ZH, Li et al, page 1 ECM1 controls T H 2 cell egress from lymph nodes through re-expression of S1P 1 Zhenhu Li 1,4,Yuan Zhang 1,4, Zhiduo Liu 1, Xiaodong Wu 1, Yuhan Zheng 1, Zhiyun Tao 1, Kairui Mao 1, Jie Wang 1, Guomei Lin 1, Lin Tian 1, Yongyong Ji 1, Meiling Qin 1, Shuhui Sun 3, Xueliang Zhu 1, and Bing Sun 1,2, *
ZH, Li et al, page 2 Supplementary Figure 1: Control of differentiation of CD4 + T cells. (a-e) T cells are cultured under different conditions and harvested on day 4. Real-time PCR assay determined the cytokines and transcription factors mrna level. Error bars indicate mean ± s.d. between three independent experiments.
ZH, Li et al, page 3 Supplementary Figure 2: Ecm1 knockout mouse design and phenotype. (a) Ecm1 knockout mouse design diagram is shown. Exon 2-11 was knocked out by homologous recombination. The whole cell (b) and the supernatant protein levels (c) were determined by immunoblot and ELISA in several CD4 + T cell subsets. (d) Thymus, LN and spleen CD4 + and CD8 + pattern is shown. Data are representative of three independent experiments. Error bars indicate mean ± s.d. between three independent experiments.
ZH, Li et al, page 4 Supplementary Figure 3: T H 2 cell phenomenon is shown using retrovirus over expression and RNAi. (a) The effect of retrovirus was analyzed using immunoblot. (b) T H 2 cells were transfected by over-expression or RNAi retrovirus twice as described in Methods. On day 4 after TCR induction, cells underwent intracellular staining with PE-IL-4 and FITC-IFN-γ. Finally GFP + cells were gated for FACS analysis. (c) T H 1 and T H 2 cytokines were detected using ELISA. (d) After transduction of overexpression vector or of RNAi encoding retrovirus, GFP + cells were sorted for the second stimulation. [ 3 H] thymidine was added to the culture when GFP + cells were restimulated. Error bars indicate mean ± s.d. between three independent experiments.
ZH, Li et al, page 5 Supplementary Figure 4: ECM1 is not necessary for T H 1 and T H 17 response in vivo. Bone marrow reconstitution mice were immunized with OVA and CFA at the base of the tail. (a) Five days later, spleen and draining lymph node cell numbers were determined. CD4 + and CD8 + cell numbers were calculated using flow cytometry. (b) Supernatants of draining lymph node cells re-stimulated with OVA for 96hrs were harvested for analysis of cytokine production using ELISA. Error bars indicate mean ± s.d. between three independent experiments.
ZH, Li et al, page 6 Supplementary Figure 5: Time course of Ecm1, Klf2 and S1P 1 mrna expression. The samples were obtained from T H 2 cells on different days. Data are representative of three independent experiments.
ZH, Li et al, page 7 Supplementary Figure 6: Responses of Ecm1 deficient cells. (a) Chemotactic response to CXCL12 and CCL21 of T H 2 cells from WT and Ecm1-deficient mice. The percentage of the input cell population that responded to the indicated concentration of CXCL12 or CCL21 is shown. (b-c) WT and Ecm1-deficient T H 1 cells were infected by Ecm1-expressing retrovirus as described in Methods. Samples were obtained 4 days after TCR signaling treatment. (b) Restoration of mrna levels after retroviral Ecm1 transduction by real-time PCR. (c) Chemotactic response of T H 1 cells is shown. Error bars indicate mean ± s.d. between three independent experiments.
ZH, Li et al, page 8 Supplementary Figure 7: Ecm1 deficiency does not affect IL-2 signaling and S1P 1 expression in T H 1 cells. (a-b) WT or KO T H 1 cells were cultured for 4 days with mil-2 neutralizing antibody added after 2 days to block IL-2 signaling. Immunoblot and real-time PCR assay were performed with T cells harvested 4 days after initiation of polarization culture. Error bars indicate mean ± s.d. between three independent experiments.
ZH, Li et al, page 9 Supplementary Figure 8: IL-7 and IL-15 are not involved in ECM1 regulating S1P 1 expression. (a-b) WT or KO T H 2 cells were cultured for 4 days with or without hil-2, mil-7 or mil-15. Immunoblot and real-time PCR assay were performed with T cells harvested 4 days after initiation of polarization culture. Error bars indicate mean ± s.d. between three independent experiments. *: p < 0.05.
ZH, Li et al, page 10 Supplementary Figure 9: Perlecan is not involved in ECM1 regulating IL-2 signaling and S1P 1 expression. (a-b) WT or KO T H 2 cells were cultured in serum-free T cell medium (which not contains perlecan) for 4 days with mil-2 neutralizing antibody added after 2 days to block IL-2 signaling. Immunoblot and real-time PCR assay were performed with T cells harvested 4 days after initiation of polarization culture. Error bars indicate mean ± s.d. between three independent experiments.
ZH, Li et al, page 11 Supplementary Figure 10: A diagram of ECM1 mediated KLF2 and S1P 1 expression. After TCR engagement, IL-4 signaling and IL-2 signaling is critical for T H 2 cell differentiation. IL-4 signaling can induce STAT6 phosphorylation and GATA-3 expression. IL-2 signaling is important for T H 2 cell proliferation but inhibit migration molecules expression (KLF2 and S1P 1 ). In the later stage, GATA-3 can bind to Ecm1 locus and initiate its expression. After ECM1 is secreted, it can directly bind to IL-2 receptor and inhibit STAT5 phosphorylation to promote KLF2 and S1P 1 expression and migration.