Beneficial and detrimental consequences of AHR activation in intestinal infection

Abstract

The ligand dependent transcription factor aryl hydrocarbon receptor (AHR) is an environmental sensor whose activation can have physiologically beneficial or detrimental consequences for host immune responses depending on the ligand. Here we investigated the hypothesis that prolonged AHR activation either due to inefficient ligand metabolism or due to genetic manipulation may underlie the distinction between beneficial and detrimental effects. Our data indicate that prolonged AHR activation caused toxic endpoints for liver and thymus but was not per se interfering with the host response to infection with the intestinal pathogen C.rodentium. Genetically driven constitutive AHR activation improved resistance to infection, whereas prolonged AHR activation by the pollutant TCDD resulted in delayed clearance of C.rodentium associated with a suppression in antibody production. Combined single cell RNAseq and ATAC-seq analysis provided evidence that TCDD, but not genetic AHR activation, negatively affected dendritic cell functions such as activation, maturation and antigen presentation. Thus, the detrimental impact of environmental pollutants such as TCDD on immune responses cannot solely be attributed to aberrantly prolonged activation of AHR.

Keywords: ATAC-seq; Aryl hydrocarbon receptor; Citrobacter rodentium; intestine; scRNAseq.

Expanded View 1 –. Prolonged AHR activation reduces DSS-induced inflammation.

(A, D) Experimental scheme, (B, E) body weight curves and (C, F) Lcn2 levels in faecal samples at the indicated timepoints in (A-C) Ahr^CAIR/CAIR and WT mice (WT, n=4; Ahr^dCAIR/dCAIR, n=2) and (D-F) TCDD and vehicle-treated mice (Vehicle, n=4, TCDD, n=5). Each dot represents the mean + SD (B, E) and bars represent the mean + SD (C, F). Two-way ANOVA with Šídák’s multiple comparisons test (C, F).

Expanded View 2 –. Delayed clearance of C. rodentium in mice exposed to TCDD.

(A-B) C. rodentium burden in faecal pellets of female mice infected with C. rodentium 6 days after an oral dose of TCDD or vehicle. Data are from 1 experiment with n=4–6 per group. (B) C. rodentium burden in faecal pellets of mice infected with C. rodentium 6 days after an oral dose of 10 μg/kg TCDD or vehicle before infection. Data are representative of 2 experiments with n=4–5 mice per group. Each dot corresponds to one mouse and bars represent the mean + SD. Two-way ANOVA with Šídák’s multiple comparisons test (A, B). *p<0.05, **p<0.01, ***p<0.001.

Expanded View 3-. TCDD suppresses production of NP-specific IgG antibodies.

(A, B) Levels of NP-Ficoll and NP-CGG-specific IgG subclasses at the indicated sera dilutions 14 days after immunization with 10 μg NP-Ficoll or NP-CGG, respectively. Mice received 10 μg/kg TCDD or vehicle 6 days before immunization. Data are representative of 2 experiments (Vehicle, n=10; TCDD, n=10, Naïve =2). Each dot represents the mean of 10 mice + SD. Two-way ANOVA with Tukey’s multiple comparisons test (A-B). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Expanded View 4-. TCDD affects immune cell composition in the c-MLN.

Cell numbers of total immune cells, and dendritic cell and B cell populations in the c-MLN of mice infected with C. rodentium 6 days after receiving 10 μg/kg TCDD or vehicle. Data analysed by flow cytometry representing 1 experiment (vehicle, n=6; TCDD, n=4). Each dot represents one mouse, and the bars show the mean + SD. Student’s t-test. *p<0.05, **p<0.01, ***p<0.001.

Expanded View 5-. Cell types identified in the single cell RNAseq and ATACseq datasets.

(A) Uniform manifold approximation and projection (UMAP) using scRNA-seq or chromatin accessibility. Cell types were assigned using the RNA-seq assay and visualised on both RNA- and ATAC-seq projections. (B) Proportion of dendritic cell subpopulations and of all other immune cells. (C) Dot plots adapted from Seurat DotPlot showing the putative marker genes across cell types defined in the vehicle sample in the MLN. The size of the dot represents the percentage of cells within a cell type expressing the marker gene, while colour represents the average scaled expression level of the marker gene in the cell type.

Figure 1 -. AHR expression in intestinal immune cell populations.

AHR-TdTomato expression across immune cell types in the colon lamina propria and colon-draining lymph node was determined by flow cytometry. Bar plots represent the geometric MFI of concatenated events from cells obtained from n=2 mice per group. Representative of 3 experiments.

Figure 2 –. Prolonged AHR activation

(A-C) Cyp1a1 gene expression was determined by qPCR from the indicated intestinal sections from Ahr^dCAIR/dCAIR mice and WT mice (A) and 6 days after a single administration of 10 μg/kg TCDD (B-C) and presented as log2 Fold change relative to Hprt. Each dot represents one mouse and the bars show the mean + SD. Data correspond to 1 experiment with n=2–4 per group. Representative of 2 experiments. (D) Cyp1a1 gene expression from the indicated organs was determined by qPCR. Timepoints correspond to the hours after a single administration of 250 mg/kg I3C or 10 μg/kg TCDD in corn oil. Expression values were normalized to Hprt and to values from the vehicle-treated mice. Each dot represents the mean of 2–3 mice + SD. (E) TCDD levels detected in the indicated organ and timepoint following a single administration of 10 μg/kg TCDD in corn oil normalized to the organ weight. Each dot represents one mouse for adipose and liver samples (n=3 per group) and a pool of samples from 3 mice for the serum, small intestine (duodenum and ileum) and colon. (F-G) Weight of thymus and liver of thymus and liver from Ahr^dCAIR/dCAIR or TCDD treated mice. Each dot represents one mouse and the bars show the mean + SD. Data are from 1 experiment with n=3–5 per group. (H) Representative Oil Red O (ORO) staining in liver sections of Ahr^dCAIR/dCAIR mice and TCDD-treated mice. Scale bars, 50 μm (left panel). Quantification of ORO stain (right panel). Each dot represents one mouse and the bars show the mean + SD. Data are from 1 experiment with n=3–5 per group. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. Student’s t-test (A, B), Two-way ANOVA with Šídák’s multiple comparisons test (C, D), One-way ANOVA with Tukey’s multiple comparisons test (F-H).

Figure 3 –. Impact of prolonged AHR activation to Citrobacter rodentium infection.

(A) C. rodentium burden in faecal pellets, (B) IL-22 protein levels in supernatants from colon explant cultures, (C) Lcn2 protein levels in faecal pellets from Ahr^dCAIR/dCAIR and WT mice (n=6–8 per group), representative of 2 experiments. (D) C. rodentium burden in faecal pellets, (E) IL-22 protein levels in supernatants from colon explant cultures and (F) Lcn2 protein levels in faecal pellets from TCDD-treated and vehicle-treated mice. Mice were infected 6 days after TCDD or vehicle administration, n=4–5 per group and representative of 2–3 experiments. Each dot represents samples collected from one mouse and the bars show the mean + SD. Two-way ANOVA with Šídák’s multiple comparisons test (A, C, D, F), Student’s t-test (B, E). *p<0.05, **p<0.01, ***p<0.001.

Figure 4 –. TCDD reduces levels of C. rodentium-specific antibodies.

(A) Levels of C.rodentium-specific Ig subclasses at the indicated sera dilutions 14 days after C. rodentium infection in mice that received 10 μg/kg TCDD or vehicle 6 days before infection. Data are and representative of 2 experiments (vehicle, n=11; TCDD, n=10, Naïve =2). Each dot represents the mean + SD. (B) Levels of indicated Ig subclasses at the indicated sera dilutions 14 days after C. rodentium infection in Ahr^dCAIR/dCAIR and WT mice. Data are from 2 experiments (WT, n=5; Ahr^dCAIR/dCAIR, n=4, Naïve =1). Each dot represents the mean + SD. Two-way ANOVA with Tukey’s multiple comparisons test (A, B). *p<0.05, **p<0.01, ***p<0.001.

Figure 5 –. Single cell RNAseq analysis

(A) Experimental setup for sc Multiome from dendritic cells and CD45+ cells. (B) Dot plot showing the putative marker genes across cell types defined in the vehicle sample in the Colon. The size of the dot represents the percentage of cells within clusters expressing the gene, while the colour scale represents the average expression level in that cluster. (C) Gene ontology enrichment analysis of biological processes for differentially expressed genes in TCDD and Ahr^dCAIR/dCAIR mice compared to vehicle for the indicated cell type. The colour scale represents the expression change bias, which represents the proportion of upregulated and downregulated genes that belong to that pathway, while the size of the dot shows the Z-score. Differentially expressed genes were defined as those with an adjusted p-value <0.01 and enriched pathways have a p-value <0.01.

Figure 6 –. Gene expression and chromatin accessibility in cDC1 of Ahr^dCAIR/dCAIR mice compared to TCDD.

ATACseq signal profiles of differentially accessible peaks and linked genes that are also differentially expressed in cDC1 populations in colon and MLN. Differentially accessible peaks are shaded in grey, while violin plots show gene expression in the respective population and condition. ATAC-seq profiles were generated from pseudo-bulk chromatin accessibility data and violin plots show normalized expression values at the single cell level.