Fig 2: Bach2 regulates the development of ILC2.(A) Representative flow plots of CLP, α-LP, and ILC2P in BM from Bach2ΔIl7r and Il7rCre/+ mice. (B) Quantification of the proportion and counts of (A). (C and D) Representative histogram (C) and quantification (D) of CD25, ICOS, GATA3, and IL-18R1 in BM ILC2P from Bach2ΔIl7r and Il7rCre/+ mice. (E) Representative flow plots and quantification of lung ILC2 from Bach2ΔIl7r and Il7rCre/+ mice. ILC2s in lung were gated on CD45+Lin−CD90.2+CD25+. (F and G) Quantification of ILC2 proportions and counts in the mLN (F) and colon tissues (G) from Bach2ΔIl7r and Il7rCre/+ mice. ILC2s in mLN and colon were defined as CD45+Lin−CD127+KLRG1+ cells. (H to J) Competitive BM chimeric experiments were performed. NCG mice intravenously inoculated with Il7rCre/+ (CD45.2) or Bach2∆Il7r (CD45.2) BM cells equally mixed WT competitive BM cells (CD45.1) and analyzed 6 to 8 weeks after transplantation (H). Representative flow plots (I) and ratios of donor chimerism (J) of lung ILC2 and colonic ILC2 were determined by flow cytometry. Data were analyzed by two-way ANOVA followed by Tukey-Kramer test for multiple comparisons [(B), (D) to (G), and (J)] from three [(B) and (D) to (G)] or two [(H) to (J)] independent experiments and n = 4 to 5 independent samples of each group. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Error bars are shown as mean ± SEM.

Fig 3: Bach2 deficiency impairs the differentiation of ILC2P in BM.(A) Two-dimensional Uniform Manifold Approximation and Projection (2D-UMAP) plots showing the distributions of CLP, ILCP, and ILC2P from Bach2ΔIl7r and Il7rCre/+ mice. (B) Scatterplot displaying the developmental trajectory from CLP to ILCP and ILC2P. (C) Heatmap showing the relative expression of signature genes in BM progenitors that are arranged along the pseudotime trajectory. (D) Violin plots comparing differential gene expression in CLP and ILC2P between Bach2ΔIl7r and Il7rCre/+ mice. (E and F) Pie plots summarizing categories of peaks with differential H3K27ac modification levels in CLPs (left) (E) and ILC2Ps (right) (F) between Bach2ΔIl7r and Il7rCre/+ mice. (G and H) Heatmap comparing relative H3K27ac modification levels of indicated loci in CLPs (G) and ILC2Ps (H) between Bach2ΔIl7r and Il7rCre/+ mice. (I and J) Bubble charts showing enriched GO terms obtained by performing GO enrichment analysis with down-regulated H3K27ac peaks in Bach2ΔIl7r CLP (I) and ILC2P (J). (K) Venn diagram illustrating common genes that are linked by loci with reduced H3K27ac levels in both CLPs and ILC2Ps from Bach2ΔIl7r mice. (L and M) Tracks of selected gene loci with signal densities of CUT&Tag-seq for H3K27ac in BM CLP (L) and ILC2P (M) from Bach2ΔIl7r and Il7rCre/+ mice.

Fig 4: Bach2 governs airway inflammation in an ILC2-dependent manner.(A) Study design for papain-induced airway inflammation. In Bach2fl/fl-CreERT2 and littermate mice, airway inflammation was induced by intranasal administration of papain for five consecutive days (20 μg per mouse per day). (B) Flow cytometric analysis and quantification of eosinophils gated as CD45+CD11c−Siglec-F+ in BALF from the indicated groups. (C) IL-5 and IL-13 concentrations in the BALF were determined by ELISA in Bach2fl/+-CreERT2 and Bach2fl/fl-CreERT2 mice. (D) Representative H&E staining and scores of lung sections from the indicated groups. Scale bars, 50 μm. (E and F) Quantification of the lung ILC2 proportions and counts (E) and cytokine production (F) from ILC2 were determined by flow cytometry in Bach2fl/+-CreERT2 and Bach2fl/fl-CreERT2 mice. (G to J) Bach2ΔIl5 and Il5Cre/+ mice were induced airway inflammation by papain administration. Study design (G). Representative flow plots and quantification of eosinophils (H), ILC2s (I), and representative flow plots and cytokine production (J) by ILC2s were determined by flow cytometry. (K to N) Purified ILC2s from Bach2ΔIl7r and Il7rCre/+ mice were transferred into NCG mice, and airway inflammation was subsequently induced by IL-33 challenge for 3 days (K). Quantification of ILC2 (L) and cytokine production from ILC2 (M) and eosinophils (N) were determined via flow cytometry. (O) Representative H&E staining and scores of lung sections from the indicated groups. Scale bars, 100 μm. The data were analyzed by two-way ANOVA followed by Tukey-Kramer test for multiple comparisons [(B), (C), (E), (F), (H) to (J), and (M) to (O)] from two [(A) to (O)] independent experiments and n = 5 [(A) to (J)] independent samples of each group. *P < 0.05, **P < 0.01, ***P < 0.001. The error bars present the mean ± SEM.

Fig 5: Impaired ILC2 responses in Bach2ΔIl7r mice.(A) Representative histogram and quantification of Bach2 expression in the indicated immune cells by using Bach2-eGFP mice, Ctrl, and naïve WT mice. (B) Illustration of the IL-33 intraperitoneal (i.p.) injection model. Bach2ΔIl5 and Il5Cre/+ mice were intraperitoneally injected with IL-33 for three consecutive days. The mice were sacrificed for flow cytometry analysis on day 4. (C and D) Representative flow plots of ILCs in the lungs (left) and ILC2 in the colon (right) in IL-33 challenge state from indicated groups (C). Quantification of ILC2 in lung, colon, and mLN from Bach2ΔIl7r and Il7rCre/+ mice were determined by flow cytometry (D). (E and F) Flow plots and quantification of IL-5tdT and IL-13 production by ILC2s from the indicated group (E). Quantification of the counts of IL-5tdT ILC2 and IL-13+ ILC2 (F) indicated tissues. (G and H) PAS staining images of distal colon (G) and lungs (H) in Bach2ΔIl5 and Il5Cre/+ mice. Scale bars, 100 μm (G) and 20 μm (H). (I and J) Representative flow plots and quantification (I) of the ILC2 proportion and cytokine production (J) in Bach2fl/+-CreERT2 and Bach2fl/fl-CreERT2 mice in the IL-33 model. (K) Cytokine concentration of ILC2 from Bach2fl/+-CreERT2 and Bach2fl/fl-CreERT2 mice was determined by enzyme-linked immunosorbent assay (ELISA) in an in vitro culture system. Data were analyzed by Student’s t test [(D) and (F) to (H)] from three (A) and two [(B) to (K)] independent experiments and n = 4 to 5 independent samples of each group. *P < 0.05, **P < 0.01, ***P < 0.001. The error bars present the mean ± SEM.