The aryl hydrocarbon receptor regulates gut immunity through modulation of innate lymphoid cells

Abstract

Innate lymphoid cells (ILCs) expressing the nuclear receptor RORγt are essential for gut immunity presumably through production of interleukin-22 (IL-22). The molecular mechanism underlying the development of RORγt(+) ILCs is poorly understood. Here, we have shown that the aryl hydrocarbon receptor (Ahr) plays an essential role in RORγt(+) ILC maintenance and function. Expression of Ahr in the hematopoietic compartment was important for accumulation of adult but not fetal intestinal RORγt(+) ILCs. Without Ahr, RORγt(+) ILCs had increased apoptosis and less production of IL-22. RORγt interacted with Ahr and promoted Ahr binding at the Il22 locus. Upon IL-23 stimulation, Ahr-deficient RORγt(+) ILCs had reduced IL-22 expression, consistent with downregulation of IL-23R in those cells. Ahr-deficient mice succumbed to Citrobacter rodentium infection, whereas ectopic expression of IL-22 protected animals from early mortality. Our data uncover a previously unrecognized physiological role for Ahr in promoting innate gut immunity by regulating RORγt(+) ILCs.

Figure 1. RORγt⁺ ILCs are the major source of intestinal IL-22 that is dependent on Ahr

Intestinal LPLs were stimulated by PMA and ionomycin for 4 hr. (A and B) RORγt, IL-22 and IL-17 expression in CD3⁻ or CD3⁺ cells were analyzed by flow cytometry. (C and D) Frequencies of IL-22 and IL-17-expressing cells in the RORγt⁺ gated cell populations of the large (LI) and small intestines (SI) were shown. Horizontal lines represented the mean. **P<0.01; ***P<0.001. Error bars represented SEM. Experiments were independently repeated at least three times.

Figure 2. Ahr is important for the accumulation of RORγt⁺ ILCs in the gut

(A) GFP(RORγt) and NKp46 expression were analyzed in intestinal LPLs by flow cytometry after gating on CD3⁻ cells. GFP(RORγt) and CD4 expression were analyzed after gating on CD3⁻NKp46⁻cells. Data were representative of at least three independent experiments. (B) Percentage of NK-22 cells [CD3⁻NKp46⁺GFP(RORγt)⁺] in the CD3⁻ population, as well as percentages of LTi₄ cells [CD3⁻NK46⁻CD4⁺GFP(RORγt)⁺] and LTi₀ cells [CD3⁻NKp46⁻CD4⁻GFP(RORγt)⁺] in the CD3⁻NKp46⁻cell population of the large (LI) and small intestinal (SI) LPLs isolated from Rorc(γ)t^gfp/+;Ahr^+/+ (WT) and Rorc(γ)t^gfp/+;Ahr^−/− (KO) mice were shown. (C) Absolute numbers of NK-22, LTi₄ and LTi₀ cells in the LI and SI LPLs of WT and KO mice were shown. (D) Intestinal LPLs were isolated from 17.5 day of embryos (E17.5) of the indicated genotypes. NK-22, LTi₄ and LTi₀ cells were analyzed by flow cytometry. Data were representative of two independent experiments. (E) Percentages of LTi₄ cells and LTi₀ cells isolated from the E17.5 intestines of Rorc(γ)t^gfp/+;Ahr^+/− (Ctrl) or Rorc(γ)t^gfp/+;Ahr^−/− (KO) were shown. (F) RORγt⁺ cells in CD3⁻ population of LI and SI LPLs isolated from littermate mice of the indicated genotypes after DMSO or FICZ injection were analyzed by flow cytometry. Data were representative of two independent experiments. (G) Percentages of NK-22, LTi₄ and LTi₀ cells from LPLs isolated from Rorc(γ)t^gfp/+;Ahr^+/+ or Rorc(γ)t^gfp/+;Ahr^+/− mice after DMSO or FICZ injection were shown. (B, C, E and G) Horizontal lines showed the mean. *P<0.05; **P<0.01; ***P<0.001. Error bars represented SEM.

Figure 3. Reduction of RORγt⁺ ILCs in Ahr^−/− mice is attributed to a hematopoietic defect

(A and B) Bone marrow cells from mice of the indicated genotypes were transferred into lethally irradiated recipient mice. NK-22 [CD3⁻NKp46⁺GFP(RORγt)⁺], LTi₄ [CD3⁻NK46⁻CD4⁺GFP(RORγt)⁺], LTi₀ [CD3⁻NKp46⁻CD4⁻GFP(RORγt)⁺] and RORγt⁺ T [CD3⁺TCRβ⁺GFP(RORγt)⁺] cells were analyzed by flow cytometry. Data were representative of three independent experiments. (C) Large intestinal LPLs were isolated from mice untreated or treated with antibiotics for 10 days or 8 weeks. Percentages of NK-22, LTi₄ and LTi₀ cells were analyzed by flow cytometry. Data were representative of two experiments. (D) Percentages of NK-22, LTi₄ and LTi₀ cells were analyzed by flow cytometry. Data were representative of three independent experiments.

Figure 4. Ahr-deficient RORγt⁺ ILCs have enhanced apoptosis

(A) Intestinal LPLs were examined by flow cytometry for CD3, GFP, and apoptosis markers (Annexin V and 7-AAD). Data were representative of three independent experiments. (B) Th17 [CD3⁺TCRβ⁺CD4⁺GFP(RORγt)⁺], NK-22 [CD3⁻TCRβ⁻NKp46⁺GFP(RORγt)⁺] and LTi [CD3⁻TCRβ⁻NKp46⁻GFP(RORγt)⁺] cells were sorted from large intestinal LPLs by flow cytometry. Bcl2 and Bcl2l1 mRNA expression in each cell population were analyzed by realtime RT-PCR. RNA samples were pooled from 3 to 4 mice in each experiment. Data were representative of two independent experiments. Error bars represented SEM of triplicate samples of realtime RT-PCR.

Figure 5. Ahr regulates the function of NK-22 and LTi cells to produce IL-22

(A) IL-22, IL-17 and CCR6 expression in large intestinal NK-22, LTi₄ and LTi₀ cells were analyzed by flow cytometry. For IL-22 and IL-17 staining, cells were stimulated for 4 hr with PMA and ionomycin. Data were derived from 3–4 pairs of littermate mice. *P<0.05; **P<0.01. Error bars represented SEM. Experiments were repeated at least three times. (B) Th17 [CD3⁺TCRβ⁺CD4⁺GFP(RORγt)⁺], NK-22 [CD3⁻TCRβ NKp46⁺GFP(RORγt)⁺] and LTi [CD3⁻TCRβ⁻NKp46⁻GFP(RORγt)⁺] cells were sorted from LI and SI LPLs by flow cytometry. IL-22 and IL-17 mRNA expression in each cell population were directly analyzed by realtime RT-PCR. RNA samples were pooled from 3 to 4 mice in each experiment. Data were representative of two independent experiments. Error bars represented SEM of triplicate samples of realtime RT-PCR. (C) AhREs and ROREs at the Il22 locus were shown. (D) RORγt binding at the Il22 locus in EL4 stable cell lines expressing either flag peptide (DFTC) or flag-tagged RORγt (DFTC-RORγt) was monitored using flag ChIP assay. The fold enrichment of RORγt binding at each locus was normalized to DFTC-empty EL4 cells. The Il12b and Il17 loci were used as negative and positive controls, respectively. Error bars represented SEM of triplicate samples of realtime PCR. Data were representative of three independent experiments. (E) Empty MIG (MIG) or RORγt-MIG were co-expressed by retroviral transduction in EL4 cell lines stably expressing either flag peptide (DFTC) or flag-tagged CA-Ahr (DFTC-CA-Ahr). Ahr binding at the Il22 locus was monitored using flag ChIP assay. Data were representative of three independent experiments. (F) Ahr physically interacted with RORγt. HEK 293T cells were transiently transfected with the indicated expression constructs. Whole cell extracts were made and subjected to anti-HA immunoprecipitation, and subsequently immunoblotted with flag or HA antibodies. Data were representative of three independent experiments. (G) EL4 cells were transduced with retroviruses encoding RORγt or/and CA-Ahr. IL-22 mRNA expression was analyzed by realtime RT-PCR. Data were representative of two independent experiments. Error bars represented SEM of triplicate samples of realtime RT-PCR.

Figure 6. IL-23-induced IL-22 expression is impaired in Ahr-deficient mice

(A) CD3⁻GFP(RORγt)⁺ cells were sorted from large intestinal LPLs by flow cytometry. Equal number of cells (1 × 10⁴) was stimulated with or without IL-23 for 24 hr. The concentration of IL-22 in the supernatant was examined by ELISA. Error bars represented SEM of triplicate samples. Data were representative of two independent experiments. (B) Large intestinal LPLs were stimulated with or without IL-23 in the absence of PMA or ionomycin overnight. Expression of IL-22 and RORγt in CD3⁻ cells were analyzed by flow cytometry. Data were representative of three independent experiments. (C) IL-23R mRNA expression in flow cytometry-sorted NK-22 [CD3⁻TCRβ⁻NKp46⁺GFP(RORγt)⁺] and LTi [CD3⁻TCRβ⁻NKp46⁻GFP(RORγt)⁺] cells from LI and SI LPLs was analyzed by realtime RT-PCR. RNA samples were pooled from 3 to 4 mice for each genotype. Data were representative of two independent experiments. (D) Splenocytes of the indicate genotypes were stimulated with or without IL-23 overnight. Expression of IL-22 in CD4⁺CD11c⁻ cells was analyzed by flow cytometry. Data were representative of two independent experiments.

Figure 7. Ectopic expression of IL-22 protects Ahr^−/− mice from C. rodentium infection

(A–C) Littermate mice of the indicated genotypes were infected with C. rodentium. (D–F) Ahr^−/− (KO) mice and Ahr^+/− or Ahr^+/+ (Ctrl) littermate mice were infected with C. rodentium. Six hours after infection, IL-22 expressing plasmid (pRK-mIL-22, Genentech) or control vector (pRK) was administered into the mice via hydrodynamic injection. (A and D) Survival rates were monitored at the indicated time points. (B and E) Body mass was monitored at the indicated time points. Results were shown as mean percentage of body weight change ± SEM. Data were pooled from 3–6 mice for each genotype (B) or for each treatment group (E). Mice were excluded from analyses after the time of death. (C and F) Colony-forming unit (CFU) counts of C. rodentium in the fecal pellets on day 5 after infection were shown. Data were pooled from 3–6 mice for each genotype (C) or for each treatment group (F). Statistical analyses were performed by Mann-Whitney test (C). *P<0.05; **P<0.01; ***P<0.001. Error bars represent SEM. Data were representative of two independent experiments.