Functional role of kynurenine and aryl hydrocarbon receptor axis in chronic rhinosinusitis with nasal polyps

Abstract

Background

Chronic rhinosinusitis with nasal polyps (CRSwNP) is associated with mast cell–mediated inflammation and heightened oxidant stress. Kynurenine (KYN), an endogenous tryptophan metabolite, can promote allergen-induced mast cell activation through the aryl hydrocarbon receptor (AhR).

Objectives

We sought to determine the role of the KYN/AhR axis and oxidant stress in mast cell activation and the development of CRSwNP.

Methods

We measured the expression of indoleamine 2,3-dioxygenase 1, tryptophan 2,3-dioxygenase, KYN, and oxidized calmodulin-dependent protein kinase II (ox-CaMKII) in nasal polyps and controls. KYN-potentiated ovalbumin (OVA)-induced ROS generation, cell activation, and ox-CaMKII expression were investigated in wild-type and AhR-deficient (AhR^−/−) mast cells. The role of ox-CaMKII in mast cell activation was further investigated.

Results

Nasal polyps in CRSwNP showed an increased expression of indoleamine 2,3-dioxygenase 1, tryptophan2,3-dioxygenase, and KYN compared with controls. AhR was predominantly expressed in mast cells in nasal polyps. Activated mast cells and local IgE levels were substantially increased in eosinophilic polyps compared with noneosinophilic polyps and controls. Furthermore, KYN potentiated OVA-induced ROS generation, intracellular Ca²⁺ levels, cell activation, and expression of ox-CaMKII in wild-type, but not in AhR^−/− mast cells. Compared with noneosinophilic polyps and controls, eosinophilic polyps showed increased expression of ox-CaMKII in mast cells. Mast cells from ROS-resistant CaMKII MMVVδ mice or pretreated with CaMKII inhibitor showed protection against KYN-promoted OVA-induced mast cell activation.

Conclusions

These studies support a potentially critical but previously unidentified function of the KYN/AhR axis in regulating IgE-mediated mast cell activation through ROS and ox-CaMKII in CRSwNP.

Chronic rhinosinusitis (CRS) with nasal polyps (CRSwNP) is an inflammatory sinonasal disease characterized by the presence of polyps in patients with the diagnosis of CRS. Most nasal polyps (NPs) in Caucasian patients with CRSwNP are eosinophilic, a state that is frequently associated with asthma.¹ In contrast, recent studies, including ours, have demonstrated that more than half of the patients with CRSwNP in East Asian countries presented with noneosinophilic inflammation.^2–4 Although genetic and/or environmental factors may contribute to the heterogeneity of CRSwNP, the mechanisms underlying the persistent and exaggerated inflammation in CRSwNP are not fully defined.

Eosinophilic CRSwNP is characterized by T_H2-skewed eosinophilic inflammation with local IgE hyperproduction and displays a poor surgical outcome.^1,5–7 Various inflammatory cells are implicated in the development of eosinophilic inflammation in CRSwNP, such as dendritic cells,^8,9 basophils,¹⁰ innate type 2 lymphoid cells,^11,12 and lymphocytes.^13,14 Recently, we and others have reported an accumulation of mast cells, particularly activated mast cells in CRSwNP, which correlates with eosinophilia in polyps.^15,16 Furthermore, locally secreted IgE can bind to mast cells in polyps and lead to mast cell activation on allergen exposure.^16,17 These findings raise the possibility that mast cells may be critical in mediating environmental antigen–induced pathogenesis of CRSwNP. Indeed, a recent study showed that mast cell deficiency may limit the development of CRS in an ovalbumin (OVA)-induced murine CRS model.¹⁸

Recent discoveries have illustrated that aryl hydrocarbon receptor (AhR), a ligand-activated transcription factor, in mast cells can mediate environmental factor–induced cell differentiation, growth, and functions.^19–21 Of interest, exposure of mouse and human mast cells to AhR ligands exacerbated allergen-induced, IgE-mediated oxidative stress, reactive oxygen species (ROS) generation, and calcium (Ca²⁺)-dependent activation of mast cells.²¹ AhR is widely expressed in barrier organs such as skin, gut, and lung²² and was originally characterized for its function in metabolizing environmental pollutants such as dioxin and many other polycyclic aromatic hydrocarbons.²³ Upon ligand binding, AhR translocates from cytosol to the nucleus, leading to changes in target gene transcription (eg, cytochrome P450 family 1 subfamily A member 1 [cyp1a1] and cytochrome P450 family 1 subfamily B member 1 [cyp1b1]) and various immunotoxicological effects.^24,25 However, recent studies found that AhR can also sense kynurenine (KYN) or other endogenous tryptophan metabolites generated by indoleamine 2,3-dioxygenase (IDO) and tryptophan 2,3-dioxygenase (TDO),^26,27 and regulate mast cell activation.²⁷ However, the potential role of KYN/AhR signaling in regulating mast cell activation in CRSwNP has not yet been studied.

The multifunctional Ca²⁺/calmodulin-dependent protein kinase II (CaMKII) is within one of the downstream signaling pathways activated by ROS.²⁸ CaMKII can be activated by oxidization at methionines 281/282 in the regulatory domain with ROS, leading to a persistently oxidative activation of CaMKII (oxidized calmodulin-dependent protein kinase II [ox-CaMKII]).²⁹ Increased ox-CaMKII was observed in the airway epithelium of patients with asthma and has been associated with asthma.^30–32 We have recently demonstrated that ox-CaMKII in mast cells play a critical role in regulating mast cell activation and the exacerbation of cockroach allergen–induced lung inflammation.³³ However, it remains unknown whether ox-CaMKII regulates chronic inflammation in CRSwNP.

The present study sought to determine the role of the KYN/AhR axis in promoting allergen-induced IgE-mediated immune responses in mast cells in the development of CRSwNP. We have made several novel findings, including increased levels of indoleamine 2,3-dioxygenase 1(IDO1)/TDO1/KYN and activated AhR signaling in mast cells of NP tissues from CRSwNP and AhR-regulated mast cell ROS generation and oxidation of CaMKII. Most importantly, we found an increased ox-CaMKII in CRSwNP, and that ox-CaMKII plays a critical role in allergen-induced, KYN-potentiated mast cell activation. Findings from these studies suggest that the KYN/AhR axis in mast cells may contribute to the distinct patterns of inflammation and possible clinical features observed in patients with eosinophilic CRSwNP.

METHODS

Study subjects

The study was approved by the Ethics Committee of Tongji Hospital of Huazhong University of Science and Technology and was conducted with written informed consent from the patients. The clinical data of patients are summarized in Table E1 in this article’s Online Repository at www.jacionline.org. The diagnosis of CRSwNP was made according to the current European Position Paper on Rhinosinusitis and Nasal Polyps and American guidelines.^34,35 CRSwNP is classified as eosinophilic when the percentage of eosinophils in the tissue exceeds 10% of total infiltrating cells as defined by our previous study.³ Control subjects are those undergoing septoplasty because of anatomical variations without sinonasal diseases. Polyp tissues from patients with CRSwNP and the inferior turbinate mucosa from control patients were harvested during surgery. Oral glucocorticoid and intranasal steroid spray were discontinued at least 3 months and 1 month before surgery, respectively. Subjects who had antrochoanal polyps, cystic fibrosis, fungal sinusitis, primary ciliary dyskinesia, or gastroesophageal reflux disease were excluded from the study.

Histological analysis

Immunohistochemical and immunofluorescence staining was performed as previously described.³ Primary antibodies (see Table E2 in this article’s Online Repository at www.jacionline.org) and additional information regarding immunohistochemistry are provided in this article’s Online Repository at www.jacionline.org.

RT-PCR

Quantitative RT-PCR was performed with appropriate primers (see Table E3 in this article’s Online Repository at www.jacionline.org) as previously reported.³ Additional information is provided in this article’s Online Repository at www.jacionline.org.

HPLC

Sample processing and HPLC analysis was performed as previously described with minor modifications.^36,37 Additional information is provided in this article’s Online Repository at www.jacionline.org.

Mice

C57BL/6, AhR-null (wild-type [WT].129-Ahr^tm1Bra/J), and Kit^W-sh/W-sh mice were purchased from the Jackson Laboratory (Bar Harbor, Me). ROS-resistant CaMKIIδ (MMVVδ) mice were generated by Dr Mark Anderson’s laboratory at the Johns Hopkins University School of Medicine. Age- and sex-matched mice were used as controls. These mice were maintained under specific-pathogen-free conditions. All experiments were approved by the Animal Care and Use Committee at Johns Hopkins University School of Medicine.

Bone marrow–derived cultured mast cells

Mouse bone marrow–derived mast cells (BMMCs) were cultured as previously described.²¹ Mast cell was confirmed by flow cytometry analysis with antibodies specific for c-Kit (1:100, 2B8, eBiosciences, San Diego, Calif) and FcεRI (1:200, MAR-1, eBiosciences) and by histochemical staining with acid Toluidine blue.

Measurements of degranulation and histamine release

Degranulation was first monitored by time-lapse microscopy. Approximately 5.0 × 10⁴ BMMCs previously sensitized with 1 μg/mL of anti-OVA IgE (E-C1, Chondrex, Redmond, Wash) were plated on fibronetic (Thermo Fisher, Halethorpe, Md)-coated Lab-Tek chambered cover glass (Thermo Fisher) in Tyrode’s buffer supplemented with 8 mg/mL of avidin-sulforhodamine 101 (Av.SRho, Sigma-Aldrich, St Louis, Mo). The cells were incubated at 37°C for 30 minutes and then stimulated with 10 μg/mL of OVA. Fluorescence was acquired every 2.3 seconds using Zeiss confocal microscope and AxioVision 4.2 software in an environmental chamber (37°C and 5% CO₂). Mast cell degranulation was quantified by flow cytometric analysis for the expression of CD107a/LAMP-1 (1:200, clone eBio1F4B, ThermoFisher, Halethorpe, Md).³⁸ Degranulation was also quantified by measuring β-hexosaminidase release in the culture supernatants as previously described.³³ Histamine release was assessed by using automated fluorimetry as previously described.³⁹

ELISA

Supernatants were collected for the measurement of IL-5 (eBiosciences), IL-13 (eBiosciences), and IL-33 (R&D Systems, Minneapolis, Minn) using ELISA kits according to the manufacturer’s instructions.

Tissue IgE measurement

Tissue samples were weighed and homogenized and the supernatants were harvested. The levels of total IgE in supernatants were detected by using the ImmunoCAP system (Phadia, Uppsala, Sweden).¹⁶

Mast cell engraftment into mast cell–deficient mice

BMMCs cultured from AhR^−/− and WT female mice were engrafted into the mast cell–deficient mice (Kit^W-sh/W-sh) as described elsewhere.⁴⁰ A total of 1.0 × 10⁶ BMMCs were engrafted into the ears by intradermal injection. After 6 weeks, mast cells’ engraftment was assessed by Toluidine blue staining.

Passive cutaneous anaphylaxis

Passive cutaneous anaphylaxis (PCA) was performed as previously described.³³ Additional information is provided in this article’s Online Repository at www.jacionline.org.

Western blotting

Equal amounts of proteins were separated on a 12% (w/v) Tris-Glycine Gel (Thermo Fisher) in NuPAGE MES SDS Running Buffer (Thermo Fisher). The separated proteins were transferred onto polyvinylidene difluoride membranes (Thermo Fisher) and probed with anti-IDO1 (1:500; Abcam, Cambridge, UK), anti–ox-CaMKII (1:1000; Millipore, Billerica, Mass), or antitotal CaMKII (1:1000, Abcam) overnight at 4°C. Blots were then probed with goat antirabbit (Santa Cruz Biotechnology, Santa Cruz, Calif) or goat antimouse IgG-horseradish peroxidase (Santa Cruz Biotechnology) for 1 hour at 37°C. Detection was performed by using the ECL Western blotting detection system (GE Life Sciences, Pittsburgh, Pa).

Detection of mitochondrial superoxide in BMMCs

For mitochondrial superoxide staining, BMMCs previously sensitized with 1 μg/mL of anti-OVA IgE were incubated on fibronectin-coated glass slides with 5 μM MitoSOX (Thermo Fisher) and 100 nM MitoTracker Green (Thermo Fisher) for 30 minutes at 37°C in Tyrode’s buffer. Cells were washed and challenged with 10 μg/mL of OVA for 30 minutes at 37°C in Tyrode’s buffer. After challenge, cells were mounted in warm buffer for imaging.

Measurement of intracellular ROS production by flow cytometry

Intracellular ROS production in BMMCs was measured by flow cytometry with CM-H2DCFDA (5 μM, ThermoFisher) as described previously.^21,33

Intracellular calcium measurement

Intracellular calcium was measured as described previously.^21,33 In brief, BMMCs sensitized with anti-OVA IgE in the presence or absence of 50 μM of KYN were loaded with 2.5 μM of Fluo-4-acetomethoxy ester (ThermoFisher) for 1 hour in the dark at 37°C and then challenged with 10 μg/mL of OVA. Fluoresecent signals at 488 nm excitation were imaged continuously for 90 to 120 seconds to detect intracellular free calcium.

Statistical analysis

The expression data generated from human sinonasal mucosal samples are presented in dot plots. Symbols represent individual samples, horizontal bars represent medians, and error bars show interquartile ranges. A Kruskal-Wallis H test was used to detect significant intergroup variability, and a Mann-Whitney U test was used for between-group comparison. The Spearman rank test was used for correlations. Statistical analysis was performed with SPSS software (SPSS, Chicago, Ill). For mouse studies, the significance of differences among groups was determined by 1-way ANOVA (nonparametric test) using GraphPad Prism statistical software program (GraphPad, Inc, La Jolla, Calif). When 2 groups were compared, an unpaired, 2-tailed Student t test was used. A P value of less than .05 was considered statistically significant.

RESULTS

Increased levels of IDO1, tryptophan2,3-dioxygenase, and KYN in CRSwNP

AhR signaling can be activated by KYN or other endogenous tryptophan metabolites generated by IDO1 and tryptophan2,3-dioxygenase (TDO2).^26,27 We first assessed whether the expression of IDO1 and TDO2 was increased in patients with CRSwNP. Compared with controls, NP tissues from patients with eosinophilic and noneosinophilic CRSwNP showed increased mRNA expression of IDO1 and TDO2 (Fig 1, A). Compared with noneosinophilic polyps, eosinophilic polyps showed a further increase in the mRNA expression of IDO1. Similar patterns were observed for IDO1- and TDO2-positive cells in lamina propria of sinonasal mucosa as analyzed by immunostaining (Fig 1, B); both eosinophilic and noneosinophilic polyps had a greater number of IDO1- and TDO2-positive cells as compared with controls (Fig 1, C). The increased expression of IDO1 in CRSwNP was further confirmed by Western blot (see Fig E1 in this article’s Online Repository at www.jacionline.org). We next analyzed whether these increased levels of IDO1 and TDO2 can lead to the generation of KYN by measuring levels of KYN in NP tissues with HPLC. Consistent with the expression of IDO1 and TDO2, the levels of KYN were much higher in both eosinophilic and noneosinophilic polyps when compared with controls, with a further increase in eosinophilic polyps compared with noneosinophilic polyps (Fig 1, D). Consistent with our previous reports,^15,16 we found that the number of activated mast cells in lamina propria (Fig 1, E and F), but not in epithelium (Fig 1, E and G), and local levels of local IgE (Fig 1, H) were markedly increased in eosinophilic polyps compared with noneosinophilic polyps and controls. Furthermore, there was a significant correlation between activated mast cells in lamina propria and levels of KYN (see Fig E2, A, in this article’s Online Repository at www.jacionline.org) and local IgE (Fig E2, B) in eosinophilic polyps. Taken together, these data demonstrated increased levels of KYN, activated mast cells, and local IgE in eosinophilic polyps.

FIG 1.

Expression of IDO1, TDO2, KYN, IgE, and mast cell activation in NPs and control tissues. A, The relative transcript levels of IDO1 and TDO2 were quantified by using PCR. B, Representative photomicrographs showing immunostaining of IDO1 and TDO2 (original magnification ×400). C, IDO1- and TDO2-positive cells were quantified by per hpf. D, Quantification of KYN protein levels by HPLC. E, Representative photomicrographs of immunohistochemical staining of eosinophilic polyp tissue section for tryptase showing activated (degranulated) mast cells (red arrows) and undegranulated mast cells (black arrows). Original magnification ×400. F and G, Number of activated (degranulated) mast cells in lamina propria (Fig 1, F) and epithelium (Fig 1, G) of NPs and control tissues. H, Local total IgE levels. Eos CRSwNP, Eosinophilic CRSwNP; Epi, epithelium; non-Eos CRSwNP, noneosinophilic CRSwNP; LP, lamina propria.

Activated AhR signaling in CRSwNP

Given the increased KYN in CRSwNP, we tested whether KYN can activate AhR signaling, particularly in mast cells, in CRSwNP. We first detected AhR⁺ cells and their relationship with mast cells in polyps. We found that about an average of 75% and 71% of AhR⁺ cells are tryptase⁺ mast cells in eosinophilic and noneosinophilic polyps, respectively (Fig 2, A). We further confirmed the predominant expression of AhR in mast cells in polyps by coimmunoflorescence staining for AhR and CD117 (c-Kit) (Fig 2, B). In addition, we detected the expression of AhR target genes in the sinonasal mucosa, including the canonical AhR target genes cyp1a1 and cyp1b1 (Fig 2, C). Both eosinophilic and noneosinophilic polyps showed increased mRNA expression for cyp1a1 and cyp1b1 when compared with controls (Fig 2, C). Compared with noneosinophilic polyps, eosinophilic polyps showed a further increase in the mRNA expression of cyp1a1 (Fig 2, C). The increased expression was further confirmed specifically in mast cells by coimmunofluorescent staining (Fig 2, D and F). CRSwNP showed an increased percentage of cyp1a1⁺ (Fig 2, E) or cyp1b1⁺ (Fig 2, G) mast cells over total mast cells in lamina propria as compared with controls, with a further increase in eosinophilic CRSwNP compared with noneosinophilic CRSwNP. Overall, these data suggest an increased activation of AhR signaling in mast cells in NPs.

FIG 2.

Mast cells are the predominant cells expressing AhR in NPs. A, Representative immunostaining of consecutive sections with anti-AhR and antitryptase antibody, respectively. Arrows with same direction indicate same cells in consecutive sections. B, Representative staining for colocalization of AhR (red) and mast cells (CD117, green) in human NP tissues from a patient with Eos CRSwNP. C, Relative transcript levels of cyp1a1 and cyp1b1 in nasal mucosa. D–G, Representative staining for colocalization of cyp1a1 (Fig 2, D) or cyp1b1 (Fig 2, F) (green) with tryptase (red) in human NP tissues. Percentage of cyp1a1⁺ (Fig 2, E) or cyp1b1⁺ (Fig 2, G) mast cells over tryptase⁺ mast cells was quantified. Original magnification ×400 for all figures. DAPI, 4′-6-Diamidino-2-phenylindole, dihydrochloride; Eos CRSwNP, eosinophilic CRSwNP; non-Eos CRSwNP, noneosinophilic CRSwNP.

KYN potentiates allergen-driven activation of AhR signaling in mast cells

To determine whether KYN can potentiate allergen-induced mast cell activation through AhR signaling in eosinophilic CRSwNP, mouse BMMCs were sensitized and challenged using our well-established protocol as illustrated in Fig 3, A.^21,33 BMMCs were isolated and identified by flow cytometry with antibodies against both c-Kit and FcεRI (Fig 3, B). The mRNA expression of AhR, cyp1a1, and cyp1b1 was significantly increased in the OVA-activated BMMCs, which could be further enhanced by KYN (Fig 3, C–E). These data support a role for KYN in potentiating allergen-activated AhR signaling in mast cells.

FIG 3.

KYN potentiates allergen-activated AhR signaling in mast cells. A, Experimental setup for KYN-potentiated IgE-mediated mast cell activation. B, Mouse BMMCs were confirmed by flow cytometry analysis. C–E, Transcript levels of AhR (Fig 3, C), cyp1a1 (Fig 3, D), and cyp1b1 (Fig 3, E) were quantified by using PCR in OVA-sensitized and -challenged mast cells with or without KYN. Results are expressed as fold changes over medium in expression of target genes to the internal control β-actin. Data are expressed as mean ± SEM of 2 independent experiments. *P < .05, **P < .01.

KYN enhances allergen-induced mast cell ROS generation, intracellular Ca²⁺ levels, and mast cell activation

ROS are important mediators that may contribute to oxidative damage and chronic inflammation in allergic diseases.^41–44 Increased ROS have also been reported in the sinus tissues of patients with CRSwNP.^34,45,46 To detect whether KYN can enhance allergen-activated ROS generation in mast cells, we measured intracellular ROS production in BMMCs. As shown in Fig 4, A, KYN promoted OVA-induced intracellular ROS generation in BMMCs, but the increased ROS was not observed in AhR-deficient (AhR^−/−) BMMCs (Fig 4, B). Next, we measured the mitochondrial ROS in these BMMCs using MitoSOX (Fig 4, C). We found higher levels of mitochondrial ROS in OVA-treated BMMCs compared with controls, and the increase was further enhanced by KYN (Fig 4, D). Similar to intracellular ROS, the KYN-potentiated mitochondrial ROS increase was not observed in AhR^−/− BMMCs. These findings support a view that KYN may potentiate ROS production, particularly mitochondrial ROS in allergen-activated mast cells through AhR.

FIG 4.

KYN enhances allergen-induced mast cell ROS, intracellular [Ca²⁺]i, and mast cell activation. A and B, Measurement of intracellular ROS by flow cytometry with CM-H2DCFDA in anti-OVA IgE-sensitized and OVA-challenged WT (Fig 4, A) and AhR-null (Fig 4, B) mouse BMMCs in the presence or absence of KYN. C, Representative MitoTracker or MitoSOX staining of these anti-OVA IgE-sensitized and OVA-challenged BMMCs derived from WT and AhR-deficient mice. Original magnification, ×20. D, Quantitative data for immunofluorescence staining. E and F, Representative Fluo-4 fluorescence heat map images (Fig 4, E) and imaging traces (Fig 4, F) of anti-OVA IgE-sensitized BMMCs with or without KYN showing changes in [Ca²⁺]i induced by OVA with or without KYN. G, Quantification of responding cells (>150 cells counted per condition). Data are presented as mean ± SEM. H and I, Representative time-lapse (Fig 4, H) and mean curves of Av.SRho MFI (Fig 4, I) of single anti-OVA IgE-sensitized BMMC activated with OVA. J, Expression of CD107α/LAMP-1 detected by flow cytometry analysis. K–M, Mast cell activation was assessed by β-hexosaminidase (Fig 4, K), histamine release (Fig 4, L), and IL-13 (Fig 4, M). Data are expressed as mean ± SEM of 3 independent experiments. AUC, Area under the curve; MFI, mean fluorescent intensity. *P < .05, **P < .01.

We had previously shown that the generation of intracellular ROS was dependent on increased intracellular calcium [Ca²⁺]i.²¹ To further determine whether KYN-increased ROS production in BMMCs is associated with changes in [Ca²⁺]i, we measured [Ca²⁺]i in OVA-activated BMMCs treated with or without KYN. Compared with untreated BMMCs, OVA-activated BMMCs exhibited elevated levels of [Ca²⁺]i as defined by heatmap images (Fig 4, E), mean fluorescent intensity (Fig 4, F), and total calcium response (Fig 4, G). Treatment with KYN further enhanced OVA-induced levels of [Ca²⁺]i. Using the same protocol, we also examined whether KYN can enhance allergen-induced mast cell activation. We first investigated the spatiotemporal dynamics of mast cell degranulation by using a soluble fluorochrome-labeled avidin (sulforhodamine 101–coupled avidin [Av.SRho]).^47,48 OVA-sensitized BMMCs were challenged with OVA (10 μg/mL) in the presence of Av.SRho and inspected by time-lapse confocal microscopy. As shown in Fig 4, H, this stimulation induced granule budding on the BMMCs. Measurement of the intensity of the Av.SRho staining using the Region Measurement function of the MetaMorph software showed that degranulation was detected approximately 20 minutes after OVA stimulation and reached a plateau approximately 55 minutes after stimulation. A similar degranulation pattern was observed with AhR^−/− BMMCs; however, this response was significantly weakened (Fig 4, I). Furthermore, staining of CD107a/LAMP-1 for degranulation in BMMCs was quantified by flow cytometry analysis (Fig 4, J). BMMCs from AhR^−/− mice showed reduced expression of CD107a/LAMP-1 after OVA challenge when compared with those from WT mice. We also quantified mast cell degranulation using β-hexosaminidase release assays. Compared with unchallenged, OVA-activated BMMCs showed elevated levels of β-hexosaminidase, which was further strengthened by KYN (Fig 4, K). Moreover, compared with WT, AhR^−/− mast cells had lower levels of β-hexosaminidase following OVA or OVA with KYN treatment (Fig 4, K). Of interest, similar results were obtained for histamine release (Fig 4, L) and IL-13 secretion (Fig 4, M). These data further support the previous findings that KYN can potentiate allergen-activated mast cell activation.

AhR is required for IgE-mediated and mast-cell–dependent allergic responses

To determine whether mast cell–expressed AhR is required for IgE-mediated and mast-cell–dependent allergic responses, we engrafted BMMCs from WT and AhR^−/− mice into the ears of mast cell–deficient mice (Kit^W-sh/W-sh) using the protocol illustrated in Fig 5, A.⁴⁰ Mast cell engraftment was confirmed by Toluidine blue staining after 6 weeks (Fig 5, B and C). Compared with Kit^W-sh/W-sh mice without engraftment, Kit^W-sh/W-sh mice engrafted with WT BMMCs showed increased OVA-induced PCA (Fig 5, D). Of interest, Kit^W-sh/W-sh mice engrafted with AhR^−/− BMMCs showed significant reduction in OVA-induced PCA when compared with those engrafted with WT BMMCs. Thus, these data suggest that mast cell–expressed AhR is required for IgE-mediated and mast-cell–dependent allergic responses.

FIG 5.

Engraftment of mast cell–deficient mice (Kit^W-sh/W-sh) with BMMCs from WT and AhR^−/− mice. A, Experimental protocol for mast cell engraftment. B, Representative Toluidine blue staining of ear sections (arrow: mast cell). C, Quantification of cells with positive staining for Toluidine blue in Fig 5, B. D, Representative images of Evans blue–stained extravasation into skin (n = 4–5/group). EB, Evan’s Blue; ns, nonsignificant. Data are expressed as mean ± SEM of 2 independent experiments. **P < .01.

Patients with CRSwNP showed increased oxidative activation of CaMKII

We have recently demonstrated that ox-CaMKII in mast cells promoted allergen-induced mast cell activation and cockroach allergen–induced lung inflammation.³³ To test whether ox-CaMKII is also involved in KYN/AhR signaling–mediated mast cell activation in CRSwNP, we first measured ox-CaMKII expression in the polyp tissues (Fig 6, A). Immunohistochemical staining showed that ox-CaMKII was mainly expressed in airway epithelial cells and infiltrating cells in sinonasal mucosa. Both eosinophilic and noneosinophilic polyps had greater numbers of ox-CaMKII–positive cells in lamina propria when compared with those from controls (Fig 6, B). The increased expression of ox-CaMKII in polyp tissues was further confirmed by Western blot (Fig 6, C). We next studied the expression of ox-CaMKII specifically on mast cells by coimmunofluorescence staining for ox-CaMKII and tryptase (Fig 6, D). We found that the percentage of ox-CaMKII⁺tryptase⁺ cells overall total mast cells in lamina propria was much higher in eosinophilic and noneosinophilic polyps as compared with controls, with a further increase found in eosinophilic polyps (Fig 6, E).

FIG 6.

Expression of ox-CaMKII in NPs and control tissues. A, Representative immunostainings of ox-CaMKII in sinonasal mucosa. B, ox-CaMKII–positive cells were quantified by per hpf. C, Measurement of ox-CaMKII protein by Western blotting. β-actin was used as a control. D, Representative immunofluores-cent staining for colocalization of ox-CaMKII (red) and tryptase (green) in human NP tissues. E, Percentage of ox-CaMKII and tryptase double-positive cells over tryptase⁺ mast cells was quantified. Original magnification × 400 for all figures. DAPI, 4′-6-Diamidino-2-phenylindole, dihydrochloride; Eos CRSwNP, eosinophilic CRSwNP; non-Eos CRSwNP, noneosinophilic CRSwNP.

AhR regulates allergen-activated mast cell activation through ox-CaMKII

We further examined whether the differential expression of ox-CaMKII in polyp’s mast cells is regulated by AhR. We specifically examined ox-CaMKII expression in OVA-activated WT and AhR^−/− BMMCs. BMMCs were sensitized and challenged following the previous protocol in Fig 3, A. Compared with unchallenged, OVA-challenged BMMCs showed increased ox-CaMKII, which was further enhanced by KYN (Fig 7, A and B). However, the increased ox-CaMKII was not seen in AhR^−/− BMMCs (Fig 7, A and B). No significant difference was detected for total CaMKII among all these groups (Fig 7, C). Next, we examined whether ox-CaMKII mediated mast cell activation. Using the same approach as above in Fig 4, H, we investigated the spatiotemporal features of mast cell degranulation by using Av.SRho^47,48 (Fig 7, D). Compared with WT, MMVVδ BMMCs showed weakened intensities in Av.SRho staining on the surface of the cells (Fig 7, E). We then measured cytokines released from BMMCs from WT and ROS-resistant CaMKII MMVVδ mice (MMVVδ). Compared with control, increased levels of IL-5, IL-13, and IL-33 were observed in OVA-activated BMMCs (Fig 7, F). However, the increase was markedly suppressed in BMMCs from MMVVδ. These findings indicate that AhR regulates allergen-induced mast cell activation though ox-CaMKII.

FIG 7.

AhR regulates allergen-activated mast cell activation through ox-CaMKII. A, Representative immunoblot of ox-CaMKII and total CaMKII (Tot-CaMKII) in anti-OVA IgE-sensitized and OVA-challenged mouse BMMCs. β-actin was used as a control. B and C, Semi-quantitative analysis of ox-CaMKII (Fig 7, B) and total CaMKII (Fig 7, C) protein levels. D and E, Representative time-lapse (Fig 7, D) and mean curves of Av.SRho MFI (Fig 7, E) of single anti-OVA IgE-sensitized BMMC activated with OVA. F, Cytokine release from mast cells was assessed by IL-5, IL-13, and IL-33. MFI, mean fluorescent intensity. Data are expressed as mean ± SEM from 2 independent experiments. *P < .05, **P < .01.

CaMKII inhibitor suppresses KYN-potentiated allergen-induced mast cell activation

KN-93 can suppress OVA-induced asthma.³⁰ To further explore the possible role of CaMKII in mast cell activation, we used KN-93 to investigate whether inhibition of CaMKII can abrogate the KYN-promoted mast cell activation induced by OVA challenge following the protocol described in Fig 3, A. We found that KYN can enhance OVA-activated mast cell activation including degranulation (Fig 8, A), histamine release (Fig 8, B), and IL-13 production (Fig 8, C). However, as expected, the KYN-enhanced mast cell activation was significantly inhibited when these BMMCs were pretreated with KN-93 (Fig 8, A–C). Furthermore, we examined whether KN-93 affects KYN-potentiated OVA-induced PCA following the protocol described in Fig 8, D. The KYN potentiated OVA-induced PCA, which was significantly suppressed when mice were pretreated with KN-93 (Fig 8, E and F). These findings further supported that CaMKII may be crucial for the KYN-potentiated allergen-induced mast cell activation and allergic responses.

FIG 8.

CaMKII inhibitor suppresses KYN-potentiated OVA-induced mast cell activation. A–C, Mast cell activation was assessed by β-hexosaminidase (Fig 8, A), histamine (Fig 8, B), and IL-13 (Fig 8, C). Data are expressed as mean ± SEM from 2 independent experiments. D, Experimental setup for KN-93 inhibition of KYN-potentiated IgE-mediated mast cell activation. E, Representative images of Evans blue–stained extravasation into skin. F, Quantification of the extravasation of Evans blue leakage into the skin (Fig 8, E, n = 3–4/group). G, Mechanistic model of the KYN/AhR axis in regulating mast cell activation through ROS and ox-CaMKII in the development of CRSwNP. EB, Evan’s Blue; LTC4, Leukotriene C₄; PDG2, prosta-glandin D2.

DISCUSSION

Eosinophilic CRSwNP is characterized by T_H2-biased inflammation with high concentrations of IgE against different allergens found in polyps.⁴⁹ The hyperproduced IgE can sensitize and activate resident mast cells, leading to mast cell activation and subsequent eosinophilic inflammation in CRSwNP.^15,16 Studies from our research group have illustrated that activated mast cells are increased in patients with eosinophilic CRSwNP, which are positively correlated with levels of local IgE.¹⁶ Thus, in this study, we extended our previous work to test our novel hypothesis that endogenous tryptophan metabolites function as AhR ligands to promote allergen-induced, IgE-mediated activation of mast cells in eosinophilic CRSwNP, and explored their possible mechanisms.

It is well established that the activation of IDO1 and TDO2 and the generation of KYN metabolites are critical in modulating the activation of various innate and adaptive immune cells through different biochemical mechanisms.⁵⁰ KYN is a novel endogenous ligand for AhR. The downstream events of AhR activation appear to occur in a cell-type and immune context–dependent manner. For example, KYN-activated AhR drives naive CD4⁺ T-cell differentiation into immunosuppressive Foxp3⁺ regulatory T cells rather than proinflammatory T_H17 cells,⁵¹ representing a potent inhibitory signaling for T-cell–mediated immune responses. In contrast, studies from our laboratory have demonstrated that AhR ligands (eg, FICZ²¹ and KYN²⁷) potentiate IgE-mediated mast cell activation and subsequently promote inflammatory immune responses. Because of those fundamental differences in findings, it is essential to further explore the mechanisms underlying the effect of KYN/AhR signaling on mast cell activation in human diseases. In this study, we provide novel evidence to support a potential role of KYN-AhR signaling to promote allergen-induced, IgE-mediated activation of mast cells in eosinophilic CRSwNP.^16,18 We found that AhR was predominantly expressed in mast cells in NPs. Most importantly, we observed differences in the levels of KYN and AhR downstream target genes cyp1a1 and cyp1b1 in NP tissues between eosinophilic and noneosinophilic polyps. Furthermore, there was a significant correlation between the levels of KYN and local IgE and the number of activated mast cells in eosinophilic polyps. Therefore, it is tempting to speculate that KYN-AhR signaling is crucial in determining mast cell activation in eosinophilic polyps with high levels of local IgE, which may contribute to the different patterns of inflammation and distinct clinical features between eosinophilic and noneosinophilic CRSwNP. Indeed, numerous studies have demonstrated significant differences in clinical features, patterns of inflammation, and tissue remodeling between eosinophilic and noneosinophilic CRSwNP.^16,52,53 We have made great efforts to identify the major risk factors that may contribute to the differences. In particular, we have previously demonstrated differences in the numbers of active mast cells and levels of local total and allergen-specific IgE.¹⁶ Also, we have displayed differences in dendritic cells and T-cell subtypes.^14,54 In this study, we made a new discovery that the AhR signaling–potentiated mast cell activation may be different between eosinophilic and noneosinophilic CRSwNP. Although results from previous and current studies did not show any differences in activated mast cells and local IgE,¹⁶ significant differences were noted for IDO1, TDO2, KYN, and AhR downstream target genes cyp1a1 and cyp1b1 between noneosinophilic CRSwNP and controls. These findings raise the possibility that the KYN-AhR signaling pathway plays a role in contributing to the different patterns of inflammation between noneosinophilic CRSwNP and controls, but may not be through mast cells. Thus, it is tempting to speculate that the KYN-AhR signaling pathway is required for CRSwNP but may play a distinct role in eosinophilic or noneosinophilic CRSwNP depending on its local immune cells (eg, mast cells vs other cell types).

We have previously demonstrated that AhR signaling mediated mast cell activation.²¹ In this study, we provided additional evidence with the spatiotemporal dynamics of mast cell degranulation in live cells from WT and AhR^−/− mice by Av.SRho, a probe that selectively stains mast cell granules.^47,48,55 This method made it possible to see that in the early step of mast cell granule exocytosis, the granule matrix is externalized and immediately bound by Av.SRho. We found that AhR^−/− BMMCs showed weakened Av.SRho staining on the surface of the cells after OVA stimulation and challenge compared with WT. The results were further supported by flow cytometry analysis of CD107a/LAMP-1. Most intriguingly, to determine the significance of AhR expressed on mast cells in allergen-induced allergic responses, we engrafted BMMCs from WT and AhR^−/− mice into Kit^W-sh/W-sh mice and tested for PCA analysis. We noted that Kit^W-sh/W-sh mice engrafted with AhR^−/− BMMCs showed significant reduction of OVA-induced PCA when compared with those engrafted with WT BMMCs, suggesting that the mast cell–expressed AhR may be required for IgE-mediated and mast-cell–dependent allergic responses.

We next explored the underlying mechanisms regarding the AhR signaling–promoted, IgE-mediated mast cell activation. Our data provided evidence that the AhR ligand–activated mast cell activation is dependent on ROS generation and intracellular Ca²⁺. Particularly, AhR ligands are able to increase ROS generation as a consequence of the increased expression of AhR target genes (eg, cyp1a1)^21,56 and several membrane nicotinamide adenine dinucleotide phosphate-oxidase complexes, p40phox.⁵⁷ Increased ROS generation in the sinus tissues of patients with CRSwNP has also been reported.^45,46 Our present studies showed that OVA induced higher levels of intracellular ROS, which was further enhanced by KYN treatment in WT, but not AhR^−/− mast cells. A similar pattern to that of intracellular ROS was discovered for the mitochondrial ROS in mast cells. These findings support that KYN may potentiate ROS generation in allergen-activated mast cells through the activation of AhR signaling. To explore the mechanisms underlying ROS generation, we analyzed intracellular Ca²⁺, which may contribute to the generation of intracellular ROS. We found that OVA-activated mast cells exhibited elevated levels of intracellular Ca²⁺, which was further promoted by KYN. Given that we have previously demonstrated the significant inhibition of the intracellular ROS by the calcium antagonist 2-aminoethoxydiphenyl borate,^21,33 we conclude that intracellular Ca²⁺ may play a critical role in the KYN-promoted generation of ROS.

CaMKII is within one of the downstream signaling pathways activated by ROS,²⁸ and the increased expression of ox-CaMKII has been associated with asthma.^30–32 Furthermore, our recent studies have indicated a role of ox-CaMKII in modulating allergen-induced mast cell activation and promoting cockroach allergen–induced asthma.³³ In this study, we demonstrated for the first time that AhR regulates oxidative activation of CaMKII through controlling ROS generation, suggesting that CaMKII may be a target of oxidation via AhR-and ROS-dependent mechanisms. Of interest, ox-CaMKII was significantly increased, particularly in mast cells of eosinophilic polyps. Our studies provided experimental evidence that ox-CaMKII modulates allergen-induced, and KYN-promoted mast cell activation, including mast cell degranulation, Leukotriene C₄, histamine release, cytokine production (eg, IL-5, IL-13, and IL-33), and PCA. In addition, previous reports found that KN-93, an inhibitor of CaMKII,⁵⁸ can suppress OVA-induced asthma³⁰ and mast cell activation.³³ Indeed, KYN-potentiated OVA-induced mast cell activation in vitro and PCA in vivo were significantly blocked when CaMKII inhibitor KN-93 was used. Those findings suggest that AhR-regulated mast cell activation may be through the regulation of ROS generation and subsequent ox-CaMKII in eosinophilic polyps. KN-93 or targeting on ox-CaMKII has the potential for the treatment of CRSwNP.

In summary, our studies tested the novel hypothesis (Fig 8, G) that KYN serves as an AhR ligand and promotes allergen-induced, IgE-mediated mast cell activation through the upregulation of ROS generation and oxidative activation of CaMKII and subsequent chronic inflammation in eosinophilic CRSwNP. However, several questions remain to be elucidated in the future, including triggers in inducing the expression of IDO1 and TDO2, and increased levels of KYN in CRSwNP, AhR-regulated mast cell–independent immune response in noneosinophilic CRSwNP, and the mechanisms underlying the ox-CaMKII–mediated mast cell activation. Interestingly, genomic analysis in evaluating nasal and sinus bacterial and fungal microbiomes in CRS has suggested associations between certain microorganisms and CRS.⁵⁹ However, there is no publication to show whether those nasal microbiomes contribute to the increased IDO/TDO/KYN and activation of AhR signaling, which is an interesting question and worth of future investigation. In addition, in contrast to Takabayashi et al’ s report,¹⁵ our present and previous study demonstrated a significant accumulation of activated mast cells in stroma, but not in glands in eosinophilic NPs.¹⁶ We postulated that the differences in the observed location of mast cells among different studies may be due to differences in the status of mast cell activation or differences in the particular mediators that activate mast cells (eg, inhaled or endogenous). Thus, future studies are clearly needed to investigate the distribution of mast cells and their corresponding phenotypes in polyp tissues of CRSwNP from different ethnic populations. Collectively, these studies could provide a conceptual framework linking the KYN/AhR axis to mast cell activation through regulating ROS generation and oxidative activation of CaMKII and new insights onto the mechanisms that underlie the pathogenic progression of CRSwNP.

Key messages.

• AhR was predominantly expressed in mast cells in NPs.

• Activated mast cells and local IgE levels were increased in eosinophilic polyps.

• KYN potentiated allergen-induced ROS generation, intracellular Ca²⁺ levels, and activation of mast cells.

• Increased expression of ox-CaMKII is critical for the KYN-promoted allergen-induced mast cell activation in eosinophilic polyps.

Abbreviations used

AhR

Aryl hydrocarbon receptor

BMMCs

Bone marrow–derived mast cells

CaMKII

Calmodulin-dependent protein kinase II

CRS

Chronic rhinosinusitis

CRSwNP

Chronic rhinosinusitis with nasal polyps

Cyp1a1

Cytochrome P450 family 1 subfamily A member 1

Cyp1b1

Cytochrome P450 family 1 subfamily B member

IDO1

Indoleamine 2,3-dioxygenase 1

KYN

Kynurenine

NP

Nasal polyp

OVA

Ovalbumin

Ox-CaMKII

Oxidized calmodulin-dependent protein kinase II

PCA

Passive cutaneous anaphylaxis

ROS

Reactive oxygen species

TDO2

Tryptophan2,3-dioxygenase

WT

Wild-type