Cadherin-26 Promotes Airway Remodeling via Stabilizing Epithelial Expression of the Epidermal Growth Factor Receptor in Asthma

Abstract

Purpose

Activation of epidermal growth factor receptor (EGFR) signaling in airway epithelial cells leads to epithelial-mesenchymal transformation (EMT) and airway remodeling in asthma. However, the mechanism regulating EGFR expression is not fully understood. We hypothesized that cadherin-26 (CDH26), a highly expressed cadherin in airway epithelium, promotes airway remodeling via regulating EGFR signaling in asthma.

Methods

The expression of CDH26 in airway brush cells of asthmatic patients was examined using quantitative polymerase chain reaction. Airway hyperresponsiveness and remodeling were assessed in the house dust mite (HDM)-sensitized and -challenged Cdh26-deficient mice. The role and mechanism of CDH26 in epidermal growth factor (EGF)-induced EMT were measured in cultured human bronchial epithelial BEAS-2B cells.

Results

The expression of CDH26 in airway brush cells was significantly correlated with the thickness of airway basement membrane in asthma patients. Cdh26 deficiency significantly reduced airway hyperresponsiveness and remodeling in HDM-sensitized and -challenged mouse model. In cultured human bronchial epithelial BEAS-2B cells, CDH26 knockdown inhibited EGF-induced EMT, whereas overexpression of CDH26 promoted EGF-induced EMT. Furthermore, CDH26 knockdown suppressed EGF-induced activation of EGFR signaling including phosphatidylinositol 3-kinase and AKT phosphorylation. By contrast, CDH26 overexpression intensified EGF-induced activation of EGFR signaling. Mechanistically, CDH26 knockdown promoted the proteasomal degradation of the EGFR.

Conclusions

CDH26 promoted airway remodeling via stabilizing the epithelial expression of EGFR and sustaining the activation of EGFR signaling in asthma. Thus, CDH26 can be a potential therapeutic target for airway remodeling in asthma.

INTRODUCTION

Asthma is a chronic inflammatory airway disease characterized by airway inflammation, mucus overproduction, airway remodeling, and hyperresponsiveness. Airway remodeling is a process including goblet cell hyperplasia, basement membrane thickening due to subepithelial collagen deposition, increased smooth muscle mass, and angiogenesis of the airways.¹,² The biological process where epithelial cells transform into cells with a mesenchymal phenotype is termed epithelial-mesenchymal transformation (EMT), which is a key driver of airway remodeling.³ Activation of epidermal growth factor receptor (EGFR) signaling plays an important role in EMT and airway remodeling.⁴

The EGFR is a glycoprotein with typical transmembrane tyrosine kinase receptors and is activated by its autophosphorylation with its ligand bindings. The receptor activation initiates downstream signaling pathways including the Ras/Raf/mitogen-activated protein kinase pathway, the phosphatidylinositol 3-kinase (PI3K)/AKT pathway, and the phospholipase C/protein kinase C pathway.⁵,⁶ Meanwhile, the E3 ubiquitin ligase is recruited to the receptor, resulting in the ubiquitination of the EGFR. The ubiquitinated EGFR is trafficked from the plasma membrane through the endosome to be degraded in lysosome, eventually leading to the termination of the downstream signaling pathway.⁷ EGFR expression in airway epithelial cells was increased in asthmatic patients, especially in patients with severe asthma.⁸ Airway epithelial EGFR expression was also higher in severe asthmatic children than in non-asthmatic or moderate asthmatic children, and was significantly correlated with basement membrane thickness, an indicator of airway remodeling.⁹ In a murine model of allergic airway disease, pretreatment with gefitinib, an EGFR tyrosine kinase inhibitor, reduces the expression of Th2 cytokines and alleviated airway eosinophilia.¹⁰ In mice exposed to house dust mite (HDM) for 6 weeks, EGFR tyrosine kinase inhibitor significantly alleviated airway remodeling.¹¹ These studies suggest that epithelial EGFR signaling drives airway remodeling in asthma. However, the mechanism by which the epithelial EGFR signaling pathway is upregulated in asthma remains unclear.

Enhanced cadherin-26 (CDH26) expression has been reported in several allergic diseases including eosinophilic gastritis, eosinophilic esophagitis, allergic rhinitis, and asthma.¹²,¹³,¹⁴,¹⁵,¹⁶,¹⁷,¹⁸,¹⁹,²⁰ Silencing Cdh26 expression in a mouse model of allergic rhinitis attenuates type 2 inflammation.¹⁶ We have recently reported that CDH26 amplifies epithelial interleukin-4 receptor (IL-4R) signaling in asthma. Cdh26 deficiency significantly reduced airway eosinophilia and mucus overproduction in a murine model of allergic airway disease.²¹ We hypothesized that CDH26 contributes to airway remodeling via regulating EGFR signaling in asthma.

In this study, we observed that elevated epithelial CDH26 expression was significantly correlated with airway basement membrane thickness in asthmatic patients. In a mouse model of chronic HDM-induced allergic airway disease, Cdh26 deficiency significantly suppressed airway remodeling and EGFR expression. Mechanistically, CDH26 inhibited EGFR degradation and intensified the activation of EGFR signaling.

MATERIALS AND METHODS

Subjects

We recruited 10 healthy control subjects and 42 asthmatic patients (Table). All subjects were Chinese and were recruited from Tongji Hospital. The inclusion criteria asthmatic patient in our study were: 1) symptoms (wheezing, breathlessness, chest tightness, and coughing) related to an asthma attack; 2) a cumulative doses of methacholine provoking a 20% fall (PD₂₀) in forced expiratory volume in the first second (FEV₁) < 2.505 mg and/or ≥ 12% increase in FEV₁ following inhalation of 200 μg of salbutamol; and 3) no history of treatment with inhaled/oral corticosteroids or leukotriene antagonists. Healthy control subjects had no respiratory symptoms, normal spirometry value, and methacholine PD₂₀ ≥ 2.505 mg. None of the subjects had ever smoked or received inhaled/oral corticosteroids or leukotriene antagonists. For each subject, we recorded demographic information and measured spirometry. We performed bronchoscopy with endobronchial brushing and biopsy as previously described.²² We used the mean of 50 measurements taken over a distance of at least 1 mm to calculate basement membrane thickness, as previously described.²³ Measurements were taken independently by two investigators, and the average values were calculated for analysis. Written informed consent was obtained from all subjects. The ethics committee of Tongji Hospital, Medical College, Huazhong University of Science and Technology approved the study (TJ-IRB-20210117), and all experimental procedures were approved by the Animal Care and Use Committee of the Tongji Hospital, Huazhong University of Science and Technology (TJ-202404002).

Table. Subject characteristics.

		Healthy controls	Asthma patients	P value
Number		10	42
Age (yr)		40.5 (30.0–47.0)	42.0 (32.0–51.0)	0.360
Sex				0.078
	Male	7	14
	Female	3	28
BMI (kg/m2)		20.94 (18.73–25.46)	22.07 (20.96–24.35)	0.350
FEV1 (%predicted)		105.70 (97.68–108.35)	85.80 (77.10–97.95)	0.016
Methacholine PD20 (mg)		2.505 (2.505–2.505)	0.035 (0.010–0.235)	< 0.0001
Basement membrane thickness (μm)		6.647 (6.104–8.475)	17.74 (14.45–21.64)	< 0.0001

Values are presented as median (range).

BMI, body mass index; FEV₁, forced expiratory volume in the first second; PD₂₀, provoking a 20% fall.

Microarray dataset acquisition

The gene expression datasets GSE63142 and GSE4302 were downloaded from the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo) using the getGEO function from the GEOquery R package (version 2.58.0) in the R software (version 4.0.4; R Foundation for Statistical Computing, Vienna, Austria). The GSE63142 dataset comprises bronchial brushing samples from control subjects (n = 27), patients with mild-moderate asthma (n = 72), and patients with severe asthma (n = 56). The GSE4302 dataset comprises bronchial brushing samples from control subjects (n = 18), asthmatic patients (n = 42), and fluticasone-treated asthmatic patients (n = 19).

Murine model of allergic airway disease

Cdh26^−/− mice and Wild Type (WT) on a C57BL/6N background were purchased from the KOMP Repository Collection (UC Davis, Davis, CA, USA) and Animal Experimental Center of Hubei Province (Wuhan, China), respectively. Exon 2 to exon 12 of Cdh26 gene was deleted in Cdh26^−/− mice. The sample size for each experimental group was determined through power analysis based on analysis of variance (ANOVA) degrees of freedom.²⁴ Specifically, a difference between the total number of animals and the total number of groups (N − k) falling within the range of 10–20 was considered statistically adequate for detecting significant effects. Cdh26^−/− and WT mice were challenged with 25 μg of HDM (Greer Laboratories, Lenoir, NC, USA) extract in 40 μL of saline intranasally 5 days/week for 5 weeks.²⁵ All mice were treated with HDM extract under isoflurane anesthesia. Seventy-two hours after the last challenge, we measured pulmonary resistance in response to a range of aerosolized methacholine concentrations using the forced oscillation technique with the FlexiVent system (SCIREQ, Montreal, Canada). Then bronchoalveolar lavage fluid (BALF) was collected. Cell counts for macrophages, eosinophils, lymphocytes, and neutrophils in BALF were performed, as previously described.²⁶ Cell counts were performed independently by two investigators, with the average of their counts used for the final analysis. Lung tissues were collected for histological analysis, quantitative polymerase chain reaction (qPCR), and immunostaining. The final group sizes were unequal due to the exclusion of some samples for the following reasons: 1) some BALF samples were of inadequate quality for analysis; 2) inaccuracies during the preparation of the quantitative real-time polymerase chain reaction (RT-PCR) amplification system resulted in missing data for several samples; and 3) insufficient imaging fields in α-smooth muscle actin (α-SMA) immunofluorescence staining precluded quantitative analysis for certain animals. All analyses were performed on the available data after these exclusions.

Cell culture and treatment

BEAS-2B cell lines were purchased from ATCC (Manassas, VA, USA). Cells were cultured in DMEM medium with 10% FBS (Biological Industries, HaEmek, Israel) and were transfected with control or CDH26 siRNA (80 nM; GeneCopoeia, Rockville, MD, USA), or empty or CDH26 cDNA expression vector (500 ng/mL) using Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA). Forty-eight hours before the end of culture, epidermal growth factor (EGF; 100 ng/mL; PeproTech, Cranbury, NJ, USA) was added to the basal medium. Finally, cells were harvested for Western blotting and immunofluorescence staining. The sequence of CDH26 siRNA is described in Supplementary Table S1.

Quantitative PCR analysis of CDH26 expression

Total RNA from bronchial epithelial brushing, mouse lungs, and BEAS-2B cells was isolated using TRIzol (Invitrogen) and reverse transcribed to cDNA using HiScript III RT SuperMix for qPCR (R323, Vazyme, Nanjing, China). The sequences of the primers for Sybr Green RT-PCR were obtained from PrimerBank. Other primer sequences were designed by Sangon Biotech (Sangon Biotech, Shanghai, China). The transcript levels of CDH26, ACTA2, COL1A1, Fn1, Vim, MUC5AC, and the EGFRs were determined by using ChamQ Universal SYBR qPCR Master Mix (Q711, Vazyme) and a CFX Connect PCR System (Bio-Rad Laboratories, Hercules, CA, USA). Fold differences were determined by the 2^-ΔΔCT method. The transcript levels of each gene are expressed relative to the median of healthy control group or the mean of the control group, and log2-transformed. Primers for qPCR were listed in the Supplementary Table S1.

Histology and immunofluorescence staining

The mouse left lungs were inflation-fixed and paraffin-embedded, and 5-μm sections were cut. Sections were used for Masson staining, Periodic acid-Schiff (PAS) staining, and immunofluorescence examination. For the Masson staining, lung sections were stained with Masson (Servicebio, Wuhan, China) for detection of fibres. For the PAS staining, lung sections were stained with PAS (Servicebio) for detecting mucus. The number of PAS-staining positive cells in airways with a luminal diameter of 170 ± 40 μm was counted in five random fields for each lung section at 200× magnification. For immunofluorescence staining, tissue sections were incubated with anti-α-SMA antibody (Proteintech, 1:100 dilution; Rosemont, IL, USA) and then incubated with IFKine red donkey anti- mouse antibody (AmyJet Scientific, 1:200, Wuhan, China). Paraffin tissue sections were deparaffinized, rehydrated, and processed with pH9.0 Tris-EDTA retrieval buffer (Servicebio) before blocking with 10% donkey serum. Paraffin tissue sections were incubated with primary antibodies at 4°C overnight and then incubated with IFKine red donkey anti- mouse antibody at 1:200 in the dark for 1 hour at room temperature. Nuclei were stained with DAPI (Servicebio). All photographs were taken by using a fluorescence microscope (Olympus Corporation, Tokyo, Japan).

Western blotting

Proteins were extracted from cells or lung tissues using RIPA buffer (Servicebio). Fifty micrograms of extracted proteins were separated using 10% SDS-PAGE, and the separated proteins were transferred onto polyvinylidene difluoride membranes (Roche, Mannheim, Germany). The membranes were first probed with indicated primary antibodies. Antibodies used in Western blotting were: CDH26 (Santa cruz, 1:1,000 dilution; Dallas, TX, USA), EGFR antibody (Abcam, 1:1,000 dilution; Cambridge, UK), total and phosphorylated AKT (Cell Signaling, 1:1,000 dilution; Danvers, MA, USA), total and phosphorylated PI3K (Affinity, 1:1,000 dilution; Cincinnati, OH, USA), and glyceraldehyde-3-phosphate dehydrogenase (GAPDH; Proteintech, 1:8,000 dilution). Then antibodies were detected using horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin G (IgG) (Aspen, 1:5,000 dilution) or goat anti-mouse IgG (Aspen, 1:5,000 dilution) or rabbit anti-goat IgG secondary antibody, followed by ECL Western blot detection reagent (Beyotime Biotech). Densitometry was assessed using ImageJ (National Institutes of Health, Bethesda, MD, USA) and calculated according to the reference bands of GAPDH.

Statistical analysis

Normally distributed data are presented as means ± standard deviation, and P values were calculated using parametric tests (Student’s t-test or one-way ANOVA, followed by Tukey’s multiple comparison test). For non-normally distributed data, we calculated medians with interquartile ranges and used non-parametric tests (Kruskal-Wallis test with Dunn intergroup comparison or Mann-Whitney test). We analyzed correlation using Spearman’s rank-order correlation. Statistical analysis was performed with GraphPad Prism version 8.0 (GraphPad Software, San Diego, CA, USA). Statistical significance was set at P < 0.05.

RESULTS

Epithelial CDH26 expression is positively correlated with airway basement membrane thickness in asthmatic patients

Airway remodeling is a key feature of asthma, and airway basement membrane thickness reflects the extent of airway remodeling. To explore the role of CDH26 in airway remodeling in asthma, we analyzed the relation between CDH26 expression and airway basement membrane thickness in asthmatic patients. CDH26 transcript levels in bronchial epithelial brushings from control subjects (n = 10) and asthmatic patients (n = 42) were determined using qPCR. Airway basement membrane thickness was measured by performing hematoxylin and eosin staining using bronchial biopsies from control subjects (n = 10) and asthmatic patients (n = 38). CDH26 transcript levels and airway basement membrane thickness were both markedly increased in asthmatic patients compared to control subjects (Fig. 1A and B). Moreover, CDH26 transcript levels were positively correlated with the airway basement membrane thickness (r = 0.6083, P < 0.0001, Fig. 1C). These results suggest that epithelial CDH26 may contribute to airway remodeling in asthma.

Fig. 1. Epithelial CDH26 expression is positively correlated with airway basement membrane thickness in asthma. (A) CDH26 transcript levels in bronchial brushings from asthma patients (n = 42) and healthy controls (n = 10) were measured by quantitative polymerase chain reaction. The transcript levels were expressed as log2-transformed and expressed relative to the median value for healthy controls (two-tailed Mann-Whitney test). (B) Statistical analysis of airway basement membrane thickness in airway biopsy specimens from asthmatic patients (n = 38) and healthy subjects (n = 10) (two-tailed Mann-Whitney test); (C) Spearman’s rank-order correlation assay between epithelial CDH26 and airway basement membrane thickness in asthmatic patients (n = 38).

CDH26, cadherin-26.

^**P < 0.01; ^****P < 0.0001.

We subsequently analyzed CDH26 expression in airway epithelial cells from patients with severe asthma and those receiving corticosteroid treatment. The GSE63142 dataset comprises transcriptomic data from bronchial brushings of 27 control subjects, 72 patients with mild-to-moderate asthma, and 56 patients with severe asthma.²⁷ Our analysis revealed that CDH26 expression was significantly elevated in both mild-to-moderate and severe asthma groups compared to controls. Interestingly, no significant difference in CDH26 transcript levels was observed between the severe and mild-to-moderate asthma groups, suggesting that CDH26 expression levels do not vary with asthma severity (Supplementary Fig. S1A). Furthermore, analysis of the GSE4302 dataset, which includes transcriptomic data from bronchial brushings of 18 control subjects and 42 asthma patients (19 of whom received one-week fluticasone propionate treatment),¹⁷ demonstrated that CDH26 expression was significantly downregulated following fluticasone treatment (Supplementary Fig. S1B). This indicates that corticosteroid therapy effectively suppresses CDH26 expression.

Cdh26 deficiency alleviates airway remodeling in a mouse model of allergic airway inflammation

To explore the role of CDH26 in airway remodeling in asthma, we intranasally stimulated WT and Cdh26^−/− mice with HDM extract for 5 days a week for 5 weeks (Fig. 2A). Masson’s trichrome staining was performed to detect subepithelial collagen deposition in mouse lungs. We found that subepithelial collagen deposition, defined as fibrosis ratio, was markedly increased in HDM-sensitized and -challenged WT mice compared to WT control mice. However, Cdh26 deficiency significantly decreased the HDM-induced subepithelial collagen deposition in mice (Fig. 2B and C). We also performed immunostaining of α-SMA to detect the smooth muscle in mouse airways. We found that HDM stimulation increased airway smooth muscle mass in WT mice, whereas Cdh26 deficiency significantly reduced mouse airway smooth muscle mass (Fig. 2B and D). Moreover, the mRNA expression of EMT markers including Col1a1, Fn1 and Vim were increased in HDM-stimulated WT mice, but were suppressed in HDM-stimulated Cdh26^−/− mice (Fig. 2E-H). Our data suggest that Cdh26 contributes to airway remodeling in the mouse model.

Fig. 2. Cdh26 deficiency attenuates airway remodeling in a mouse model of allergic airway inflammation. (A) Experimental schedule. (B) Representative images of Masson staining and α-SMA immunofluorescence staining in mouse lung sections. Scale bar, 50 μm. (C) Quantification of Masson staining of WT and Cdh26^−/− mice treated with saline or HDM. The area of collagen fibers around the airway (shown in blue) as well as the percentage was calculated using ImageJ. (D) Quantification of α-SMA immunofluorescent staining of WT and Cdh26^−/− mice treated with Saline or HDM. Fluorescent intensity assay was performed using ImageJ. (E-H) The mRNA levels of ACTA2 (D), Col1a1 (E), Fn1 (F) and Vim (G) in mouse lungs were determined by quantitative polymerase chain reaction. The transcript level was relative to the mean of control group. n = 6–9 mice per group. Data are presented as mean ± standard deviation.

α-SMA, α-smooth muscle actin; WT, Wild Type; HDM, house dust mite; DAPI, 4′,6-diamidino-2-phenylindole.

^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ^****P < 0.0001 (one-way analysis of variance followed by Tukey’s multiple comparison test).

Furthermore, HDM stimulation led to enhanced airway hyperresponsiveness to methacholine, as shown by increased airway resistance in WT mice. However, HDM-induced airway hyperresponsiveness was significantly attenuated in Cdh26^−/− mice (Fig. 3A and B). Additionally, HDM stimulation increased the number of PAS-staining-positive mucous cells and the expression of Muc5ac in WT mice, but this was significantly attenuated in Cdh26^−/− mice (Fig. 3C-E). Cdh26 deficiency also inhibited HDM-induced airway eosinophilia as shown by the number of eosinophils in mouse BALF (Fig. 3F). These findings suggest that Cdh26 also contributes to airway inflammation in a chronic model of asthma.

Fig. 3. Cdh26 deficiency suppresses airway hyperresponsiveness, mucus overproduction, and lung inflammation in the mouse model. (A, B) Pulmonary resistance in response to different concentrations of aerosolized methacholine in WT mice and Cdh26^−/− mice that were administered intranasally with HDM or saline. (C) Representative images of PAS staining in mouse lung sections. Scale bar, 50 μm. (D) The numbers of PAS-staining-positive cells were counted in four random fields for each lung section at 200× magnification. (E) The mRNA levels of Muc5ac in mouse lungs were determined by quantitative polymerase chain reaction. The transcript level was relative to the mean of control group. n = 6–9 mice per group. Data are presented as mean ± standard deviation. (F) Cell counts for macrophages, eosinophils, lymphocytes and neutrophils in BALF.

WT, Wild Type; HDM, house dust mite; PAS, Periodic acid-Schiff; BALF, bronchoalveolar lavage fluid.

^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ^****P < 0.0001 (one-way analysis of variance followed by Tukey’s multiple comparison test).

CDH26 promotes epithelial-mesenchymal transition in airway epithelial cells

Given the important role of EMT in airway epithelial cells, we asked whether CDH26 regulates the expression of the EMT markers including fibronectin and vimentin. We found that EGF treatment for 48 hours upregulated CDH26 protein expression in BEAS-2B cells (Supplementary Fig. S2A and C). Furthermore, EGF treatment upregulated vimentin protein expression in BEAS-2B cells (Supplementary Fig. S2B and D). However, CDH26 knockdown significantly suppressed EGF-induced fibronectin and vimentin expression (Fig. 4A-C). Consistently, immunostaining of vimentin revealed that CDH26 knockdown inhibited EGF-induced vimentin expression in BEAS-2B cells (Fig. 4G). In contrast, CDH26 overexpression further enhanced EGF-induced fibronectin and vimentin expression (Fig. 4D-F, Supplementary Fig. S3). Our data suggest that CDH26 promotes EGF-induced epithelial-mesenchymal transition in airway epithelial cells.

Fig. 4. CDH26 promotes EGF-induced epithelial-mesenchymal transition in airway epithelial cells. (A) The protein levels of fibronectin, vimentin in BEAS-2B cells after transfection with negative control or CDH26 siRNA with or without EGF treatment for 48 hours were detected by western blotting. (B, C) Densitometry assay of the western blotting results was performed using ImageJ. The protein levels of fibronectin and vimentin were indexed to GAPDH. (D) The protein levels of fibronectin and vimentin in BEAS-2B cells after transfection with either the empty factor or the CDH26 cDNA expression vector and with or without EGF treatment for 48 hours, were analyzed by Western blotting. (E, F) Densitometry assay of the western blotting results was performed using ImageJ. The protein levels of fibronectin and vimentin were indexed to GAPDH. (G) Representative images of vimentin immunofluorescence staining and DAPI staining in BEAS-2B cells after transfection with negative control or CDH26 siRNA with or without EGF treatment for 48 hours. Scale bar, 50 μm. The data are representative of three independent experiments. Data are presented as mean ± standard deviation.

CDH26, cadherin-26; EGF, epidermal growth factor; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; DAPI, 4′,6-diamidino-2-phenylindole.

^*P < 0.05; ^**P < 0.01; ^****P < 0.0001 (one-way analysis of variance followed by Tukey’s multiple comparison test).

CDH26 inhibits both baseline and EGF-induced degradation of the EGFR

We next examined whether CDH26 affects airway epithelial EGFR expression. In the lung tissue from mouse model of allergic airway inflammation, the protein levels of the EGFR were significantly reduced in Cdh26^−/− mice compared to WT mice (Fig. 5A and B). We further examined whether CDH26 can inhibit the downregulation of EGFR expression. In the presence of cycloheximide, an inhibitor of protein synthesis, the protein levels of EGFR were significantly decreased in BEAS-2B cells transfected with CDH26 siRNA compared to control cells (Fig. 5C and D). It has been reported that higher concentration of EGF (100 ng/mL) stimulation can significantly promote the degradation of the EGFR.²⁸ We found that CDH26 knockdown intensified EGF-induced EGFR down-regulation in BEAS-2B cells (Fig. 5E and F). Moreover, EGF-induced AKT and PI3K phosphorylation were also reduced after CDH26 knockdown (Fig. 5G-I). By contrast, CDH26 overexpression suppressed EGF-induced EGFR down-regulation and intensified AKT and PI3K phosphorylation (Supplementary Fig. S4). Our data suggest that CDH26 inhibits baseline and EGF-induced degradation of the EGFR in airway epithelial cells.

Fig. 5. CDH26 inhibits baseline and EGF-induced EGFR degradation. (A) EGFR protein levels in mouse lungs were measured by western blotting. (B) Densitometry assay was performed using ImageJ, and the protein levels of the EGFR were indexed to GAPDH. (C) The protein levels of the EGFR in BEAS-2B cells, following transfection with either a negative control or a CDH26 siRNA and subsequent treatment with or without CHX treatment for 6, 12, and 24 hours, were analyzed by Western blotting. (D) Densitometry assay of the western blotting results was performed using ImageJ. The protein levels of EGFR were indexed to GAPDH. (E) The protein levels of EGFR in BEAS-2B cells, following transfection with a negative control or a CDH26 siRNA and subsequent treatment with or without EGF for 1, 4, and 6 hours were analyzed by Western blotting. (F) Densitometry assay was performed using ImageJ, and the protein levels of EGFR were indexed to GAPDH. (G) The protein levels of phosphorylated and total AKT, phosphorylated and total PI3K in BEAS-2B cells, following transfection with a negative control or a CDH26 siRNA and subsequent treatment with or without EGF for 15, 30, and 60 minutes, were analyzed by Western blotting. (H, I) Densitometry assay of phosphorylated AKT relative to total AKT and phosphorylated PI3K relative to total PI3K. The data are representative of three independent experiments. (J) Binding of CDH26 with the EGFR was examined using immunoprecipitation assay. (K, L) The protein levels of the EGFR were determined in BEAS-2B cells transfected with control or CDH26 siRNA and treated with MG132 or CQ using Western blotting. Densitometry assay was performed using ImageJ, and EGFR protein levels were indexed to GAPDH. Data are presented as mean ± standard deviation.

CDH26, cadherin-26; EGF, epidermal growth factor; EGFR, epidermal growth factor receptor; CHX, cycloheximide; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; CQ, chloroquine; HDM, house dust mite; WT, Wild Type; ns, not significant.

^*P < 0.05; ^**P < 0.01; ^***P < 0.001 (one-way analysis of variance followed by Tukey’s multiple comparison test).

To further investigate the mechanism by which CDH26 inhibits EGFR degradation, we transfected BEAS-2B cells with a Flag-tagged CDH26 overexpression plasmid. Co-immunoprecipitation assays demonstrated that CDH26 does not directly interact with the EGFR (Fig. 5J). We then transfected BEAS-2B cells with CDH26 siRNA and treated them with either the proteasomal inhibitor MG132 or the lysosomal inhibitor chloroquine (CQ) for 48 hours. Western blot analysis showed that MG132, but not CQ, significantly restored EGFR protein levels. While CDH26 knockdown substantially reduced EGFR expression, this reduction was largely rescued by MG132 co-treatment. However, the restored EGFR levels remained lower than those in cells treated with MG132 alone (Fig. 5K and L). These results suggest that CDH26 at least partially regulates EGFR stability through the proteasomal degradation pathway.

Epithelial CDH26 expression is correlated with EGFR expression in subjects with asthma

We further examined the association between airway CDH26 and EGFR expression in asthma patients. EGFR transcript levels in bronchial epithelial brushings were markedly increased in asthma patients compared to healthy controls (Fig. 6A). EGFR transcript levels were significantly correlated with basement membrane thickness (Fig. 6B). Moreover, there was a positive correlation between epithelial CDH26 and EGFR transcript levels in asthma patients (Fig. 6C). This is consistent with our finding in the mouse model and in vitro experiments.

Fig. 6. Epithelial CDH26 expression is correlated with EGFR expression in asthma patients. (A) EGFR transcript levels in bronchial brushings from asthmatic patients (n = 42) and healthy controls (n = 10) were measured by quantitative polymerase chain reaction. The transcript levels were expressed as log2-transformed and expressed relative to the median value for healthy controls (two-tailed Mann-Whitney test). (B) Spearman’s rank-order correlation assay between epithelial EGFR and airway basement membrane thickness in asthmatic patients (n = 38). (C) Spearman’s rank-order correlation assay between epithelial EGFR and CDH26 transcript levels in asthma patients (n = 38). Data are presented as mean ± standard deviation.

CDH26, cadherin-26; EGFR, epidermal growth factor receptor.

^*P < 0.05.

DISCUSSION

Airway remodeling in asthma causes the narrowing and stiffening of the airway lumen, leading to irreversible airflow limitation and worsening of symptoms in patients. However, mechanisms driving airway remodeling in asthma remain unclear. Our study reveals that CDH26 is highly expressed in the airway epithelial cells of asthmatic patients, showing a significant correlation with airway basement membrane thickness. Airway remodeling was significantly suppressed in HDM-challenged Cdh26-deficient mice compared to control mice. Mechanistically, CDH26 inhibited EGFR degradation, enhanced PI3K and AKT phosphorylation, and subsequently facilitated EGF-induced epithelial-mesenchymal transition (Fig. 7).

Fig. 7. Schematic diagram for the mechanism by which CDH26 promotes airway remodeling. CDH26 promotes airway remodeling in asthma by inhibiting EGFR proteasomal degradation, thereby activating the PI3K-AKT signaling pathway and facilitating epithelial-mesenchymal transition.

CDH26, cadherin-26; EGFR, epidermal growth factor receptor; EGF, epidermal growth factor; EMT, epithelial-mesenchymal transformation; HDM, house dust mite.

CDH26 belongs to the non-classical cadherin family. We have recently reported that CDH26 promotes airway inflammation via amplifying IL-4R signaling.²¹ The IL-4R signaling pathway plays a critical role in airway inflammation and mucus hypersecretion. However, it cannot fully account for the characteristic features of airway remodeling in asthma, such as EMT, subepithelial fibrosis, or basement membrane thickening.²⁹ We demonstrated a strong positive correlation between epithelial CDH26 expression and basement membrane thickness, an indicator of airway remodeling. These findings suggest that CDH26 may promote airway remodeling in asthma through alternative signaling pathways. In our previous study, we established an acute model of asthma by sensitizing and challenging mice with HDM and collected samples on day 24. This model exhibited more severe airway inflammation. By contrast, the present study utilized an extended HDM exposure protocol with sample collection on day 35, which resulted in more pronounced airway hyperresponsiveness and remodeling features. In this chronic model of asthma, Cdh26 deficiency inhibited HDM-induced airway hyperresponsiveness and remodeling in mice, suggesting that CDH26 contributes to airway remodeling in asthma. Meanwhile, Cdh26 deficiency also attenuated HDM-induced airway inflammation in this model, consistent with our previous observations in the acute model of asthma.

Airway remodeling is linked to repetitive injury and repair of the airway epithelial tissue induced by airway inflammation.³⁰ The airway epithelium can trigger remodeling by releasing alarmins including thymic stromal lymphopoietin, IL-25, and IL-33 when activated by allergens and other stimuli.³¹,³² These epithelial cytokines foster airway remodeling by facilitating communication between epithelial cells and fibroblasts, mast cells, and airway smooth muscle cells, as well as interacting with immune cells such as macrophages.² EMT promotes airway remodeling via leading to extracellular matrix accumulation, subepithelial fibrosis, and airway wall thickening.³³,³⁴,³⁵ Cytokines like transforming growth factor-β and EGF are pivotal in governing EMT.³⁶,³⁷ In our in vitro experiments, we observed that EGF stimulated CDH26 expression in airway epithelial cells. Furthermore, CDH26 knockdown significantly impeded EGF-induced EMT, indicating a potential role for CDH26 in EGF-induced EMT.

EGF binds to the EGFR, leading to downstream activation of the PI3K-AKT pathway and promotion of EMT.³⁸ Meanwhile, the EGFR is internalized and degraded upon binding with EGF, thus terminating downstream signaling.³⁹ Increased expression of the EGFR and the EGF in the airway epithelium has been reported in asthma patients.⁴⁰ Moreover, airway EGFR expression was significantly correlated with airway remodeling in children with severe asthma.⁴¹ In the HDM-induced chronic airway inflammation mice model, EGFR inhibitor erlotinib improved airway remodeling.¹¹ However, targeting EGFR for asthma therapy remains challenging due to several limitations. First, the pathological features of asthma, including airway inflammation, mucus hypersecretion, and remodeling, are not exclusively dependent on EGFR signaling. Second, the EGFR exerts broad physiological functions, and its inhibition is associated with adverse effects such as skin rash, myelosuppression, gastrointestinal disturbances, and diarrhea.⁴² In this study, we demonstrated that CDH26 stabilizes the EGFR by inhibiting its proteasomal degradation, leading to the activation of the PI3K-AKT pathway and subsequent promotion of EMT. In contrast to the EGFR, CDH26 expression is specifically elevated in the airway epithelium of asthmatic patients. Importantly, the genetic ablation of Cdh26 in mice simultaneously attenuated airway inflammation, mucus overproduction, and airway remodeling. These results position CDH26 as a potential and more specific therapeutic target for asthma treatment.

The EGFR undergoes degradation through either the ubiquitin-proteasome pathway or the lysosomal pathway.⁴³,⁴⁴ Our findings indicate that CDH26 likely inhibits EGFR degradation via the proteasomal pathway. However, co-immunoprecipitation experiments revealed no direct interaction between CDH26 and the EGFR.

During ubiquitin-proteasome-mediated EGFR degradation, ubiquitin molecules are conjugated to the EGFR through the sequential actions of E1 ubiquitin-activating, E2 ubiquitin-conjugating, and E3 ubiquitin ligase enzymes, leading to the recognition and degradation of ubiquitinated EGFR by the proteasome.⁴⁵ Since E3 ubiquitin ligases determine substrate specificity, we hypothesize that CDH26 may interact with a specific E3 ubiquitin ligase, thereby inhibiting its ability to ubiquitinate EGFR and subsequent proteasomal degradation. Nevertheless, the precise mechanism by which CDH26 regulates EGFR proteasomal degradation requires further investigation.

This study has several limitations. First, the asthmatic patients included in this study had mild to moderate asthma; therefore, the expression and function of CDH26 in the airway epithelium of severe or corticosteroid-treated asthma patients remain to be further investigated. Second, we cannot rule out the possibility of Cdh26 affecting other cell types beyond epithelial cells in the mouse model because global Cdh26 knockout mice were used. Third, the mechanism by which CDH26 inhibits EGFR degradation in bronchial epithelial cells requires further exploration.

In summary, CDH26 promotes airway remodeling via stabilizing epithelial EGFR expression and sustaining the activation of EGFR signaling in asthma. CDH26 can be a potential therapeutic target for airway remodeling in asthma in the future.

SUPPLEMENTARY MATERIALS

Supplementary Table S1

Primers for quantitative polymerase chain reaction

Supplementary Fig. S1

CDH26 expression in severe and corticosteroid-treated asthma bronchial epithelium from GEO datasets. (A) CDH26 expression levels in bronchial brushings from controls (n = 27), mild-moderate asthma (n = 72), and severe asthma (n = 56) subjects (GSE63142). (B) CDH26 expression levels in bronchial brushings from controls (n = 18), asthma patients (n = 42), and fluticasone-treated asthmatic patients (n = 19) (GSE4302).

Supplementary Fig. S2

EGF enhances CDH26 and vimentin expression in BEAS-2B cells. (A) The protein levels of CDH26 in BEAS-2B cells after exposure to EGF for 48 hours were determined by Western blotting. (B) The protein levels of vimentin in BEAS-2B cells after exposure to EGF for 48 hours were determined by Western blotting. (C, D) Densitometry assay was performed using ImageJ, and CDH26 and vimentin protein levels were indexed to GAPDH. The data are representative of three independent experiments. Data are presented as mean ± standard deviation.

Supplementary Fig. S3

CDH26 promotes EGF-induced epithelial-mesenchymal transition in airway epithelial cells. Representative images of Vimentin immunofluorescence staining and DAPI staining in BEAS-2B cells, following transfection with either an empty vector or CDH26 cDNA expression vector and subsequent treatment with or without EGF for 48 hours. Scale bar, 50 μm.

Supplementary Fig. S4

CDH26 inhibits EGFR degradation and promotes PI3K-AKT phosphorylation. (A) The protein levels of EGFR in BEAS-2B cells, following transfection with either an empty vector or CDH26 cDNA expression vector and subsequent treatment with or without EGF for 1, 4, and 6 hours were analyzed by Western blotting. (B, C) Densitometry assay was performed using ImageJ, and the protein levels of EGFR were indexed to GAPDH. (D) The protein levels of phosphorylated and total AKT, and phosphorylated and total PI3K in BEAS-2B cells, following transfection with either an empty vector or a CDH26 cDNA expression vector and subsequent treatment with or without EGF for 15, 30, and 60 minutes were analyzed by Western blotting. (E, F) Densitometry assay of phosphorylated AKT relative to total AKT and phosphorylated PI3K relative to total PI3K. The data are representative of three independent experiments. Data are presented as mean ± standard deviation.