 3), and ratiometric FRET signals are presented in the heatmap images (columns 2 and 4). Scale bar, 20 μm. See corresponding Videos S5 and S6. (B and C) Mean ΔR_t/R₀ (B) and time-integrated calcium index (C) of individual contacts between Bcl6^+/+ (n = 65) or Bcl6^+/− (n = 64) OT-II T cells and MD4 B cells. Lines denote the means. Data are pooled from four experiments. ^∗∗p < 0.01.

Figure 4

BCL6 is required for efficient CD40L delivery OT-II T cells of indicated genotypes, 7–8 days after being activated in vivo in adoptive B6 hosts immunized with NP-OVA, were subjected to the CD40L mobilization assay as described in the STAR Methods. (A) Representative histograms of cell surface CD40L (left) or total CD40L measured by intracellular staining (right) of Bcl6^+/+ and Bcl6^+/− OT-II T cells. (B) MFIs of CD40L staining for conditions in (A); each line represents one independent experiment for each comparison setup. (C and D) CD40L mobilization by anti-CD3 in Bcl6^+/+ and Bcl6^+/− OT-II T cells; representative histogram overlays (C) and changes in surface CD40L MFI (ΔMFI) following indicated stimulation (D). Each line represents one independent experiment. n.s. not significant, ^∗∗p < 0.01.

Figure 5

Haploinsufficiency of BCL6 impairs a polyclonal response (A) The experimental scheme. (B and C) Representative flow-cytometry profiles and frequencies of FAS^hi GL7^hi GC B cells in total B220⁺ B cells (B) or CXCR5^hiPD-1^hi Tfh cells in CD44^hi CD4 T cells of the donor origin (C) 13 days after NP-KLH immunization in CD45.1 Sap^−/− recipients. All donors are on the CD4-Cre background, and genotypes of the Bcl6 allele are indicated. Each symbol represents one mouse, and lines denote the means. Data are pooled from 2 independent experiments. n.s., not significant. ^∗p < 0.05.

Figure 6

Maintenance of BCL6 and other Tfh cell markers requires constant T-B cell interactions (A–C) CXCR5, PD-1, and BCL6 expression after acute antigen stimulation and T-B cell cooperation blockade in vivo. (A) Histograms and MFI statistics of indicated markers on transferred OT-II T cells 15 h after PBS, αDEC-OVA, or αCD40L injection. (B) Frequencies of OT-II Tfh cells gated as CXCR5^hi PD1^hi cells. (C) BCL6 expression by CXCR5^hi PD1^hi OT-II Tfh cells as gated in (B). Each symbol in scatterplots denotes one mouse. One of four experiments with similar results is shown. ^∗p < 0.05, ^∗∗∗p < 0.001, ^∗∗∗∗p < 0.0001.

Figure 7

Rescue of BCL6-insufficient Tfh cells secondary to GC rectification (A) Representative flow-cytometry profiles (left) and frequencies of IgM^a+ MD4 GC B cells in total CD19⁺ B cells (right) 5 days after HEL-OVA immunization of Sap^−/− mice that received transfer of MD4 B cells and Bcl6^+/+ or Bcl6^+/− OT-II T cells transduced with a control GFP or CD40L-expressing vector. Each symbol in the scatterplot (right) represents one mouse, and lines denote the means. One of three independent experiments with similar results is shown. ^∗∗p < 0.01, ^∗∗∗p < 0.001. (B) Representative flow-cytometry profiles (left) and frequencies of CXCR5^hi PD1^hi Tfh cells in OT-II T cells as in (A). Each symbol shape represents one mouse, and lines denote the means. One of three independent experiments with similar results is shown. ^∗p < 0.05.