Systemic lupus erythematosus favors the generation of IL-17 producing double negative T cells

Abstract

Mature double negative (DN) T cells are a population of αβ T cells that lack CD4 and CD8 coreceptors and contribute to systemic lupus erythematosus (SLE). The splenic marginal zone macrophages (MZMs) are important for establishing immune tolerance, and loss of their number or function contributes to the progression of SLE. Here we show that loss of MZMs impairs the tolerogenic clearance of apoptotic cells and alters the serum cytokine profile, which in turn provokes the generation of DN T cells from self-reactive CD8⁺ T cells. Increased Ki67 expression, narrowed TCR V-beta repertoire usage and diluted T-cell receptor excision circles confirm that DN T cells from lupus-prone mice and patients with SLE undergo clonal proliferation and expansion in a self-antigen dependent manner, which supports the shared mechanisms for their generation. Collectively, our results provide a link between the loss of MZMs and the expansion of DN T cells, and indicate possible strategies to prevent the development of SLE.

Fig. 1. Marginal macrophage depletion promotes the expansion of DN T cells.

a–g Age-matched female B6.lpr mice were treated with either PBS-loaded control liposomes (PBSLs) or Clodronate liposomes (CLs, 100 µg/mouse) every other week for 2 months starting at 10 weeks of age. Naive B6 mice were used as controls. n = 5–6 mice per group for two independent experiments. a ELISA analysis of anti-dsDNA IgG in sera from indicated mice. b Flow cytometry quantization of the percentage of germinal center (GC) B cells and double-negative (DN) T cells in spleens from indicated mice. Upper: PNA⁺FAS⁺ GC B cells (gated in CD19⁺); lower: CD4⁻CD8⁻NK1.1⁻ DN T cells (gated in CD3⁺TCRβ⁺). c Flow cytometry quantitation of the percentage of IL-17⁺ and IFN-γ⁺ CD4 T cells (upper) and DN T cells (lower) in spleens of indicated mice. d Bar graphs show absolute cell numbers of indicated cell population of spleens from mice with indicated treatment. e Representative immunofluorescent staining of MARCO⁺ marginal zone macrophages (blue) and CD3⁺ (green) T cells that also showed negative staining for CD4, CD8, and NK1.1 (red) in the spleens from the indicated mice. Upper: magnification, ×20. Scale bar: 50 µm; Lower: digitally magnified views of the boxed areas in the upper panels. f Flow cytometry quantitation of infiltrating T cells in kidneys of indicated mice. g Bar graphs show absolute cell numbers of indicated cell population of kidneys from mice with indicated treatment. Data represent the mean ± SEM, *P < 0.05, **P < 0.01, **P < 0.005 vs. indicated control, two-tailed Student’s t test.

Fig. 2. Exposure to apoptotic cell debris induces loss of CD8, but not CD4 expression.

CD8 T cells from Cd45.1 OT-I TCR Tg Rag1^−/− B6 mice (a) or CD4 T cells from Cd90.1OT-I TCR Tg Rag1^−/− B6 mice (b) were labeled with CFSE and i.v. transferred. Recipients were administered with CLs or control liposomes, apoptotic m-OVA Tg thymocytes, 12 h and 16 h later sequentially. Mice were sacrificed after an additional 72 h. a Left: flow cytometry analysis of cell surface CD8 expression on transferred OT-I CD8 T cells; middle: flow cytometry analysis of in vivo OVA-antigen-specific responses indicated by attenuation of CFSE intensity. Right: flow cytometry quantitation of the percentage of IL-17⁺ and IFN-γ⁺ cells in CD8⁻ and CD8⁺ populations. b Flow cytometry analysis of cell surface CD4 expression on transferred OT-II Vα₂⁺Vβ₅⁺ CD4 T cells. In vivo OVA responses were indicated by attenuation of CFSE intensity. CD8 from Cd45.1 OT-I Rag1^−/− B6 mice (c, e, g) or CD4 T cells from Cd90.1 OT-II Rag1^−/− B6 mice (d, f) were labeled with CFSE and i.v. transferred into m-OVA Tg B6 mice administered with CLs or control liposome 4 h before and sacrificed after an additional 72 h. c Left: Flow cytometry analysis of cell surface CD8 expression on transferred OT-I T cells. Middle: flow cytometry analysis of in vivo OVA responses indicated by attenuation of CFSE intensity. Right: flow cytometry quantitation of the percentage of IL-17⁺ and IFN-γ⁺ cells in CD8⁻ and CD8⁺ populations. d Flow cytometry analysis of cell surface CD4 expression on transferred OT-II T cells. In vivo OVA responses were indicated by attenuation of CFSE intensity. e, f Immunofluorescent staining of transferred OT-II (e) or OT-I T cells (f). Upper: magnification, ×20. Scale bar: 50 µm; Lower: digitally magnified views of the boxed areas in the upper panels. g Flow cytometry analysis of infiltrating T cells from both donors and recipients in the kidneys from indicated mice. Data represent the mean ± SEM, *P < 0.05, **P < 0.01, ***P < 0.005 vs. indicated control, two-tailed Student’s t test. n = 4–5 mice per group for two independent experiments.

Fig. 3. Autoreactive Polyclonal CD8 T cells differentiate into DN T cells.

A mixed population of purified CD8 T cells (TCRβ⁺NK1.1⁻CD8⁺, 2 × 10⁶/mouse, 1:1 ratio) from Cd45.1 B6 mice and Cd45.2 B6.lpr.DsRed (a, b) or from normal Cd45.2 B6.DsRed mice and Cd45.2 B6 Aire^−/− mice (c, d) were transferred into either Cd45.2 B6 (a, b) or Cd45.1 B6 (c, d) recipients. Twelve hours later, recipients were administered with control liposome, CLs, CLs plus apoptotic thymocytes prepared from B6 mice. Mice were euthanized after an additional 72 h. n = 5 mice per group for two independent experiments. a Left: flow cytometry analysis of the percentage of the transferred CD8 T cells in the spleens of Cd45.2 B6 recipients with the indicated treatment. Right: flow cytometry analysis of CD8 expression on the surface of the transferred CD8 T cells in the spleens of Cd45.2 B6 recipients with indicated treatment. b Flow cytometry quantitation of the percentage of IL-17⁺ and IFN-γ⁺ cells in CD8⁻ (upper) and CD8⁺ (lower) populations derived from transferred CD8 T cells. c Left: flow cytometry analysis of the percentage of the transferred CD8 T cells in the spleens of Cd45.1 B6 recipients with indicated treatment. Right: flow cytometry analysis of CD8 expression on the surface of the transferred CD8 T cells in the spleens of Cd45.1 B6 recipients with indicated treatment. d Flow cytometry quantitation of the percentage of IL-17⁺ and IFN-γ⁺ cells in CD8⁻ (upper) and CD8⁺ (lower) populations derived from transferred CD8 T cells. Data represent the mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.005 vs. PBSLs-treated control, two-tailed Student’s t test.

Fig. 4. IL-23 favors DN T-cell formation.

B6 mice were i.p. injected with apoptotic thymocytes prepared from B6 mice 4 h after control liposome or CLs i.v. administration, and blood was collected after indicated time. a ELISA analysis of indicated cytokines in the serum from indicated mice after indicated time, n = 5 per group. b, c CD8 T cells from Cd45.1 OT-I TCR Tg Rag1^−/− B6 mice were i.v. transferred into B6 mice. Twelve hours later, recipients were administered with apoptotic thymocytes prepared from m-OVA Tg B6 mice plus PBS, LPS (2 mg/kg) and Curdlan (5 mg/kg) separately. Mice were euthanized after an additional 72 h. n = 6 per group. b Flow cytometry analysis of CD8 expression on transferred OT-I CD8 T cells from spleens of indicated recipients. c Bar graphs show quantitation of the absolute cell numbers of both CD8⁺ and CD8^- CD45.1⁺OT-I T cells from the spleens of indicated recipients. d Flow cytometry quantitation of the percentage of CD3⁺TCRb⁺CD4⁻CD8⁻ DN T cells in spleens or lymph nodes from mice with the indicated MC administration. e Representative H&E images of kidneys from mice with indicated MC administration. Upper: magnification, ×4, scale bar: 200 μm; Lower: digitally magnified views of the boxed areas in the upper panels. f Fluorescence microscopic images of IgG staining in the kidneys of mice with indicated MC administration. Magnification, ×10, scale bar: 100 μm. g IL-23 MC was administered to 2-months-old IL-17 GFP B6.lpr mice. Three months after administration, splenic T cells were purified and co-cultured with or without irradiated wild-type B6 splenocytes pre-loaded with U1-70 peptide. Top: flow cytometry quantitation of the percentage of GFP⁺ cells CD4, CD8 and DN T cells individually. Bottom: flow cytometry analysis of the production of IL-17 and IFNγ in GFP⁺ populations from indicated T cells. h Flow cytometry quantitation of the percentage of U1-70:I-A^b tetramer-positive cells in GFP⁺ population from indicated T cells. Data represent the mean ± SEM (*P < 0.05, ***P < 0.005 vs. control, Student’s t test; n = 4 mice per group for two independent experiments for (e–h).

Fig. 5. Deficiency of Tgfb1 in macrophages promotes DN T-cell formation.

a, b CFSE-labeled CD4 T cells from OT-II Rag1^−/− or CD8 T cells from OT-I Rag1^−/− B6 mice were co-transferred with apoptotic m-OVA thymocytes into the indicated recipients. Recipients were euthanized after an additional 72 h. n = 4 per group for two independent experiments. Flow cytometry analysis of proliferation (a) or surface CD4/CD8 expression (b) on transferred OT-I or OT-II T cells. c, d Apoptotic thymocytes were administered into mice pre-treated without (c) or with CL (d) for 2 h. MZMs, red pulp macrophages (RPMs or RPs), CD4⁺DCs, and CD8⁺DCs were FACS sorted 30 min later. qRT-PCR analysis of indicated gene expression in indicated cell populations sorted from mice administered with apoptotic thymocytes (c) or mice administrated with CLs plus apoptotic thymocytes (d). n = 4 per group. e–g CFSE-labeled apoptotic thymocytes were administered into mice with or without CL pre-treatment. n = 4 per group. Magnification, ×20, scale bar: 50 μm. e Confocal microscopic images show that in the absence of MZMs, administration of apoptotic thymocytes leads to follicular translocation of RPMs (CD68⁺). f Confocal microscopic images show the uptake of apoptotic cell debris by CD8⁺ dendritic cells in the absence of MZMs. g Flow cytometry shows the apoptotic cell uptake (CFSE⁺) by indicated populations with indicated treatment. h, j–l Mice were at 12 months of age. h–l n = 5 per group for two independent experiments. h Representative photos of spleens and lymph nodes from indicated mice. i Flow cytometry quantitation of CD3⁺TCRβ⁺CD4⁻CD8⁻ DN T cells in spleens or lymph nodes from indicated mice at indicated ages. j Fluorescence microscopic images of IgG staining in the kidneys of indicated mice. Magnification, ×10, scale bar: 100 μm. k Representative H&E images of kidneys from indicated mice. Upper: magnification, ×4, scale bar: 200 μm; Lower: Digitally magnified views of the boxed areas in the upper panels. l Flow cytometry analysis of infiltrating T cells in the kidneys from indicated mice. Data represent the mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.005 vs. control, two-tailed Student’s t test).

Fig. 6. DN T cells display proliferating or proliferated phenotype.

a–c Sixteen week-old female MPJ or MRL.lpr mice were injected i.p. with BrdU (1 mg/mouse) 12 and 0.5 h prior to the euthanasia. Data represent the mean ± SEM (*P < 0.05, ***P < 0.005 vs. control, Student’s t test; n = 6 mice per group). a Flow cytometry analysis of different subset T cells from indicated mouse spleens. b Flow cytometry analysis of cell cycle phases of different subset T cells from indicated mouse spleens (gated in CD3⁺TCRβ⁺Ly49b⁻). BrdU incorporation is used for characterization of cells that are actively synthesizing DNA and 7-AAD staining intensities were further applied to define related cell phase in the cell cycle (G0/G1, G2/M, or S phases). c Representative immunofluorescent staining of BrdU⁺ (magenta) DN T cells (green) in kidneys from indicated mice. Upper: magnification, ×10. Scale bar: 100 µm; Lower: digitally magnified views of the boxed areas in the upper panels. d–g n = 23 for healthy controls and n = 26 for SLE patients. d Representative flow cytometry analysis of CD4, CD8, and DN T cells in total T cells in PBMC from either healthy subjects or SLE patients. e Scatter plots show the percentage of CD4, CD8, and DN T cells in total circulating T cells (gated in CD3⁺TCRβ⁺CD56⁻) from either healthy controls or SLE patients. f Flow cytometry analysis of Ki67⁺ CD4, CD8, and DN T cells in total T cells from PBMC from healthy controls or SLE patients. g Scatter plots show the quantitation of Ki67⁺ CD4, CD8, and DN T cells of human PBMC (gated in CD3⁺TCRvβ⁺CD56⁻) from either healthy controls or SLE patients. h Representative immunofluorescent staining of Ki67⁺ (magenta) DN T cells (green) in kidney biopsy samples from patients with SLE. Upper: magnification, ×100. Scale bar: 5 µm; Lower: digitally magnified views of the boxed areas in the upper panels. n = 10. Data represent the mean ± SEM (*P < 0.05, ***P < 0.005 vs. indicated control, Student’s t test).

Fig. 7. Skewed TCR Vβ repertoires and diluted TREC of DN T cells.

a Bar graphs show quantitation of TCR Vβ usage by CD4, CD8, and DN T cells of spleens from indicated mice. CD4 T cells: CD3⁺TCRβ⁺NK1.1⁻CD4⁺, CD8 T cells: CD3⁺TCRβ⁺NK1.1⁻CD8⁺, DN T cells: CD3⁺TCRβ⁺NK1.1⁻CD4⁻CD8⁻, bottom: female B6.lpr mice were treated with clodronate liposomes (CLs, 100 ug/mouse) every other week for 2 months total starting at 3 months of age. n = 4 mice per group. b Scatter plots show the quantitation of TCR Vβ usage by DN, CD4, and CD8 T cells in PBMC from either healthy controls or SLE patients. CD4 T cells: CD3⁺TCRβ⁺CD56⁻CD4⁺, CD8 T cells: CD3⁺TCRβ⁺CD56⁻CD8⁺, DN T cells: CD3⁺TCRβ⁺CD56⁻CD4⁻CD8⁻. n = 12 each group. c Left: quantitative real-time PCR analysis of TCR excision circles in CD4, CD8, and DN T cells in spleens from B6.lpr with indicated ages. Right: quantitative real-time PCR analysis of TCR excision circles in CD4, CD8, and DN T cells from either spleens or kidneys of 12-months-old female B6.lpr mice (n = 4 mice per group). d Quantitative real-time PCR analysis of TCR excision circles in CD4, CD8, and DN T cells (gated in CD3⁺TCRαβ⁺CD56⁻) in PBMC from five SLE patients and five matched healthy controls with three technical replicates for each sample. Data represent the mean ± SEM (*P < 0.01, **P < 0.05, ***P < 0.005 vs. indicated controls). Two-tailed unpaired Student’s t test was used when only two groups were compared, and the one-way ANOVA was applied with comparison of more than two groups.