Fig 2: HIF-1a in NKp46+ cells favors an ILC1 phenotype in the SI. (A) Top: Representative FACS plots showing the frequency of small intestinal ILC1s, NKp46+ ILC3s in the alive Lin- CD127+ cells from HIF-1a KO and HIF-1a WT mice, and the ratio of NKp46+ ILC3 to ILC1 from HIF-1a KO and HIF-1a WT mice. Bottom: Quantitative analysis of frequency and absolute number of small intestinal ILC1s and NKp46+ (pooled data of three independent experiments, n = 11); data are mean values ± SEM; **, P < 0.01; ***, P < 0.001 by two-tailed Student’s t test. (B) Left: Representative FACS plots showing the frequency of IFN-?–expressing small intestinal ILC1s from HIF-1a KO and HIF-1a WT mice stimulated with PMA/ionomycin and corresponding quantitative analysis (pooled data of two independent experiments, n = 5); data are mean values ± SEM; *, P < 0.05 by two-tailed Student’s t test. Right: Representative FACS plots showing the frequency of IL-22–expressing small intestinal NKp46+ ILC3s from HIF-1a KO and HIF-1a WT mice stimulated with PMA/ionomycin and corresponding quantitative analysis (pooled data of two experiments, n = 5), two-tailed Student’s t test. SSC, side scatter. (C) Gene expression of Il22 and Ifng in small intestinal tissues from HIF-1a KO and HIF-1a WT mice, as fold-change to the WT samples (pooled data of three independent experiments, n = 8); data are mean values ± SEM; *, P < 0.05 by two-tailed Student’s t test. (D) Gene expression analysis of IL-22–dependent gut-homeostatic factors such as Reg3a, Reg3b, Reg3g, Defa21, and Muc2 in small intestinal tissues from HIF-1a KO and HIF-1a WT mice (pooled data of three independent experiments, n = 10); data are mean values ± SEM; *, P < 0.05; **, P < 0.01 by multiple two-tailed Student’s t test.

Fig 3: HIF-1a in NKp46+ cells controls the microbiome and susceptibility to colitis. (A) The frequency and absolute number (Nb) of large intestinal NKp46+ ILCs in CD45+ population from HIF-1a KO and HIF-1a WT mice (pooled data of four independent experiments, n = 13); data are mean values ± SEM; two-tailed Student’s t test. (B) Flow cytometric analysis of alive CD45+ cells, neutrophils (NPs), macrophages (M?), eosinophils (Eos), B cells, total T cells, CD4 cells, CD8 cells, and dendritic cells at the steady state and after DSS treatment on day 4 after treatment and day 7 after treatment from large intestinal tissues from HIF-1a KO and HIF-1a WT mice (pooled data of four independent experiments for steady-state analysis, n = 10; one experiment for DSS midpoint analysis, n = 3; two experiments for DSS endpoint analysis, n = 6); data are mean values ± SEM; two-way ANOVA test. (C) Ratio of Il22 expression to Ifng expression in large intestinal tissues from HIF-1a KO and HIF-1a WT mice (pooled data of two independent experiments, n = 6); data are mean values ± SEM; **, P < 0.01 by two-tailed Student’s t test. (D) Representative H&E staining of large intestinal tissues from HIF-1a KO and WT mice; scale bar, 100 µm.

Fig 4: Imiquimod increases IL-17 and IL-22 in both ?d and aß T cells in skin.Skin and ear of wild type C57BL/6 mice were treated by topical application of imiquimod cream (Fougera, n=4) or control cream (Vaseline, n=3) for 6 consecutive days. (a) Weight loss of imiquimod and control cream treated mice monitored daily. (b) Photographs of imiquimod and control cream treated skin and ear; photos were taken at day 6. H&E stained ear and back skin sections of imiquimod treated and control mice. Scale bar, 200 µm. (c) Thickness of skin measured by Digimatic Caliper at day 6 in control and imiquimod treated mice. Data showed represents average of at least two measurements. (d) Ear thickness of imiquimod and control cream treated mice monitored daily. Ear thickness was measured using Digimatic Caliper. (e) Quantitative PCR analysis of Th17 associated cytokines and Foxp3 in skin after 6 days of imiquimod and control cream treatment. (f) Representative flow cytometric analysis of TCR?d+V?4+IL-17+ (upper raw) and TCR?d+V?4+IL-22+ cells (lower row) in the skin of control (n=3) and imiquimod treated mice (n=4). (g) Representative flow cytometric analysis of single-positive TCRß+IL-17+ (upper raw) or TCRß+IL-22+ (lower raw) cells in the skin of control (n=3) and imiquimod treated mice (n=4). Data representative of more than three experiments, results are shown as mean±s.e.m., significance determined by unpaired two-tailed Student's t-test (*P<0.05; **P<0.01; ***P<0.001).

Fig 5: HIF-1a promotes the expression of Tbx21 in NKp46+ cells. (A) Left: T-bet mean fluorescence intensity of HIF-1a KO and HIF-1a WT ILC1s (pooled data of three experiments, n = 9), two-tailed Student’s t test. Right: T-bet and ROR?t mean fluorescence intensity of HIF-1a KO and HIF-1a WT ILC1s (pooled data of three independent experiments, n = 9); data are mean values ± SEM; *, P < 0.05; **, P < 0.01 by two-tailed Student’s t test. MFI, mean fluorescence intensity. (B) Sorting strategy for NKp46+ ILCs, ILC1s (defined as EYFP+/NKp46+/CD45high/CD90low), and NKp46+ ILC3s (defined as EYFP+/NKp46+/CD45low/CD90high). (C) Gene expression of Hif1a on sorted ILC1s and NKp46+ ILC3s from the SI of EYFP-WT and EYFP-HIF-1a KO mice (n = 4; each point represents a pool of four mice); data are mean values ± SEM; *, P < 0.05; ***, P < 0.001 by two-tailed Student’s t test. (D) Left: Gene expression of Tbx21 and Ifng on sorted ILC1s from SI of EYFP-WT and EYFP-HIF-1a KO mice (n = 4; each point represents a pool of four mice); data are mean values ± SEM; *, P < 0.05 by two-tailed Student’s t test. Right: The gene expression of Tbx21, Rorc, Il22, and Ifng on sorted NKp46+ ILC3s from SI of EYFP-WT and EYFP-HIF-1a KO mice (n = 4; each point represents a pool of four mice); data are mean values ± SEM; *, P < 0.05 by two-tailed Student’s t test. (E) Schematic diagram of HREs within promoter region of murine Tbx21. (F) HIF-1a binding to HREs ofTbx21 promoter in sorted EYFP+ ILCs upon in vitro IL-12–mediated stimulation.