DOCK8-expressing T follicular helper cells newly generated beyond self-organized criticality cause systemic lupus erythematosus

Abstract

Pathogens including autoantigens all failed to induce systemic lupus erythematosus (SLE). We, instead, studied the integrity of host's immune response that recognized pathogen. By stimulating TCR with an antigen repeatedly to levels that surpass host's steady-state response, self-organized criticality, SLE was induced in mice normally not prone to autoimmunity, wherein T follicular helper (Tfh) cells expressing the guanine nucleotide exchange factor DOCK8 on the cell surface were newly generated. DOCK8⁺Tfh cells passed through TCR re-revision and induced varieties of autoantibody and lupus lesions. They existed in splenic red pulp and peripheral blood of active lupus patients, which subsequently declined after therapy. Autoantibodies and disease were healed by anti-DOCK8 antibody in the mice including SLE-model (NZBxNZW) F1 mice. Thus, DOCK8⁺Tfh cells generated after repeated TCR stimulation by immunogenic form of pathogen, either exogenous or endogenous, in combination with HLA to levels that surpass system's self-organized criticality, cause SLE.

Keywords: Cell biology; Immune response; Immunology.

Graphical abstract

Figure 1

Identification of DOCK8 as marker of aiCD4 T cell and induction of SLE by DOCK8⁺CD4 T cells (A) Expression of DOCK8 on CD45RB^lo CD122^lo PD-1⁺ CD4 T cells. Membrane and cytosolic fractions of either whole T cells (Lanes 1 and 5), CD45RB^hi CD122^hi CD4 T cells (Lanes 2 and 6), CD45RB^lo CD122^lo PD-1^- CD4 T cells (Lanes 3 and 7), or CD45RB^lo CD122^lo PD-1⁺ CD4 T cells (Lanes 4 and 8) from 12x OVA-immunized BALB/c mice, were subjected to 10% of polyacrylamide gel electrophoresis, transferred to Immobilon-P membrane, and stained with anti-DOCK8 Ab. (B) Increase of DOCK8⁺CD4 T cells after repeated immunization with OVA without an adjuvant. Flow cytometry analyses of DOCK8⁺CD4 T cell from splenic CD4 T cells of BALB/c mice after repeated 12x stimulation with PBS, 6x with OVA, or 12x with OVA. (C) DOCK8⁺CD4 T cells transferred into recipient mice induced autoantibodies. Rheumatoid factor (RF), anti-Sm Ab, and anti-dsDNA Ab were quantified in the sera of BALB/c mice that had been pre-immunized 8x with OVA, depleted of CD4 T cells by anti-CD4 Ab, and then inoculated with DOCK8⁻CD4 T cells or DOCK8⁺CD4 T cells from 12x OVA-immunized BALB/c mice. The control group was BALB/c mice pre-immunized 8x with OVA, CD4 T cell-depleted by anti-CD4 Ab, and inoculated with CD4 T cells from 12x PBS-immunized BALB/c mice. Anti-dsDNA Ab and anti-Sm Ab titers were represented by arbitrary unit (AU). Data were represented as mean ± SEM. Statistical assessment was by Student's t-test. ∗p<0.05. (D) Generation of CTL, i.e., IFNγ⁺ CD8 T cells, after transfer of DOCK8⁺CD4 T cells. BALB/c mice were immunized 12x with OVA, and CD4 T cells were isolated. Cells were transferred into the anti-CD4 Ab-treated recipient mice immunized 8x with OVA. Matured CTL, i.e., IFNγ⁺ CD8 T cells, were measured 2 weeks after booster immunization 1x with OVA (n = 5). Data were represented as mean ± SEM. Statistical assessment was by Student's t-test. ∗p<0.05. (E) Skin lesion. a: Moderate dermatitis with dermal fibrosis, b: liquefaction degeneration (arrow) in the basal cells of epidermis, and c: panniculitis in the deep area of the dermis of mice pre-treated 8x with OVA, depleted of CD4 T cells, and inoculated with DOCK8⁺ T cells from 12x OVA-immunized mice. d: Discoid lupus-like advanced dermatitis observed in mice immunized 12x with OVA. e: Healed dermatitis with some fibrosis in 12x OVA-immunized mice treated with anti-DOCK8 Ab 24 h each before the 6x, 8x, 10x, and 12x OVA immunizations with OVA. H & E stain. (bar = 30μm; original magnification a, b, ×400; c, x300; d, e, x100). (F) Lesion other than kidney or skin. a: lung interstitial pneumonitis, b: interstitial pneumonitis accompanied by angiitis, c: liver pericholangitis, d: onion-skin lesion with amyloid deposition classical to lupus in the spleen, and f: thyroiditis, accompanied by giant cells, in the BALB/c mice 8x pre-immunized with OVA, CD4 T cell-depleted, and then inoculated with DOCK8⁺ CD4 T cells from 12x OVA-immunized BALB/c mice. e: Classical onion-skin-like lesion with amyloid deposits in the spleen, and g: perineuritis observed in the dermis of the BALB/c mice immunized 12x with OVA. H & E stain. (bar = 30μm; original magnification ×200). See also Figures S1–S3.

Figure 2

DOCK8⁺ CD4 T cell as T fh cell located at splenic red pulp, and its increase in peripheral blood of patients with SLE (A) a: Morphology of DOCK8⁺CD4 T cell observed by immunoelectron microscopy, stained with rabbit anti-DOCK8 Ab and Au-tagged ant-rabbit IgG Ab. (bar = 1μm; original magnification ×12,000). b,c: Nuclear membrane portions of DOCK8⁺ CD4 T cell, stained likewise. (bar = 1μm; original magnification ×18,000; x24,000). (B) Markers of splenic DOCK8⁺CD4 T cells. Flow cytometry for ICOS, CXCR5, and PD-1 expression in splenic DOCK8⁺CD4 T cells taken 9 days after the final immunization of BALB/c mice immunized 12x with PBS versus OVA (Left). Flow cytometry analysis of ICOS, CXCR5, PD-1, GATA3, RORγt, T-bet, Bcl6, Ly6C, LFA1, FR4, and GL7 expression in splenic DOCK8⁻CD4 T cells and DOCK8⁺CD4 T cells isolated from 12x OVA-immunized BALB/c mice. (C) Cytokines released in culture supernatants of 1x10⁶/ml DOCK8⁺ or DOCK8^- CD4 T cells following incubation with 2 μg/mL of anti-CD3 and 5 μg/mL of anti-CD28 Ab at 37°C for 24hr. Data were represented as mean ± SEM. Statistical assessment was by Student's t-test. ∗p<0.05, ∗∗p<0.01, ∗∗∗p<0.005, ∗∗∗∗p<0.001. (D) DOCK8⁺CD4 T cells in the autopsied spleen of untreated patient with active SLE who died before starting therapy. Tissues were fixed in 10% formalin, deparaffinized by graded xylene, and graded alcohol sequentially, and serially treated with biotinylated anti-DOCK8 monoclonal Ab and rabbit anti-IFNγ Ab (for T cell staining), alkaline phosphatase avidin D, and anti-rabbit Novolink, followed by reaction with Fast Red (Red for DOCK8⁺ cells) and Histogreen (Blue for T cells). S: red pulp sinus. A; central artery. P: penicillar artery. (bar = 100μm; original magnification ×100). Inset; Magnified DOCK8⁺CD4 T cells (Arrows). (bar = 20μm; original magnification ×400). (E) Circulating DOCK8⁺CD4 T cells within CD4 T cells in the peripheral blood of patients with SLE and its relationship to the SLE disease activity index (SLEDAI). Data of control patients were depicted as; ■ rheumatoid arthritis, ▲ mixed connective tissue disease, △ polymyositis, ◆ systemic sclerosis, □ microscopic polyangiitis. Also shown was quantification of interferonα (IFNα) in sera of the same patients with SLE measured by ELISA. See also Figures S2 and S4–S6.

Figure 3

TCR revision in DOCK8⁺CD4 T cells (A) Expression of V(D)J recombinase complex and related molecules in splenic CD4 T cells from 12x PBS-immunized BALB/c mice, and in DOCK8⁻CD4 T cells and DOCK8⁺CD4 T cells from 12x OVA-immunized BALB/c mice. Two mice were analyzed for each group. (B) Western blot analysis of TCR signal transduction molecules in splenic CD4 T cells from 12x PBS-immunized BALB/c mice, and in DOCK8⁻CD4 T cells and DOCK8⁺CD4 T cells from 12x OVA-immunized BALB/c mice. Three mice were analyzed for each group. GAPDH, control housekeeping glyceraldehyde 3-phosphate dehydrogenase gene. (C) TCR repertoire analysis. Heatmap visualization of TCR usage of a combination of TRV and TRJ genes in the splenic DOCK8^- CD4 T cells and DOCK8⁺ CD4 T cells from 12x OVA-immunized mice. See also Figures S7 and S8.

Figure 4

Autoantibody microarray study Autoantibodies produced in BALB/c mice pre-immunized 8x with OVA, depleted of CD4 T cells, and inoculated with DOCK8⁻CD4 T cells or DOCK8⁺CD4 T cells from 12x OVA-immunized BALB/c mice. Control was autoantibodies produced in BALB/c mice pre-immunized 8x with OVA, depleted of CD4 T cells, and inoculated with CD4 T cells from 12x PBS-immunized BALB/c mice. Two mice were analyzed for each group.

Figure 5

Expression of transcriptional regulators in DOCK8⁺CD4 T cells (A) Transcriptional regulators responsible for TCR revision. Western blot analysis of DOCK8 and transcriptional regulators in splenic CD4 T cells from 12x PBS-immunized BALB/c mice, and in DOCK8⁻CD4 T cells and DOCK8⁺CD4 T cells from 12x OVA-immunized BALB/c mice. Three mice were analyzed for each group. (B) Upregulated genes in DOCK8⁺CD4 T cells. Results of gene expression microarray analysis, abstract from Supplemental Information (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE159240). Ratio of mRNA expression of splenic DOCK8⁺CD4 T cells from 12x OVA-immunized BALB/c mice in comparison with splenic DOCK8⁻CD4 T cells from 12x OVA-immunized mice, given as a log₁₀ scale, wherein 2, 3, 4, 5, and 6 indicated x10², x10³, x10⁴, x10⁵, and x10⁶, respectively.

Figure 6

Enhanced chromatin accessibility in DOCK8⁺CD4 T cells (A) Chromatin-immunoprecipitation sequencing (ChIP-seq) assay for V(D)J chromatin accessibility. Left panel: Histone acetylation (AcH3) and H3K4 methylation (H3K4me3) of the TCRA gene of splenic aiCD4 T cells from 8x SEB-stimulated BALB/c mice. Splenic CD4 T cells of 8x SEB-stimulated BALB/c mice which produced amply high titers of autoantibodies were assayed. TEA, TCRα promoter. Eα, TCRα enhancer. MyoD, myoblast determination 1 as negative controls. Right panel: Histone acetylation (AcH3) and H3K4 methylation (H3K4me3) of the TCRB gene of splenic aiCD4 T cells from 8x SEB-stimulated BALB/c mice as above. Eβ, TCRβ enhancer. (B) TCRα chain revision in the splenic CD4 T cells from the mice immunized 8x with SEB producing high titers of autoantibodies was determined by LM-PCR detection of dsDNA breaks at the RSS flanking the TCRAJ12. WT, wild-type.

Figure 7

Treatment with anti-DOCK8 Ab (A) Effect of anti-DOCK8 mAb treatment on autoantibody production in12x OVA-immunized mice. Autoantibodies, rheumatoid factor (RF), anti-Sm Ab, and anti-dsDNA Ab, were quantified in the sera of 12x PBS-immunized or 12x OVA-immunized mice: Mice received anti-DOCK8 Ab (100μg) or control rabbit IgG (100μg) 24 h each before the 6x, 8x, 10x, and 12x immunizations with OVA. Anti-Sm Ab and anti-dsDNA Ab titers were represented by arbitrary unit (AU). Data were represented as mean ± SEM. Statistical analysis was by Student's t-test; ∗p<0.05, ∗∗p<0.01, ∗∗∗p<0.005. (B) Effects of anti-DOCK8 Ab treatment on renal disease. Proteinuria measured 9 days after the final immunization in 12x PBS-immunized or 12x OVA-immunized BALB/c mice, graded with a score of 0 (<30 mg/dL); 1 (30–99 mg/dL); 2 (100–299 mg/dL); or 3 (300–999 mg/dL). (C) Effects of anti-DOCK8 mAb treatment on autoantibody production and renal disease in (NZBxNZW) F1 female mice. Autoantibodies, anti-dsDNA Ab, anti-Sm Ab, and rheumatoid factor (RF), and proteinuria quantified by ELISA in (NZBxNZW) F1 female mice, which were treated either with anti-DOCK8 mAb (100 μg) or control rabbit IgG (100 μg) every week from 15 to 24 weeks (arrow). Data were represented as mean ± SEM. Statistical assessment performed by Student's t-test; ∗p<0.05, ∗∗p<0.01, ∗∗∗∗p<0.001.