The regulation of T cell homeostasis and autoimmunity by T cell-derived LIGHT

Abstract

Costimulatory molecules on antigen-presenting cells (APCs) play an important role in T cell activation and expansion. However, little is known about the surface molecules involved in direct T-T cell interaction required for their activation and expansion. LIGHT, a newly discovered TNF superfamily member (TNFSF14), is expressed on activated T cells and immature dendritic cells. Here we demonstrate that blockade of LIGHT activity can reduce anti-CD3-mediated proliferation of purified T cells, suggesting that T cell-T cell interaction is essential for this proliferation. To test the in vivo activity of T cell-derived LIGHT in immune homeostasis and function, transgenic (Tg) mice expressing LIGHT in the T cell lineage were generated. LIGHT Tg mice have a significantly enlarged T cell compartment and a hyperactivated peripheral T cell population. LIGHT Tg mice spontaneously develop severe autoimmune disease manifested by splenomegaly, lymphadenopathy, glomerulonephritis, elevated autoantibodies, and severe infiltration of various peripheral tissues. Furthermore, the blockade of LIGHT activity ameliorates the severity of T cell-mediated diseases. Collectively, these findings establish a crucial role for this T cell-derived costimulatory ligand in T cell activation and expansion; moreover, the dysregulation of T cell-derived LIGHT leads to altered T cell homeostasis and autoimmune disease.

Figure 1

T cell–derived LIGHT functions as a costimulatory ligand in a T-T cell–dependent fashion. (a) Splenocytes were collected from WT B6 mice (5–6 weeks old) and cultured with CTLA4-Ig, LTβR-Ig, HVEM-Ig, or control Ig at various concentrations in the presence of ConA (1.5 μg/ml). (b) Splenocytes and LN cells were pooled and T cells were purified and stained with anti-CD3-FITC. The percentages of CD3-positive cells are indicated. (c) Highly purified T cells were stimulated with immobilized anti-CD3 mAb (1 μg/ml) in the presence of HVEM-Ig, CTLA4-Ig, or control Ig. Proliferation was measured by [³H]thymidine incorporation. The results are representative of three experiments.

Figure 2

T cell–derived LIGHT is sufficient to cause the expansion of peripheral T cells in vivo. (a) Splenomegaly and lymphadenopathy were observed in LIGHT Tg mice at 5–8 months of age. Spleen and axillary LNs were shown. Pictures (×4) are representative of seven mice analyzed for each group. (b) The total cell number was increased in the spleens (left panel, *P < 0.01) and peripheral LNs (right panel, **P < 0.001) of Tg mice. The results were representative of seven mice analyzed in each group. (c) FACS analysis of the ratio of T (CD3⁺) to B cells (CD19⁺) in the spleen (left panel) and LNs (right panel) of WT (diamonds) and Tg (squares) mice (n = 5). (d) The enlargement of the T cell zone in LIGHT Tg LNs. Immunohistochemical staining was performed using anti-Thy1.2-Bio for T cells (blue) and anti-B220-FITC for B cells (brown). Representative pictures (×10) are shown. PLN, peripheral LN.

Figure 3

Hyperactivation of T lymphocytes mediated by T cell–derived LIGHT. (a) Splenocytes or LN cells from WT and Tg mice were stained with antibodies against CD3 and CD69. Numbers indicate the percentage of CD3⁺CD69⁺ cells. (b) Splenocytes from WT and Tg mice were stained with antibodies against CD3, CD62L, and CD44. The dot plots represent cells gated on the CD3⁺ population. The results are representative of five experiments. WT and Tg mice at the age of 5–8 months were used. MLN, mesenteric LN.

Figure 4

Enhanced cytokine production and expanded macrophage and granulocyte populations in LIGHT Tg mice. (a) Surface CD3 and intracellular IFN-γ expression was determined by double color staining of LN cells from WT and Tg mice after 4 hours of in vitro stimulation with PMA (50 ng/ml) and ionomycin (500 ng/ml). Numbers indicate the percentage of IFN-γ–producing T cells (CD3⁺). Data are representative of three experiments. (b) GM-CSF–responding precursors were increased in Tg mice. Splenocytes from WT and Tg mice were cultured with different concentrations of GM-CSF; proliferation was measured by [³H]thymidine incorporation. (c) FACS analysis of macrophage and granulocyte populations in Tg mice. Splenocytes (spl) from WT and Tg mice were stained with antibodies against CD11b or Gr-1. Percentage of CD11b⁺ and Gr-1⁺ cells is indicated. (d) Expanded populations of macrophages and granulocytes in Tg mice. Splenocytes were isolated from WT and Tg mice and cell number was determined, and then cells were stained with CD11b-Bio or Gr-1-Bio antibodies. Absolute cell numbers were calculated. These results are representative of three experiments. WT littermates and Tg mice at 5–8 months of age were used.

Figure 5

Induction of autoimmunity by T cell–derived LIGHT. Histology of intestine (a) and skin (b) from WT and Tg mice at 5–8 months of age. (a) Sections of colon from WT and Tg mice were stained with H-E. A severalfold increase of the thickness of intestine wall was prominently observed in Tg mice compared with control littermates, as well as dense inflammatory cell infiltration. (b) In the skin lesions of Tg mice, conspicuous mixed acute and chronic inflammatory cell infiltrate extending from epidermis to subcutis was observed in LIGHT Tg mice, accompanied by the destruction of skin appendage and aberrant hair follicular proliferation. Representative pictures are shown. Original magnification: ×10.

Figure 6

Renal pathological analysis and the elevated autoantibodies and rheumatoid factor levels in LIGHT Tg mice. Representative pictures from WT (a, c, e, and g) and Tg (b, d, f, and h) mice at 5–8 months of age are shown. (a and b) H-E staining; (c and d) PAS staining; (e–h) immunofluorescence staining. Original magnification is as follows: a–d, ×40; e–h, ×63. Glomeruli of Tg mice at 5–8 months of age were enlarged and lobulated with increased cellularity and inflammatory cell infiltration (b). The deposition of PAS-positive material was observed along the capillary wall and in mesangium in Tg mice (d). Sections of kidneys from WT and Tg mice were stained with FITC-conjugated goat anti-mouse IgG or total Ig. Strong Ig deposits were observed in the glomeruli of Tg mice (f and h). The serum levels of anti-DNA autoantibody (i, bottom panel) and total IgG (i, top panel) were determined in parallel by ELISA using the same set of WT (open circles) and Tg (filled squares) mice (n = 5). The serum level of rheumatoid factor in WT and Tg mice (n = 7) was determined by ELISA (j). Serum dilutions are indicated and data are means ± SD.

Figure 7

Blockade of LIGHT activity ameliorated the severity of spontaneous autoimmune diabetes. A soluble receptor of LIGHT prevented the development of IDDM. NOD female mice (5–6 weeks old) were treated with HVEM-Ig weekly for 2 weeks. Blood glucose levels were measured weekly starting from 9 weeks of age (n = 8), and animals were considered diabetic after two consecutive measurements of ≥250 mg/dl.

Figure 8

Proposed model for the LIGHT-induced autoimmunity. Question mark means other unidentified receptor(s).