Fig. 1|. The BH4 pathway is indispensable for effective T-cell proliferation in vitro and in vivo.

a, Immunoblot of GCH1 after 24 hours of TCR stimulation with anti-CD3/CD28 antibodies in CD4⁺ T cells. The experiment was repeated three times with similar results. Staining for actin acts as a control. b, BH4 production upon 24 hours of anti-CD3/CD28 stimulation in purified CD4⁺ control and Gch1-null T cells. Individual data (dots and squares; n = 5 mice in each case) are shown as means ± s.e.m. c, d, Representative fluorescence-activated cell sorting (FACS) blot depicting early activation markers (CD62L and CD25; c) and IL-2 secretion (d) before and after T-cell stimulation (16 hours). Data (n = 5 independent samples) are shown as means ± s.e.m. The experiment was repeated two independent times with similar results. FITC, fluorescein isothiocyanate. Naive T cells, CD25^low, CD62L^hi; activated T cells, CD25^hi, CD62L^low. e, Proliferation of CD4⁺ T cells after three days of stimulation of control and Gch1;Lck mice. Cell Trace Violet gets diluted in proliferating cells (see Methods). Representative data are shown from more than 15 experiments with similar results. f, Quantification of CD4⁺ T cell proliferation from individual Gch1;Lck (left; n = 10) and Gch1;RORc (right; n = 7) mice. Data are shown as means ± s.e.m. g, h, Transfer colitis model of intestinal autoimmunity. g, Schematic outline (top) and colitis scores of transferred control and Gch1- ablated CD4⁺ T cells into Rag1^−/−hosts (bottom). D, day. Data (n = 10 mice) are shown as means ± s.e.m. h, Representative immunofluorescence depicting intestinal infiltration of various immune cells (CD3ε⁺, CD4⁺, CD11c⁺ and myeloperoxidase (MPO)⁺ cells). Scale bar represents 200 μm. i, Allergic airway inflammatory disease model (top) and quantification of inflammatory cells in bronchoalveolar lavage fluids (BALFs; bottom). Data are shown as means ± s.e.m.; n = 35 for control mice; n = 31 for Gch1;Lck mice. OVA, ovalbumin. j, Percentage increase in ear swelling after re-challenge using the 2,4,6-trinitrochlorobenzene (TNCB)-dependent skin hypersensitivity model. Data are shown as means ± s.e.m.; n = 8 for control mice; n = 9 for Gch1;Lck mice. NS, not significant; *P< 0.05; **P< 0.01; ***P < 0.001 (two-tailed Student’s t-test for panels b, d, f, i, j; two-way analysis of variance (ANOVA) with Dunnett’s comparison for panel g).