FNDC5/Irisin Attenuates TGF-β-induced Fibroblast Differentiation Via AMPK/Nrf2 Signaling and Mitochondrial Stabilization in Subepithelial Fibrosis

Abstract

Severe asthma is characterized by subepithelial fibrosis and mitochondrial dysfunction in bronchial fibroblasts. Fibronectin type III domain-containing 5 (FNDC5)/irisin modulates metabolism and protects mitochondria during inflammation, but its role in fibroblast differentiation remains unclear. In this study, we assessed FNDC5 expression in bronchial tissues from house dust mite (HDM)–induced asthmatic mice and controls by immunohistochemistry, analyzed fibrotic proteins by Western blotting, performed RNA sequencing to identify antioxidant pathway–associated genes, evaluated mitochondrial function using JC-1 staining, and measured airway hyperresponsiveness (AHR) with the FlexiVent system. FNDC5 levels were reduced in asthma-derived lung fibroblasts and in HDM-induced asthmatic mice compared to controls. Treatment with irisin decreased α-SMA expression, improved mitochondrial function, and suppressed TGF-β–induced fibrotic protein expression and Smad3 phosphorylation in human lung fibroblasts. RNA sequencing confirmed that irisin enhanced antioxidant-related gene expression and activated the AMPK pathway, which in turn reduced TGF-β–induced Smad3 phosphorylation and fibronectin expression. Moreover, irisin induced Nrf2 phosphorylation, and Nrf2 knockdown diminished its protective effects on fibrosis markers. In vivo, irisin treatment alleviated HDM-induced airway fibrosis, inflammation, and lung function impairments. Together, these findings demonstrate that FNDC5/irisin protects against airway fibrosis in patients with asthma by activating AMPK and Nrf2 signaling pathways, thereby preserving mitochondrial function and reducing fibroblast differentiation, highlighting FNDC5/irisin as a potential therapeutic target for airway remodeling in asthma.

Introduction

Asthma is a chronic lung disease that affected an estimated 262 million individuals and resulted in 455,000 deaths in 2019 [1]. It is an allergic airway disease, characterized by features such as airway hyperresponsiveness (AHR), mucus hyperproduction, eosinophilic inflammation, and airway remodeling [2]. Airway remodeling in asthma is characterized by subepithelial fibrosis, extracellular matrix (ECM) deposition, smooth muscle cell hypertrophy, and epithelial damage [3–5]. During the fibrotic process, lung fibroblasts are activated by a multitude of mediators, including transforming growth factor (TGF)-β, platelet-derived growth factor (PDGF), and fibroblast growth factor (FGF), among others. These activated fibroblasts, known as myofibroblasts, express α-smooth muscle actin (α-SMA) and elevate levels of ECM components such as fibronectin, and various types of collagen (I, III, V, and VI) [6]. Furthermore, TGF-β, a common inflammatory mediator secreted by inflammatory cells, plays a pivotal role in the progression of organ fibrosis [7]. TGF-β effectively induces the transition of fibroblasts into myofibroblasts and instigates expressions of essential fibrosis-related genes through diverse downstream pathways, notably the Smad 2/3 signaling pathway [6]. Moreover, mitochondrial dysfunction promotes fibrogenesis, while mitochondrial ROS alters TGF-β signaling in fibroblasts [8–10]. Despite extensive research on asthma pathophysiology, subepithelial fibrosis remains a major challenge in asthma management.

Fibronectin type III domain-containing 5 (FNDC5) encodes irisin, a cleaved and secreted peptide hormone primarily produced by muscle and adipose tissue [11, 12]. Irisin plays a crucial role in protecting mitochondrial function, enhancing glucose uptake, promoting lipolysis, and regulating metabolic activity by binding to its receptor integrin α_Vβ₁ or α_Vβ₅ and subsequently activating the mitogen-activated protein kinase (MAPK) or adenosine monophosphate (AMP)-activated protein kinase (AMPK) signaling pathways [12–14]. Previous studies revealed that irisin downregulates inflammatory responses by reducing reactive oxygen species (ROS) levels in macrophages and neutrophils, and attenuates TGF-β-induced metabolic reprogramming in a mouse model of kidney injury [15, 16]. These results suggest that FNDC5/irisin plays a protective role in inflammatory processes by promoting antioxidant production and preserving mitochondrial function. However, whether FNDC5/irisin can downregulate subepithelial fibrosis in asthma by modulating mitochondrial function remains unknown.

Nuclear factor erythroid 2-related factor 2 (Nrf2) is a key transcription factor that regulates intracellular redox homeostasis and protects against oxidative stress [17, 18]. A recent study showed that irisin ameliorates cigarette-induced emphysema through the Nrf2/heme oxygenase (HO)−1 pathway [19]. However, whether FNDC5/irisin can decrease airway subepithelial fibrosis of asthma through activation of Nrf2 still needs further investigation.

Subepithelial airway fibrosis is a characteristic pathological feature of asthma and a major contributor to airway remodeling [5]. Mitochondrial dysfunction also plays a crucial role in fibroblast differentiation in inflammatory diseases [8]. However, no studies have explored whether irisin can regulate TGF-β-induced lung fibroblast activation or mitigate subepithelial fibrosis in asthma through Nrf2 activation and mitochondrial stabilization.

In this study, we investigated the role of FNDC5/irisin in the differentiation of lung fibroblasts in asthma. We further examined whether irisin modulates TGF-β signaling, mitochondrial function, and its protective effects on house dust mite (HDM)-induced airway fibrosis in mice. These findings provide new insight into the therapeutic potential of irisin in mitigating airway remodeling in asthma.

Materials and methods

Materials

Minimum essential medium (MEM) (10-009-CV) and fetal bovine serum (FBS) (35-015-CV) were obtained from Corning (Corning, NY, USA). Opti medium (31985070), non-essential amino acids (100×) (NEAAs; 11140050), sodium pyruvate (100 mM; 11360070), L-glutamine (200 mM; 25030081), and antibiotic-antimycotic (100×) (15240062) were purchased from Gibco (Waltham, MA, USA). Control small interfering (si)RNA (scrambled), Nrf2 siRNA, an α-tubulin antibody (T6199), and methacholine chloride (A2251) were obtained from Sigma-Aldrich (St. Louis, MO, USA). Lipofectamine 3000 reagent, and Lipofectamine Plus reagent (L3000015), were purchased from Invitrogen Life Technologies (Carlsbad, CA, USA). Recombinant Human/Mouse/Rat Irisin Protein (8880-IR-025/CF) was from R&D Systems (Minneapolis, MN, USA). AICAR and rotenone were purchased from MedChemExpress (MCE) (Middlesex, NJ, USA) A fibronectin (ab2413) antibody, an α-SMA antibody (ab124964), a phosphorylated (p)-Nrf2 (Ser40) antibody (ab76026), an FNDC5 antibody (ab174833) a SMAD3 antibody (ab84177), a goat anti-rabbit immunoglobulin G (IgG) H&L (Alexa Fluor^® 488) antibody (ab150077), an anti-α smooth muscle actin (α-SMA)-conjugated Alexa Fluor^® 555 antibody (ab202509), and mounting medium with 4’,6-diamidino-2-phenylindole (DAPI; ab104139) were acquired from Abcam (Cambridge, UK), while p-SMAD3 (Ser423/425) (#9520), AMPKα Antibody (#2532) and Phospho-AMPKα Thr172 (#2531) were purchased from Cell Signaling Technology (Danvers, MA, USA). Four-well cell culture slides (30404) were bought from SPL Life Sciences (Pocheoni, Gyeonggi, Korea). An Nrf2 antibody (GTX103322) was purchased from Genetex (Irvine, CA, USA). A β-actin antibody (AM1021B) was purchased from Abgent (San Diego, CA, USA). Goat anti-rabbit IgG (H + L)-horseradish peroxidase (HRP) (C04003) and goat anti-mouse IgG (H + L)-HRP (C04001) were acquired from Croyez Bioscience (Taipei, Taiwan). Recombinant human TGF-β (100 − 21) was obtained from PeproTech (Cranbury, NJ, USA). A JC-1 mitochondrial membrane potential assay kit (10009172) was purchased from Cayman Chemical (Ann Arbor, MI, USA). The Novolink Polymer Detection System (RE7280-K) was purchased from Leica (Wetzlar, Germany). A mix of Dermatophagoides pteronyssinus and D. farinae HDMs were obtained from CITEQ biologics (Groningen, the Netherlands).

Cell Culture

Normal human lung fibroblasts (NHLFs) (Catalog #: CC-2512) and lung fibroblasts from an asthma patient (Catalog #: 00194912) were obtained from Lonza (Basel, Switzerland). Cells were cultured in MEM nutrient mixture, supplemented with 10% FBS, 100 units/ml penicillin, 100 µg/ml streptomycin, 250 ng/mL of Gibco Amphotericin B, and 0.1 mM NEAAs, in a humidified incubator at 37 °C with 5% CO₂ and 95% N₂. Once the NHLFs reached 80% confluence, they were seeded into 6-cm dishes for immunoblotting or onto four-well culture slides for immunofluorescence (IF) staining or the JC-1 assay.

HDM-induced Asthma in Mice

BALB/c mice were provided by the National Laboratory Animal Center (NARLabs, Taiwan). BALB/c mice were intratracheally administered HDM extract at 100 µg on days 0 and 7 and then at 30 µg on days 21, 22, and 23. They were sacrificed on day 24; mice treated with phosphate-buffered saline (PBS) were used as controls. Mice were treated with HDM to induce asthma, followed by intraperitoneal injection of irisin (10 mg/kg) from days 8 to 20, based on a previously published study demonstrating its efficacy in diabetic cardiomyopathy [20] (Fig. 6A). Mice were sacrificed on day 24.

Fig. 6.

Irisin’s therapeutic effect on house dust mite (HDM)-induced airway fibrosisA, BALB/c mice were sensitized with 100 µg HDM/50 µl through two intratracheal administrations on days 0 and 7. Irisin therapy was administered from days 8 to 20 via intraperitoneal injections. Mice were then challenged with 30 µg HDM/50 µl on days 21, 22, and 23, and sacrificed on day 24. B, Lungs from control and irisin mice exposed to HDM, as described in Fig. 6A. Lung homogenates from each group of mice were harvested and analyzed for fibronectin and β-actin. Each lane represents a single animal, and β-actin was used as a loading control. Samples from three animals are shown. C, Quantification of fibronectin expression normalized to β-actin from Fig. 6B. Data are presented as the mean ± SEM, n=3, *p<0.05, relative to HDM mice. Lungs from each group of mice were fixed and analyzed by (D) hematoxylin and eosin (HE) staining, (E) Masson’s trichrome staining, (F) periodic acid-Schiff (PAS) staining, and (G) IHC staining for α-SMA (original magnification, 20×). (PBS, n=3, irisin, n=2, HDM, n=5, HDM+irisin, n=5)

Lung Function

Lung function was performed as described previously [21]. Pulmonary function in mice was measured using the FlexiVent ventilator system (SCIREQ, Montréal, Canada). Mice were anesthetized with a diluted solution of Zoletil 50 (1 ml Zoletil 50 + 5 ml Rompun) and connected to the FlexiVent system via an intratracheal tube. Mice were ventilated with a tidal volume of 10 ml/kg at a frequency of 150 breaths/min to reach a mean lung volume similar to that of spontaneous breathing. Each mouse was nebulized with PBS or increasing concentrations of methacholine (6.25, 12.5, 25 mg/ml). Testing of lung mechanical properties, including compliance (Crs), resistance (Rrs), Forced Expiratory Volume in 0.1 s (FEV_0.1), peak expiratory flow (PEF), and the ratio of Forced Expiratory Volume in 0.1 s to Forced Vital Capacity (FEV_0.1/FVC), was carried out by a software-generated script that took three readings per animal.

Bronchoalveolar Lavage Fluid (BALF) Cell Counting

A cannula was inserted into the trachea, and the lungs were lavaged three times with 1 ml of fresh PBS. The average recovery of the lavage fluid was approximately 50–60%. The lavage samples were centrifuged at 400 ×g for 5 min, and pellets were resuspended in 100 µl of PBS. Concentration of inflammatory cells in BAL including neutrophils, eosinophils, white blood cells, and lymphocytes was conducted using a hematology analyzer (ProCyte Dx; IDEXX Laboratories, Westbrook, ME, USA).

Immunohistochemistry (IHC)

Serial 4-µm paraffin sections were processed using the NOVA Histo system (Bremerton, WA, USA) for deparaffinization, rehydration, and antigen heat retrieval. The antigen retrieval step involved heating the sections to 100 °C for 10 min, followed by cooling to room temperature for 30 min. A 10-min peroxidase block was then applied. After blocking, sections were incubated overnight at 4 °C with primary antibodies diluted in protein block reagent. For permanent staining, the Novolink Polymer Detection System (Novolink, Novocastra, Newcastle, UK) was used, followed by development with the diaminobenzidine (DAB) chromogen. In the case of anti-FNDC5 or α-SMA, biotinylated anti-rat secondary antibodies (Abcam, Cambridge, UK) were applied at a 1:500 dilution in protein block reagent. Finally, slides were counterstained with hematoxylin and cover-slipped using Permount.

IF Staining

IF staining was performed as described previously [22]. In brief, NHLFs were seeded onto four-well slides. When cells had reached 80% confluency, they were treated with irisin for 30 min, followed by TGF-β treatment for another 24 h. Slides were then incubated with PBS containing 0.25% Triton X-100 for permeabilization. After blocking with 5% BSA, slides were incubated overnight with specific primary antibodies such as fibronectin in 1% BSA (1:1000). The following day, the slides were incubated with a fluorescein isothiocyanate (FITC)-conjugated secondary antibody (1:500) for 1 h in the dark at room temperature. This was followed by incubation with an anti-α-SMA antibody conjugated with Alexa Fluor^® 555 (1:500) for another hour. Finally, slides were counterstained with a 4’,6-diamidino-2-phenylindole (DAPI) staining solution. Fluorescent images were captured using a fluorescence microscope (TCS SP5, Leica, Wetzlar, Germany).

SiRNA Transfection

Cells were transfected for 24 h with control siRNA or Nrf2 siRNA using Lipofectamine 3000 reagent with Opti medium. Twenty-four hours after siRNA transfection, cells were treated with TGF-β or irisin (100 ng/ml) for indicated time intervals. Results were analyzed by Western blotting.

Western Blotting

The Western blot analysis was as previously described [22]. Cell lysates were extracted from NHLFs, separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), and transferred to polyvinylidene difluoride (PVDF) membranes. PVDF membranes were blocked with 5% bovine serum albumin (BSA), followed by incubation with specific primary antibodies and HRP-conjugated secondary antibodies. Enhanced chemiluminescence (ECL) was used to detect immunoreactivity, and the signal intensity was quantified using Image-Pro software (Eastman Kodak, Rochester, NY, USA).

RNA Sequencing (RNA-Seq)

RNA-Seq was performed as described in a previous study [23]. RNA was extracted utilizing the Trizol^® reagent (Invitrogen, Dallas, TX, USA). The concentration of the extracted RNA was measured at OD 260 nm on an ND-1000 spectrophotometer (Nanodrop Technology, Wilmington, DE, USA). The quality of the RNA samples, each with a minimum RNA integrity number (RIN) value of 7, was verified with a Bioanalyzer 2100 (Agilent Technology, Santa Clara, CA, USA) along with an RNA 6000 LabChip kit (Agilent Technology). For library preparation, the SureSelect XT HS2 mRNA Library Preparation kit (Agilent Technology) was employed, followed by size selection using AMPure XP beads (Beckman Coulter, Brea, CA, USA). Sequencing was conducted using Illumina’s sequencing-by-synthesis technology (Illumina, San Diego, CA, USA) on an Illumina NovaSeq instrument. Post-sequencing data (FASTQ reads) were produced through Welgene Biotech’s (Taipei, Taiwan) pipeline, utilizing Illumina’s basecalling bcl2fastq v2.20 software. Adapter clipping and sequence quality trimming were carried out with Trimmomatic v0.36, employing a sliding-window strategy. For mapping next-generation sequencing reads to genomes, the sensitive HISAT2 alignment program was used. A differential expression analysis was executed using DEseq (v1.39.0), incorporating genome bias detection and correction, within the Welgene Biotech’s proprietary pipeline.

JC-1 Assay

The mitochondrial membrane potential (MMP, Δψm) is detected by JC-1 assay. The JC-1 assay was performed as described in a previous study [23]. NHLFs were pretreated with irisin (100 ng/ml) for 30 min, followed by treatment with TGF-β (10 ng/ml) for an additional 24 h. After the treatment period, 100 µl of the JC-1 staining solution was added to each 1 ml of culture medium in each well of the culture slides, and cells were incubated in 5% CO₂ and 95% N₂ at 37 °C for 30 min. Following incubation, cells were directly analyzed using a fluorescence microscope. Healthy cells (Active mitochondria) were identified by fluorescence signals detected in the wavelength range of 540/570 nm, while unhealthy cells (mitochondria with lower membrane potential) were identified by fluorescence signals detected in the wavelength range of 485/535 nm.

ROS Measurement

NHLFs were pretreated with irisin (100 ng/mL) for 30 min, followed by stimulation with TGF-β (10 ng/mL) or rotenone (500 nM) for an additional 24 h. Intracellular reactive oxygen species (ROS) levels were assessed by flow cytometry using the CellROX reagent. After treatment, cells were incubated with the CellROX reagent at a final concentration of 5 µM for 30 min at 37 °C in the dark, then washed 3 times with PBS prior to analysis by flow cytometry.

Study Approval

All animal protocols were approved by the Animal Ethics Committee of Taipei Medical University (approval nos. LAC-2023-0294).

Statistical Analysis

Results were analyzed using a one-way analysis of variance (ANOVA) with Dunnett’s test as a post-test to compare at least three groups. An unpaired t-test was performed to compare two groups. A two-way ANOVA was used to assess the effect of Nrf2 siRNA on irisin treatment in TGF-β-induced lung fibroblast differentiation or the effects of irisin, HDM treatment, and their interaction on lung function, with multiple comparisons conducted using Tukey’s post-hoc test. Data are presented as the mean ± standard error of the mean (SEM), and a p value of < 0.05 was considered statistically significant.

Results

Expression of FNDC5 decreased in lung fibroblasts from asthmatic fibroblasts and in mice with HDM-induced asthma

To investigate the role of FNDC5 in subepithelial fibrosis of asthma, we first examined FNDC5 expression in lung fibroblasts from asthma patients and in normal human lung fibroblasts. FNDC5 expression was significantly reduced in fibroblasts from asthma patients compared with normal controls (Fig. 1A, B). Treatment of asthma-derived fibroblasts with increasing concentrations of irisin (0–100 ng/ml) led to a dose-dependent downregulation of the myofibroblast marker α-SMA (Fig. 1C, D). In addition, irisin at 100 ng/ml stabilized the mitochondrial membrane potential (MMP, Δψm) in fibroblasts from asthma patients (Fig. 1E, F). In vivo, immunohistochemical analysis revealed decreased FNDC5 expression in lung tissues of house dust mite (HDM)–treated mice compared with PBS-treated controls (Fig. 1G). Consistently, Western blot analysis confirmed reduced FNDC5 protein levels in lung tissues of HDM-treated mice relative to PBS controls (Fig. 1H, I).

Fig. 1.

Downregulation of fibronectin type III domain-containing protein 5 (FNDC5) in bronchial tissues from asthmatic mice. A, Cell lysates were prepared from normal human lung fibroblasts (NHLFs; Lonza, Catalog #CC-2512) and lung fibroblasts derived from an asthma patient (Lonza, Catalog #00194912). Samples were subjected to immunoblotting with antibodies against fibronectin, α-SMA, and α-tubulin. B,Quantification of FNDC5 expression normalized to α-tubulin from Fig. 1A. Data are presented as the mean ± SEM of three experiments.* p<0.05, compared to normal human lung fibroblasts (NHLFs). C, Lung fibroblasts from asthma patients were treated with irisin (0–100 ng/ml) for 24 h, and then α-SMA and α-tubulin protein levels were determined by immunoblotting. D,Quantification of α-SMA levels from Fig. 1C, Bars show the mean± SEM of three independent experiments. * p<0.05, vs. non-stimulated cells. E, Lung fibroblasts from asthma patients were treated with irisin (100 ng/ml) for 24 h. The mitochondrial membrane potential (Δψm) was assessed using a JC-1 assay. F, Quantification of the green/red fluorescence intensity ratio from Fig. 1E. Data are presented as the mean ± SEM from three independent experiments. * p<0.05 vs. untreated asthma-derived fibroblasts (n = 3). G, Representative images of IHC staining specific for FNDC5 in HDM- and PBS-treated BALB/c mice (PBS,n=3 and HDM, n=5). Images were captured at an original magnification of 20×. H, Western blotting was performed to assess levels of FNDC5 and β-actin in lung tissue lysates from HDM mice. I, Quantification of FNDC5 expression normalized to β-actin from Fig. 1H. Data are presented as the mean ± SEM, n=4. *p<0.05, PBS vs. HDM group

Irisin downregulated TGF-β-induced human lung fibroblast differentiation

To determine whether TGF-β downregulates FNDC5 expression in NHLFs, we observed that FNDC5 (Fig. 2A, green) expression was upregulated, while α-SMA (Fig. 2A, red) expression increased after TGF-β (10 ng/ml) stimulation in human lung fibroblasts by IF staining. We also detected similar results by Western blotting, showing that TGF-β caused a decrease in FNDC5 expression, while increasing expressions of fibrotic proteins such as fibronectin and α-SMA (Fig. 2B). When investigating the therapeutic effect of irisin on TGF-β-induced human lung fibroblast differentiation, we found that irisin downregulated α-SMA and fibronectin expressions in dose-dependent manners, with maximum effects observed at 100 ng/ml (Fig. 2C, D). Furthermore, irisin (100 ng/ml) also reduced TGF-β-induced Smad3 phosphorylation in NHLFs (Fig. 2E).

Fig. 2.

Fibronectin type III domain-containing protein 5 (FNDC5) expression decreased in normal human lung fibroblasts (NHLFs) with transforming growth factor (TGF)-β stimulation. A, NHLFs were treated TGF-β (10 ng/ml) for 24 h. FNDC5 (green) and α-SMA (red) expressions were determined by IF staining. Nuclei were visualized through DAPI staining, highlighted in blue. Images were captured at an original magnification of 20× (n=3). B, NHLFs were treated with TGF-β (10 ng/ml) for indicated times and then fibronectin, α-SMA, FNDC5, and α-tubulin protein were determined by immunoblots. Bars show the mean ± SEM of three independent experiments. * p<0.05, vs. non-stimulated cells. Cells were treated with irisin for 30 min, and then incubated with TGF-β (10 ng/ml) for 24 h, Cells were lysed and then immunoblotted with antibodies specific for (C) α-tubulin, α-SMA, or (D) fibronectin. Values are presented as the mean ± SEM (n = 3).* p<0.05, compared to the TGF-β-treated group. E, NHLFs was treated with irisin for 30 min, and then stimulated with TGF-β for another 30 min. Cells were lysed and then immunoblotted with antibodies specific for phosphorylated (p)-Smad3 and Smad3. Bars show the mean ± SEM of three independent experiments. * p<0.05, vs. non-stimulated cells

By comparing TGF-β treatment and combined irisin plus TGF-β treatment, mitochondrion-related differentially expressed genes (DEGs) were observed

In comparisons of irisin plus TGF-β treatment vs. TGF-β treatment, among DEGs with fold changes of > 1.5, the mitochondrion-associated gene, PAM16, MSS51, and TSFM were upregulated in the irisin plus TGF-β treatment group (Fig. 3A). Gene Ontology (GO) enrichment analysis of DEGs (|log₂ fold change| >1.5) further demonstrated that irisin significantly regulated mitochondrion-associated biological processes (BPs), cellular components (CCs), and molecular functions (MFs) in TGF-β–treated NHLFs. Enriched BPs included mitochondrial fission, protein import into the mitochondrial matrix, mitochondrial transcription, and negative regulation of cytochrome c release from mitochondria. Enriched CCs were primarily associated with the extrinsic component of the mitochondrial inner membrane and the TIM23 mitochondrial inner membrane translocase complex. Enriched MFs included fatty acid ligase activity, long-chain fatty-acid–CoA ligase activity, and very-long-chain fatty-acid–CoA ligase activity (Fig. 3B). Moreover, flow cytometric analysis demonstrated that TGF-β and rotenone (a mitochondrial electron transport chain inhibitor) significantly increased the proportion of ROS⁺ cells. Notably, irisin treatment significantly reduced the TGF-β– and rotenone-induced increases in the ROS⁺ cell population (Fig. 3C–F). To confirm whether irisin can stabilize the MMP (Δψm), we examined the MMP (Δψm) using the JC-1 assay. We found that irisin could stabilize TGF-β-induced MMP (Δψm) deterioration in NHLFs (Fig. 3G, H).

Fig. 3.

RNA-sequencing analysis comparing transcriptomes of cells treated with transforming growth factor (TGF)-β alone and cells treated with TGF-β in the presence of irisin. A, Significant alterations in gene expression levels between groups treated with TGF-β (10 ng/ml) for 24 h and those treated with TGF-β (10 ng/ml)+irisin (100 ng/ml) for 24 h. Significant differentially expressed gene (DEG) heatmap of expressions between the groups treated with TGF-β and those treated with TGF-β+irisin. (n=3)B, Enriched cellular component (CC), biological process (BP), and molecular function (MF) gene ontology (GO) terms associated with mitochondrial functions. These results were obtained from a three-experiment (n=3). C,Flow cytometric analysis and quantitative summary of the ROS⁺ cell population (E)in NHLFs treated with control, irisin, TGF-β (10 ng/ml), or TGF-β (10 ng/ml)+irisin (100 ng/ml). TGF-β, Bars show the mean ± SEM of four independent experiments. *p<0.05, vs. TGF-β-stimulated cells (n=4).D, Flow cytometric analysis and quantification of the ROS⁺ cell population(F) in NHLFs treated with control, irisin, rotenone (500 nM), or rotenone (500 nM)+irisin (100 ng/ml). Bars show the mean ± SEM of four independent experiments. * p<0.05, vs. rotenone-stimulated cells (n=4).G, Irisin downregulated the mitochondrial membrane potential (Δψm) in NHLFs. NHLFs were pretreated with 10 ng/ml irisin for 30 min followed by treatment with TGF-β (10 ng/ml) for 24 h. The mitochondrial membrane potential (Δψm) was assessed using a JC-1 assay. Images were captured at an original magnification of 20×. H,Quantification of the green/red fluorescence intensity ratio from Fig. 3G. Data are presented as the mean ± SEM from three independent experiments. *p<0.05 vs. TGF-β–stimulated cells (n = 3)

Irisin downregulated TGF-β-induced fibrotic protein expression via AMPK signaling pathway

To investigate the role of irisin in modulating fibrotic signaling in asthma, we first conducted pathway analysis. Heatmap clustering revealed that irisin treatment upregulated genes associated with the AMPK and FOXO pathways, which are linked to antioxidant responses. In contrast, TGF-β–related transcripts were strongly activated in TGF-β-treated groups, but not by control and irisin (Fig. 4A). To validate the activation of AMPK by irisin, we examined AMPK phosphorylation by Western blot. Figure 4B showed that irisin (100 ng/ml) increased phosphorylation of AMPK at Thr172, reaching a peak at 5 min. To further determine whether AMPK activation modulates fibrotic signaling, we pretreated cells with the AMPK activator AICAR in the presence of TGF-β. AICAR significantly attenuated TGF-β-induced Smad3 phosphorylation and fibronectin expression in NHLFs (Fig. 4C-D).

Fig. 4.

Irisin-induced adenosine monophosphate (AMP)-activated protein kinase (AMPK) activation was involved in down-regulation of transforming growth factor (TGF)-β-induced normal human lung fibroblast (NHLF) differentiation.A, Heatmap showing RNA-seq analysis of associated with the AMPK, FoxO, and TGF-β signaling pathways following irisin (100 ng/ml) or TGF-β (10 ng/ml) treatment for 24 h in NHLFs. These results were obtained from a three-experiment.B, NHLFs were exposed to irisin (100 ng/ml) for indicated time. Levels of phosphorylated (p)-AMPK and AMPK in cell lysates were assessed by Western blotting. Data are shown as the mean ± SEM, n=3. * p<0.05, relative to non-stimulated cells. C, NHLFs was treated with AICAR (0.5 mM) for 30 min, and then stimulated with TGF-β for another 30 min. Cells were lysed and then immunoblotted with antibodies specific for phosphorylated (p)-Smad3 and Smad3. Bars show the mean ± SEM of three independent experiments. * p<0.05, vs. non-stimulated cells. D, NHLFs was treated with AICAR (0.5 mM) for 30 min, and then stimulated with TGF-β for another 24h. Cells were lysed and then immunoblotted with antibodies specific for phosphorylated fibronectin and α-tubulin. Bars show the mean ± SEM of three independent experiments. * p<0.05, vs. non-stimulated cells

Irisin decreased TGF-β-induced human lung fibroblast differentiation through the Nrf2 pathway

Previous study showed that Nrf2 attenuates inflammatory response in COPD crosstalk with AMPK pathways [24]. Next, we investigated the mechanism through which irisin decreases TGF-β-induced human lung fibroblast differentiation. We found that irisin increased Nrf2 expression and phosphorylation in a time-dependent manner (Fig. 5A, B). The addition of Nrf2 siRNA diminished the effect of irisin on TGF-β-induced α-SMA and fibronectin expressions by human lung fibroblasts (Fig. 5C-F).

Fig. 5.

Nuclear factor erythroid 2-related factor 2 (Nrf2) is involved in the effects of irisin on transforming growth factor (TGF)-β-induced normal human lung fibroblast (NHLF) differentiation.NHLFs were exposed to irisin (100 ng/ml) for indicated time. Levels of (A) Nrf2 and α-tubulin, and (B)phosphorylated (p)-Nrf2 and Nrf2 in cell lysates were assessed by Western blotting. Data are shown as the mean ± SEM, n=3. *p<0.05, relative to non-stimulated cells. C, NHLFs were transfected with control siRNA or Nrf2 siRNA for 24 h, first stimulated with irisin for 30 min, and then stimulated with TGF-β for another 24 h. Western blotting was performed to assess levels of Nrf2 and α-tubulin, and fibronectin, α-SMA, α-tubulin in cell lysates. Data are presented as the mean ± SEM, n=6. * p<0.05, relative to TGF-β-stimulated cells. Quantification of (D) fibronectin and (E) α-SMA levels from Fig. 5C. F, NHLFs were transfected with 50 nM of control siRNA, Nrf2 siRNA, for 24 h. Immunoblotting was performed to determine the expression levels of Nrf2 and α-tubulin. Data are presented as the mean ± SEM, n=3,* p<0.05, relative to control siRNA transfected cells

Irisin decreased HDM-induced airway fibrosis in asthmatic mice

To investigate the effect of irisin on HDM-induced asthma in mice, we treated HDM-challenged mice with irisin (Fig. 6A). We found that HDM-induced fibronectin expression by lung lysates of mice significantly decreased after irisin treatment (Fig. 6B, C). Moreover, a pathological analysis showed that HDM-induced cell infiltration decreased after irisin treatment (Fig. 6D). Masson’s trichrome staining also indicated that collagen deposition was reduced compared to levels in HDM-treated mice (Fig. 6E). Additionally, HDM-induced mucus secretion was attenuated by irisin treatment as observed by periodic acid-Schiff (PAS) staining (Fig. 6F). HDM-induced α-SMA expression in the subepithelial layer decreased after irisin treatment (Fig. 6G).

Irisin improved HDM-induced airway infiltration and lung function decline in asthmatic mice

Finally, we examined the effect of irisin on lung function following HDM treatment. Irisin administration attenuated the HDM-induced increase in airway hyperresponsiveness (AHR), as evidenced by improved lung compliance and reduced airway resistance compared with HDM-treated mice. (Fig. 7A, B). Moreover, we mimicked a clinical methacholine provocation test and compared results before and after methacholine administration to evaluate FEV_0.1, PEF, and FEV_0.1/FVC. We found that these parameters had worsened after HDM treatment, but irisin treatment prevented their decline (Fig. 7C-E). HDM-induced inflammatory cells showed a decreasing trend in the BALF inflammation profile after irisin treatment (Fig. 7F-I).

Fig. 7.

Irisin’s therapeutic effect on the decline of lung function and airway inflammation in mice with house dust mite (HDM)-induced asthma.Airway hyperresponsiveness (AHR) was recorded after administration of different concentrations of methacholine (0–25 mg/ml). (A) Compliance (Crs) and (B) resistance (Rrs) were measured. n=3–5 for each group. Data are presented as the mean ± SEM, n=3–5. * p<0.05, relative to HDM-treated control mice. Asthma-relevant lung function parameters of (C)FEV_0.1,(D) PEF, and (E) FEV_0.1/FVC were compared between before methacholine and after methacholine administration. Data are presented as the mean ± SEM, n=3–5. * p<0.05, relative to HDM-treated control mice. Airway inflammatory cells were detected in bronchoalveolar lavage fluid (BALF) including (F) neutrophils (NEUT), (G) eosinophils (EO), (H) white blood cells (WBC) and (I) lymphocytes (LYMPH). Data are presented as the mean ± SEM, n=3–5. * p<0.05, relative to HDM-treated control mice

Discussion

Airway fibrosis in severe asthma resists steroid or bronchodilator therapy [2]. Therefore, exploring a therapeutic strategy for airway fibrosis may improve clinical management of asthma exacerbations. Previous studies showed that FNDC5/irisin is an antioxidant mediator that prevents the progression of inflammation [12]. Additionally, mitochondria and oxidative stress also contribute to airway fibrosis in severe asthma [25]. Based on these findings, we examined the effect of irisin on airway fibrosis in an asthmatic animal model. In this study, we first demonstrated that irisin decreased airway fibrotic protein expression and the MMP (Δψm) in lung fibroblasts from asthma patients. Moreover, irisin downregulated TGF-β-induced fibroblast differentiation through activation of AMPK and Nrf2. Finally, irisin reduced HDM-induced airway remodeling and lung function decline. These findings indicated that irisin has a therapeutic role in airway remodeling and fibrosis through AMPK/Nrf2 activation and regulation of mitochondrial function.

Airway inflammation is a common feature that drives asthma deterioration, inducing immunoglobulin E (IgE) production, eosinophilia, and tissue remodeling [26]. HDM, air pollution, and other environmental stimuli can cause airway epithelial cell damage and subsequently release alarmins such as interleukin (IL)−25, IL-33, and thymic stromal lymphopoietin (TSLP) [27]. The release of these cytokines triggers a strong ripple effect, including increases in type 2 T helper (Th2) inflammation and eosinophils, and sometimes increases in Th1 or Th17 inflammation and neutrophils. Asthmatic inflammation amplifies human lung fibrosis differentiation and causes airway remodeling and fibrosis progression [2, 26, 28–30].

Irisin is a cleavage product of FNDC5, a transmembrane precursor protein controlled by proliferator-activated receptor-gamma coactivator (PGC)−1α, usually in response to exercise [12, 31]. Irisin is well-known for regulating metabolic processes, muscle tissue remodeling, and bone resorption [12]. Moreover, several studies reported that irisin is involved in downregulating inflammation. Irisin decreases the release of inflammatory cytokines from macrophages through Toll-like receptor 4 (TLR4)/nuclear factor (NF)-κB pathway [32]. Additionally, irisin decreases neuroinflammation by inhibiting p38, signal transduction and activator of transcription 3 (STAT3), and NF-κB [31, 33]. Irisin also regulates development and reduces the browning of fat tissues, indicating that it may play a role in obesity-induced inflammation [34]. On the other hand, irisin regulates cellular metabolism and oxidative stress to decrease fibroblast activation. For instance, previous studies showed that irisin can attenuate fibrosis in different organs through AMPK phosphorylation signaling or by decreasing endoplasmic reticular stress [11, 13, 35]. In this study, irisin decreased fibrotic protein expressions through AMPK/Nrf2 activation. Also, irisin improved HDM-induced airway remodeling and fibrotic protein expressions. These results may indicate its therapeutic effect on airway fibrosis.

AMPK plays a crucial role in cellular senescence and accumulation of ECM components [36]. The AMPK activator AICAR inhibited TGF-β-induced differentiation of human renal mesangial cells into myofibroblasts, and this inhibition was dependent on AMPKα2-mediated suppression of Smad3 transcriptional activity [37]. AMPK signaling may also inhibit myofibroblast differentiation by alleviating endoplasmic reticulum stress, promoted myofibroblast differentiation and contributes to renal fibrosis in mice [38, 39]. Moreover, AMPK signaling inhibits the differentiation of myofibroblasts via JAK/STAT3, NF-κB/NLRP3, YAP/TAZ, and c/EBPβ signaling [40]. These indicated that AMPK signaling not only regulated oxidative stress but fibrotic signaling in fibrotic process. In this study, irisin improved TGF-β-induced mitochondrial dysfunction. RNA-seq analysis further revealed that irisin upregulated genes associated with the AMPK signaling pathway and increased AMPK phosphorylation. Moreover, treatment with AICAR reduced TGF-β-induced Smad3 phosphorylation and fibronectin expression in NHLFs. These findings suggest that the anti-fibrotic effects of irisin may occur through inhibition of TGF-β signaling and regulation of mitochondrial function.

Nrf2 plays a crucial role in reducing oxidative damage and mitochondrial dysfunction in neuromuscular diseases [17, 18, 41]. Nrf2 activation induces the expression of heme oxygenase (HO)−1, an antioxidant enzyme, which in turn improves mitochondrial function and prevents oxidative damage [42, 43]. Additionally, a recent study showed that irisin exerts antioxidative and anti-inflammatory effects through the Nrf2/HO-1/high-mobility group box 1 (HMGB1) pathway in RAW 264.7 mouse macrophages [44]. Moreover, several studies have shown that AMPK activation promotes Nrf2 signaling and subsequently contributes to the suppression of oxidative stress [24, 45, 46]. These findings may suggest that the antioxidant and mitochondrial-protective effects of irisin against TGF-β–induced oxidative stress and mitochondrial dysfunction may be mediated, at least in part, through the AMPK/Nrf2 pathway in NHLFs. Beyond a unidirectional relationship, accumulating evidence from other experimental systems suggests that AMPK and Nrf2 may form a bidirectional, context-dependent regulatory network. For instance, prolonged Nrf2 activation has been reported to suppress AMPK expression and autophagy, potentially serving as a safeguard mechanism under chronic oxidative stress [47]. In addition, given the redox sensitivity of AMPK, Nrf2-mediated regulation of cellular redox homeostasis, mitochondrial function, and energy metabolism may indirectly influence AMPK activity by modulating reactive oxygen species levels and cellular AMP/ATP ratios [48, 49]. Previous studies have also suggested that irisin may promote Nrf2 nuclear translocation and HO-1 expression via upregulation of p62 in irisin-induced browning of 3T3-L1 adipocytes, further supporting the existence of AMPK-independent regulatory routes for Nrf2 activation [50]. In the present study, irisin was shown to increase mitochondrial function and Nrf2 phosphorylation in time-dependent manners in NHLFs. Additionally, Nrf2 siRNA decreased TGF-β-induced fibrotic protein expression by NHLFs. These results suggest that Nrf2 plays a role in TGF-β-induced human lung fibroblast differentiation and HDM-induced airway inflammation by regulating mitochondrial function and oxidative damage. Therefore, AMPK-Nrf2 crosstalk should be regarded as a potential mechanism rather than a confirmed pathway, and further studies are required to clarify the directionality and dependency of these signaling events.

In the present study, RNA-seq analysis revealed that irisin significantly influenced mitochondria-associated cellular components, biological processes, and molecular functions. These findings suggest that irisin reduces TGF-β-induced fibrotic protein expression by modulating mitochondrial dysfunction. Mitochondrial dysfunction is caused by structural impairments to mitochondria, a reduction in mitochondrial numbers, defects in the respiratory chain, and decreased oxidative protein activity within cells and tissues. ROS, inflammatory cytokines, and TGF-β were identified as key contributors to oxidative damage to mitochondria [51]. A previous study demonstrated that mitochondria play important roles in airway remodeling and subepithelial fibrosis [30]. TGF-β-induced ROS production through complex IV within the mitochondrial inner membrane leads to structural damage and functional abnormalities in lung fibroblasts’ mitochondria [4, 52]. Moreover, TGF-β-induced oxidative stress leads to myofibroblast differentiation and resistance to apoptosis, which in turn exacerbates airway remodeling [7, 51, 52]. Therefore, mitochondrial dysfunction contributes to subepithelial fibrosis in asthma, and targeting the antioxidative process is a novel effective strategy for managing severe asthma. In this study, irisin downregulated the MMP (Δψm) in TGF-β-stimulated NHLFs, and irisin had an impact on mitochondrial-associated cellular components, biological processes, and molecular functions compared to TGF-β-treated NHLFs. This suggests that irisin attenuates TGF-β-induced lung fibroblast differentiation, possibly through regulating mitochondrial function.

FNDC5, the precursor of irisin, is commonly found in many organs and is implicated in various metabolic, inflammatory, and cardiovascular diseases, such as obesity, diabetes, gestational diabetes, non-alcoholic fatty liver disease, lipid metabolism disorders, and metabolic bone diseases [53]. Previous studies showed that the lack of FNDC5 gene expression leads to reduced uncoupling protein 1 (UCP1) expression, increased adipogenesis, and more-aggressive phenotypes of gastric cancer [54, 55]. This may indicate that FNDC5 expression serves as a protective factor against the aggression of certain diseases. Our results showed that FNDC5 expression was decreased in airway tissues from patients with asthma, as well as in HDM-induced asthmatic mice. This suggests that FNDC5 may have a negative association with asthma pathogenesis.

The mechanism of TGF-β downregulating FNDC5 expression through Smad3 phosphorylation in skeletal muscle was reported [56]. Our results showed similar findings of TGF-β decreasing FNDC5 expression, followed by increases in α-SMA and fibronectin expressions in normal human lung fibroblasts. Moreover, irisin reduced TGF-β-induced Smad3 phosphorylation. This may indicate that Irisin may also downregulate the TGF-β/Smad3 signaling pathway, thereby enhancing antifibrotic and anti-inflammatory processes.

In the present study, although irisin tended to reduce inflammatory cell infiltration in BALF, these changes did not reach statistical significance in some experimental settings. This lack of significance may be attributed to several factors. First, irisin’s predominant effects may be directed toward modulating fibrotic remodeling rather than directly suppressing airway inflammation. Previous studies have shown that irisin attenuates fibrosis by counteracting doxorubicin (DOX)-induced perivascular fibrosis through inhibition of endothelial-to-mesenchymal transition (EndMT) [57] and ameliorates vascular dysfunction by enhancing nitric oxide (NO) bioavailability, reducing oxidative and nitrosative stress, and limiting endothelial inflammation [58]. These findings support the notion that irisin primarily exerts anti-fibrotic and mitochondrial-protective effects. Second, the sample size in certain in vivo groups may have limited the statistical power to detect modest anti-inflammatory effects. Third, the timing and dosage of irisin treatment may have been sufficient to modulate airway remodeling and mitochondrial dysfunction but insufficient to fully suppress established airway inflammation. Collectively, these findings suggest that irisin exerts a protective effect on airway fibrosis and structural remodeling than on inflammatory cell infiltration.

There are several limitations to the present study. First, the animal sample size was insufficient, which may explain why we observed only a partial improvement in the inflammatory response. However, we found that irisin ameliorated HDM-induced decline in lung function and reduced the expression of fibrotic proteins. Second, the mechanism by which TGF-β downregulates FNDC5 expression in human lung fibroblasts remains unclear and warrants further investigation. We propose that a TGF-β-dependent Smad3 classical pathway may be a potential mechanism.

In conclusion, irisin downregulated TGF-β-induced lung fibroblast differentiation and alleviated HDM-induced airway fibrosis and inflammation. The mechanism involved AMPK and Nrf2 phosphorylation and the upregulation of mitochondrial function (Fig. 8). These results provide new insights into strategies for managing asthma-related airway fibrosis.

Fig. 8.

Simplified image shows the effect of irisin treatment, which decreases transforming growth factor (TGF)-β-induced fibrotic protein expression and mitochondrial dysfunction by activating AMPK and Nrf2 in normal human lung fibroblasts (NHLFs).This image was created with BioRender (https://biorender.com/)

Abbreviations

ADAM

a disintegrin and metalloproteinase

AHR

airway hyperresponsiveness

AMPK

adenosine monophosphate-activated protein kinase

ANOVA

analysis of variance

α-SMA

α-smooth muscle actin

BALF

bronchoalveolar lavage fluid

BP

biological process

BSA

bovine serum albumin

CC

cellular component

CTGF

connective tissue growth factor

DAPI

4’,6-diamidino-2-phenylindole, DEG, differentially expressed gene

ECM

extracellular matrix

FBS

fetal bovine serum

FITC

fluorescein isothiocyanate

FNDC5

fibronectin type III domain-containing protein 5

GO

gene ontology

HDM

house dust mite

HO

heme oxygenase

HRP

horseradish peroxidase

IF

immunofluorescence

IHC

immunohistochemical

IgG

immunoglobulin G

IL

interleukin

mAb

monoclonal antibody

MAPK

mitogen-activated protein kinase

MEM

minimum essential medium

MMP

mitochondrial membrane potential

p

phosphorylated

MF

molecular function

Nrf2

nuclear factor erythroid 2-related factor 2

NEAAs

nonessential amino acids

NF-κB

nuclear factor-κB

NHLF

normal human lung fibroblast

PBS

phosphate-buffered saline

PDGF

platelet-derived growth factor

PVDF

polyvinylidene difluoride

SEM

standard error of the mean

SDS-PAGE

sodium dodecylsulfate polyacrylamide gel electrophoresis

siRNA

small interfering RNA

TGF

transforming growth factor

Th2

T helper 2

TSLP

thymic stromal lymphopoietin

Author Contributions

Conceptualization: WHC and CMWData curation: WHC and HTL.Formal analysis: WHC and YCWFunding acquisition: WHC, BCC.Investigation: WHC and HTL.Methodology: BCC, and YCW.Project administration: WHC.Resources: BCC and CHL.Software: WHC and BCC.Supervision: WHC.Validation: WHC, CMW, YCW, LYL, FSY, and YJW.Visualization: WHC, CMW, YCW, LYL, FSY, and YJW.Writing - original draft: HTL and WHC.Writing - review & editing: WHC.

Funding

This study was supported by a grant (NSTC114-2314-B-038-153-MY3, and NSTC114-2320-B-038-070, 114-2813-C-038-106-B) from the National Science and Technology Council of Taiwan, R.O.C., and by a grant from Taipei Medical University (TMU) under the framework of the Higher Education Sprout Project (DP2-TMU-113-T-03) funded by the Ministry of Education (MOE) of Taiwan, R.O.C. Additional support was provided by grants (TMU 113-5431-002-400, and 112-5431-006-400) from the Chen Wei-Tien Research Foundation for Thoracic Medicine.

Data Availability

The RNA-seq RDS files generated in this study are publicly available in the Zenodo repository at [https://doi.org/10.5281/zenodo.17831773](https:/doi.org/10.5281/zenodo.17831773).

Declarations

Ethics Approval and Consent to Participate

All animal experiments were approved by the Animal Ethics Committee of Taipei Medical University (Approval Nos. LAC-2023-0294).

Consent for Publication

All participants provided written informed consent for the publication of study findings.

Competing interests

The authors declare no competing interests.