Respiratory syncytial virus downregulates the airway aryl hydrocarbon receptor pathway: implication for the development of a novel therapeutic target

Abstract

Despite recent advances in preventative options, respiratory syncytial virus (RSV) infection is still a major cause of hospitalizations of young children and older adults, with no specific treatment available. The aryl hydrocarbon receptor (AHR) is a transcription factor originally identified as the mediator of the toxic effects of environmental pollutants but later shown to be also activated by dietary and endogenous ligands. AHR is involved in various physiological and pathophysiological processes, including host response to infections. Many clinically relevant viruses have been shown to induce AHR activation as a strategy to evade antiviral immunity and promote replication, including the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). It is currently not known whether RSV infection affects the AHR pathway. In this study, we investigated the effects of RSV infection on the AHR signaling pathway by using in vitro and in vivo experimental models. We found that RSV infection led to inhibition of the AHR-dependent gene transcription in human airway epithelial cells and in lungs of mice. Human lung epithelial cells lacking AHR showed upregulation of genes related to inflammatory response and airway remodeling, as well as increased production of pro-inflammatory mediators in response to RSV infection. In contrast, administration of the dietary AHR ligand indole-3-carbinol (I3C) to mice led to beneficial effects on RSV-associated disease, including anti-inflammatory and antiviral activity. Collectively, our results suggest that the AHR has a protective role during RSV infection and therefore its modulation can be explored as a novel therapeutic target for RSV-induced disease.

Graphical Abstract

NEW & NOTEWORTHY

Our study reveals that respiratory syncytial virus (RSV) downregulates the aryl hydrocarbon receptor (AHR) pathway in human airway epithelial cells and mice lungs. Loss of the AHR in lung cells led to exacerbated inflammatory response and the AHR ligand indole-3-carbinol (I3C) showed in vivo anti-inflammatory and antiviral activity during RSV infection. Our data suggest that AHR plays a protective role during RSV infection and can be explored as a novel therapeutic target.

INTRODUCTION

Respiratory syncytial virus (RSV) is the single most important virus causing acute respiratory tract infections in children and is a major cause of severe respiratory morbidity and mortality in the elderly (1, 2). There is currently no approved therapeutic for RSV infection (ribavirin is not recommended due to limited efficacy) and vaccines are available only for pregnant women and older individuals, but not for the pediatric population (3). Severe infection with RSV is associated with recurrent wheezing, reduced lung function, and a two-fold increased risk of premature adult death from respiratory disease (4). In animal models of infection, RSV is associated with long-term airway changes including expanded goblet cell populations, enlarged alveolar spaces (5) and enhanced collagen deposition (6), pointing to possible lifetime structural alterations of the lung. Despite this significant burden, multiple knowledge gaps remain and advancing our understanding will help to better address prevention of severe RSV disease and treatment for vulnerable populations.

The aryl hydrocarbon receptor (AHR) represents a complex system that provides a major line of defense against environmental toxins, although it can also bind endogenous and dietary ligands (7). Although chronic activation of the AHR pathway by pollutants can lead to detrimental health outcomes, the scientific literature supports a role for AHR in normal physiological homeostasis, which includes regulation of reproductive, cardiovascular, renal, and immune systems, as AHR, in association with the AHR nuclear translocator (ARNT), modulates the expression of genes involved in proliferation, apoptosis, cell growth/differentiation and cellular stress (8, 9). AHR is expressed in different types of lung cells, including club cells, alveolar type II cells, and endothelial cells, and it has been shown to play a protective role in allergic asthma (10) and in the context of different types of lung injury such as hyperoxia-induced lung inflammation and damage, both in adult and neonatal mice (8). In models of chronic obstructive pulmonary disease, AHR-deficient lung cells exhibit increased reactive oxygen species generation and decreased expression of several antioxidant enzymes when acutely exposed to cigarette smoke (8). In addition, cellular depletion of AHR can act as a switch for epithelial-mesenchymal transition (EMT) (11) and decreased expression of AHR plays a detrimental role in in vitro and in vivo models of lung fibrosis (12). While multiple studies have found that AHR plays an important role during infections, whether AHR signaling improves or worsens host resistance and clinical outcomes depends on the pathogen and target organ. Several viruses, such as Zika and severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), have been shown to induce AHR activation as a strategy to evade antiviral immunity and promote virus replication (13-15). It is currently not known whether RSV infection affects the AHR pathway, and this understanding could provide insights into if and how this pathway could be explored as a therapeutic target.

In this study, using in vitro and in vivo experimental models of infection, we found that RSV led to decreased AHR-dependent gene expression and signaling. Cells lacking AHR showed a significant increase in the expression of pro-inflammatory genes and genes associated with airway remodeling. In addition, administration of the dietary AHR ligand indole-3-carbinol (I3C) to mice infected with RSV significantly reduced pro-inflammatory mediator secretion and viral replication, improving clinical disease.

MATERIALS AND METHODS

Virus stocks

RSV Long strain was grown in HEp-2 cells (ATCC Cat# CCL-23, RRID:CVCL_1906) and purified by ultracentrifugation in sucrose density gradient. Virus pools were aliquoted and stored at −80 °C. The titer of viral pools was determined by methylcellulose plaque assay and ranged from 8 to 9 log₁₀ plaque-forming units (PFU)/mL.

Cells and in vitro RSV infection

Primary human small airway epithelial (SAE) cells (Lonza Cat# CC-2547), isolated from normal human lung tissue of cadaveric donors, were cultured according to the manufacturer’s instructions in Small Airway Epithelial Cell Growth Medium, which consists of Small Airway Epithelial Basal Medium (Lonza Cat# CC-3119) with supplements and growth factors (Lonza Cat# CC-4124). SAE cells from two donors (one female 25 years old Lot# 18TL179344, and one male 56 years old Lot# 20TL107065, both White) were used in this study. A549 cells (ATCC Cat# CCL-185, RRID:CVCL_0023), a lung epithelial cell line isolated from the lung tissue of a White, 58-year-old male with lung cancer, and A549 wild-type (WT) and AHR knockout (KO) cells generated by CRISPR/Cas9 (16) were maintained in F-12K medium supplemented with antibiotics and 10% FBS. HEp-2 cells, an epithelial cell line derived via HeLa contamination (HeLa was isolated from the cervix of a 31-year-old, Black female with adenocarcinoma), were maintained in Minimum Essential Medium (MEM) (Thermo Fisher Scientific Cat# 11095-080) supplemented with antibiotics and 10% FBS. For all graphs, data collected from female-derived cells were denoted by circles, while those from male-derived cells were denoted by squares. RSV infection in cells was done as previously described (17-20). SAE and A549 cells were infected with RSV at a multiplicity of infection (MOI) of 3 and HEp-2 cells were infected with RSV at an MOI of 0.01. Mock-infected cells were used as a control.

Mice and in vivo RSV infection

All studies involving mice were approved by the UTMB Institutional Animal Care and Use Committee (protocol# 9001002). Female BALB/cJ mice (Strain #:000651, RRID:IMSR_JAX:000651) were purchased from The Jackson Laboratory (Bar Harbor, ME). Animals were housed in groups of four and had unlimited access to food and water. Mice were 11 to 12 weeks old at the start of experiments and were randomly assigned to experimental groups. Under light anesthesia, the mice were infected intranasally with 5×10⁶ PFU of RSV in 50μl of phosphate-buffered saline (PBS). Controls were mock inoculated with same volume of PBS. The mice were euthanized at different experimental days following RSV infection and samples were collected.

Measurements of viral titers

RSV titers in cell culture medium and lung tissue of mice were determined by methylcellulose plaque assay on HEp-2 cells (21). Mouse lung tissue samples were homogenized in 1 mL of MEM and centrifuged twice at 14,000 rpm for 1 min at 4 °C.

AHR ligands

Stock solutions of 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) (Thermo Fisher Scientific Cat# NC9311376), Benzo[a]pyrene (BaP) (Sigma-Aldrich Cat# B1760), 6-formylindolo[3,2-b] carbazole (FICZ) (Sigma-Aldrich Cat# SML1489), and I3C (Sigma-Aldrich Cat# I7256) were prepared in DMSO and further diluted to working concentration in cell culture medium. The final concentration of DMSO in the cell culture medium was 0.01%. When given to mice, I3C dissolved in DMSO was diluted in corn oil (5% DMSO + 95% corn oil) and administered by oral gavage (100 μL).

Luciferase reporter assay

A549 cells were transfected with pGL4.43[luc2P/XRE/Hygro] (Promega Cat# E4121 Part# E412A), a designed vector that uses an optimized luc2 firefly luciferase gene in the pGL4 backbone. The pGL4.43[luc2P/XRE/Hygro] Vector contains three copies of a xenobiotic response element (XRE) that drives transcription of the luciferase reporter gene luc2P (Photinus pyralis, a synthetically derived luciferase sequence with humanized codon optimization (Transcription factor: AHR; Binding site: XRE; GenBank accession #: JQ858513). After transfection, cells were incubated overnight (18 h), and in the next day were either infected with RSV at MOI of 3 or stimulated with the AHR ligand TCDD (10 nM) as positive control. At 15 and 24 hours postinfection (hpi) or 3 h after AHR ligand stimulation, cells were lysed with Firefly Luciferase Lysis Buffer (Signosis Cat# LS-001) and the luciferase reporter activity was measured using Firefly Luciferase Substrate (Signosis Cat# LUC015). Luminescence was detected using a SpectraMax iD3 microplate reader (Molecular Devices) and luciferase activity is shown in relative light units (RLU).

Western blot

Whole-cell lysates were prepared with RIPA buffer (Cell Signaling Cat# 9806) following the manufacturer’s protocol. For nuclear extraction, performed as described previously with some modifications (22), cells were collected, pelleted by centrifugation (1500 × g for 5 min at 4°C), resuspended with cold buffer A (10 mM HEPES-pH 7.4, 10 mM NaCl, 0.1 mM EDTA, 0.1 mM EGTA, and protease and phosphatase inhibitor cocktail) and incubated on ice for 15 min. NP40 was added to a final concentration of 0.5%, samples were vortexed for 30 s, and incubated on ice for 3 min. After centrifugation at 6000 rpm for 2 min at 4°C, the supernatant (cytoplasmic fraction) was collected, and the pellet (containing nuclei) was resuspended in cold buffer B (buffer A supplemented with 1.7 M sucrose) and centrifuged at 12,000 rpm for 30 min at 4°C. The nuclei pellet was resuspended in cold buffer C (20 mM HEPES, 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, and protease and phosphatase inhibitor cocktail) and placed in a shaker for 1 h at 4°C. After centrifugation at 12,000 rpm for 5 min at 4°C, the supernatant was saved as nuclear fraction (effectiveness of the nuclear/cytoplasm fractionation is shown in the Supplemental Fig. S1). Protein concentration was determined with Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, 23225). Protein samples were separated by SDS-PAGE and transferred onto polyvinylidene difluoride (PVDF) membrane. Nonspecific binding was blocked by immersing the membrane in Tris-buffered saline with 0.1% Tween 20 (TBST) blocking solution containing 5% skim milk powder. After blocking, the membranes were incubated with the primary antibody overnight at 4°C, followed by the appropriate secondary antibody for 1 h at room temperature. Proteins were detected using enhanced chemiluminescence (ECL) reagents. The western blot imaging was performed on an Odyssey Fc Imager (LI-COR). Bands were quantified using Image Studio software (LI-COR) and the relative normalized ratio was calculated as described in (23). The antibodies used are as follows: AHR rabbit monoclonal (Cell Signaling Technology Cat# 83200, RRID:AB_2800011; 1:1000), HIF-1β/ARNT rabbit monoclonal (Cell Signaling Technology Cat# 5537, RRID:AB_10694232; 1:1000), HDAC1 rabbit monoclonal (Cell Signaling Technology Cat# 34589, RRID:AB_2756821; 1:1000), and β-actin mouse monoclonal (Santa Cruz Biotechnology Cat# sc-47778 HRP, RRID:AB_2714189; 1:2000). Specificity of the AHR antibody was verified by us in AHR KO cells and is shown in Fig. 4A and the specificity of the HIF-1β/ARNT antibody was verified by others using siRNA (24).

RNA extraction and reverse transcription-quantitative PCR (RT-qPCR)

RNA from cells was extracted using the RNeasy mini kit (Qiagen Cat# 74104). Lung tissue was homogenized in TRIzol reagent (Thermo Fisher Scientific Cat# 15596018) and after phase separation with chloroform, the top aqueous layer was mixed with ethanol and further processed using the RNeasy Mini kit spin columns by following manufacturer’s protocol. DNase digestion was performed using RNase-Free DNase Set (Qiagen Cat# 79254). RNA was measured using a SpectraDrop Micro-Volume Microplate (SpectraMax iD3, Molecular Devices). cDNA was prepared using iScript Reverse Transcription Supermix (Bio-Rad, 1708841) and qPCR was performed with specific primers and SYBR Green master mix (ABclonal Cat# RK21203). Data were analyzed using the delta-delta Ct method and 18S rRNA was used as endogenous control. Primer sequences are available upon request.

Chromatin Immunoprecipitation (ChIP)

ChIP assays were performed using a combination of a two-step cross-linking protocol (25), and the ChIP-IT High Sensitivity kit (Active Motif Cat# 53040). Chromatin (15 μg) was immunoprecipitated with 10 μL of AHR rabbit monoclonal antibody (Cell Signaling Technology Cat# 83200, RRID:AB_2800011) or 5 μL of ARNT/HIF-1β rabbit polyclonal antibody (Novus Cat# NB100-110, RRID:AB_10003150). A sample immunoprecipitated with Normal Rabbit IgG polyclonal antibody (Cell Signaling Technology Cat# 2729, RRID:AB_1031062) was used as negative control. After IP, cross-links were reversed, proteins were digested with Proteinase K, and DNA was purified. ChIP DNA analysis was done by qPCR using SYBR Green master mix (ABclonal Cat# RK21203) and primers spanning the human cytochrome P450 family 1 subfamily A member 1 (CYP1A1) proximal (XRE1) (26) and central (XRE4-6) promoter regions (27). The results were normalized to the input samples and expressed in fold change relative to negative control (DNA immunoprecipitated with IgG).

Cytokines and chemokines analysis

Cytokines and chemokines levels were measured in cell culture medium by using the Bio-Plex Pro Human Cytokine Screening Panel 48-Plex (Bio-Rad Cat# 12007283) and in bronchoalveolar lavage fluid (BALF) of mice by using the Bio-Plex Pro mouse cytokine 23-plex assay (Bio-Rad Cat# M60009RDPD). The levels of human interferon (IFN)-β in cell culture medium were measured by ELISA (PBL Cat# 41410).

Microarrays

RNA was extracted from A549 WT and AHR KO cells using the RNeasy mini kit (Qiagen Cat# 74104). Samples from one experiment done in triplicate were pooled, and microarray analysis was performed by Arraystar Inc. (Rockville, MD, USA). Differentially expressed genes (DEGs) were identified through fold change and volcano filtering. Pathway analysis was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG, RRID:SCR_012773). P values were calculated by Fisher’s exact test and enrichment scores are −log₁₀(P value). Excel files are provided as supplemental material (Supplemental Excel files E1 and E2).

Statistical analysis

Statistical analyses were performed using GraphPad Prism (RRID:SCR_002798). After analyzing the data using conventional tests that assume normal distribution, t test for data in two groups and one- or two-way ANOVA with a post hoc test for multiple group comparisons, the normality of residuals was assessed. When residuals did not pass a normality test, the data were log-transformed, and the transformed values were analyzed, which in all cases restored the normality. The data shown are the original non-transformed values. Sample size (n) and the statistical tests used are described in the figure legends. P ≤ 0.05 was considered statistically significant.

RESULTS

AHR-dependent transcription in human airway epithelial cells and in lungs of mice following RSV infection

AHR is a transcription factor that can be activated by different ligands, including endogenous (host/gut microbiota metabolites) and exogenous (dietary, environmental pollutants, drugs) (7). Upon ligand binding, AHR translocates from the cytoplasm to the nucleus, where it dimerizes with ARNT. The AHR-ARNT heterodimer then binds DNA at XRE sites to regulate the expression of target genes, including CYP1A1 (28, 29), which is often used as a marker of AHR activation (30) (Fig. 1A). To investigate the effect of RSV infection on the AHR-dependent gene expression, we used primary human SAE cells derived from different donors and A549 cells, a model of lung alveolar type II-like cells well characterized in RSV infection (31) (Fig. 1B). In the first set of experiments, SAE cells were mock or infected with RSV for 6, 15, and 24 h, total RNA was extracted, and CYP1A1 gene expression was quantified by RT-qPCR. We found that RSV infection induced a significant time-dependent reduction of CYP1A1 mRNA levels (Fig. 1C). To investigate specifically the effect of RSV infection on AHR transcriptional activity, A549 cells were transfected with a synthetic XRE-driven promoter, linked to a luciferase reporter gene, and infected with RSV for 15 and 24 h. In agreement with the reduction observed in CYP1A1 mRNA levels, the XRE luciferase activity was significantly decreased by RSV infection. Results of cells stimulated with the AHR ligand TCDD (10 nM) are shown as positive control (Fig. 1D).

Figure 1.

Aryl hydrocarbon receptor (AHR)-dependent transcription in airway epithelial cells and in lungs of mice following respiratory syncytial virus (RSV) infection. A: Schematic representation of the canonical AHR signaling pathway. Upon ligand binding, AHR translocates from the cytoplasm to the nucleus, where it dimerizes with AHR nuclear translocator (ARNT). The AHR-ARNT heterodimer then binds DNA at xenobiotic response element (XRE) sites to regulate the expression of target genes, such as CYP1A1. B: Primary human small airway epithelial (SAE) cells from 2 donors and A549 cells were used in the experiments shown in C and D. Circles represent cells from the female, squares represent cells from male, open symbols represent uninfected cells, and filled symbols represent RSV-infected cells. C: SAE cells mock or infected with RSV were harvested 6, 15, and 24 hours postinfection (hpi), RNA was extracted, and CYP1A1 gene expression was quantified by RT-qPCR (n = 6; pooled data from three independent experiments performed in duplicate; *** P < 0.001 by two-way ANOVA followed by Bonferroni’s test). D: A549 cells were transfected with the pGL4.43[luc2P/XRE/Hygro] vector and either infected with RSV or stimulated with the AHR ligand TCDD (10 nM) as positive control. At 15 and 24 hpi (or 3 h after stimulation with TCDD), the luciferase reporter activity was measured (n = 6; pooled data from two independent experiments performed in triplicate; *** P < 0.001 by one-way ANOVA followed by Dunnett's test). E: SAE cells mock or infected with RSV for 15 h were exposed to vehicle (0.01% DMSO) or the AHR ligands TCDD (5 nM), BaP (1 μM), FICZ (1 nM), or I3C (20 μM) for 3h. Cells were harvested at 18 hpi, RNA was extracted, and CYP1A1 gene expression was quantified by RT-qPCR (n = 6, pooled data from three independent experiments performed in duplicate, *** P < 0.001 by two-way ANOVA followed by Bonferroni’s test). F: Female BALB/c mice infected with RSV (or mock inoculated with PBS) were euthanized at 1, 2, 4, 7, and 10 days postinfection (dpi), the lungs were harvested for RNA extraction, and Cyp1a1 gene expression was determined by RT-qPCR (n = 4-9 mice per group, biological replicates; *** P < 0.001 by lognormal one-way ANOVA followed by Dunnett's test). Each point or symbol represents data obtained from one mouse. Open circles represent control/mock inoculated mice and filled circles represent RSV-infected mice. Bar graphs show means ± SD.

As activation of AHR by different ligands can exert different effects (7, 32, 33), we next investigated how exposure to various AHR agonists would affect CYP1A1 gene expression during RSV infection. SAE cells mock or infected with RSV for 15 h were exposed to DMSO as vehicle control or the AHR ligands TCDD, BaP, FICZ, or I3C for 3h. Cells were harvested at 18 hpi for RNA extraction and CYP1A1 gene expression was quantified by RT-qPCR. The results showed that RSV infection inhibited activation of the AHR pathway, as measured by expression of the CYP1A1, regardless of the agonist used (Fig. 1E).

We next investigated whether RSV infection would affect the AHR expression in vivo, as we observed in the airway epithelial cells. Mice were infected intranasally with RSV (or mock inoculated with PBS) and euthanized at 1, 2, 4, 7, and 10 days postinfection (dpi). The lungs were harvested, RNA was extracted, and Cyp1a1 mRNA levels were determined by RT-qPCR (Fig. 1F). The results showed statistically significant downregulation of Cyp1a1 expression in lungs of RSV-infected mice at 1, 2, 4, and 7 dpi (Fig. 1F). Interestingly, Cyp1a1 expression in lungs of RSV-infected mice was downregulated during the acute phase of infection (RSV peaks at day 4), with levels returning to baseline when virus is cleared and mice no longer exhibit signs of disease.

Taken together, our results showed that the AHR-dependent gene expression is suppressed in human airway epithelial cells and in lungs of mice during RSV infection.

AHR and ARNT cellular levels in primary human SAE cells following RSV infection

To determine whether the inhibition of CYP1A1 was due to changes in activation of AHR and its DNA binding partner ARNT in response to RSV infection, we next examined AHR and ARNT cellular and nuclear protein levels. SAE cells were mock or RSV infected and harvested at 6, 15, and 24 hpi to prepare whole-cell lysates and nuclear fractions. AHR and ARNT protein levels were assessed by western blot. The results showed a progressive time-dependent decrease of AHR for both total and nuclear levels starting at 15 hpi, with a statistically significant reduction observed at 24 hpi, when compared to uninfected cells and cells infected with RSV for 6 h (Fig. 2, A and B). While ARNT total levels were not significantly affected by RSV infection (Fig. 2C), there was a trend towards decreased nuclear levels at 24 hpi (P = 0.06) when compared to uninfected cells (Fig. 2D). In addition, after an initial increase in nuclear translocation of ARNT at 6 hpi (not significant when compared to mock), the results showed a time-dependent decrease in ARNT nuclear levels in RSV-infected cells at 15 and 24 hpi when compared to cells infected for 6 h (Fig. 2D). The full length immunoblots are shown in Supplemental Fig. S2.

Figure 2.

AHR and ARNT total and nuclear protein levels in primary human SAE cells following RSV infection. SAE cells mock or infected with RSV for 6, 15, and 24 h were harvested to prepare whole-cell lysates and nuclear fractions. The total cellular and nuclear protein levels of AHR (A and B) and ARNT (C and D) were assessed by western blot. Graphs show relative protein fold change determined by quantification of bands and normalization to the loading control (β-actin or HDAC1). Each point/symbol represents data obtained from a different experiment. A and C (total cellular levels): image representative of 4 independent experiments. B and D (nuclear levels): image representative of 3 independent experiments. Circles represent results of experiments with cells from the female donor (25 years old) and squares represent results of experiments with cells from the male donor (56 years old). Open symbols represent mock infected cells and filled symbols represent RSV-infected cells. Bar graphs show mean ± SD (n = 3-4, independent experiments, * P ≤ 0.05 and ** P < 0.01 by two-way ANOVA followed by Bonferroni’s multiple comparisons test). Full length western blot images are shown in Supplemental Fig. S2.

AHR and ARNT binding to the CYP1A1 gene promoter in primary human SAE cells infected with RSV and exposed to TCDD

To investigate changes in AHR and ARNT binding to the XRE sites of the CYP1A1 endogenous gene promoter in response to RSV infection and AHR agonists, SAE cells were mock or RSV infected for 15 h and then exposed to TCDD (5 nM) or DMSO, as vehicle control, for 1h. Cells were harvested at 16 hpi for ChIP analysis. After immunoprecipitation with AHR or ARNT antibodies, qPCR was performed with immunoprecipitated DNA using primers spanning proximal (XRE1) and central (XRE 4-6) regions of the CYP1A1 promoter (Fig. 3A). The results showed that TCCD-inducible binding of both AHR and ARNT to both XRE sites of the CYP1A1 promoter was significantly reduced when cells were infected with RSV when compared to mock-infected (Fig. 3, B-E).

Figure 3.

AHR and ARNT binding to the CYP1A1 gene promoter in SAE cells infected with RSV and exposed to an AHR ligand. A: Schematic representation of XRE sites in the human CYP1A1 promoter. Primers spanning proximal (XRE1) and central (XRE4-6) regions were used. SAE cells (female donor, 25 years old) mock or infected with RSV for 15 h were exposed to vehicle (0.01% DMSO) or TCDD (5 nM) for 1h. Cells were harvested at 16 hpi and Chromatin Immunoprecipitation (ChIP) assay was performed. The chromatin was immunoprecipitated with IgG antibody (negative control) or AHR antibody (B and C) or ARNT antibody (D and E). qPCR was performed with specific primers for proximal and central CYP1A1 promoter regions. Fold changes over IgG control are expressed as means ± SD (n = 6, pooled results from three independent experiments performed in duplicate), * P = 0.02, ** P ≤ 0.003, *** P < 0.001 by two-way ANOVA followed by Bonferroni's multiple comparisons test. Open symbols represent control/mock infected cells and filled symbols represent RSV-infected cells.

Viral replication and differential gene expression analysis in AHR-deficient cells following RSV infection

To better understand the role of the AHR signaling during RSV infection, we next conducted experiments with AHR-deficient airway epithelial cells. We used A549 cells wild-type (WT) and AHR knockout (KO) generated by CRISPR/Cas9 technique, as described in (16). We first analyzed whole cell extracts from A549 WT and AHR KO cells by western blot to confirm the AHR deficiency in the KO cells (Fig. 4A, and full length immunoblots are shown in Supplemental Fig. S3). Next, we assessed CYP1A1 gene expression in response to AHR ligands and RSV infection. A549 WT and AHR KO cells mock or infected with RSV for 15 h were exposed to vehicle (0.01% DMSO) or TCDD (5 nM) or I3C (20 μM) for 3 h. Cells were harvested at 18 hpi, RNA was extracted, and CYP1A1 mRNA levels were assessed by RT-qPCR. Similar to the response observed in SAE cells, RSV infection inhibited both the TCDD- and I3C-induced expression of CYP1A1 in A549 WT cells, while there was no increase in CYP1A1 mRNA levels in the AHR KO cells, confirming the loss of the functional AHR signaling in those cells (Fig. 4, B and C). To determine whether the AHR deficiency would affect viral replication, A549 WT and AHR KO cells were infected with RSV for 20 h and the RSV titers in the cell culture medium were quantified by plaque assay. Our results showed no significant difference in RSV replication in A549 AHR KO cells when compared to WT cells (Fig.4D). To further investigate the role of AHR during RSV infection, we used gene arrays to identify DEGs between A549 WT and A549 AHR KO cells in basal conditions and following infection. We found that lack of AHR expression resulted in 1225 DEGs in mock-infected cells, 695 upregulated and 530 downregulated, and in 3380 DEGs in RSV infection, 1342 upregulated and 2038 downregulated (Fig. 4, E and F, and Supplemental Excel file E1). The KEGG pathway enrichment analysis was performed to better understand relevant biological pathways these DEGs are involved in. Significantly enriched pathways of upregulated DEGs in infected A549 AHR KO cells vs. WT cells included toll-like receptor signaling pathway, viral protein interaction with cytokine and cytokine receptor, IL-17 signaling pathway, nuclear factor kappa B (NF-κB) signaling pathway, and cytokine-cytokine receptor interaction, while the downregulated DEGs were enriched in transforming growth factor beta (TGF-β), mitogen-activated protein kinase (MAPK), AMP-activated protein kinase (AMPK), phosphatidylinositol 3-kinase-protein kinase B (PI3K-Akt), autophagy, and mechanistic target of rapamycin (mTOR) signaling pathways. Some of the enriched pathways of up- and downregulated DEGs relevant to RSV disease pathogenesis are summarized in Tables 1 and 2, while the full list is presented in Supplemental Excel file E2.

Figure 4.

Viral replication and differential gene expression analysis in lung epithelial AHR-deficient cells following RSV infection. A: Whole-cell lysates from A549 wild-type (WT) and AHR knockout (KO) cells were analyzed by western blot to detect the AHR protein levels. β-actin was used as a loading control. Representative images are from one experiment performed in triplicate. Full length western blot images are shown in Supplemental Fig. S3. B and C: A549 WT and AHR KO cells mock or infected with RSV for 15 h were exposed to vehicle (0.01% DMSO) or TCDD (5 nM) or I3C (20 μM) for 3 h. Cells were harvested at 18 hpi, RNA was extracted, and CYP1A1 mRNA levels were assessed by RT-qPCR. Data are presented as scatter plots with bars and means ± SD. Open symbols represent mock infected cells and filled symbols represent RSV-infected cells. Each point/symbol represents data obtained from one cell culture well (n = 3; technical replicates; data from one experiment performed in triplicate). Data were analyzed by two-way ANOVA followed by Bonferroni's multiple comparisons test (*** P < 0.001). D: A549 WT and AHR KO cells were infected with RSV for 20 h and viral titers in the cell culture medium were quantified by plaque assay (n = 10-11; pooled results from five independent experiments performed at least in duplicate). Data were analyzed by unpaired t test, not significant. E: A549 WT and AHR KO cells were mock or infected with RSV and at 18 hpi RNA was extracted. Number of differentially expressed genes (DEGs) were identified by microarray analysis. F: Scatter plot of DEGs in RSV-infected cells shows upregulated genes with red dots and downregulated genes with green dots.

Table 1.

Relevant enriched KEGG pathways of upregulated differentially expressed genes (DEGs) in RSV-infected A549 AHR KO vs. WT cells

Pathway	Genes
Toll-like receptor signaling pathway P = 0.0003 Enrichment score = 3.545	CCL4, CD86, CXCL10, CXCL11, CXCL9, IFNA2, IL12A, IL1B, IL6, MAP2K1, NFKB1, PIK3R3, TLR2, TLR6, TNF, TRAF6
Glyoxylate and dicarboxylate metabolism P = 0.001 Enrichment score = 2.881	ACO1, ACSS1, CAT, GLDC, GLUL, HOGA1, PCCA
Arginine and proline metabolism P = 0.002 Enrichment score = 2.636	AGMAT, ALDH1B1, ALDH9A1, ARG2, CKMT1A, HOGA1, OAT, ODC1, PRODH2
Tryptophan metabolism P = 0.002 Enrichment score = 2.614	AANAT, ALDH1B1, ALDH9A1, AOX1, CAT, DDC, EHHADH, INMT
Viral protein interaction with cytokine and cytokine receptor P = 0.005 Enrichment score = 2.317	CCL13, CCL4, CSF1R, CXCL10, CXCL11, CXCL9, CXCR2, CXCR5, IL10RA, IL18R1, IL6, TNF, TNFRSF10A
Retinol metabolism P = 0.005 Enrichment score = 2.271	ADH6, AOX1, DHRS9, PNPLA4, RDH12, RDH5, RETSAT, UGT1A6, UGT1A8, UGT2A3
Valine, leucine and isoleucine degradation P = 0.006 Enrichment score = 2.241	ABAT, ALDH1B1, ALDH9A1, AOX1, BCAT1, EHHADH, HIBADH, PCCA
IL-17 signaling pathway P = 0.008 Enrichment score = 2.116	CXCL10, IL1B, IL6, MAPK4, MMP1, MMP3, NFKB1, S100A8, S100A9, TAB3, TNF, TRAF6
Complement and coagulation cascades P = 0.01 Enrichment score = 2.024	C4BPB, C8A, CFB, CFI, F12, F13B, F2, F5, FGA, PROS1, SERPINA1
Cytosolic DNA-sensing pathway P = 0.01 Enrichment score = 2.011	AIM2, CCL4, CXCL10, IFNA2, IL1B, IL6, NFKB1, NFKBIB, POLR3B
PPAR signaling pathway P = 0.011 Enrichment score = 1.972	ACOX2, ADIPOQ, APOA1, EHHADH, FABP1, GK, MMP1, PCK1, SLC27A2, SORBS1
NF-kappa B signaling pathway P = 0.016 Enrichment score = 1.782	BCL2A1, BTK, CCL13, CCL4, GADD45G, IL1B, LYN, NFKB1, TAB3, TNF, TNFSF11, TRAF6
NOD-like receptor signaling pathway P = 0.02 Enrichment score = 1.706	AIM2, ANTXR2, ATG16L1, CASP5, GBP2, GBP5, GBP7, IFNA2, IL1B, IL6, JAK1, NFKB1, NFKBIB, TAB3, TNF, TRAF6, TXNIP, VDAC1
Cytokine-cytokine receptor interaction P = 0.02 Enrichment score = 1.689	ACVR1C, BMPR1A, CCL13, CCL4, CSF1R, CXCL10, CXCL11, CXCL9, CXCR2, CXCR5, IFNA2, IL10RA, IL12A, IL18R1, IL1B, IL1R2, IL36G, IL6, TGFB3, TNF, TNFRSF10A, TNFRSF11B, TNFRSF9, TNFSF11, TNFSF15, TNFSF4
Propanoate metabolism P = 0.035 Enrichment score = 1.456	ABAT, ACSS1, ECHDC1, EHHADH, PCCA
Arginine biosynthesis P = 0.035 Enrichment score = 1.453	ARG2, GLS, GLUL, OTC

P values were calculated by Fisher’s exact test and enrichment scores are −log₁₀(P value).

Table 2.

Relevant enriched KEGG pathways of downregulated differentially expressed genes (DEGs) in RSV-infected A549 AHR KO vs. WT cells

Pathway	Genes
TGF-β signaling pathway P = 0.003 Enrichment score = 2.530	ACVR2B, BMP7, BMPR1A, BMPR2, E2F4, GREM2, ID4, LEFTY1, MAPK3, PITX2, PPP2R1A, RPS6KB1, SMAD2, SMAD6, TGFBR1, ZFYVE16
MAPK signaling pathway P = 0.004 Enrichment score = 2.437	AKT2, ATF2, BDNF, CACNA1A, CACNA1D, CACNG4, CACNG6, CACNG7, CASP3, CSF1, EPHA2, ERBB2, FGF2, FGFR1, FLNA, GNG12, IGF2, IRAK1, JMJD7-PLA2G4B, JUND, MAP3K12, MAPK12, MAPK3, MET, NFATC1, NGF, NGFR, NR4A1, PDGFC, PPP5C, PRKCG, PTPRR, RPS6KA5, TAB1, TGFBR1, TP53, TRAF2
AMPK signaling pathway P = 0.007 Enrichment score = 2.171	ACACA, AKT1S1, AKT2, EEF2K, PCK2, PFKFB1, PFKFB2, PFKFB3, PFKL, PIK3R2, PPP2R1A, PPP2R3B, RAB11B, RAB14, RPS6KB1, TBC1D1, TSC1, TSC2
PI3K-Akt signaling pathway P = 0.016 Enrichment score = 1.808	AKT2, ATF2, BCL2L1, BDNF, CCNE2, CSF1, EPHA2, ERBB2, FGF2, FGFR1, FN1, GNB2, GNB4, GNG12, IFNA16, IFNA7, IGF2, IL3RA, IL4R, ITGA3, ITGA4, JAK3, LAMB2, LPAR1, MAPK3, MET, NGF, NGFR, NR4A1, PCK2, PDGFC, PIK3R2, PKN3, PPP2R1A, PPP2R3B, RPS6KB1, TP53, TSC1, TSC2, YWHAE
Autophagy P = 0.017 Enrichment score = 1.779	AKT1S1, AKT2, ATG12, ATG13, ATG2B, BCL2L1, HIF1A, LAMP2, MAPK3, MTMR3, PIK3R2, RAB39B, RAB7A, RPS6KB1, SUPT20H, TSC1, TSC2, ULK2, WIPI1
mTOR signaling pathway P = 0.023 Enrichment score = 1.63	AKT1S1, AKT2, ATP6V1C1, CASTOR1, CASTOR2, FLCN, FNIP1, FZD2, LRP6, MAPK3, MAPKAP1, PIK3R2, PRKCG, RPS6KB1, SEC13, SLC7A5, TSC1, TSC2, ULK2, WNT7B
Phospholipase D signaling pathway P = 0.048 Enrichment score = 1.321	ADCY7, AGPAT5, AKT2, CYTH1, CYTH3, DNM1, JMJD7-PLA2G4B, LPAR1, MAPK3, PDGFC, PIK3R2, PLCB4, PTPN11, RAPGEF3, SHC1, SHC3, TSC1, TSC2

P values were calculated by Fisher’s exact test and enrichment scores are −log₁₀(P value).

Some DEGs identified by microarrays were selected for further analysis by RT-qPCR. In these validation studies we confirmed that cells lacking AHR exhibited a significant increase in the expression of C-X-C motif chemokine ligand 9 (CXCL9), CXCL11, collagen type XIII alpha 1 chain (COL13A1), collagen type IV alpha 1 chain (COL4A1), matrix metallopeptidase 1 (MMP1), and matrix metallopeptidase 3 (MMP3) in response to RSV infection (Fig. 5A). Given the increased expression of genes involved in the pro-inflammatory response in the AHR-deficient cells following RSV infection, and previous data indicating that AHR plays a suppressive role in the control of the inflammatory response (16), we next measured levels of cytokines and chemokines by Bio-Plex and levels of IFN-β by ELISA in the cell culture medium collected from A549 WT and AHR KO cells that were mock or infected with RSV for 20 h. The AHR-deficient cells showed significantly higher levels of interleukin-6 (IL-6), granulocyte-colony stimulating factor (G-CSF), C-C motif chemokine ligand 3 CCL3 (MIP-1α), CXCL9 (MIG), CXCL10 (IP-10), and IFN-β (Fig. 5B), as well as IL-12 (p40), hepatocyte growth factor (HGF), and leukemia inhibitory factor (LIF) (data not shown) following RSV infection, when compared to WT cells. Taken together, these results demonstrated that in the context of RSV infection, the loss of AHR in airway epithelial cells led to upregulation of pro-inflammatory and airway remodeling-related genes, as well as increased production of pro-inflammatory mediators.

Figure 5.

Validation of DEGs by RT-qPCR and assessment of cytokines and chemokines in AHR-deficient cells following RSV infection. A: A549 WT and AHR KO cells were mock or infected with RSV and at 18 hpi RNA was extracted. The expression of C-X-C motif chemokine ligand 9 (CXCL9), CXCL11, collagen type XIII alpha 1 chain (COL13A1), collagen type IV alpha 1 chain (COL4A1), matrix metallopeptidase 1 (MMP1), and matrix metallopeptidase 3 (MMP3) was measured by RT-qPCR. Each point/symbol represents data obtained from one cell culture well (n = 3; technical replicates; data from one out of two independent experiments with similar results). B: A549 WT and AHR KO cells were mock or infected with RSV for 20 h and the levels of cytokines and chemokines in cell culture medium were measured by Bio-Plex or ELISA (IFN-β) (mock n = 3: from one experiment performed in triplicate; RSV n = 7-9: pooled data from three independent experiments performed at least in duplicate). Open symbols represent mock infected cells and filled symbols represent RSV-infected cells. Bar graphs show means ± SD. * P ≤ 0.05, ** P < 0.01, *** P < 0.001 by two-way ANOVA followed by Bonferroni's multiple comparisons test.

Effects of the dietary AHR ligand I3C on experimental RSV infection

Previous studies indicate that severe RSV infection develops because of altered host immune responses (34, 35). As AHR is involved in the regulation of both innate and adaptive immunity (36) and the loss of AHR in lung epithelial cells led to exacerbated inflammatory response in response to RSV infection, we speculated whether AHR activation with a non-toxic ligand prior to the infection would have beneficial effects in vivo. To test this, we used I3C, a natural AHR ligand found in cruciferous vegetables with reported anti-inflammatory properties in the context of other diseases (37-39). Based on the commercially available formulations and in previous preclinical studies (40, 41), we selected doses of 35 and 70 mg/kg, which are equivalent to human doses of 200 and 400 mg, respectively (42). BALB/c mice received a daily dose of I3C or vehicle control (5% DMSO + 95% corn oil) by oral gavage for different durations up to 5 days, starting one day prior to RSV infection (d–1) (Fig. 6, A and B). To evaluate the effects of I3C on the inflammatory response to RSV infection, mice were treated at d–1 and d0 (2 h after mock or RSV inoculation), euthanized at d1, and pro-inflammatory mediator production (levels of cytokines and chemokines) in the BALF were determined by using a Bio-Plex assay. Both doses of I3C were associated with significantly lower levels of IL-6, Ccl3 (MIP-1α), Ccl4 (MIP-1β), and Ccl5 (RANTES), and treatment with the higher dose led to lower secretion of tumor necrosis factor-α (TNF-α) in response to RSV infection (Fig. 6A). For improved clarity regarding the effects of I3C on the RSV-infected animals, comparisons with the uninfected controls (mock) are shown in Supplemental Fig. S4. When all groups were analyzed together, the results indicated that treatment with the higher dose of I3C also led to significant lower secretion of G-CSF in response to RSV infection (Supplemental Fig. S4)

Figure 6.

Effects of the dietary AHR ligand indole-3-carbinol (I3C) on experimental RSV infection. A: BALB/c female mice received vehicle (5% DMSO + 95% corn oil) or I3C by oral gavage once a day for 2 days, starting one day prior to RSV infection (d–1). The second dose was given 2 h after mock or RSV inoculation (d0). At d1, bronchoalveolar lavage fluid (BALF) was collected and cytokines and chemokines levels were measured by Bio-Plex. Each point/symbol represents data obtained from one mouse (n = 4 mice per group, biological replicates; * P < 0.05, ** P < 0.01, by one-way ANOVA followed by Dunnett's test). Comparisons with the uninfected controls (mock) are shown in Supplemental Fig. S4. B: Schematic of the I3C treatment for figures C and D. BALB/c female mice received a daily dose of vehicle or I3C by oral gavage for 3 days (for lung viral titers determination at d2) or 5 days (for body weight change analysis and lung viral titers determination at d4), starting one day prior to RSV infection (d–1). C: Percent of body weight change following I3C treatment and RSV infection. Open symbols represent mock infected mice and filled symbols represent RSV-infected mice (n = 8-16 mice per group, biological replicates). ** P < 0.01 and *** P < 0.001 vs. I3C 35 and 70 mg/kg + RSV by mixed model for repeated measures followed by Tukey's test). D: Lung viral titers at 2 and 4 days postinfection (dpi) determined by plaque assay. Each point/symbol represents data obtained from one mouse (n = 6-8 mice per group, biological replicates; * P = 0.05, *** P < 0.001 by two-way ANOVA followed by Tukey’s test). E: SAE and HEp-2 cells were infected with RSV (multiplicity of infection of 3 for SAE and of 0.01 for HEp-2 cells) in the presence or absence of I3C (20 μM) for 18 or 48 h, respectively. Cell supernatants were harvested at 18 hpi (SAE) or 48 hpi (HEp-2) and viral titers were determined by plaque assay (n = 4, pooled data from two independent experiments performed in duplicate). ** P < 0.01 by unpaired t test. Data are shown as means ± SD.

We next evaluated body weight loss and viral replication. BALB/c mice received a daily dose of vehicle or I3C by oral gavage for 3 days (for lung viral titers determination at d2) or 5 days (for body weight loss analysis and lung viral titers determination at d4), starting one day prior to RSV infection (d–1) (Fig. 6B). We observed a protective and similar effect of both I3C doses in RSV-infected mice, which lost significantly less body weight than the vehicle/RSV group (Fig. 6C). We next assessed whether I3C supplementation would affect RSV replication. Mock and RSV-infected mice treated with I3C were euthanized at 2 and 4 dpi, and lung viral titers were determined by plaque assay. The results showed no significant differences at 2 dpi but a statistically significant reduction at 4 dpi, corresponding to peak viral titer, which was more pronounced for the 70 mg/kg dose (Fig. 6D). The antiviral effect of I3C was confirmed in primary human SAE cells and in HEp-2 cells (Fig. 6E), a cell line naturally permissive to RSV infection and commonly used for virus propagation. Altogether, our results show that activating the AHR pathway with the dietary ligand I3C right before the beginning of infection has beneficial effects on RSV-associated disease, including attenuation of RSV-induced body weight loss, anti-inflammatory and antiviral activity.

DISCUSSION

The AHR is a well-known environmental sensing molecule that participates in multiple physiological and pathological processes (43, 44). AHR is widely expressed in the lung and it is an important modulator of the intrinsic, innate and adaptive immune responses to infections, with different effects on host resistance based on the virus and experimental system used (14). Many RNA and DNA viruses have been shown to induce AHR activation as a strategy to evade antiviral immunity and promote virus replication (13-15). Among the respiratory viruses, coronaviruses such as the SARS-CoV-2 were shown to activate the AHR signaling (45), although downregulation of AHR was reported in mouse endothelial cells following influenza virus infection in vivo (46). To our knowledge, there are no reports on whether RSV, another clinically relevant virus, affects the AHR pathway. Considering that RSV infections remain a major health burden with no specific treatment available, this understanding could provide valuable insights into whether (and how) the AHR pathway could be explored as therapeutic target.

In this study, we demonstrated that RSV infection, different from most viruses (13-15), leads to inhibition of AHR transcriptional activity and subsequent expression of the canonical AHR-dependent gene CYP1A1 both in primary human SAE cells and in mouse lungs. . As AHR activation occurs in response to a variety of different classes of molecules, both endogenous (host/gut microbiota metabolites) and exogenous (dietary, environmental pollutants, drugs), we exposed SAE cells to different types of ligands and found that RSV also inhibited ligand-induced AHR-dependent signaling.

There are many nuclear receptors and transcriptional factors controlling transcriptional regulation of the CYPs (e.g. HNF4α, PPARα, GRα), however regulation of CYP1 genes in human lung is primarily mediated by AHR and ARNT (47). Upon ligand binding, by either endogenous or exogenous agonists, AHR translocates from the cytoplasm to the nucleus, where it dimerizes with ARNT. The AHR-ARNT heterodimer then binds DNA at XRE sites to regulate the expression of target genes, including CYP1A1 (7, 28, 44). Assessment of cellular levels of AHR following RSV infection revealed a time-dependent reduction in AHR total cellular and nuclear protein levels. The cellular content of AHR can be controlled by its degradation via the ubiquitin-proteasome system, usually in a ligand-dependent manner, but also through other pathways such as lysosome-dependent autophagy, which can be ligand independent, the latter via serine phosphorylation by the glycogen synthase kinase beta 3 (GSK3)(48). Post-translational modifications can also affect AHR stability and function, particularly SUMOylation (49), a process we found to play a key role in RSV-induced degradation of another transcription factor, NRF2, a key regulator of antioxidant gene expression (18). Different from AHR, we found no significant changes in ARNT total cellular levels, though there was a trend in reduction in nuclear levels when compared to uninfected cells and a significant reduction when comparing RSV infected cells across time, because of an initial increase in nuclear translocation of ARNT at 6 hpi (not significant when compared to mock).

Although we clearly saw changes in AHR-dependent gene expression and transcriptional activity following RSV infection we did not detect significant differences in AHR and ARNT binding to CYP1A1 XRE sites by ChIP assay at the time points assayed. In addition to reduced sensitivity in detecting changes by ChIP assay when binding is at low level, AHR transcriptional activity is regulated by several post-translational modifications, which can affect protein-protein interaction and assembly of the transcriptional machinery required for initiation of gene transcription, independent from its DNA-binding activity. AHR can be phosphorylated on both serine and tyrosine residues, which positively regulate AHR activity, while SUMOylation has been found to repress the transcriptional activity of AHR, in addition to affecting its stability and degradation (reviewed in 50).

On the other hand, there was significantly decreased inducible binding of both transcription factors to the CYP1A1 XRE sites in response to the canonical agonist TCDD. While the significantly lower AHR cellular and nuclear levels following RSV infection could be responsible for the lower AHR binding to the XRE sites, there was a much less prominent reduction in ARNT nuclear levels, suggesting that other mechanism(s) could be responsible for the observed finding. ARNT, also known as hypoxia-inducible factor (HIF)-1β, is a common binding partner for AHR and HIF-α proteins (44). As we have recently shown that RSV infection is a strong activator of HIF-1α and −2α (51), there could be competition for the recruitment of HIF-1β to XRE promoter sites (52, 53).

To further investigate the role of AHR during RSV infection, we assessed DEGs between WT and AHR-deficient lung cells in basal conditions and following infection. Our data showed that AHR deficiency in infected cells results in upregulation of pro-inflammatory genes such as CXCL9 and CXCL11. In agreement, the AHR KO cells showed increased secretion of multiple cytokines and chemokines, following RSV infection, including IFN-β. AHR has been shown to play a suppressive role in the control of the inflammatory response (16) and our study confirmed this finding in the context of RSV infection. The AHR signaling can also negatively regulate type I IFN response during viral infections (54). Increased IFN-β production and reduced viral titers were reported in AHR-deficient cells and mice after several viral infections including influenza (54), however, we did not find any significant effect on RSV replication.

Pro-inflammatory responses to RSV may be associated with epithelial-mesenchymal plasticity (EMP), which is linked to airway remodeling (55). AHR activation has been shown to limit inflammation and preserve lung epithelial cell integrity (56), while decreased expression of AHR was reported to play a detrimental role in in vitro and in vivo models of lung fibrosis (12). In agreement, we found that cells lacking AHR had a significant increase in the expression of COL13A1, COL4A1, MMP1, and MMP3 genes associated with EMP and airway remodeling in response to RSV infection.

Among the pathways upregulated in infected AHR KO cells, in addition to cytokine/chemokine/IFNs, there were the NF-κB and toll-like receptor signaling pathways, which play a key role in RSV-induced proinflammatory responses (57, 58), as well as the IL-17 signaling pathway. Severe RSV infection in infants is partly attributed to skewed CD4⁺ T cell differentiation as consequence of secreted combinations of cytokines and metabolites by infected lung epithelium. Higher populations of certain T cell subsets including pro-inflammatory Th17 cells and a concurrent depletion of inflammation-suppressing regulatory T cells is associated with increased inflammation and pathology (34). Interestingly, among the downregulated pathways there was the TGF-β, which is also associated with EMP in epithelial cells (59), as well as the autophagy and mTOR signaling pathways, which are involved in RSV replication (60, 61).

Increasing evidence points to a protective role of the AHR against different diseases, via anti-inflammatory and antioxidant mechanisms (62, 63), including the recent approval of a topical AHR-activating drug for the treatment of psoriasis and atopic dermatitis (64). In this study, the reduction in AHR levels and AHR-dependent transcription and the exacerbated inflammatory response observed in absence of AHR, also pointed to a possible protective role of AHR during RSV infection. Although exposure to the prototypical AHR ligand TCDD, a halogenated aromatic hydrocarbon found in the environment as a contaminant with immunotoxic and carcinogenic properties (65), has been shown to increase morbidity of influenza virus in a mouse model of infection (66), dietary and microbiota-derived AHR ligands, such as I3C and indole-3-acetic acid (IAA) can positively modulate inflammation and viral infections (33, 67, 68). I3C, a natural and nontoxic AHR ligand found in cruciferous vegetables, was previously found to have in vitro antiviral activity against SARS-CoV-2 (69). In animal models of disease, mice fed an I3C-supplemented diet and infected with influenza virus exhibited similar levels of immune cell recruitment, cytokine, and chemokine production but were protected against influenza-induced tissue damage and vascular leakage (46). Similarly, I3C administration led to attenuated lipopolysaccharide-induced acute respiratory distress syndrome in a mouse model through activation of AHR and regulation of immune cell trafficking in the lungs (70). In this study, we found that I3C (administered to mice shortly before RSV infection) has beneficial effects on disease, resulting in significant attenuation of RSV-induced body weight loss, anti-inflammatory and antiviral activity. I3C also showed antiviral effect against RSV in our in vitro studies with epithelial cells. Although we observed an antiviral activity of the dietary AHR ligand I3C in epithelial cells and in lungs of mice, our results did not show an increase in viral titers in the AHR KO cells. A possible explanation for lack of change in viral titers is the increased production of IFN-β in the AHR KO cells vs. the WT, which can affect viral replication.

This study has some limitations. Although primary SAE cells from both female and male donors were used, they were derived from a limited cohort of two donors. The in vivo experiments were conducted only in female mice to simplify the initial design and test I3C doses, and future work should include males. Future studies are also needed to investigate the specific mechanism(s) involved in the RSV-induced downregulation of the AHR signaling, and to better characterize the disease modulating, immunomodulatory, and antiviral properties of I3C (and possibly other nontoxic AHR ligands) against RSV infection, including the effects on innate and adaptive immune cells and viral clearance. However, despite these limitations, our study provides novel insights into the role of AHR during RSV infection and lay the groundwork for future studies.

In summary, our study showed that: (a) RSV infection downregulates the AHR pathway in human airway epithelial cells and in lungs of mice, (b) loss of the AHR led to exacerbated inflammatory response during in vitro RSV infection, and (c) administration of the dietary AHR ligand I3C to mice led to beneficial effects, including anti-inflammatory and antiviral activity. Collectively, our results suggest that AHR plays a protective role during RSV infection and indicate it can be explored as a novel therapeutic target for RSV-induced disease.

Supplementary Material

Supplemental Excel file E1.

Supplemental Excel file E2.

Supplemental Figs. S1-S4.

[https://figshare.com/s/e836bd202fefa23128e7]