Contribution of Pseudomonas aeruginosa − mediated club cell necroptosis to the bias of type 17 inflammation and steroid insensitivity in asthma

Graphical abstract

The role of opportunistic bacteria infection in the progression of asthma that PA infection promotes cellular CYP450 activation to mediate club cell necroptosis for DAMPs release and subsequent type 17 immune response and steroid resistance. In the rest of club cells, E-cadherin / β-catenin signaling was activated to regulate goblet metaplasia and mucus hypersecretion. Both mechanisms contribute to asthma exacerbation. Necroptosis inhibitor Nec-1 and CCSP administration could improve the pathology of asthma.

Highlights

• •Opportunistic PA infection is the risk factors for asthma exacerbation.
• •PA infection promotes club cell necroptosis, resulting in type 17 immune responses and steroid insensitivity.
• •Rest of club cells involves in mucus hypersecretion through E-cadherin / β-catenin signaling pathway.
• •Necroptosis inhibition and CCSP supplementation can alleviate the pathology and restore steroid sensitivity in asthma.

Abstract

Introduction

Opportunistic pathogen infection is one of the important inducements for asthma exacerbation. Pseudomonas aeruginosa (PA) is a kind of dominant pathogenic bacteria in the respiratory tract that is associated with severe asthma, but the underlying mechanisms still remains unclear.

Objectives

To examine the role of PA infection in the bias of the inflammatory endotype in asthma and its effect on the sensitivity to steroid therapy.

Methods

An adjusted HDM (House Dust Mite) −induced asthma model with PA inoculation in the airway was utilized to mimic the process of opportunistic PA infection in asthma, focusing on the interaction between bacteria and epithelium. Dexamethasone administration in vivo was used to test the sensitivity to steroid therapy.

Results

It was uncovered that PA could promote the loss of club cells in the necroptosis pattern through cellular CYP450 activation, leading to an imbalance of inflammatory response and steroid insensitivity. Club cell loss results in the activation of cellular E-cadherin/β-catenin axis in the rest of club cells for goblet metaplasia and mucus hypersecretion, as well as epithelial damage and GR downregulation for steroid resistance. For clinical applications, the necroptosis inhibitor Nec-1 can effectively relieve the pathological symptoms of asthma in vivo. Meanwhile, CCSP administration in the airway can regulate the pulmonary inflammation and restore the steroid sensitivity in asthma.

Conclusion

These experiments provide a novel mechanism of concurrent PA infection in asthma through club cell necroptosis and the pathological consequences. Nec-1 treatment and CCSP supplementation may be possible therapeutic strategies for asthma treatment.

Introduction

Asthma is a chronic respiratory disease characterized by recurrent attacks of breathlessness, dyspnea, stuff chest cough, and chronic airway inflammation. Its prevalence continues to increase globally [1]. According to Global Initiative for Asthma (GINA), the long-term therapeutic goal for asthma is stated as “overall control”, that is, relieving and controlling the current symptoms while reducing the risk of severe exacerbations [2]. Currently, the GINA-recommended medicines mainly include inhalation of bronchial extensions (to alleviate hyperthyroid symptoms) and steroid medications (to reduce airway inflammation). Although these drugs have some effects in controlling asthmatic symptoms, long-term usage of steroids may cause the imbalance of pulmonary inflammation (type 2/type 17 inflammation), to increase the risk of severe asthma [3]. Meanwhile, about 10 % of asthma patents are steroid-resistant and defined as severe asthma. Up to 50 % of severe asthma patients present as neutrophilic and type 17 dominant inflammation, all of which are major challenges for clinic treatment [4], [5]. An in-depth investigation of the molecular mechanism underlying asthma is imperative, to explore the key regulatory targets for its pathogenic processes and to develop more specific therapies beyond basic symptomatic relief, as beneficial complements to the current clinical treatments for asthma.

The occurrence of asthma is considered to be related to an imbalance between protective and deleterious environmental factors, including exposure to dominant opportunistic pathogens from adversely altered symbiotic microorganisms, which together cause damage of respiratory epithelial cells and impaired airway barrier integrity [6]. Pulmonary inflammation could reduce the ability of lung resistance to bacterial infections, so that opportunistic microorganisms such as Pseudomonas aeruginosa (PA) in the airway become dominant pathogenic bacteria for adverse outcomes [7]. PA is one of the most important pathogenic bacteria in the respiratory tract [8], [9], that is associated with severe asthma based on airway metagenome analysis [10]. PA infection may be an important risk factor for promoting the development of asthma/severe asthma, but its pathogenesis is unclear.

It has been reported that the airway epithelial defects in asthma could increase the sensitivity to environmental stimuli, ultimately leading to chronic inflammation in the airway [11]. Club cell is a kind of non-ciliated epithelial cell, which mainly exists in bronchioles. It is an important secretory cell to synthesize and secrete special CCSP protein (Club cell Secretory Protein, also known as CC16 or SCGB1A1) [12], [13]. CCSP is one of the most abundant secretions in normal airway, to participate in airway immune regulation. Furthermore, club cell is the progenitor cell in the trachea to involve in epithelial wound repair and remodeling.

In our work, we utilized PA as the representative of airway opportunistic microorganisms and mainly focused on their interaction with epithelial cells, evaluating their impact on inflammatory endotypes and steroid sensitivity, to explore the pathological mechanisms of opportunistic bacteria in asthma. We further investigated the effect of necroptosis inhibitor and CCSP supplementation in vivo, to provide alternative therapeutic strategies for asthma treatment.

Material and methods

Animal experiments

For HDM-induced asthma model with PA or ANF (Airway Normal Flora) inoculation, 6–8 weeks old male C57BL/6 mice were intraperitoneally sensitized on day 0 and day 7 with a mixture of 25 μg HDM and 2 mg aluminum hydroxide melted in a final 250 μL solution in PBS. On day 14, 16 and day 20, 22, mice were intratracheally challenged with HDM (10 μg). On day 18, mice were inoculated intratracheally with PA or ANF (1 × 10⁶ CFU). On day 23, all mice were euthanatized for analysis.

Clinical patient samples

Patients with asthma diagnosis based on the Global Initiative for Asthma (GINA guidelines 2021) were recruited by Respiratory Medicine Department of the First Affiliated Hospital of Guangzhou Medical University.

Ethics statement

All experiments involving animals were conducted according to the ethical policies and procedures approved by the Animal Ethical Committee of Second Affiliated Hospital of Guangzhou Medical University (Reference No. A2021-031).

For clinical patient samples, the project was approved by the ethics review committees of the First Affiliated and the Second Affiliated Hospital of Guangzhou Medical University (Ethical Review of Medical Research 2021 No. 32), and obtained the consent of all participants (Reference No. KY-EC2024-008–01).

Statistical analysis

All Data were presented as mean ± SEM. Statistical analysis was performed with GraphPad Prism 8 software. Multiple comparisons were conducted using one-way ANOVA with post-hoc test. Correlations were performed using Linear regression and Pearson correlation coefficient test. p < 0.05 is considered statistically significant in all experiments.

Results

PA infection promotes the bias of type 17 inflammation in asthma

To replicate the pathophysiology of opportunistic PA infection in asthmatic patients, an adjusted HDM-induced asthma model with PA inoculation was utilized. As ideograph presented (Fig. 1A), before PA inoculation in the airway (airway normal flora (ANF) was used as control strain [14]), sensitized mice were challenged with HDM twice to induce the asthmatic immune microenvironment for PA infection. After bacterial inoculation, mice were challenged with HDM twice to simulate acute asthmatic attacks. Histopathological test showed a certain deterioration of pulmonary inflammation in the PA inoculated group compared with ANF treatment and simple HDM-induced model mice (Fig. 1B and C). To distinguish different inflammatory endotypes, flow cytometry analysis indicated a greater ratio of neutrophils but a lower number of eosinophils in the pulmonary single-cell suspension from PA inoculated group (Fig. 1D). Additionally, multiple cytokines analysis in bronchoalveolar lavage fluid (BALF) revealed higher level of IL-17A but less type 2 cytokines (Fig. 1E) in the PA inoculated group. Flow cytometry analysis showed that the PA inoculated group presented as fewer T_H2 and ILC2 and more T_H17, but no difference in ILC3 compared with other two groups (Fig. 1F to 1H). All of these results indicated a dominate type 17 inflammatory response in the asthma associated with PA infection.

Fig. 1.

PA infection promotes the bias of type 17 inflammation in asthma. C57BL/6 mice were performed for HDM induced asthmatic models with PA or ANF inoculation in the respiratory tract (Detail seen “Full supplementary material and methods”). (A) Schematic diagram of mice model. (B) H&E staining of the representative lung sections. (C) At least five random fields of each slide were analyzed for inflammatory scores, and each scatter plot represents the average value of each mouse (n = 5). (D) Frequencies of neutrophils and eosinophils in CD45⁺ cells from pulmonary single-cell suspension through flow cytometry (n = 5). (E) Normalized cytokine levels in BALFs through multiplexed cytokines (Luminex) assay (n = 3). (F and G) Frequencies of T_H2 (IL-4, IL-5 or IL-13) (F) and T_H17 (IL-17A) (G) cells in CD3/CD4 double positive cellular groups from pulmonary single-cell suspension through flow cytometry (n = 5). (H) Frequencies of ILC2 and ILC3 in ILC cellular group through flow cytometry (n = 5). Data in (C, D, and F to H) are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, as analyzed by one-way ANOVA with post-hoc test.

In order to confirm the deviation of inflammatory endotype, the PA inoculation scheme was further performed in the conventional eosinophilic asthma model induced by OVA/Alum hydroxide (Fig. S1A). Similarly, type 2 inflammation was converted to type 17 phenotype including IL-17A upregulation, neutrophils recruitment and induction of neutrophil related chemokines in the PA inoculated group (Fig. S1B to S1G). To exclude the direct impact of acute response to PA infection, we examined the airway inflammatory response to bacteria-inoculation before HDM challenge. Our results showed that after two days of inoculation, inflammatory cytokines returned to their basal levels (Fig. S1H). Taken together, opportunistic PA infection could induce the predisposition of type 17 inflammation in asthma.

PA infection impairs the effectiveness of steroid treatment for asthma

We further evaluated the impact of PA infection on the sensitivity to steroid treatment. In the HDM-induced asthma model with PA or ANF inoculation, dexamethasone (DXMS) was administered to model mice, to mimic the clinical steroid therapy for asthma (Fig. S2A). Histopathological test showed that DXMS significantly reduced the pulmonary inflammation in the ANF group, whereas no improvement was detected in the PA inoculated group (Fig. 2A and B). Cytological and cytokine examinations revealed that the ratio of eosinophils and type 2 cytokines (IL-4 and IL-13) in lungs were suppressed, but no effect on neutrophils and IL-17A levels (Fig. 2C to 2E, S2B). Meanwhile, DXMS treatment dramatically improved the airway hyperresponsiveness (AHR) in the ANF group, but no effect in PA inoculated group (Fig. 2F). All of these results suggested that the sensitivity to steroid therapy for asthma was reduced after PA infection.

Fig. 2.

PA infection impairs the effect of steroid treatment for asthma. (A to F) HDM-induced model mice were inoculated with PA or ANF same as (Fig. 1). After bacterial inoculation, mice were i.p injected with DXMS (2 mg/kg) 1 h before each excitation (Detail seen “Full supplementary material and methods” and Fig. S2A). H&E staining of the representative lung sections and inflammatory score analysis (A); Flow cytometry analyzed the number of total cells in BALFs (B), and the frequencies of neutrophils/eosinophils in CD45⁺ cells from pulmonary single-cell suspension (C); ELISA analysis of IL-4, IL-13 (D) and IL-17A (E) in lung lysates; AHR to methacholine (F) (n = 5). (G and H) HDM-induced asthmatic model mice with PA or ANF inoculation same as (Fig. 1). Real-time QPCR analysis of Nr3c1, Hdac2 and Hsd11b1 mRNA levels in lungs (n = 5) (G); Immunoblotting analysis of GRα, HDAC2 and HSD11B1 in lungs with β-actin as internal control, and the grayscales of relative protein levels through ImageJ (n = 3) (H). Data are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, ns. p ≥ 0.05, as analyzed by one-way ANOVA (F) and with post-hoc test (A to E, G and H).

Because glucocorticoid receptor (GR) is positively correlated with steroid sensitivity [15], we detected the expression of GR and other essential genes in lungs. As data presented, expression levels of Nr3c1 (Nuclear receptor subfamily 3 group C member 1, encoding gene of GR), Hdac2 (Histone deacetylase 2, functional recruited by GR to suppress NF-ĸB transcriptome activator complex for anti-inflammatory effects [16]) and Hsd11b1 (11 β-hydroxysteroid dehydrogenase type 1, the enzyme to convert inactive cortisone to its receptor-active form cortisol for intracellular glucocorticoid action [17]) were suppressed after PA inoculation (Fig. 2G and H). As reference reported [18], one of the crucial targets for glucocorticoids in asthma are airway epithelial cells. Our tissue immunofluorescence staining results showed that the expression of GRα in airway epithelium was obviously suppressed in PA inoculated group compared with ANF group and normal mice (Fig. S2C). Taken together, opportunistic PA infection in asthma could influence the expression of GR-related genes in the lung to impair the responsiveness of steroid therapy.

PA infection exacerbates the loss of club cells in asthmatic epithelium

Since epithelium is the direct interacting sites in the airway for PA infection, we further investigated the essential target epithelial cells. Epithelial functional examination showed that PA inoculation in asthmatic models enhanced the epithelial permeability in vivo (Fig. 3A), indicating more severe damage to the epithelium caused by PA infection. Immunoblotting analysis in total lung lysis showed the downregulation of adherent junction protein E-cadherin was more obvious in the PA inoculated group than ANF group or simple asthmatic models (Fig. 3B and 3C). For the other classic tight junction proteins including Occludin, Claudin and ZO-1, no difference was detected between the PA and ANF inoculated groups (Fig. S3A), indicating E-cadherin to be the target transmembrane molecule for PA to disrupt cell–cell adhesion. Meanwhile, the PA inoculated group showed a decreased level of the club cell marker (CCSP), whereas no change was detected in FOXJ1, the ciliated cell marker (Fig. 3B and C). Same phenomenon was also presented in OVA/Alum hydroxide induced model mice (Fig. S3B and S3C). Immunofluorescence staining and SEM analysis in pulmonary slides both showed less ratio of club cells in the epithelium from PA inoculated group (Fig. 3D to F). Furthermore, ELISA analysis revealed that the expression of CCSP in the lungs was upregulated in a simple asthma model or with ANF inoculation. However, in the PA-inoculated group, CCSP expression was suppressed to levels below those observed in normal mice (Fig. 3G). It suggested that the loss of club cells and CCSP deficiency was caused by PA infection. All of these results indicated that PA infection in the airway is primarily responsible for the loss of club cells, and subsequently affects their function in the epithelium.

Fig. 3.

PA infection exacerbates the loss of club cells in asthmatic epithelium. HDM-induced asthmatic model mice were inoculated with PA or ANF same as (Fig. 1). (A) Epithelial barrier assay through Evans blue i.t injection and bye intensity detection (Detail seen “Full supplementary material and methods”) (n = 5). (B) Immunoblotting analysis of SCBG1A1, FOXJ1 and E-cadherin levels in lungs with β-actin as internal control. (C) The grayscales of relative protein levels through ImageJ as indicated (n = 3). (D and E) Immunofluorescence staining of lung slides with anti-CCSP antibody (upper panel) and anti-CCSP plus anti-Ac-α-Tubulin antibodies in-combination (lower panel) (D); The percentages of club cells (CCSP⁺) and ciliated cells (Ac-α-Tubulin⁺) in the airway epithelium were calculated based on at least five random fields per slide in (D), and presented as the average value of each mouse (n = 5) (E). (F) Representative SEM images of lung sections as indicated. (G) ELISA analysis of CCSP level in lung lysates (n = 5). Data in (A, C E and G) are presented as mean ± SEM. * p < 0.05, ** p < 0.01, ns. p ≥ 0.05, as analyzed by one-way ANOVA with post-hoc test. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

PA infection aggravates goblet cell metaplasia and mucus hypersecretion in asthma

Mucus hypersecretion is one of the important pathological phenotypes in asthma. In our results, PAS staining in the pulmonary slides revealed more PAS positive cells in the epithelium of PA inoculated group than ANF treatment or simple asthma models (Fig. 4A and 4B), indicating more goblet metaplasia induced by PA infection. Meanwhile, immunoblotting demonstrated that the MUC5AC (a large gel-forming glycoprotein in the respiratory tract that protects against infection, but its overproduction contributes to pathological hypersecretion of airway mucus and airway hyperreactivity) synthesis was increased after PA infection (Fig. 4C, the first panel). Because of club cell as the progenitor cell in epithelium, we examined the contribution of club cell to goblet cell metaplasia. Immunofluorescence staining showed that more club cells (CCSP positive cells) were also MUC5AC positive stained in the PA inoculated group (Fig. 4D and S3D), indicating the differentiation of club cells to goblet cells by PA induction. For the molecular mechanism of goblet cell metaplasia and mucus hypersecretion, the E-cadherin protein level in the lung was detected to be much less after PA infection (The second panel in Fig. 4C, 4E and S3E). It was also demonstrated in vitro that E-cadherin was downregulated while MUC5AC elevated in C22 cells after PA induction (Fig. 4F and G). According to literature report [19], E-cadherin/β-catenin axis involves in mucin expression. It was further confirmed in our results that PA treatment in vitro induced the nuclear translocation of β-catenin (Fig. 4H). Furthermore, β-catenin depletion could disrupt the PA-induced MUC5AC expression in vitro (Fig. 4I and S3F), all of which suggesting that PA infection aggravates goblet cell metaplasia and mucus hypersecretion via the E-cadherin/β-catenin molecular axis.

Fig. 4.

PA infection aggravates goblet metaplasia and mucus hypersecretion in asthma. (A to E) HDM-induced asthmatic model mice were inoculated with PA or ANF same as (Fig. 1). PAS staining of the representative lung sections (A) and the percentage of PAS positive cells in the airway epithelium based on at least five random fields per slide, and presented as the average value of each mouse (n = 5) (B); Immunoblotting analysis of MUC5AC and E-cadherin levels in lungs with GAPDH as internal control (C); Immunofluorescence staining of lung slides with anti-CCSP and anti-MUC5AC antibodies, and DAPI staining for nuclei (D); The number of CCSP positive, MUC5AC positive or double positive cells was calculated based on at least five random fields per slide, and the average value was presented as Venn diagram (D); Immunofluorescence staining of lung slides with anti-CCSP and anti-E-cadherin antibodies, and DAPI staining for nuclei (E). (F to H) C22 cells were infected with PA (5 moi) for different time as indicated. Immunofluorescence staining of cells with anti-E-cadherin antibody, and DAPI staining for nuclei (F); Immunoblotting analysis of MUC5AC and E-cadherin levels in total cellular lysis with GAPDH as internal control (G); Cellular lysis were fractionated into cytosol and nucleus fractions. Immunoblotting analysis of β-catenin in different factions with GAPDH and Histone H3 as cytosol and nucleus markers (H). (I) C22 cells were transfected with scramble siRNA (si Ctrl) or si catenin beta, followed by PA infection (5 moi) for 6 hs as indicated. Immunoblotting analysis of MUC5AC and β-catenin with GAPDH as internal control. Data in (B) is presented as mean ± SEM. * p < 0.05, ** p < 0.01, as analyzed by one-way ANOVA with post-hoc test.

PA infection induces necroptosis of club cells

Biochemical examination demonstrated that BALFs from the PA inoculated group had greater levels of DAMPs (LDH, HMGB-1) (Fig. 5A), and the necroptosis signaling was detected to be upregulated in lungs after PA infection (Fig. 5B). Data from clinical asthmatic patients showed HMGB-1 levels in BALFs were positively correlated with serum IL-17A (Fig. 5C, Demographic information and results of clinical examination were summarized in Table S1). These results suggest that inflammatory tissue damage and inflammatory cell death in the airway epithelium after PA infection can further promote the IL-17A inflammatory response in asthma. Immunofluorescence staining showed that in the large airway epithelium from PA inoculated group, approximately 39 % of club cells were p-MLKL positive stained, but only a small amount of p-MLKL⁺ cells were detected in ciliated cells; Meanwhile, in the total necroptotic cells, about 97 % of cells were CCSP positive stained (Fig. 5D and 5E), all of which indicated PA infection specifically induced club cell necroptosis but no effect on ciliated cells in the epithelium. Data in mlkl deficient mice showed PA induced less inflammatory response in mlkl^-/- mice than WT mice (Fig. 5F to H), suggesting necroptosis to be the important pathogenesis of asthma with PA infection.

Fig. 5.

PA infection induces necroptosis of club cells. (A, B, D and E) C57BL/6 mice were performed for HDM-induced asthma model with PA or ANF inoculation same as (Fig. 1). LDH and HMGB1analysis in BALFs (n = 5) (A); Immunoblotting analysis of p-MLKL and MLKL levels in lungs with β-actin as internal control (B); Immunofluorescence staining of lung slides with anti-FOXJ1, anti-CCSP and anti-p-MLKL antibodies, and DAPI staining for nuclei (D); The percentage of p-MLKL positive cells in total club cell group (CCSP⁺) or ciliated cell group (FOXJ1⁺), as well as CCSP or FOXJ1 positive cells in total necroptotic cell group (p-MLKL⁺) were calculated based on at least five random fields per slide, and the average value was presented as Venn diagram (E). (C) Concentration of HMGB1 in BALFs and IL-17A in serums from 52 asthmatic patients were detected by ELISA, and their Linear correlations were analyzed through Pearson correlation coefficient test. (F to H) Mlkl-deficient mice (Mlkl^-/-) and WT littermates were performed for HDM-induced asthma model with PA inoculation same as (Fig. 1). H&E staining of the representative lung sections (F); Number of total cells in BALFs through flow cytometry (G); ELISA analysis of IL-17A level in lungs (H), n = 5. (I) C22 cells were infected with PA (5 moi) for 12 hrs, followed with PI staining. The PI positive ration were analyzed through flow cytometry (n = 3). (J and K) C22 cells were pre-treated with 5 µM 5P1P for 1 hr, followed by PA infection (5 moi) for 12 hrs. CYP450 activity in cells were analysis through P450-Glo^TM assay (n = 5) (J); Cells were stained with PI and the percentage of PI positive cells were analyzed through flow cytometry (n = 3) (K). Data in (A, G to K) are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, as analyzed by one-way ANOVA with post-hoc test.

Regarding the molecular mechanism of PA interacting with club cells, our results revealed that PA could directly induce club cell necroptosis in dose- and time- dependent patterns in vitro (Fig. 5I and S4). Because of high levels of cytochrome P450 monooxygenases (such as CYP2F2) in club cells, they are susceptible to toxicant-induced cell death [20], [21]. PA treatment in vitro could induce the activation of CYP450, which was suppressed by 5-phenyl-1-pentyne (5P1P, the inhibitor of CYP2F2) (Fig. 5J). Furthermore, 5P1P administration could dramatically suppress PA-induced club cell death (Fig. 5K). Taken together, these results indicated that PA infection could promote necroptosis of club cells through CYP450 activation.

RIPK1 inhibitor Nec-1 alleviates inflammation and asthmatic features in vivo

Considering the potential clinical utility, we tested the effect of Necrostatin-1 (Nec-1, RIPK1 inhibitor to suppress necroptosis) on the asthmatic pathophysiology in model mice (Fig. S5A). Immunofluorescence staining showed that after Nec-1 treatment, PA-induced necroptosis of club cells was significantly restricted (Fig. S5B). Furthermore, Nec-1 treatment reversed the reduction of E-Cadherin in asthmatic lungs after PA-inoculation (Fig. S5C). Pathological examination showed that Nec-1 administration in the airway could relieve the pulmonary inflammation and mucin synthesis in the epithelium (Fig. 6A to 6C, S5D). AHR analysis showed that airway reactivity to methacholine was obviously reduced after Nec-1 treatment (Fig. 6D). Biochemical examination further confirmed that Nec-1 treatment could inhibit the necroptosis signaling in lungs, and decrease the LDH level in BALFs (Fig. 6E and F). Cytokine analysis showed that both type 2 (IL-4, IL-13) and type 17 (IL-17A), as well as other relative pro-inflammatory cytokines (TNF-α, IL-6, IL-1β, IFN-γ) and neutrophilic chemokines (CXCL-1, CXCL-2) were significantly decreased after Nec-1 administration (Fig. 6G, H and S5E). Meanwhile, Nec-1 treatment could also reverse PA − induced CCSP downregulation in lungs (Fig. 6I). All of these results indicated that pharmacological inhibition of necroptosis in vivo could directly and effectively alleviate the inflammation and pathological features of asthma with PA infection.

Fig. 6.

RIPK1 inhibitor Nec-1 alleviates inflammation and asthmatic features in vivo. HDM-induced model mice with PA inoculation same as (Fig. 1). After bacterial inoculation, mice were i.t injected with Nec-1 (4 mg/kg body weight) 1 h before each excitation (Detail seen “Full supplementary material and methods” and Fig. S5A). (A) H&E and PAS staining of the representative lung sections as indicated. (B) Inflammatory scores and percentages of PAS positive cells were calculated based on at least five random fields per slide, and presented as the average value of each mouse (n = 5). (C) Number of total cells in BALFs through flow cytometry. (D) AHR to methacholine (n = 5). (E) LDH analysis in BALFs. (F) Immunoblotting analysis of p-MLKL and MLKL levels in lungs with β-actin as internal control (n = 3). (G) Normalized cytokine levels in BALFs through multiplexed cytokines (Luminex) assay (n = 4). (H and I) ELISA analysis of IL-17A (H) and CCSP (I) in lungs (n = 5). Data in (B to E, H and I) are presented as mean ± SEM. ** p < 0.01, *** p < 0.001, as analyzed by one-way ANOVA (D) and with post-hoc test (B, C, E, H and I).

CCSP administration restores steroid sensitivity in asthma

Pearson correlation coefficients analysis of data from clinical asthmatic patients showed the serum CCSP level was negatively correlated with serum IL-17A (Fig. 7A, Table S1), indicating its role in inflammatory balance in asthma. We therefore directly supplemented CCSP protein in the airway to investigate the pathological changes in asthma (Fig. S6A). Cytological examination showed that although CCSP administration could not directly suppress the total cell amount in BALFs, it was conducive to decrease the neutrophils infiltration (Fig. 7B). Meanwhile, the DAMP level in BALFs was also reduced after CCSP treatment (Fig. 7B). Cytokine analysis revealed that CCSP treatment could restrain IL-17A and relative pro-inflammatory cytokines (IL-23 TNF-α, IL-6, IL-1β, G-SCF), as well as neutrophilic chemokines (CXCL-1, CXCL-2) (Fig. 7C and S6B). In contrast, type 2 cytokines (IL-4, IL-13) and eosinophilic chemokines (CCL-11, CCL-24) were aggravated after CCSP treatment (Fig. 7C and S6B), all of which demonstrated that CCSP could reverse PA-induced type 17 inflammatory tendency in asthma. Histopathological test and AHR analysis both revealed that while CCSP treatment was unable to directly restrain the inflammation or airway hyperresponsiveness, it significantly improved their response to steroid treatment (Fig. 7D to F).

Fig. 7.

CCSP administration restores steroid sensitivity in asthma. (A) Concentration of CCSP and IL-17A in serums from 52 asthmatic patients were detected by ELISA, and their Linear correlations were analyzed through Pearson correlation coefficient test. (B to H) HDM-induced model mice were inoculated with PA same as (Fig. 1). After bacterial inoculation, mice were i.t injected with CCSP solution (total 20 µg) 1 h before each excitation (Detail seen “Full supplementary material and methods” and Fig. S6A). Number of total cells in BALFs, percentages of eosinophils and neutrophils in CD45⁺ cells from pulmonary single-cell suspension through flow cytometry, as well as LDH analysis in BALFs (n = 5) (B); Normalized cytokine levels in BALFs through multiplexed cytokines (Luminex) assay (n = 5) (C); H&E staining of the representative lung sections (D); Inflammatory scores based on at least five random fields and presented as the average value of each mouse (n = 5) (E); AHR to methacholine (n = 5) (F); Real-time QPCR analysis of Nr3c1, Hdac2 and Hsd11b1 mRNA levels in lungs (n = 5) (G); Immunofluorescence staining of lung slides with anti-CCSP and anti-p-MLKL antibodies, and DAPI staining for nuclei (H). Data in (B, E to G) are presented as mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, ns. P ≥ 0.05, as analyzed by one-way ANOVA (F) and with post-hoc test (B, E and G).

To further clarify the underlying mechanism, real-time QPCR analysis showed that CCSP treatment could restore the expression of GR-related genes (Fig. 7G). Additionally, CCSP administration could restrain necroptosis signaling in lungs (Fig. 7H and S6C), and improve the E-cadherin level to preserve the epithelial barrier (Fig. S6D). Taken together, CCSP treatment could balance the inflammatory response, enhance the expression of GR-related genes and maintain the epithelial barrier, ultimately restore the steroid sensitivity in asthma.

Discussion

Overall, our research demonstrated the role of opportunistic bacteria infection in the progression of asthma, promoting type 17 inflammation and impairing the sensitivity to steroid therapy. It was uncovered that PA could promote the loss of club cells in a necroptosis pattern through cellular CYP450 activation, to enhance DAMPs release and the imbalance of inflammatory response. Club cell loss results in the activation of cellular E-cadherin/β-catenin axis in the rest of club cells for goblet metaplasia and mucus hypersecretion, as well as epithelial damage and GR downregulation for steroid resistance. For clinical applications, the necroptosis inhibitors Nec-1 can effectively relieve the pathological symptoms of asthma in vivo. Meanwhile, CCSP administration in the airway can regulate the pulmonary inflammation and restore the steroid sensitivity in asthma (Graphical abstract).

In this article, we used the PA-inoculation animal models to simulate opportunistic pathogen infection during asthma. It is confirmed that club cell necrosis is caused by the upregulation of intracellular CYP450 activity (Fig. 5J and K). However, in terms of pathogenic bacteria, we have no data to prove which bacterial-derived components (such as bacterial exotoxins, metabolites, etc.) induced cell death. Therefore, we cannot assume that this is the unique pathological mechanism underlying PA-induction. Additional opportunistic microorganisms, such as Klebsiella pneumoniae, as long as they are able to induce the CYP450 activity in club cells for their necroptosis, can lead to similar pathological phenotypes like PA-inoculation. Further experiments are required to validate the pathological function of other opportunistic pathogens.

In this study, we focused on the pathological mechanism of PA interacting with the epithelium, and verified club cells as the essential target for PA-induced epithelial damage. There are two aspects of pathological consequences associated with club cell loss. First of all, PA-induced club cell loss is in the form of necroptosis, a kind of pro-inflammatory death pattern for subsequent immunological stimulation. Secondly, as important precursor cells in the epithelium, club cell damage not only leads to destruction of the epithelial barrier, but also promotes goblet metaplasia and mucus hypersecretion. There is also an intrinsic correlation between these two pathological effects. PA infection exacerbates the loss of club cells in epithelium, resulting in the disruption of the epithelial barrier. Our data confirmed the downregulation of E-cadherin in the epithelium and the disruption of cell–cell adhesion. A possible reason for the mechanism of E-Cadherin reduction is that PA-induced necroptosis of club cells leading to the breakdown of cell–cell interactions in the rest of the epithelium, and the membrane-localized E-Cadherin can be internalized for degradation [22]. On the other hand, disruption of cell–cell adhesion and E-cadherin downregulation can promote cellular E-cadherin/β-catenin axis in the rest of club cells. It has been reported that β-catenin signaling involves in cell fate regulation [23]. Our results suggested that the rest of club cells involve in goblet metaplasia and mucus hypersecretion, which is mediated by epithelial disruption and cellular β-catenin signaling. For other epithelial cell types, such as basal cells, tuft cells and pulmonary neuroendocrine cells, their pathological role in asthma regulation still need further investigation.

As a kind of heterogeneous disease, asthma is regulated by different mechanisms at cellular and molecular levels, to manifest as dynamic drug sensitivity [24]. In many asthma patients, the chronic airway inflammation is characterized as type 2 inflammation, but there are about 10 % of asthma patients exhibit high ratio of neutrophils and T_H17 mediated IL-17 accumulation (Type 17 inflammation), which are usually related to severe disease phenotypes and steroid resistance [25], [26]. IL-17 cytokine can regulate neutrophil accumulation by promoting epithelium-derived chemokines such as CXCL1 and CXCL2, to recruit neutrophils. Recent evidence reveals that severe neutrophil asthma is positive correlated with significant IL-17 upregulation, highlighting the role of IL-17 in the development of neutrophilic or severe asthma [27]. Therefore, further studies are needed to completely understand the mechanism of IL-17 inflammatory bias in asthma and to develop more effective strategies for asthma/severe asthma treatment.

Our data confirmed that opportunistic PA infection in the airway can promote the bias of type 17 inflammation including IL-17, IL-1β and IFN-γ expression (Fig. 1E). It is reported that septum IL-1β levels from severe asthma patients are much higher than that from mild asthma, that is associated with neutrophil recruitment and the reduction of lung functions [28]. Our previous publication has confirmed the cross-talking between necroptosis and inflammasome mediated IL-1β release in epithelial cells [14], indicating the role of PA-induced club cell necroptosis in promoting type 17 inflammation in asthma. According to the latest classification standards and research reports, in addition to type 2 inflammation, non-type 2 cytokines, including IL-17A and IFN-γ, are also implicated in asthma pathogenesis. Meanwhile, there is direct association between IFN-γ, a kind of type 1 inflammation, with T_H17 responses. Due to the considerable plasticity of T_H17, it is readily induced for IFN-γ production [29]. It has been reported that some steroid-resistant severe asthma patients present as high levels of IL-17A and IFN-γ, but there is a lack of direct evidence for IFN-γ involving in steroid insensitivity [30]. Considering the direct evidence reports on IL-17A and steroid-resistance in asthma [31], we focused on IL-17A as the representative indicator in steroid-resistant asthma. Further studies are needed to clarify the regulatory mechanisms of non-type 2 cytokines, including IL-17A, IL-1β and IFN-γ, in steroid-resistant asthma.

In terms of steroid resistance, GR-related genes were detected to be downregulated (Fig. 2G and S2C), which could partially explain the causes of steroid insensitivity. The mechanism for GR downregulation is still unclear. Additional studies are required to figure out whether it is related to PA-mediated club cell loss, or if some other pathways also contribute to GR downregulation and steroid resistance.

CCSP has been demonstrated to be a negative inflammatory regulator in acute and chronic pulmonary diseases, but its target cells are not clear. A recent study reported that the integrin VLA-4 (Very Late antigen-4) can act as the receptor of CCSP, implying that CCSP can directly antagonize neutrophil adhesion to lung endothelium and thus inhibit neutrophil recruitment [32]. CCSP can also interact with the N-formyl-Met-Leu-Phe receptor on neutrophils and eosinophils to suppress granulocyte infiltration in the airway during lung inflammation [33]. In this study, we confirmed that CCSP administration in the airway could effectively protect the epithelial barrier, enhance the expression of GR-related genes to restore the steroid sensitivity (Fig. 7), but the full spectrum of target cells that respond to CCSP need further investigation.

As a kind of opportunistic pathogens in the respiratory tract, Pseudomonas aeruginosa has complex drug resistance mechanisms that it is listed in the most formidable antibiotic-resistant pathogens. The purpose of our mice model is to mimics the clinical asthma patients with PA colonization (including antibiotic-resistant situation). We mainly focus on PA-mediated epithelial damage and its subsequent pathological readout, rather than the acute response to bacterial infection. It is important to develop novel intervention medicines instead of relying solely on antibiotic application. Taken together, our study revealed the pathological effect of opportunistic PA infection, to induce club cell necroptosis for subsequent type 17 inflammatory bias and steroid resistance in asthma. Pharmacological intervention of necroptosis or CCSP supplementation in the airway epithelium could be possible therapeutic approaches for asthma.

Conclusions

PA infection promotes club cell necroptosis, resulting in type 17 immune responses and steroid insensitivity. Rest of club cells involves in goblet cell metaplasia and mucus hypersecretion through E-cadherin/β-catenin signaling pathway. Necroptosis inhibition and CCSP supplementation in the airway effectively alleviate the pathology and restore steroid sensitivity in asthma.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: