Aryl hydrocarbon receptor activation reduces dendritic cell function during influenza virus infection

Abstract

It has long been known that activation of the aryl hydrocarbon receptor (AhR) by ligands such as 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) suppresses T cell-dependent immune responses; however, the underlying cellular targets and mechanism remain unclear. We have previously shown that AhR activation by TCDD reduces the proliferation and differentiation of influenza virus-specific CD8(+) T cells through an indirect mechanism; suggesting that accessory cells are critical AhR targets during infection. Respiratory dendritic cells (DCs) capture antigen, migrate to lymph nodes, and play a key role in activating naive CD8(+) T cells during respiratory virus infection. Herein, we report an examination of how AhR activation alters DCs in the lung and affects their trafficking to and function in the mediastinal lymph nodes (MLN) during infection with influenza virus. We show that AhR activation impairs lung DC migration and reduces the ability of DCs isolated from the MLN to activate naive CD8(+) T cells. Using novel AhR mutant mice, in which the AhR protein lacks its DNA-binding domain, we show that the suppressive effects of TCDD require that the activated AhR complex binds to DNA. These new findings suggest that AhR activation by chemicals from our environment impacts DC function to stimulate naive CD8(+) T cells and that immunoregulatory genes within DCs are critical targets of AhR. Moreover, our results reinforce the idea that environmental signals and AhR ligands may contribute to differential susceptibilities and responses to respiratory infection.

FIG. 1.

Suppression of CD8⁺ T cell activation requires the ligand-activated AhR complex binds to DNA. Naive wild-type (AhR^+/+) and AhR^dbd/bdd mutant mice received a single oral dose of either peanut oil vehicle control (Veh) or TCDD (100 μg/kg) 1 day before i.n. x31 infection. A 10-fold greater amount of TCDD is used because these mice have the d allele of AhR and therefore require a higher dose of TCDD than wild-type C57Bl/6 mice (ahr b allele). For these experiments, wild-type mice that do not bear a mutated ahr gene (and express ahr^d allele) are generated in house. They have been backcrossed > 10 times onto a C57Bl/6 genetic background. Importantly, the B6.Ahr^d/d mice dosed with 100 μg/kg TCDD present the same magnitude defect in their CD8⁺ T cell response to infection as B6.Ahr^b/b mice treated with 10 μg/kg, indicating that they are sensitive to this immunomodulatory effect (data not shown). MLN cells were collected 9 days after infection and stained with fluorochrome-conjugated reagents for flow cytometric analysis, as described in the Materials and Methods. (A) The average number of CD8⁺ T cells specific for an immunodominant influenza virus peptide, NP_366–374, was determined using MHC class I–restricted tetramers and anti-CD8 Abs. (B) Cells were restimulated in vitro and then stained with anti-CD8 and anti-IFN-γ. Graphs depict the average number of IFN-γ⁺CD8⁺ cells in the MLN. Data are representative of two separate experiments with similar results (five to seven mice in each group). An asterisk indicates a significant difference compared with vehicle control mice of the same genotype (p ≤ 0.05).

FIG. 2.

TCDD suppresses the response of CD8⁺ T cells if it is administered on or before the fourth day of infection. Naive 6- to 8-week-old B6 mice were given a single oral dose of TCDD (10 μg/kg) on the indicated day relative to x31 infection (day 0). Vehicle control mice received a single oral dose of peanut oil on day 1. MLN were removed 7 days after infection. Cells were processed and stained with fluorochrome-conjugated phenotypic markers. CTL are defined as CD8⁺ T cells bearing a CD44^hiCD62L^lo phenotype. An asterisk indicates a significant difference compared with vehicle-treated mice (n = 6 per group, p ≤ 0.05).

FIG. 3.

Trafficking of DCs from the lung to MLN. B6 mice (n = 8–10 per group) were given a single dose of TCDD (10 μg/kg) or peanut oil (Veh) 1 day prior to x31 infection. (A) CFSE (8mM) was instilled (i.n.) before or after infection. Mice were sacrificed 18 h after CFSE treatment. MLN cells were processed and analyzed by flow cytometry. DCs are defined as CD11c⁺MHCII⁺ cells, and lung DCs that have migrated to the MLN are defined as CFSE⁺CD11c⁺MHCII⁺ cells. (B) Bar graphs show that ≥ 95% of leukocytes in the lung are CFSE labeled regardless of TCDD treatment. CFSE⁺ cells at distal sites (e.g., spleen and other lymph nodes) were negligible (data not shown). (C) Graphs show the average total number of DCs (top row) and number of CFSE⁺ DCs (bottom row) in the MLN during the very early, early, and middle phases of infection. As a control for infection, naive mice were administered PBS (i.n.). This experiment was repeated at least three times with similar results. An asterisk indicates p value ≤ 0.05.

FIG. 4.

Isolated CD11c⁺ MLN cells from infected mice activate naive CD8⁺ T cells in vitro. (A) B10 mice were infected (i.n.) with Mem/102, and 3 days later, CD11c⁺ cells from MLN were isolated as described in the Materials and Methods. Isolated CD11c⁺ cells were serially diluted from 1 × 10⁵ per well (T:DC ratio = 2:1) to 0.06 × 10⁵ per well (T:DC = 32:1) and used to stimulate CFSE-labeled naive (CD44^lo) F5 CD8⁺ T cells (2 × 10⁵ per well). Cells were collected after 3 days of in vitro coculture and stained with Abs for analysis by flow cytometry. F5 CD8⁺ T cells were analyzed for the decay of CFSE intensity and upregulation of CD44 expression. (B) The graph shows the percentage of F5 CD8⁺ T cells that had proliferated (CFSE^decay). Isolated CD11c^neg MLN cells from mice infected with Mem/102 were employed as a negative control. CD11c⁺ MLN cells isolated from mice infected with x31 were used as a control for the specificity of response.

FIG. 5.

AhR activation impairs the function of DCs from MLN. B10 mice were given a single dose of TCDD (10 μg/kg) or peanut oil (Veh) 1 day prior to Mem/102 infection. Isolated CD11c⁺ MLN cells from Mem/102-infected mice were cocultured with naive CD8⁺ T cells as described in Fig. 4. To obtain enough purified DCs from MLNs of infected mice requires pooling of MLN from ∼30 mice per group. Therefore, this type of ex vivo functional assay is not performed with a large number of experimental replicates. Instead, diminution of the measured end point with dilution of cells is used to validate specificity and reproducibility of the data. Data shown are representative of three separate experiments with similar results. (A) Histograms represent the decay of CFSE density and cell generation (G1–G6). Gray filled histograms depict the CFSE density of naive F5 CD8⁺ T cells cultured without CD11c⁺ cells. The bar graph shows the percentage of F5 CD8⁺ T cells that have undergone successive cell divisions. Similar findings were observed at other T cell:DC ratios (not shown). (B) Gated events in the dot plots (4:1 T cell:DC ratio) show the frequency of activated F5 CD8⁺ T cells, as defined by downregulation of CFSE intensity (CFSE^decay) and upregulation of CD44 expression (CD44^hi). There were no cells within this gate if naive F5 CD8⁺ T cells were cultured without CD11c⁺ cells. (C) Culture supernatants were collected and IFN-γ levels were measured by ELISA. Wells in which CD11c⁺ cells were cultured for 3 days without F5 T cells or in which F5 T cells were cultured without CD11c⁺ cells did not have detectable IFN-γ (not shown).