TRAF3 is critical for initial T follicular helper cell specification via coordination of the IL-6R/IL-2R-BCL6 signaling nexus

Abstract

CD4⁺ T follicular helper (T_FH) cells are essential for orchestrating robust humoral immunity, yet the signals that initiate T_FH cell differentiation are not fully understood. We identified that the adapter protein TRAF3 was required for T_FH cell differentiation and function during systemic inflammatory infections. Loss of CD4⁺ T cell-intrinsic TRAF3 impaired chromatin remodeling and transcriptional programming essential for T_FH cell initiation and instead augmented T_H1 development and function. TRAF3-deficient CD4⁺ T cells exhibited altered interleukin-6 (IL-6) and IL-2 responsiveness, which were coupled to failures in BCL6 expression. Enforced expression of either IL-6 receptor or BCL6 or blockade of IL-2 signaling was sufficient to rescue T_FH cell differentiation. Human CD4⁺ T cells lacking TRAF3 exhibited impaired T_FH polarization, supporting a conserved mechanism by which TRAF3 regulates CD4⁺ T cell fate determination. Thus, TRAF3 functions at the nexus of cytokine, transcriptional, and epigenetic nodes that promote the T_FH cell specification during infection.

Fig. 1.. TRAF3 deficiency in CD4 T cells impairs T_FH cell differentiation and GC responses.

(A,B) Representative fluorescence-activated cell sorting (FACS) plots (left panels) and summary data (right panels) showing CXCR5⁺CD150^lo T_FH cells (A) and CXCR5⁺BCL6^hi GC T_FH cells (B), gated on splenic CD4⁺CD44^hiCD11a^hi T cells from WT and T-Traf3^−/− mice on day 8 p.i. (C) Summary data showing the normalized geometric mean fluorescence intensity (gMFI) of T_FH cell related markers. (D) Representative FACS plots (left panels) and summary data (right panels) showing splenic GL7⁺CD95⁺ GC B cells on day 8 p.i. (E) Representative confocal images showing B220⁺ B cell follicle (green) and GL7⁺ GC (red) of spleens from P. yoelii-infected mice on day 12 p.i. (Scale bars, 100 μm). White dashed line marks the follicular area. Each symbol represents a single GC structure analyzed from a single section from one of four independent recipient spleens. (F,G) Summary data showing the anti-parasite serum total IgG and IgG2b antibody end-point titers (EPT) on day 24 p.i. (F) and the kinetics of parasite burdens in male (top) and female mice (bottom) (G). (H) Experimental design. (I) Representative FACS plots and summary data showing the proportion of donor-derived WT and T-Traf3^−/− CD4⁺ T cells recovered from Tcrα^−/− recipient mice on day 16 p.i. (J) Representative FACS plots and summary data showing the proportion of splenic CD95⁺GL-7⁺ B cells induced by WT and T-Traf3^−/− CD4⁺ T cells. Data are representative of two (E,F,G,I,J) or three (A-D) independent experiments. Each symbol represents an individual mouse or structure, and summary data represent the mean ± SEM. P values were calculated by unpaired two-tailed t test. NS, not significant.

Fig. 2.. CD4⁺ T cell-intrinsic role of TRAF3 in T_FH cell differentiation and function.

(A) Experimental design for adoptive co-transfers of PbTII cells to congenic B6 recipient mice. (B,D) Representative FACS plots and summary data showing CXCR5⁺CD150^lo T_FH cells (upper panels) and CXCR5⁺BCL6^hi GC T_FH cells (lower panels), gated on splenic CD44^hiCD11a^hi WT and i-Traf3^−/− PbTII cells on day 8 and day 15 p.i., respectively. (C,E) Summary data showing the normalized gMFI of T_FH cell related markers on day 8 (C) and day 15 p.i. (E), respectively. (F) Representative confocal images showing B220⁺ B cell follicle (green), GL7⁺ GC (blue), and CD90.1/CD90.2⁺ PbTII cells (red) of spleens from Tcrα^−/− recipients on day 12 p.i. (Scale bars, 50 μm). White dashed line marks the follicular area. Each symbol represents a single GC structure analyzed from a single section from one of four independent recipient spleens. (G) Plots showing the frequency of WT and i-Traf3^−/− PbTII cells on day 8 and day 15 p.i., respectively. (H) Representative FACS plots and summary data showing CD4⁺Flica⁺ WT or i-Traf3^−/− PbTII cells on day 8 p.i. (I) Representative FACS plots and summary data showing CD4⁺Ki67⁺ WT and i-Traf3^−/− PbTII cells on day 8 p.i. Data are representative of two (D-F,H,I) or three (B,C,G) independent experiments. Each symbol represents an individual mouse and summary data represent the mean ± SEM. P values were calculated by unpaired two-tailed t test.

Fig. 3.. TRAF3 is required for early T_FH lineage determination and late T_FH cell maturation.

(A) Representative FACS plots showing CD44⁺ PbTII cells and CXCR5⁺BCL6⁺ PbTII T_FH precursor cells. WT or i-Traf3^−/− PbTII cells were labeled with cell trace violet (CTV) and adoptively transferred at 1×10⁶ cells/recipient, followed by P. yoelii infection. PbTII cells were identified in the recipient spleens and assessed for the activation marker CD44 expression and CTV dilution (division 0–9) on day 4 p.i. Proliferation index and expansion index of WT or i-Traf3^−/− PbTII cells were shown on the right panel. (B) Plots showing the proportion of early CXCR5⁺BCL6⁺ WT or i-Traf3^−/− PbTII T_FH cells from division 5 to 8 p.i. (C) Representative offset histograms showing the gMFI of WT or i-Traf3^−/− PbTII T_FH cell markers at the division 8. The dotted vertical lines mark signal strength by the staining of naive cells, and values denote gMFI. (D) Scatter plots showing the summary data of means ± SD of normalized gMFI of T_FH cell markers at the division 8. (E) Experimental design. (F) Scatter plots showing the proportion of WT or TRAF3^−/− PbTII cells on day 8 p.i during early deletion and late deletion of Traf3 by tamoxifen administration. (G,I) Representative FACS and summary data showing CXCR5⁺CD150^lo T_FH cells (G) and CXCR5⁺BCL6^hi GC T_FH cells (I) on day 8 p.i. during early or late deletion of Traf3 by tamoxifen administration, gated on splenic CD44^hiCD11a^hi PbTII cells. (H,J) Scatter plots showing the normalized gMFI of CXCR5 (H) and BCL6 (J) of WT and TRAF3^−/− PbTII cells during early and late deletion of Traf3. Data are representative of two independent experiments. Each symbol represents an individual mouse and summary data represent the mean ± SEM. P values were calculated by two-way ANOVA with Bonferroni multiple comparison test (B), unpaired two-tailed t test (D) or one-way ANOVA with Tukey’s multiple comparison test (F-J).

Fig. 4.. TRAF3 is critical for the T_FH transcriptional program.

(A) Venn diagram of RNA-Seq data showing overlap between SMARTA and PbTII DEGs of WT and i-Traf3^−/− CXCR5⁺CD150^lo T_FH cells sort-purified on day 8 p.i. by LCMV Armstrong and P. yoelii infection, respectively. (B) Heatmaps of selected common DEGs with normalized log₂ (fold change) ≥1 of WT and i-Traf3^−/− SMARTA and PbTII T_FH cells. The black lines denote gene expression of transcription factors (TFs). (C) Volcano plot showing select DEGs of T_FH and T_H1 lineage genes in i-Traf3^−/− versus WT SMARTA T_FH cells, with genes upregulated in TRAF3^−/− T_FH cells shown in red and those downregulated shown in blue. (D) Top enriched GO terms for DEGs among i-Traf3^−/− T_FH cells. The color scale indicates the enrichment adjusted P value of each GO term. (E) GSEA enrichment plots showing T_FH- and GC T_FH-associated gene set negatively enriched in i-Traf3^−/− T_FH cells. Data are representative of three independent experiments.

Fig. 5.. TRAF3 differentially regulates T_FH and T_H1 genes by either limiting or increasing chromatin accessibility.

(A) Venn diagram showing differential chromatin accessibility between WT and TRAF3-deficient T_FH cells, as determined by ATAC-seq. (B) Scatter plots displaying correlation of RNA-seq and ATAC-Seq datasets of the commonly regulated genes induced by P. yoelii and LCMV infection. (C). Graphs showing ATAC-seq tracks of naïve, WT and i-Traf3^−/− T_FH cells at select regulated gene loci and their normalized read counts determined by RNA-seq. (D) GSEA analysis of the DEG-associated chromatin accessibility (ChrAcc) sites. (E) Motif enrichment analysis of TF binding motifs, with indicated percentage of target genes containing differential accessibility peaks and enriched in consensus motifs of the top four TF family members shown. (F) Comparison of the average profiles of ATAT-seq signals at the top four TFs of WT and i-Traf3^−/− T_FH cells. Data are representative of three independent experiments.

Fig. 6.. Enforced IL-6 signaling rescues the defect of T_FH cells in TRAF3 deficient CD4⁺ T cells.

(A) Plots showing the proportion of IL6RA⁺ and gp130⁺ WT and i-Traf3^−/− PbTII cells on day 8 p.i. by P.yoelii. (B) Representative FACS plots and summary data showing the proportion of p-STAT3⁺ subsets in WT or i-Traf3^−/− PbTII cells incubated with medium alone (unsti.) or with 5 ng/ml recombinant human IL-6 (rhIL-6). (C) Representative FACS plots and summary data showing the gMFI of p-STAT3 in unstimulated or IL-6 stimulated WT or i-Traf3^−/− PbTII cells. (D) Representative FACS plots and summary data showing the proportion of p-STAT1⁺ subsets in WT or i-Traf3^−/− PbTII cells incubated with medium alone (unsti.) or 5 ng/ml rhIL-6. (E) Representative plot showing the proportion of p-STAT3⁺ subsets among empty vector (EV), IL-6RA- or gp130-overexpressing vector transduced WT, i-Traf3^−/− PbTII cells responding to rhIL-6 stimulation. (F,G) Representative FACS plots (F) and cumulative data (G) showing the proportion of WT and i-Traf3^−/− CXCR5⁺CD150^lo T_FH cells and CXCR5⁺BCL6^hi GC T_FH cells, which were gated on mcherry⁺ (pMSCV) and GFP⁺ (MigR1) PbTII cells. Data are representative of two (B-G) or three (A) independent experiments. Each symbol represents an individual mouse and summary data represent the mean ± SEM. P values were calculated by unpaired two-tailed t test (A,C) or one-way ANOVA with Tukey’s multiple comparison test (B,D,E,G). NS, not significant.

Fig. 7.. Overexpression of BCL6 rectifies T_FH cell defects in TRAF3 deficient PbTII cells.

(A) Representative FACS plots showing the CXCR5⁺CD150^lo T_FH cells, CXCR5^hiBCL6^hi and CXCR5^hiPD-1^hi GC T_FH cells gated on mcherry⁺ WT or i-Traf3^−/− PbTII cells that were transduced with empty vector (EV) or BCL6-overexpressing retrovirus on day 7 p.i. by P. yoelii. (B) Scatter plots showing the proportion of T_FH cells and GC T_FH cells gated on mcherry⁺ WT or i-Traf3^−/− PbTII cells. (C) Scatter plots showing the normalized gMFI of BCL6, CXCR5 and ICOS gated on EV or BCL6-overexpressing WT or i-Traf3^−/− PbTII cells. (D) Representative FACS plots and summary data showing the splenic GL7⁺CD95⁺ GC B cells on day 18 p.i. The ratio of frequency of GC B cells to PbTII CD4⁺ T cells is shown on the right panel. Data are representative of two independent experiments. Each symbol represents an individual mouse and summary data represent the mean ± SEM. P values were calculated by one-way ANOVA with Tukey’s multiple comparison test. NS, not significant.

Fig. 8.. Impaired T_FH polarization in TRAF3-deficient human CD4⁺ T cells.

(A) Experimental design. (B) TIDE assay on T_FH polarized human CD4⁺ T cells transfected with control sgRNA (sgCD8a) or TRAF3-targeting sgRNA (sgTRAF3). (C,D) Representative FACS plots and summary data showing the CXCR5⁺ICOS^hi, CXCR5⁺PD1^hi and CXCR5⁺BCL6^hi CD4⁺ T cells transfected with sgCD8a or sgTRAF3. (E) Graphs showing the gMFI of selected genes on T_FH polarized human CD4⁺ T cells transfected with sgCD8a or sgTRAF3. (F,G) Graphs showing the frequency of CD126⁺ CD4⁺ T cells (F) and the gMFI of CD126 (G). Each symbol represents an individual donor. P values were calculated by paired two-tailed t test.