Fig 2: Flow cytometry data for TH17 cell identification. The same groups of mice were analyzed as for LTi cell identification. The gates show CD3+CD5+CD4+ROR?t+ TH17 cells in the cerebellum. The comparison between a non-immunized, b MP4-immunized mice during acute and chronic EAE and c MOG:35–55-immunized mice during acute and chronic EAE is shown

Fig 3: Targeting LSD1 enhances long-lasting response to anti-PD-1 treatment.a Experimental design of administering GSK2879552 and PD-1 blocking antibodies. b Tumor growth curves of wildtype mice inoculated with MC38 tumor cells and treated with GSK2879552 or vehicle control in combination with anti-PD-1 or isotype control. c Individual tumor growth curves over time in response to anti-PD1 treatment alone or in combination with GSK2879552. d, e Percentages of TCF1+PD-1int and TCF1-PD-1hi cells (d) or GzmB+ cells (e) in CD8+ TILs (Vehicle + Isotype, n = 6; GSK + Isotype, n = 6; Vehicle + anti-PD-1, n = 6; GSK + anti-PD-1, n = 7). f, g, Percentages (f, n = 9, 10, 6, 9, respectively) and cell numbers (g, n = 8, 9, 6, 5, respectively) of tumor-infiltrating CD3+CD4+ T cells and CD3+CD8+ T cells in gated CD45+ leukocytes. h, i Frequencies (h, n = 9, 10, 6, 9, respectively) and cell numbers (i, n = 8, 9, 6, 5, respectively) of CD8+TCF1-TIM-3+ TILs. Data represent two independent experiments (b, d–i) or are pooled from two independent experiments (c), and are presented as mean ± SEM (b, d–i). Sample sizes are as indicated. Statistical significance was determined by two-way ANOVA (b) or two-sided unpaired t test (d–i). Source data are provided as a Source data file.

Fig 4: LSD1/CoREST complex physically interacts with TCF1 and antagonizes its transcriptional activity.a MFI of TCF1 in CD8+TCF1+PD-1int TILs analyzed by flow cytometry (n = 4 per group). b The physical interaction between LSD1 and TCF1 isoforms as well as other proteins was examined by co-immunoprecipitation (co-IP) assays using whole-cell extract (WCE) of HEK293T cells transfected with FLAG-HA-tagged LSD1. c Schematic representation of different isoforms of mouse TCF1. d Co-IP assays precipitating HA-tagged TCF1 isoforms and detecting LSD1 in eluted IP fractions. e Co-IP assays precipitating Flag-tagged TCF1 (referring to p45 unless otherwise specified) and detecting components of the LSD1/CoREST complex in eluted IP fractions. f–j TCF/LEF reporter assay measuring relative luciferase units (RLU) in response to transfection with plasmids expressing indicated proteins (f–j, n = 4 per condition) or shRNA targeting LSD1 (i) and the treatment with iCRT3 (i) (n = 4 per condition). Protein expression for plasmid transfection in f is shown by immunoblotting (g). k, l Representative flow plots (k) and Slamf6 MFI (l) of splenic CD8+ T cells activated in vitro for 3 days with anti-CD3/anti-CD28 in the presence of vehicle, GSK2879552 or iCRT3 (n = 3 per treatment). Data represent two (a, b, d–j) or three (k, l) independent experiments and are presented as mean ± SEM (a) or mean ± SD (f, h–j, l). Statistical significance was determined by two-sided unpaired t test (a, f, h–j, l). Source data are provided as a Source data file.

Fig 5: Flow cytometry gating strategy for identification of ILCs, LTi, and TH17 cells in EAE mice. Forward scatter area (FSC-A) and height (FSC-H) were used to exclude cells, which had formed doublets. The lymphocyte gate was determined by using FSC-A and sideward scatter area (SSC-A). Viable cells were separated from dead cells. Cells were divided into CD3-CD5- and CD3+CD5+ cells. CD3-CD5- cells were further analyzed for the presence of LTi cells and ILCs, and CD3+CD5+ cells were further analyzed for TH17 cells. CD3-CD5- cells being positive for CD4 and ROR?t represented CD3-CD5-CD4+ROR?t+ LTi cells. Moreover, CD3-CD5-CD4-ROR?t+ cells appeared in the ILC gate. CD3+CD5+ cells, which were also positive for CD4 and ROR?t, occurred in the gate of TH17 cells