Follicular helper T cells are required for systemic autoimmunity

Abstract

Production of high-affinity pathogenic autoantibodies appears to be central to the pathogenesis of lupus. Because normal high-affinity antibodies arise from germinal centers (GCs), aberrant selection of GC B cells, caused by either failure of negative selection or enhanced positive selection by follicular helper T (T(FH)) cells, is a plausible explanation for these autoantibodies. Mice homozygous for the san allele of Roquin, which encodes a RING-type ubiquitin ligase, develop GCs in the absence of foreign antigen, excessive T(FH) cell numbers, and features of lupus. We postulated a positive selection defect in GCs to account for autoantibodies. We first demonstrate that autoimmunity in Roquin(san/san) (sanroque) mice is GC dependent: deletion of one allele of Bcl6 specifically reduces the number of GC cells, ameliorating pathology. We show that Roquin(san) acts autonomously to cause accumulation of T(FH) cells. Introduction of a null allele of the signaling lymphocyte activation molecule family adaptor Sap into the sanroque background resulted in a substantial and selective reduction in sanroque T(FH) cells, and abrogated formation of GCs, autoantibody formation, and renal pathology. In contrast, adoptive transfer of sanroque T(FH) cells led to spontaneous GC formation. These findings identify T(FH) dysfunction within GCs and aberrant positive selection as a pathway to systemic autoimmunity.

Figure 1.

Heterozygosity for Bcl6 reduces the magnitude of the GC response in Roquin^+/+ and Roquin^san/san mice and ameliorates the lupus-like phenotype of Roquin^san/san mice. (a) Flow cytometric contour plots (left) and graphical analysis (right) of B220⁺GL-7⁺CD95⁺ GC B cells in 10-wk-old wild-type (Bcl6^+/+) and Bcl6^+/− mice 8 d after SRBC immunization (P = 0.0011). Data are representative of four independent experiments (n = 4 per group). (b) Flow cytometric contour plots (left) and dot plots (right) showing B220⁺GL-7⁺CD95⁺ GC B cells from 10-wk-old naive Roquin^san/san Bcl6^+/+ and Roquin^san/san Bcl6^+/− mice. Data are representative of five independent experiments (n ≥ 4 per group). (c) Representative determination of serum IgG anti-dsDNA from 6-mo-old female Roquin^+/+ Bcl6^+/+, Roquin^san/san Bcl6^+/+, and Roquin^san/san Bcl6^+/− mice, determined by immunofluorescence staining of C. luciliae substrate. Data shown reflect the occurrence (n ≥ 6 mice per group); three out of six Roquin^san/san Bcl6^+/− mice had low intensity staining (illustrated in the fourth panel from left), and three out of six were negative (illustrated in the third panel from left). (d) Representative images of kidney sections stained with H&E (left) or viewed under an electron microscope (right) from 6-mo-old mice of the indicated genotypes. Roquin^san/san animals show widespread mesangial proliferative lesions with moderate interstitial infiltrate. There are multiple electron-dense deposits (arrows). Histological changes in Roquin^san/san Bcl6^+/− mice were mild, with occasional electron-dense deposits visible on electron microscopy in two individuals (far right). Images are representative (n ≥ 4 per group). Bars: (H&E) 100 µm in all panels; (electron microscopy) 5 µm in the Roquin^+/+ Bcl6^+/+ and Roquin^san/san Bcl6^+/+ panels, 2 µm in the left Roquin^san/san Bcl6^+/− panels, and 10 µm in the right Roquin^san/san Bcl6^+/− panel. (e) Nephritis severity score of 6-mo-old female Roquin^+/+, Roquin^san/san, and Roquin^san/san Bcl6^+/− mice as determined by histological analysis according to the criteria given in Table S1 (available at http://www.jem.org/cgi/content/full/jem.20081886/DC1). Horizontal bars indicate medians. In a, b, and e, each symbol represents one mouse; p-values are indicated on the graphs, and the numbers in the plots represent percentages.

Figure 2.

The autoimmune phenotype of Roquin^san/san mice requires T cell activation through CD28, and T_FH cells are expanded cell autonomously. (a, left) Detection of IgG-ANA using Hep-2 slides in sera from 8-wk-old female Roquin^san/san and Roquin^san/san Cd28^−/− mice (n = 5 per group). (right) Detection of anti-dsDNA IgG serum antibodies in 6-mo-old female Roquin^san/san and Roquin^san/san Cd28^−/− mice determined by staining C. luciliae slides. Data are representative of three independent experiments (n ≥ 5 per group). (b) Score of nephritis severity in 6-mo-old female Roquin^+/+ Roquin^san/san and Roquin^san/san Cd28^−/− mice as determined by histological analysis defined by the criteria given in Table S1 (available at http://www.jem.org/cgi/content/full/jem.20081886/DC1). Each symbol represents one mouse. Horizontal bars indicate medians. (c) Representative images of kidney sections stained with H&E (left) or viewed under an electron microscope (right). Histology from Roquin^san/san Cd28^−/− animals was much less severe, with normal H&E appearances and few electron-dense deposits (arrows) in the mesangium. Bars: (left) 100 µm; (right) 10 µm. (d) Representative flow cytometric contour plots of CD4⁺CXCR5⁺PD-1^high T_FH cells in 10-wk-old Roquin^san/san mice and control littermates. Data are representative of five independent experiments (n = 4 per group), and the numbers in the plots represent percentages. (e) Dot plots representing percentages of CD4⁺CXCR5⁺PD-1^high T_FH cells from SRBC-immunized chimeric mice generated by reconstituting sublethally irradiated mice with a 1:1 mix of either Roquin^+/+.Ly5^a/Roquin^+/+.Ly5^b (left) or Roquin^+/+.Ly5^a/ Roquin^san/san.Ly5^b (right). Data are representative of three independent experiments (n = 4 per group). Each symbol represents the Ly5^a or Ly5^b population derived from one mouse. (f) ELISA was used to determine culture supernatant IL-21 levels from 24-h splenocyte cultures from Roquin^san/san mice and littermate controls in the presence of PMA and ionomycin. Error bars indicate means ± SEM. Data are representative of three independent tests where each sample was run in triplicate.

Figure 3.

Spontaneous GC and T_FH formation are corrected by loss of SAP in Roquin^san/san mice. (a) CD4⁺CXCR5⁺PD-1^high T_FH cells in unimmunized 10-wk-old Roquin^san/san and Roquin^san/san Sap^−/− mice (P = 0.0056). Representative flow cytometric contour plots are shown (right). Data are representative of four independent experiments (n = 4 per group). (b) B220⁺GL-7⁺CD95⁺ GC B cells in unimmunized 10-wk-old Roquin^san/san Sap^+/+ and Roquin^san/san Sap^−/− mice (P = 0.0007). Representative flow cytometric contour plots are shown (right). Data are representative of four independent experiments (n = 4 per group). (c) Photomicrographs of frozen spleen sections from unimmunized 6-mo-old mice of the indicated genotypes stained with IgD (brown; all panels), PNA (blue; left), TCRβ (blue; middle), and PD-1 (blue; right). Bars, 200 µm.

Figure 4.

Th1 and Th2 cells are present in Roquin^san/san mice in the absence of SAP and T_FH cells, but not non-T_FH effector cells, induce a GC response in wild-type mice. (a) Representative flow cytometric contour plots and (b) graphical analysis of GATA3⁺CD44^highCD4⁺ and Tbet⁺CD44^highCD4⁺ cells in mice of the indicated genotypes. Data are representative of two independent experiments (n ≥ 4 per group). (c) Representative dot plots of lymph node CD4⁺PD-1^highCXCR5⁺ (left), Tbet⁺CD44 ^highCD4⁺ (middle), and GATA3⁺CD44^highCD4⁺ (right) cells. (d) Experimental outline for adoptive transfer of Roquin^san/san CD4⁺ CD45.2 CD44^high PD-1^high CXCR5⁺ or CD4⁺ CD45.2 CD44^high PD-1⁻ CXCR5⁻ T cells into CD45.1 C57BL/6 mice. (e) Flow cytometric contour plots and (f) dot plots of B220⁺ GL-7⁺ CD95⁺ GC B cells from CD45.1 C57BL/6 recipients 3 wk after adoptive transfer of the indicated cell type. Data were generated from three mice per group (**, P > 0.001). In a, d, and e, the numbers in the plots represent percentages.

Figure 5.

Serum autoantibodies and renal pathology in Roquin^san/san Sap^−/− mice and reduction of the GC response in SAP-deficient Roquin^san/san mice is caused by B cell–extrinsic factors. (a, left) Representative staining of Hep-2 slides for detection of IgG ANA in the serum of 8-wk-old female Roquin^san/san and Roquin^san/san Sap^−/− mice (n = 5 per group). (right) Detection of IgG anti-dsDNA serum antibodies in 6-mo-old female Roquin^san/san and Roquin^san/san Sap^−/− mice determined by immunofluorescence staining using C. luciliae substrate. Data are representative of three independent experiments (n ≥ 5 per group). (b) Representative images of kidney sections stained with H&E (top) or viewed under an electron microscope (bottom) from 6-mo-old female Roquin^san/san Sap^−/− mice. Roquin^san/san Sap^−/− mice show slight mesangial expansion on H&E staining but no electron-dense deposits. Bars: (top) 100 µm; (bottom) 5 µm. (c) Score of nephritis severity in 6-mo-old female Roquin^+/+, Roquin^san/san, and Roquin^san/san Sap^−/− mice as determined by histological analysis according to the criteria given in Table S1 (available at http://www.jem.org/cgi/content/full/jem.20081886/DC1). Each symbol represents one mouse. Horizontal bars indicate medians. (d) Gating strategy for assessing the HEL-specific GC response (top) and graphic representation (bottom) of the total number of HEL-specific GC cells per spleen in mice with the indicated genotypes 7 d after cotransfer of SW_HEL B cells and HEL^2x-conjugated SRBCs. Data are representative of three independent experiments (n ≥ 4). Each symbol represents one mouse. (e) Gating strategy for determining HEL-specific extrafollicular plasma cells (top) and graphical representation (bottom) of the number of HEL-specific plasma cells 5 d after cotransfer into mice with the indicated genotypes. Data are representative of two independent experiments (n ≥ 3 per group). p-values are indicated on graphs.

Figure 6.

T_FH cell–associated molecules are decreased in Roquin^san/san Sap^−/− mice. (a) Representative flow cytometric histograms and (b and c) graphical analysis showing ICOS mean fluorescence intensity (MFI) of splenic naive CD44^low (b) and activated/memory CD44^high (c) CD4⁺ T cells from 10-wk-old unimmunized mice of the indicated genotypes. Data are representative of three independent experiments. Each symbol represents one mouse. (d) ELISA quantification of IL-21 in supernatant from an overnight culture of splenocytes in the presence of PMA and ionomycin from mice of the indicated genotypes. Data are representative of three experiments. (e) Flow cytometric contour plots of CD40L expression on splenocytes, 5 h after stimulation with anti-CD3 and anti-CD28, derived from mice of the indicated genotypes. (left) Staining with an isotype control; (right) CD40L staining. (f) Histograms of the percentage of CD4⁺ cells that express CD40L (gated as shown in e) 5 h after CD3 and CD28 stimulation on splenocytes from mice with the indicated genotypes. In d and f, error bars indicate means ± SEM.

Figure 7.

Lack of IL-21 does not affect the phenotype, T_FH cell accumulation, or GC formation of Roquin^san/san mice. (a) IgG ANAs in the serum of mice of the genotypes indicated, detected by immunofluorescence using Hep-2 substrate. Data are representative of three independent experiments (n ≥ 5 mice per group). (b) Score of ANA staining intensity by confocal microscopy from sera taken from mice of the indicated genotypes. Data are representative of three independent experiments (n ≥ 5 mice per group). (c) Basal serum total IgG, IgG1, and IgE measured by ELISA. Data are representative of two independent experiments (n = 5 mice per group). (d) Lymph node and spleen weight in grams for mice of the indicated genotypes. Data are representative of two independent experiments (n ≥ 5 per group). (e) Flow cytometric contour plots and dot plots of PD-1^highCXCR5⁺CD4⁺ T_FH cells and (f) GL-7⁺CD95⁺B220⁺ GC cells from mice of the indicated genotypes. Data are representative of two independent experiments (n ≥ 5 mice per group). In e and f, the numbers in the plots represent percentages.