The AHR repressor limits expression of antimicrobial genes but not AHR-dependent genes in intestinal eosinophils

Abstract

Intestinal eosinophils express the aryl hydrocarbon receptor (AHR), an environmental sensor and ligand-activated transcription factor that responds to dietary or environmental ligands. AHR regulates tissue adaptation, survival, adhesion, and immune functions in intestinal eosinophils. The AHR repressor (AHRR) is itself induced by AHR and believed to limit AHR activity in a negative feedback loop. We analyzed gene expression in intestinal eosinophils from wild-type and AHRR knockout mice and found that AHRR did not suppress most AHR-dependent genes. Instead, AHRR limited the expression of a distinct small set of genes involved in the innate immune response. These included S100 proteins, antimicrobial proteins, and alpha-defensins. Using bone marrow-derived eosinophils, we found that AHRR knockout eosinophils released more reactive oxygen species upon stimulation. This work shows that the paradigm of AHRR as a repressor of AHR transcriptional activity does not apply to intestinal eosinophils. Rather, AHRR limits the expression of innate immune response and antimicrobial genes, possibly to maintain an anti-inflammatory phenotype in eosinophils when exposed to microbial signals in the intestinal environment.

Keywords: aryl hydrocarbon receptor; eosinophils; intestine; mucosal immunology; transcription factor.

Fig. 1.

AHRR is highly expressed in intestinal eosinophils. (A) Representative flow cytometry plots of sorted cell populations from the small intestine. Unselected cells were used to sort epithelial cells, CD45⁺ cells, and CD45⁻EpCam⁻ cells. Myeloid cells were first enriched by positive magnetic-activated cell sorting selection with anti-CD11b and anti-CD11c microbeads. All populations were pregated on live, single cells. (B) Gene expression was determined by qPCR in sorted cells and normalized to Hprt. Data are pooled from 3 independent experiments for n = 8 to 14 biological replicates. Data from one experiment were previously reported in Diny et al. (C) Representative flow cytometry plots of splenic and small intestinal eosinophils from WT and Ahrr^EGFP/EGFP mice. (D, E) Frequency of AHRR-EGFP⁺ eosinophils and macrophages. Data are pooled from 2 independent experiments for n = 6 mice. (F, G) Mice were injected with 3-MC and analyzed after 3 d. (F) Representative flow cytometry plots of small intestinal eosinophils and epithelial cells from WT and Cyp1a1^Cre/+Rosa26^YFP mice. (G) Frequency of YFP⁺ eosinophils and epithelial cells in the small intestine. Data are pooled from 2 independent experiments for n = 7 mice. (H–M) BMDEos were generated from Ahrr^EGFP/EGFP (H–J) or Cyp1a1^Cre/+Rosa26^YFP mice (K–M) and WT controls. (I, J) BMDEos at day 13 of culture were treated with 5 nM FICZ or dimethyl sulfoxide control for 24 h, and the frequency of GFP⁺ cells was determined by flow cytometry. Data are pooled from 2 independent experiments for n = 6 biological replicates; mean ± SD is shown. (L, M) BMDEos were cultured to day 11 and then cultured for 3 d with 5 nM FICZ or dimethyl sulfoxide control. The frequency of YFP⁺ cells was determined by flow cytometry on day 14 of culture. Data are pooled from 3 independent experiments for n = 6 biological replicates. (J, M) Mean ± SD is shown. Eos = eosinophils; Mac = macrophages; MHC-II = major histocompatibility complex class II; SI = small intestine.

Fig. 2.

AHRR does not repress AHR-dependent genes in intestinal eosinophils. (A) Gating for small intestinal (SI) eosinophils following enrichment with CD11b microbeads. Cells are pregated on single, live, CD45⁺ cells. (B–E) RNA sequencing (RNA-Seq) of WT, Ahr^−/−, and Ahrr^−/− eosinophils. n = 6 mice (3 males, 3 females) were analyzed per genotype. (B) Heatmap of 56 genes with differential expression (|log₂(FC)|≥1 and adjusted P value <0.05) between eosinophils from WT and Ahrr^−/− mice. (C) Number of differentially expressed (DE) genes in Ahr^−/− vs. WT and Ahrr^−/− vs. WT eosinophils. Only 9 genes were found in both comparisons. (D) Heatmap of 9 genes with differential expression in both Ahr^−/− and Ahrr^−/− eosinophils. (E) Gene expression of canonical pathway genes in eosinophils from Ahr^−/−, WT, and Ahrr^−/− mice. Data were analyzed by t test comparing Ahrr^−/− with WT eosinophils. (F, G) BMDEos were generated from WT and Ahrr^−/− mice. (F) Gene expression in BMDEos was determined by qPCR and normalized to Hprt. Data are from n = 4 to 5 biological replicates (mean ± SEM) and were analyzed by 2-way analysis of variance with Sidak correction. **P < 0.01; ***P < 0.001; ****P < 0.0001. (G) EROD activity in BMDEos from WT and Ahrr^−/− mice following treatment with FICZ or dimethyl sulfoxide (DMSO) control. Data were pooled from n = 3 independent experiments (mean ± SEM) and analyzed by t test with Holm-Sidak correction for multiple comparisons. F = female; M = male; MHC-II = major histocompatibility complex class II; ns = not significant; SSC-A = side scatter area; TPM = transcripts per million.

Fig. 3.

AHRR limits expression of antimicrobial genes in intestinal eosinophils. (A) Volcano plot of differentially expressed genes in Ahrr^−/− vs. WT intestinal eosinophils. (B) Gene Ontology analysis of differentially expressed genes was conducted using DAVID. (C) Heatmap of example genes within each enriched pathway. (D) Gene expression was determined by qPCR in sorted small intestinal eosinophils from WT and Ahrr^−/− mice and normalized to Hprt. Data are from n = 8 mice per genotype and were analyzed by t test. *P < 0.05. (E–J) Small intestinal eosinophils from WT and Ahrr^−/− mice were quantified and analyzed by flow cytometry. Data are pooled from 2 independent experiments for n = 6 to 7 mice per genotype. (K–P) Production of ROS in response to PMA stimulation in BMDEos from WT and Ahrr^−/− BMDEos. (K–M) Combined extracellular and intracellular ROS was measured with luminol. (N–P) Extracellular ROS was measured with isoluminol. (K, N) Representative plots of ROS production. (L, M, O, P) Maximal and total ROS production was determined in n = 3 independent experiments and normalized to WT BMDEos. Data were analyzed by 1-sample t test. *P < 0.05. (Q) Gene expression in sorted intestinal eosinophils was determined by qPCR and normalized to Hprt. Data are from 1 experiment with n = 4 mice per genotype and were analyzed by t test. *P < 0.05; **P < 0.01. AU = arbitrary units; gMFI = geometric mean fluorescent intensity; MFI = mean fluorescent intensity; SSC = side scatter.