IL-1β Derived Th17 Immune Responses Are a Critical Factor for Neutrophilic-Eosinophilic Airway Inflammation on Psychological Stress-Induced Immune Tolerance Breakdown in Mice

Bronchial asthma is a disease in which the mucous membrane thickens due to chronic inflammation of the bronchi and narrowing of the airways, leading to wheezing and dyspnea. In 2017, there were 273 million people with asthma (prevalence rate 3.57%), and the number of deaths due to asthma was 490,000 (mortality rate 0.006%) [1]. Aspirin- and exercise-induced asthma are two common phenotypes of asthma with distinct clinical, physiological, and morphological characteristics. Asthma phenotypes are challenging to treat due to the heterogeneity of the disease manifestations [2]. To facilitate treatment, it is necessary to classify asthma into endotypes based on the underlying molecular pathology [2].

In our previous studies, we reported that exposure to psychological stress during the induction phase of respiratory tolerance, but not during the sensitization phase, promoted the development of type 2 helper T (Th2)-biased sensitization and, consequently, induced asthmatic airway responses to inhaled allergens. In contrast, restraint stress exposure did not affect sensitization and allergic airway inflammation [3‒5]. Tolerance to ovalbumin (OVA) before sensitization in asthmatic mice increases the levels of T regulatory (Treg) cells [6, 7]. Treg cells maintain immunological unresponsiveness to self-antigens and suppress excessive immune responses that are detrimental to the host [8]. Previously, we reported that stress disrupts immune tolerance [5, 9]. The level of Treg cells, confirmed by a transcription factor Foxp3, was increased in tolerized mice compared to stressed/tolerized mice [5, 10, 11].

In addition, we previously reported elevated neutrophils in psychologically stressed mice with a compromised immune tolerance [5, 9]. Th17 cells are involved in the neutrophil immune responses. For naive T cells to differentiate into Th17 cells, cytokines, such as IL-1β, and transcription factors, such as retinoid-related orphan receptor γt (RORγt), are required [12, 13]. Furthermore, it has been pointed out that Th17 response enhancement through induction of IL-1β production is involved in immune modulation by stress exposure. Exacerbation of symptoms by stress exposure IL1-β production has been implicated in multiple sclerosis [14, 15]. These findings suggest that psychological stress has the potential to activate the IL-17 immune response. Therefore, IL-1β-associated Th17 cell immune responses may play a role in the breakdown of immune tolerance due to stress exposure.

Stress is inevitable in modern-day life. However, suppression of stress-induced effects could prevent the exacerbation of asthma [3]. Therefore, in this study, we focused on the stress-IL-1β-Th17 axis to evaluate the impact of allergen-induced pre-sensitization stress on the development of asthma. Furthermore, we examined the potential of an interleukin-1 receptor antagonist (IL-1RA) to suppress stress-related asthma.

Specific pathogen-free female BALB/c mice were purchased from CLEA Japan (Tokyo, Japan). Mice were housed under a 12-h light/dark cycle at a constant temperature of 23 ± 1°C. Sterilized food and water were provided ad libitum. All experiments described below were approved by the Committee of Animal Experiments at Tohoku Medical and Pharmaceutical University (approval numbers: 16002-cn, 17004-cn, A2040, A2207). All experiments were performed in compliance with relevant institutional guidelines.

Seven-week-old mice were sensitized and exposed to inhalation antigens as described previously [5] (shown in Fig. 1a). Briefly, respiratory tolerance was induced by two instances of inhalation of chicken (grade V; Sigma-Aldrich, St. Louis, MO, USA) without aluminum hydroxide on days 6 and -3. In the tolerance induction phase for the first 30 min of stress exposure time (described below), aerosolized OVA (5 mg/mL in saline) flowed through the chambers for OVA exposure. For the control group, saline alone was inhaled through the chambers (asthmatic mice). To induce restraint stress, each mouse was placed in a 50-mL conical centrifuge tube with multiple ventilation holes [5]. This restraint allowed the mice to rotate from a supine to a prone position but not turn their heads toward their tails or consume food and water. Restraint stress is generally considered to be able to induce psychological stress in animals [16]. During the tolerance phase, stressed/tolerized and stressed mice in 50-mL conical centrifuge tubes were left in plastic chambers for 6 h for 6 consecutive days (−7 to −2) at the same time each day. Non-stressed mice were deprived of food and water during the period when the stressed mice were exposed to stress. Then, mice were sensitized via intraperitoneal injection of OVA (8 μg/mouse) adsorbed onto aluminum hydroxide (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan; 4 mg/mouse) on day zero and 5. After the challenge, bronchoalveolar lavage (BAL) was performed [17].

BALB/c mice were divided into four groups: asthmatic mice, stressed mice, tolerized mice, and stressed/tolerized mice. Bronchoalveolar lavage fluid (BALF) was obtained from mice in each treatment group 5 days after the last OVA challenge (day 22). BALF was recovered by injecting 0.25 mL of 4°C cold phosphate-buffered saline (PBS) twice to wash the lungs, and approximately 0.4 mL of the recovered PBS was pooled and obtained from each mouse. BAL cells (2 × 10⁵) were stained with DiffQuik solution (Sysmex Co., Kobe, Japan), and cell differential percentage was determined by counting a minimum of 200 cells via light microscopy [5].

The concentrations of OVA-specific IgE and IgG1 in serum collected on day 22 were measured by ELISA, as described previously [17]. The sensitivity of detection was 1.9 × 10⁻² EU/mL for IgE and 1.9 × 10⁻² EU/mL for IgG1.

The entire lungs of mice were homogenized in chilled 0.1% Triton-X PBS with 1% protease inhibitor (Sigma-Aldrich). After centrifugation at 15,000×g for 15 min at 4°C, the supernatants were harvested and stored at −80°C until cytokine analysis [18].

The IL-1β and IL-17A levels were measured using ELISA kits (R&D Systems, Minneapolis, MN, USA), and the detection limits were 2.31 pg/mL and 5 pg/mL, respectively. Total protein levels in lung homogenates were measured using a detergent-compatible protein assay kit (Bio-Rad Laboratories, Hercules, CA, USA). The cytokine concentration in the lungs was adjusted for the protein level in each lung [18].

Bronchial lymph nodes (BLN) were collected on day −2 from the stressed/tolerized mice and tolerized mice. RNA was extracted using the RealiaPrep RNA Cell Miniprep System (Promega, Madison, WI, USA) according to the manufacturer’s protocol.

Microarray hybridization was performed by Kurabo Industries, Ltd. RNA quality was first assessed using an Agilent 2100 bioanalyzer, and only samples with RNA integrity number values >8 were subsequently used. A mouse Clariom S microarray (Thermo Fisher Scientific, Waltham, MA, USA) was used to analyze only the fully annotated genes. The Clariom S Array can analyze more than 22,100 genes in mice, and the gene transcript levels were determined using the Microarray Analysis Suite Software Transcriptome Viewer (KURABO Industries Ltd.).

Total RNA was extracted from BLN. RNA was extracted as previously described. cDNA was synthesized using the PrimeScript RT reagent Kit (Takara Bio Inc. Kusatsu, Shiga, Japan) following the manufacturer’s protocol. PCR was performed using Power SYBR Green Master Mix and the Step One Real-Time PCR System (Thermo Fisher Scientific). HPRT expression was used as the internal control. The following primers synthesized by Eurofines Genomics (Tokyo, Japan) were used: Hprt forward 5′-CTT CCT CCT CAG ACC GCT TT-3′, reverse 5′-CAT CATC GCT AAT CAC GAC GC-3′; Rorc forward 5′-GAC CCA CAC CTC ACA AAT TGA-3′, reverse 5′-AGT AGG CCA CAT TAC ACT GCT-3′; Icos forward 5′- ATG AAG CCG TAC TTC TGC CG-3′, reverse 5′-AGT AGG CCA CAT TAC ACT GCT-3′. qPCR analysis cycling conditions were as follows: an initial denaturing step at 95°C for 10 min, followed by 40 cycles of a denaturation step (95°C for 10 s) and an annealing/extension step (60°C for 30 s). Fold changes in gene expression were determined using standard curve method.

BLN was collected on day 20 from the mice in the four groups: asthma, stress, tolerized, and stress/tolerized groups. The BLN cells were preincubated with anti-CD16/CD32 (FC gamma III/II receptor; BD Biosciences, Franklin Lakes, NJ, USA) to reduce the nonspecific binding of the subsequent antibodies. Dead cells were excluded using the LIVE/DEAD Fixable Blue Dead Cell Stain Kit (Thermo Fisher Scientific). The cells were stained for surface antigens with anti-CD3ε-PerCP-Cy5.5 (clone 145-2C11; Miltenyi Biotec, Bergisch Gladbach, Germany), anti-CD4-AF700 (clone GK1.5; BD Biosciences), or isotype control antibodies. For intracellular staining, cells were stimulated with phorbol 12-myristate 13-acetate (50 ng/mL; Sigma-Aldrich), ionomycin (1,000 ng/mL; Sigma-Aldrich), and monensin (2 μm; BioLegend, San Diego, CA, USA) for 4 h before surface antigen staining. After fixation and permeabilization, the cells were incubated with anti-Foxp3-PE-Cy7 (clone FJK-16s; Thermo Fisher Scientific), anti-IL-17A-APC-Cy7 (clone.TC11-18H10.1; BioLegend), anti-IL4-PE-CF594 (clone 11B11; BioLegend), or an isotype control antibody. We considered CD3⁺CD4⁺Foxp3⁺ cells as Treg cells, CD3⁺CD4⁺IL-4⁺ cells as Th2 cells, and CD3⁺CD4⁺IL-17a⁺ cells as Th17 cells. The cells were counted on a FACSAria II flow cytometer (BD Biosciences), and the analyses were performed using the FACSDiva software (BD Biosciences).

IL-1RA (140 μg/kg i.p.) (BioLegend) was administered for 60 min before stress exposure between day −7 and −2.

Data are expressed as the mean ± standard deviation from multiple independent experiments (as indicated by n values). Significant differences between the two groups were determined using the nonparametric Mann-Whitney U test. The analyses were performed using Prism 6 (GraphPad Software, San Diego, CA, USA). Statistical significance was set at p < 0.05.

We investigated the mechanism underlying psychological stress in the development of neutrophilic and eosinophilic airway inflammations and the induction of IgE and IgG1 synthesis. First, we evaluated the effect of psychological stress on airway inflammation after antigen challenge in the four groups of mice (shown in Fig. 1b). Airway inflammation levels were analyzed by determining eosinophil and neutrophil infiltration in BALF. On day 22, BALF was collected from the mice in the four groups. The number of eosinophils in the tolerized mice was significantly lower than in the other groups. The number of neutrophils in the stress/tolerized group was significantly higher than that in the other groups. Next, we evaluated the effect of psychological stress on OVA-specific IgE and IgG1 synthesis after antigen challenge in the four groups of mice (Fig. 1c, d). As expected, tolerized mice showed significantly lower OVA-specific IgE and IgG1 levels in serum than asthmatic mice. In contrast, The IgE and IgG1 levels were significantly higher in stressed/tolerized mice than in tolerized mice. However, no significant differences were observed in airway responses or IgE and levels between asthmatic and stressed/asthmatic mice. This likely reflects disruption of immune tolerance by stress exposure. Having shown that neutrophil infiltration was detected in stress/tolerized mice, we investigated whether neutrophil infiltration was dependent on IL-17A expression in the lungs after stress exposure. IL-17A levels were measured in the lungs of the mice in the four groups on day 20 (shown in Fig. 2). The mice in the stress/tolerized group showed increased IL-17A levels. This result suggests the increased activation of Th17 cells in the lungs of stress/tolerized mice.

Based on the observed eosinophil and neutrophil airway infiltration in stress/tolerized mice, we analyzed the T-cell subsets in the BLN lymphocytes on day 20 using flow cytometry (the gating strategy shown in Fig. 3a). As expected, we observed that the Treg cell population in tolerized mice was significantly higher than that in the other groups (shown in Fig. 3b, c). These results suggest that the increased Treg cell level characterizes the occurrence of tolerance. Notably, Treg cell levels in tolerized mice were suppressed by stress exposure. Th17 cells were predominant in the stressed/tolerized group compared to other groups. These results suggest that infiltration of neutrophils in the airways of stress/tolerized mice is induced by inhibiting Treg cells and activating Th17 cells. Based on these results, we considered the possibility that Th17 cells are involved in Treg cell depletion in stress/tolerized mice. To examine this possibility, we used IL-17A knockout (KO) mice. We investigated airway inflammation induced by OVA in stress/tolerized mice. The number of neutrophils was significantly decreased in the IL-17A KO mice. However, the number of eosinophils was not significantly different between the WT and IL-17AKO mice (shown in online suppl. Fig. S1; for all online suppl. material, see www.karger.com/doi/10.1159/000529108). These results suggest that the IL-17A is not directly involved in the reduction of Treg cell development, and conversely, the Treg depletion induced IL-17A production due to Th17 development and neutrophilic airway inflammation. Therefore, we considered that additional upstream immunological changes might have contributed to these observations.

Based on our results, we hypothesized that exposure to stress triggers immune changes. Therefore, we comprehensively investigated the gene expression following stress exposure. The results of comprehensive microarray analysis of BLN of mice from the stressed/tolerized and tolerized groups with immune tolerance on day 6 and day 3 are shown in a heatmap (shown in Fig. 4a). Genes satisfying log ratio >0.9 and log ratio < −0.9 were extracted from the microarray data in the heatmap. From there, we extracted genes that affect eosinophils, neutrophil inflammation, and immune tolerance, and may be involved in this experimental system. Other genes were also altered, but this time, we focused on eosinophilic and neutrophil inflammation. Interestingly, we found that Rorc gene expression in BLN increased, whereas Icos expression decreased during stress exposure. To confirm the reproducibility of these results, we analyzed the difference in gene expression using qPCR, and obtained results similar to those of the microarray (shown in Fig. 4b). The results obtained from the comprehensive microarray and qPCR analysis demonstrated that transcription factors and co-stimulatory receptor were associated with the pathophysiology of Th2- and Th17-dominant asthma.

Rorc gene expression is elevated upon stimulation with the pro-inflammatory cytokine IL-1β [19, 20]. Furthermore, IL-1β is a key cytokine that contributes to Th17 cell differentiation [21]. Because stress exposure increased Rorc gene expression in the BLN, we analyzed whether IL-1β is also involved in airway inflammation in stress/tolerized mice. We first compared IL-1β secretion in the lungs immediately after final stress exposure (shown in Fig. 5a). OVA exposure in the tolerized mice group had no effects on the expression of IL-1β, whereas stress exposure increased IL-1β expression in the tolerized mice but not in the non-tolerized mice (shown in Fig. 5b). This result suggests that IL-1β expression need the stress and antigen exposures. Therefore, an increase of IL-1β in stressed/tolerized mice may contribute to Th17 differentiation and stress-induced immune tolerance breakdown.

In the lung environment, IL-1 receptor (IL-1R) signaling is required for producing IL-17A by mouse CD4⁺ T cells [21‒23] and facilitates IL-1β in maintaining the Th17 response. To investigate the relationship between IL-1β expression in stress/tolerized mice and the development of neutrophilic and eosinophilic airway inflammation, we administered an IL-1R antagonist (IL-1RA) before each stress exposure. Mice were divided into four groups: asthma, tolerance, stressed/tolerized, and stressed/tolerized/IL-1RA mice. Therefore, we evaluated the effects of IL-1RA on airway inflammation. Interestingly, we observed that neutrophilic and eosinophilic airway inflammation in stress/tolerized mice was abolished upon IL-1RA administration (shown in Fig. 6b). The CD4 population change in the stressed/tolerized group was abolished by IL-1RA administration and returned to that of the tolerized group (shown in Fig. 6c). This result strongly indicates that psychological stress-induced expression of IL-1β suppressed the induction of immune tolerance, resulting in neutrophilic and eosinophilic airway inflammation.

In this study, we discovered that persistent psychological stress hinders immunological tolerance induction and triggers the development of Th17- and Th2-dominant asthma. Our previous studies have shown that stress impairs the establishment of immune tolerance and causes asthma in mouse models. The expression of eosinophils and neutrophils in each group revealed increased eosinophils in the asthma mice group, indicating a Th2-dominant airway inflammation. Similarly, stressed mice also developed Th2-dominant asthma. In tolerized mice, eosinophils were significantly reduced compared to that in asthmatic mice. This indicates that immune tolerance suppresses the onset of Th2-dominant asthma. However, both eosinophils and neutrophils increased in stressed/tolerized mice compared to tolerized mice. This suggests the suppression of immune tolerance and promotion of the onset of Th2-dominant asthma and neutrophilic asthma, in which Th17 cells are increased. There was no difference in both eosinophil and neutrophil levels between asthmatic and stressed mice. This suggests that stress does not affect Th2-dominant asthma. However, when comparing tolerized and stressed/tolerized mice, eosinophils and neutrophils were predominantly increased in stressed/tolerized mice. This indicates the development of Th2-dominant asthma in addition to Th17-dominant asthma, which further implies that stress does not induce immune tolerance but causes the development of neutrophilic asthma.

To confirm these results, T-cell subsets in BLN were confirmed in post-sensitized mice. The confirmed subsets were Th2 (IL-4), Treg (FOXP3), and Th17 (IL-17A) cells. The results showed that in tolerized mice, Treg cells were the highest, Th2 cells were significantly low, and Th17 cells were similar compared to the non-tolerized mice. Therefore, eosinophil inflammatory reaction is suppressed by immune tolerance. Stress/tolerized mice exhibited decreased Treg cell and increased Th2 and Th17 cell count. This indicates that stress suppresses immune tolerance and exacerbates inflammatory responses in neutrophils and eosinophils.

These results were obtained by comprehensively performing gene analysis using a microarray (shown in Fig. 4). Based on the results of the analysis, the following can be considered: ICOS is involved in the production, proliferation, and survival of Treg cells [24]. Stress exposure significantly decreased Icos expression. These results indicate that stress suppresses Treg cell differentiation, resulting in an environment where immune tolerance is less likely to occur. The mechanism underlying Th17 lymphocyte differentiation involves the binding of IL-6 to signaling and transcriptional activator 3 (STAT3), which directly binds to the chromatin of Th17 cells and activates RORγt [25, 26]. The expression of Rorc gene significantly increases, indicating an increase in the Th17 cell level. Since Th17 cells characterize a neutrophilic inflammatory response, stress is considered to cause neutrophilic asthma [26] by eliciting a neutrophilic inflammatory response. The fact that Ccl24 and Rorc expression is increased by stress suggests neutrophilic and eosinophilic inflammatory reactions.

IL-1β is involved in early Th17 cell differentiation [27]. As described above, the differentiation of dendritic cells into Th17 cells makes it difficult for immune tolerance to occur and creates an environment where eosinophil and neutrophil inflammation are likely to occur. In this study, IL-1β was produced only when stress and OVA exposure were combined, but not in cases of stress exposure alone or OVA exposure alone (Fig. 5). The results suggest that the increased IL-1β expression in lung may be caused by a synergistic effect between brain cell activity in response to stress exposure and peripheral cell activity in response to antigen exposure, as psychological stress has been reported to activate microglia in the brain and increase IL-1β production [28]. In addition, antigen stimulation has also been reported to induce IL-1β production from lung tissue macrophages and airway epithelial cells [29]. Therefore, we investigated the effects of stress by administering IL-1RA. Using IL-1β as an activator increases the activity of the transcription factor RORγt [26]. Therefore, the action of IL-1 could be canceled by administering IL-1RA, which has an antagonistic action. We found that the number of inflammatory cells and the development of Th17 decreased in stressed/tolerized/IL-1RA mice. This suggests that stress-induced action was inhibited by IL-1RA, which suppresses Th17 cell differentiation.

This current study found that stress before sensitization increases cytokines involved in eosinophil and neutrophil inflammation, thereby worsening asthma symptoms. Furthermore, the administration of IL-1RA reduces inflammatory substances that exacerbate eosinophilic and neutrophilic asthma symptoms. Currently, research is being conducted in animal models; however, further clinical trials involving humans are warranted to determine the precise effects of these cytokines in the pathophysiology of human asthma phenotypes.

All experimental procedures involving animals were approved by the Committee of Animal Experiments at Tohoku Medical and Pharmaceutical University (approval numbers: 16002-cn, 17004-cn, A2040, A2207). All experiments were performed in compliance with relevant institutional guidelines.

The authors declare that they have no conflict of interest.

This study was supported by a Japan Society for the Promotion of Science (JSPS) Grant-in-Aid for Young Scientists (B) (No. 17K16212). The funders played no role in the study design, collection, analysis, interpretation of the data, or preparation of the manuscript.

S.S., T.K., and T.T. planned the experiments, wrote the manuscript, and performed the majority of the experiments. S.S. and E.I. conducted the qPCR and flow cytometric analysis. J.S. conducted the ELISA assay. K.T. and T.M. induced asthma models and conducted tissue processing and BAL. Y.M., F.I., and M.T. discussed the experiments with T.K.

All data generated or analyzed during this study are included in this article and its supplementary material files. Further inquiries may be directed to the corresponding author.

Edited by: H.-U. Simon, Bern.