Autoreactive helper T cells alleviate the need for intrinsic TLR signaling in autoreactive B cell activation

Abstract

T cells play a significant role in the pathogenesis of systemic autoimmune diseases, including systemic lupus erythematosus; however, there is relatively little information on the nature and specificity of autoreactive T cells. Identifying such cells has been technically difficult because they are likely to be rare and low affinity. Here, we report a method for identifying autoreactive T cell clones that recognize proteins contained in autoantibody immune complexes, providing direct evidence that functional autoreactive helper T cells exist in the periphery of normal mice. These T cells significantly enhanced autoreactive B cell proliferation and altered B cell differentiation in vivo. Most importantly, these autoreactive T cells were able to rescue many aspects of the TLR-deficient AM14 (anti-IgG2a rheumatoid factor) B cell response, suggesting that TLR requirements can be bypassed. This result has implications for the efficacy of TLR-targeted therapy in the treatment of ongoing disease.

Figure 1. Generation of IC-specific T cells using AM14 B cells and autoAb ICs.

(A) RF B cells have the potential to interact with a large repertoire of T cell specificities due to the complexity of autoAb ICs. (B) Experimental strategy. Hybridoma T cell lines originally stimulated with PL2-3 (C) or Y2:Sm (D) were cultured with SP APC or AM14 B cells in the presence the indicated Ab. IL-2 in the culture supernatant was measured 18 hours later by ELISA. Each graph is representative of at least 2 independent experiments.

Figure 2. Primary IC-specific T cells respond to autoAb ICs and enhance AM14 B cell proliferation and differentiation in vitro.

AM14 B cells were cultured with T cells from DO11.10 BALB/c mice or from 13C2 Rg BALB/c mice in the presence of PL2-3 (1 μg/ml) or BWR4 (10 μg/ml). All cells were labeled with VPD450. AM14 B cells were identified with the anti-idiotype Ab, 4-44. The percentage of divided 4-44⁺ cells (A) was determined by flow cytometry on day 4. (B) Representative flow cytometry plots of AM14 B cells on day 3. Cells were first gated as live, surface, and intracellular 4-44⁺. 4-44⁺ IgM (C) and IgG2a (E) AFCs on day 4 were determined using ELIspot assay. The concentration of 4-44⁺ IgM (D) and IgG2a (F) in the culture supernatant on day 4 was measured by ELISA. (G) Representative flow cytometry plots of DO11.10 T cells (TCRVβ6^–GFP^–) and 13C2 T cells (TCRVβ6⁺GFP⁺) from the same culture on day 5. Cells are first gated as live, 4-44 CD3e⁺. The concentrations of IL-2 (H) and IFN-γ (I) in the culture supernatant on day 4 were measured by ELISA. Data are combined from 3 or 4 independent experiments. Data are represented as the percentage of maximum for each experiment; mean ± SEM. Statistics were calculated with 2-way ANOVA (A and C–F) or 1-way ANOVA (H and I); multiple testing was corrected with Holm-Sidak’s. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 3. 13C2 T cells alleviate the need for intrinsic TLR signaling for AM14 proliferation but not Ab production.

VPD450-labeled AM14 B cells (WT, Tlr7^–/–, Tlr9^–/–, Tlr7/9^–/–) were cultured with VPD450-labeled T cells from DO11.10 BALB/c mice (top, A–C) or 13C2 Rg BALB/c mice (bottom, A–C, and D–F) in the presence of PL2-3 (1 μg/ml) or BWR4 (10 μg/ml). Analysis was performed on day 4. (A) The percentage of divided 4-44⁺ cells was determined by flow cytometry. (B) Total 4-44⁺ IgM and (C) IgG2a concentration in the culture supernatant. (D) The percentage of divided 13C2 T cells was determined by flow cytometry. (E) Concentrations of IL-2 in the culture supernatant. Data are represented as the percentage of maximum for each of 4 independent experiments; mean ± SEM. Statistics were calculated with 3-way ANOVA (A–C) or 2-way ANOVA (D–F); multiple testing was corrected with Holm-Sidak’s. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.

Figure 4. IC-specific T cells enhance AM14 B cell proliferation and increase AFC formation in vivo.

(A) Experimental design. (B and C) Representative flow cytometry plots as indicated. Left columns, cells were first gated as live. (D and E) Total 4-44⁺ cells (D) and 13C2 T cells (E) enumerated by flow cytometry. (F) Total 4-44⁺ PBs (CD138⁺CD44⁺intracellular 4-44^hi) enumerated by flow cytometry. 4-44⁺ IgM (G) and IgG2a (H) AFCs were determined by ELIspot. Data are from 1 experiment; bars indicate the mean, each point represents a mouse. Statistics were calculated with 1-way ANOVA; multiple testing was corrected with Holm-Sidak’s. *P < 0.05; **P < 0.01.

Figure 5. 13C2 T cells can adopt a Tfh cell phenotype and induce AM14 B cells to differentiate into germinal center cells.

(A) Representative flow cytometry plots from the experiment described in Figure 4; percentages reflect the mean. Cells were first gated as live and 4-44⁺. (B and C) Experimental design is as depicted in Figure 4A, except 5 × 10⁶ AM14 B cells and 2.5 × 10⁵ T cells were transferred. (B) Representative flow cytometry histograms of PBs (CD138⁺CD44⁺intracellular 4-44^hi) and germinal center (GC) cells (CD38^loCD95⁺) from mice that received AM14 B cells and PL2-3, without or with 13C2 T cells. Percentages reflect means from 4 experiments, with 14 and 16 total mice per group, respectively. (C) Representative flow cytometry plots and histograms gated on Rg T cells from mice that received AM14 B cells and 13C2 T cells, without and with PL2-3. Percentages reflect means from 6 experiments, with 8 and 25 total mice per group, respectively. All cells were first gated as live. (D–G) Immunofluorescent histology from the experiment described in Figure 4; representative of 2 similar experiments. (D and F) Representative images from mice that received AM14 B cells and PL2-3. (E and G) Representative images from mice that received AM14 B cells, 13C2 T cells, and PL2-3. Each column of images is from a different mouse. White arrows indicate 4-44⁺ PNA⁺ GC cells. Original magnification, ×20. Scale bar: 100 μm.

Figure 6. The effects of cognate T cell help for AM14 B cells depend on a threshold number of T cells and on T cell clone identity.

(A) Experimental design. (B) Representative flow cytometry plots. Cells were first gated as live and 4-44⁺. 4-44⁺ germinal center (GC) cells (B and C) and PBs (D and E) enumerated by flow cytometry. 4-44⁺ IgM (F) and (G) IgG2a AFCs determined by ELIspot. Data are presented as mean ± SEM from 2 independent experiments; each point represents a mouse, with 5–8 total mice per group. Statistics were calculated with 1-way ANOVA; multiple testing was corrected with Holm-Sidak’s. *P < 0.05; ***P < 0.001; ****P < 0.0001.

Figure 7. AM14 GC B cell development is not dependent on early T cell help.

(A) Experimental design. (B) Representative flow cytometry plots. Cells were first gated as live and 4-44⁺. 4-44⁺ germinal center (GC) cells (C) and PBs (D) enumerated by flow cytometry. Data are presented as mean ± SEM from 2 independent experiments; each point represents a mouse, with 3–7 total mice per group. Statistics were calculated with 1-way ANOVA; multiple testing was corrected with Holm-Sidak’s. *P < 0.05.

Figure 8. T cell clone 1B9 dramatically alters AM14 B cell accumulation and differentiation.

(A) Experimental design. (B) Representative flow cytometry plots as indicated. Cells were first gated as live and 4-44⁺. 4-44⁺ germinal center (GC) cells (C and F), PBs (D and G), and total 4-44⁺ cells enumerated by flow cytometry. 4-44⁺ IgM (H) and (I) IgG2a AFCs were determined by ELIspot. Data are presented as mean ± SEM from 4 independent experiments; each point represents a mouse, with 6–16 total mice per group. Statistics were calculated with 1-way ANOVA; multiple testing was corrected with Holm- Sidak’s. *P < 0.05; **P < 0.01; ****P < 0.0001.

Figure 9. 13C2, but not 1B9, T cells require AM14 B cells for activation with PL2-3 in vivo.

Experimental design was the same as described in Figure 6A using both T cell clones, 13C2 and 1B9. (A and B) Representative flow cytometry plots and histograms as indicated; percentages reflect the mean from 4 experiments. Cells were first gated as live. (C) Representative flow cytometry plots and histograms as indicated from 4 experiments. Cells were first gated as live and CD4⁺. (D) 13C2 and 1B9 Tfh cells (PD1⁺CXCR5⁺) enumerated from flow cytometry. Data are shown as mean ± SEM from 4 experiments, with 6–18 total mice per group. Statistics were calculated with 1-way ANOVA; multiple testing was corrected with Holm-Sidak’s. ****P < 0.0001.

Figure 10. IC-specific T cells partially bypass the requirement for intrinsic TLR signaling in the AM14 in vivo response.

(A) Representative flow cytometry plots and histograms from mice that received WT or Tlr7/9^–/– AM14 B cells. In the left columns for each genotype, cells were first gated as live. (B) Total 4-44⁺ cells enumerated from flow cytometry. 4-44⁺ germinal center (GC) cells (C) and PBs (D) enumerated by flow cytometry. 4-44⁺ IgM (E) and (F) IgG2a AFCs were determined by ELIspot. (G) Total 13C2 and 1B9 T cells enumerated by flow cytometry. Cells were first gated as live and CD4⁺. (H) 13C2 and 1B9 Tfh cells (PD1⁺CXCR5⁺) enumerated from flow cytometry. Data are presented as mean ± SEM from 4 independent experiments, with 7–16 total mice per group; each point represents a mouse. Statistics were calculated with 2-way ANOVA; multiple testing was corrected with Holm-Sidak’s. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001.