NRP1 instructs IL-17-producing ILC3s to drive colitis progression

Abstract

Group 3 innate lymphoid cells (ILC3s) control tissue homeostasis and orchestrate mucosal inflammation; however, the precise mechanisms governing ILC3 activity are fully understood. Here, we identified the transmembrane protein neuropilin-1 (NRP1) as a positive regulator of interleukin (IL)-17-producing ILC3s in the intestine. NRP1 was markedly upregulated in intestinal mucosal biopsies from patients with inflammatory bowel disease (IBD) compared with healthy controls. Genetic deficiency of NRP1 reduces the frequency of ILC3s in the gut and impairs their production of IL-17A in an NF-κB signaling-dependent and cell-intrinsic manner. The diminished IL-17A production in ILC3s altered the composition of the microbiota and improved the outcome of dextran sodium sulfate (DSS)-induced colitis. Furthermore, pharmacological inhibition of NRP1 with EG00229 alleviated the severity of colitis. These observations demonstrated the critical role of NRP1 in the control of intestinal ILC3s, suggesting that NRP1 is a potential therapeutic target for IBD.

Keywords: Group 3 innate lymphoid cells (ILC3s); Inflammatory bowel disease (IBD); Mucosal Immunity; NRP1.

Fig. 1

NRP1 expression was induced in ILC3s following colitis. A mRNA expression of NRP1 in mucosal biopsies from human IBD patients (n = 135) and non-IBD healthy controls (n = 55). The data were obtained from available RNA-seq data (GSE117993). B Correlation analysis of the indicated proinflammatory genes with NRP1 in the dataset from (A). C Dot plot showing the expression of NRP1 in distinct immune cells from healthy human donors. Data were obtained from scRNA-seq of the innate lymphoid cell and T cell compartments (GSE184291). D Representative images of CD3^- NRP1⁺ RORγt⁺ ILC3s in colons from remittent and active human UC patients. Scale bars, 20 μm. NRP1 (green), RORγt (red), CD3 (white) and DAPI (blue) are shown. E Quantification of CD3^- NRP1⁺ RORγt⁺ ILC3 numbers from (D) (n = 15 fields per group). F Heatmap of NRP1 and ILC-related genes from the RNA-seq dataset of intestinal ILC subsets. G Flow cytometric analysis of NRP1 expression in distinct immune cells from the intestines; both a representative histogram and MFI (n  =  4) are shown. H Flow cytometric analysis of NRP1 expression in ILC3s from the colons of DSS-induced colitis or control mice (n  =  4). The data in (E, G and H) are representative of two independent experiments. The data are presented as the means ± SEMs; the statistics shown in (A, E and H) were obtained via unpaired t tests. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 2

NRP1 deficiency impairs IL-17A production by ILC3s and causes remission of colitis. A–J Nrp1^fl/fl and Nrp1^ΔIl7r mice were challenged with 3% (weight per volume) DSS to induce colitis (n = 4). A Experimental strategy. B Frequencies and quantification of ILC3s in the colon from Nrp1^fl/fl and Nrp1^ΔIl7r mice, as evaluated by flow cytometry (pregated live Lin^- CD45⁺ cells). C Total numbers of colonic ILC3 subsets from Nrp1^fl/fl and Nrp1^ΔIl7r mice as measured by flow cytometry. Representative flow cytometry and quantification analysis of IL-17A production by colonic ILC3s (D) and CD4⁺ T cells (E) from Nrp1^fl/fl and Nrp1^ΔIl7r mice after ex vivo restimulation with PMA/ionomycin for 4 h. The severity of colitis was monitored, including daily weight loss (F), representative images of the colons and colon length (G), clinical disease score (H), representative H&E staining of colon sections and quantification of histological scores (I), representative PAS staining of colon sections and statistical analysis of PAS⁺ goblet cells per crypt (counted from 50 villi per field of vision per mouse) (J). Scale bar, 50 µm. K‒O Nrp1^fl/fl mice, Nrp1^ΔIl7r mice and Nrp1^ΔIl7r mice were pretreated with rmIL-17A (625 ng per mouse), followed by DSS challenge (n = 6). K Experimental strategy. The severity of colitis was evaluated, including daily weight loss (L), shortened colon length (M), clinical disease scoring (N), representative H&E staining of colon sections and quantification of histological scores (O). Scale bar, 50 µm. The data are representative of three independent experiments. The data are presented as the means ± SEMs; unpaired Student’s t tests (B, D–E, G–J, M–O) and two-way ANOVA (C, F, L) were used. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 3

NRP1 regulates IL-17A⁺ ILC3s in a cell-intrinsic manner. A–C Mixed BM chimera experiments. A Illustration of mixed BM chimera experiments: BM cells from Nrp1^fl/fl (CD45.1⁺ CD45.2⁺) or Nrp1^ΔIl7r (CD45.2⁺) donor mice were mixed (1:1) and transferred into lethally irradiated WT (CD45.1⁺) mice. After 8–10 weeks of reconstitution, the recipients were treated with DSS or water (n = 4). Representative flow cytometric analysis of the proportions of the indicated populations derived from CD45.1⁺ CD45.2⁺ Nrp1^fl/fl versus CD45.2⁺ Nrp1^ΔIl7r donors, including colonic ILC3s (B) and IL-17A⁺ ILC3s (C). D–J ILC3s adoptive transfer experiment. D Experimental design. Intestinal ILC3s from Nrp1^fl/fl and Nrp1^ΔIl7r mice were adoptively transferred into NCG mice. After 10 d, the mice were given water containing 3% DSS for 7 days, followed by 2 days with regular water. The mice were sacrificed and analyzed on day 19. The severity of colitis, including daily weight loss (E), representative images of the colons and colon length (F), and the clinical disease score (G), was monitored. Representative H&E staining of colon sections and histological scores (H), representative PAS staining of colon sections and statistical analysis of PAS⁺ goblet cells per crypt (I). Scale bar, 50 µm. J Representative flow cytometry data and quantification of colonic IL-17A⁺ ILC3s from DSS-treated mice. The data are representative of three independent experiments. The data are presented as the means ± SEMs; unpaired Student’s t test (F-J) and two-way ANOVA (B–C, E) were used. **P < 0.01; ***P < 0.001

Fig. 4

NRP1 deficiency in ILC3s alleviates colitis. A Experimental design: Nrp1^fl/fl mice and Nrp1^fl/flRorc^cre/+ mice were treated with anti-CD3e (50 μg per mouse once a day) or anti-IgG control intraperitoneally for 9 days (n = 5). B Deletion efficiency of CD3e⁺ cells by flow cytometry. The severity of colitis, including daily weight loss (C), representative images of the colons and colon length (D), and the clinical disease score (E), was monitored. F Frequencies of ILC3s in the colon of Nrp1^fl/fl mice and Nrp1^fl/flRorc^cre/+ mice, as evaluated by flow cytometry. Representative flow cytometry data and quantification of IL-17A production (G) and IL-22 production (H) in colonic ILC3s from Nrp1^fl/fl mice and Nrp1^fl/flRorc^cre/+ mice. Representative H&E staining of colon sections and histological scores (I), representative PAS staining of colon sections and statistical analysis of PAS⁺ goblet cells per crypt (J). Scale bar, 50 µm. The data are representative of three independent experiments and are presented as the means ± SEMs. Unpaired Student’s t test (D-J) and two-way ANOVA (C) were used. *P < 0.05; **P < 0.01; ***P < 0.001

Fig. 5

NF-κB signaling mediates the effect of NRP1 on ILC3s. A-E SMART-seq2 analysis of colonic ILC3s from DSS-treated Nrp1^fl/fl and Nrp1^ΔIl7r mice. A, B GSEA results showing the indicated pathways enriched in Nrp1^fl/fl and Nrp1^ΔIl7r mouse colonic ILC3s. C Heatmap of IL-17A pathway-related genes in colonic ILC3s from Nrp1^fl/fl and Nrp1^ΔIl7r mice. D KEGG analysis of pathways associated with the DEGs. E GSEA of the NF-κB pathway enriched in colonic ILC3s from Nrp1^fl/fl and Nrp1^ΔIl7r mice. F Flow cytometric analysis of p-P65 in colonic ILC3s from Nrp1^fl/fl and Nrp1^ΔIl7r mice after stimulation with PMA or DMSO for 25 min. G Colonic ILC3s from Nrp1^fl/fl and Nrp1^ΔIl7r mice were stimulated with IL-23 with or without an NF-κB inhibitor (JSH23, 10 µM) and the NF-κB agonist betulinic acid (BA, 10 µM) for 16 hours. IL-17A production was evaluated by flow cytometry. H qRT‒PCR analysis of the indicated NRP1 ligands in colonic ILC3s under steady-state conditions and colitis conditions. The values were normalized to those of the housekeeping gene β-actin. I Colonic ILC3s from Nrp1^fl/fl and Nrp1^ΔIl7r mice were stimulated with IL-23 with or without VEGF164 (120 nM) for 16 hours. IL-17A production was evaluated by flow cytometry. J Colonic ILC3s from WT mice were stimulated with IL-23 in combination with DMSO, EG00229 (30 µM) or the mixture of EG00229 and NF-κB agonist betulinic acid (BA, 10 µM) for 16 h. IL-17A production was evaluated by flow cytometry. The data are representative of three independent experiments and are presented as the means ± SEMs. Unpaired Student’s t test (G, I-J) and two-way ANOVA (F, H) were used. *P < 0.05; **P < 0.01

Fig. 6

The microbiota of NRP1-deficient mice protects against colitis. A-F Antibiotic treatment experiment. A Experimental schematic: Nrp1^fl/fl and Nrp1^ΔIl7r mice were treated with antibiotics for 10 days and then challenged with DSS to induce colitis (n = 6). The severity of colitis, including loss of body weight (B), colon length (C), clinical disease scoring (D), representative H&E staining of colon sections and quantification of the histological score (E), representative PAS staining of colon sections and quantification of PAS⁺ goblet cells per crypt (F). Scale bar, 50 µm. G-L Fecal microbiota transplantation experiment. G Experimental schematic: WT recipients were pretreated with antibiotics, followed by transplantation with feces from DSS-treated Nrp1^fl/fl and Nrp1^ΔIl7r mice as indicated. The mice were then challenged with DSS to induce colitis (n = 5). The severity of colitis, including loss of body weight (H), colon length (I), clinical disease scoring (J), representative H&E staining of colon sections and quantification of histological scoring (K), representative PAS staining of colon sections and quantification of PAS⁺ goblet cells per crypt (L), is shown. Scale bar, 50 µm. The data are representative of three independent experiments. The data are presented as the means ± SEMs; an unpaired Student’s t test (C-F, I-L) and two-way ANOVA (B, H) were used. *P < 0.05; **P < 0.01

Fig. 7

NRP1 deficiency alters the microbiota composition following colitis. Box plots representing the richness (Shannon diversity index; A) and evenness (Pielou’s index; B) of microbiome compositions from Nrp1^fl/fl and Nrp1^ΔIl7r mice (n = 4). Boxes represent the interquartile range, and the horizontal line inside the box represents the median. Whiskers span the most extreme data points within 1.5X the interquartile range. C Representation of the relative abundances of specific bacteria at the phylum level in Nrp1^fl/fl and Nrp1^ΔIl7r mice. D Linear discriminant analysis (LDA) showing the differentially expressed species between Nrp1^fl/fl and Nrp1^ΔIl7r mice. E qPCR analysis of Lachnospiraceae, Clostridia and Erysipelotrichales in the feces of Nrp1^fl/fl, Nrp1^ΔIl7r and Nrp1^ΔIl7r mice pretreated with rmIL-17A (625 ng per mouse), followed by DSS challenge (n = 6). F Volcano plot showing differentially altered metabolites in Nrp1^fl/fl and Nrp1^ΔIl7r mice with DSS-induced colitis. (G, H) Correlation analysis between the indicated metabolites and the abundance of Clostridia or Lachnospiraceae in feces from Nrp1^fl/fl and Nrp1^ΔIl7r mice (n = 5 per group). The data are representative of three independent experiments. The data are presented as the means ± SEMs; an unpaired Student’s t test was used. *P < 0.05; **P < 0.01

Fig. 8

Therapeutic effects of NRP1 inhibitors on colitis. A Experimental design: Nrp1^fl/fl and Nrp1^ΔIl7r mice were treated with the NRP1 inhibitor EG00229 (10 mg/kg body weight every other day) or the DMSO control intraperitoneally for 9 days during DSS-induced colitis (n = 4). The severity of colitis, including loss of body weight (B), colon length (C) and clinical disease score (D), is shown. E Representative H&E staining of colon sections and quantification of histological scores. F Representative PAS staining of colon sections and analysis of PAS⁺ goblet cells per crypt. Scale bar, 50 µm. Representative flow cytometry and quantification analysis of IL-17A⁺ ILC3s (G) and IL-22⁺ ILC3s (H) after ex vivo restimulation of ILC3s with PMA/ionomycin for 4 h. I Colonic ILC3s from Nrp1^fl/fl and Nrp1^ΔIl7r mice were stimulated with IL-23 in combination with DMSO or EG00229 (30 µM) for 16 h, respectively. IL-17A production was evaluated by flow cytometry. The data are representative of three independent experiments and are presented as the mean ± SEM; unpaired Student’s t test (C-I) and two-way ANOVA (B) were used. *P < 0.05; **P < 0.01; ***P < 0.001