Bifidobacterium pseudolongum enriched by arctigenin ameliorates cough variant asthma via the acetate-aryl hydrocarbon receptor-interleukin-33 axis

Abstract

Gut microbiota dysbiosis is significantly associated with the onset and progression of cough variant asthma (CVA). Targeting and improving gut microbiota dysbiosis may serve as a promising strategy for CVA. Suhuang antitussive capsule (SH) is the sole clinically approved traditional Chinese patent medicine specifically indicated for managing CVA. Herein, we initially identified arctigenin (AG) as one of the effective components detected and preliminarily screened from SH for its potential in treating CVA. AG significantly ameliorates ovalbumin (OVA)-induced CVA symptoms, alleviates gut dysbiosis, and improves intestinal barrier function. Specifically, AG alleviates CVA symptoms in a gut microbiota-dependent manner, as demonstrated by antibiotic treatment and fecal microbiota transplantation (FMT). 16S rDNA sequencing revealed that AG treatment could notably enrich the commensal bacterium Bifidobacterium pseudolongum (B. pseudolongum), which is typically reduced in CVA. The mono-colonization with B. pseudolongum also showed a reduction in CVA severity. Gas chromatography-mass spectrometry (GC-MS) metabolomic analysis identified acetate as the key and main serum metabolite induced by B. pseudolongum. We demonstrated that acetate enters pulmonary circulation and activates the aryl hydrocarbon receptor (AHR). Validation using AHR antagonists, siRNA knockdown, chromatin immunoprecipitation (ChIP) assays, luciferase reporter experiments, and molecular docking confirmed that acetate-mediated AHR activation suppresses interleukin-33 (IL-33) secretion from lung epithelial cells, thereby preventing CVA progression. These findings suggest AG, one of the effective components in SH, exhibits prebiotic-like effects to alleviate CVA and highlight the potential of targeting specific gut microbiota species for CVA treatment.

Graphical abstract

Highlights

• •AG, one of the efficacious components of SH, alleviates CVA.
• •AG alleviates CVA by increasing the abundance of B. pseudolongum and promoting acetate production.
• •Acetate restores lung function by targeting and activating AHR to decrease the secretion of IL-33.

1. Introduction

Asthma represents a complex heterogeneous disorder pathologically defined by the triad of bronchial hyperresponsiveness, airway remodeling and chronic inflammation [1], clinically presenting with characteristic respiratory symptoms including repeated episodes of wheezing, dyspnea, and chronic cough, which severely affects the patient's quality of life [2]. Cough variant asthma (CVA) refers to the main symptom of cough, accompanied by eosinophilic inflammation, bronchial hyperresponsiveness and airway remodeling [[3], [4], [5]]. CVA is a common cause of prolonged or chronic cough and is considered a precursor to classic asthma [6,7]. As usual in classical asthma, inhaled bronchodilators, corticosteroids, and leukotriene receptor antagonists are the main clinical treatments for CVA [8]. However, these drugs can only alleviate the patient's condition, cannot achieve the purpose of a radical cure, and bring certain side effects. Hence, it is crucial to develop novel therapeutic strategies or drugs for CVA treatment.

Recent studies have revealed that asthma is closely connected with body metabolism and gut microbiota [9]. Asthma may cause functional and structural changes in the intestinal mucosa [10], for example, gut microbiota disturbances driven by antibiotics or diet may heighten susceptibility to allergic airway diseases and compromise respiratory infection defense [11]. Mechanistically, gut microbiota-derived metabolites, particularly short-chain fatty acids (SCFAs), exert their functions through some receptors such as G protein-coupled receptors (GPCRs) and histone deacetylases (HDACs), and have become potential mediators of the gut-lung axis, affecting the severity of allergic inflammation [12]. However, the specific effect and mechanism by which the gut microbiota regulates asthma, particularly CVA, need to be further elucidated.

As a member of the interleukin-1 (IL-1) cytokine family, IL-33 mediates its effects through interaction with the membrane-bound ST2 receptor (IL-1RL1). This binding event triggers downstream activation of both nuclear factor kappa-B (NF-κB) and mitogen-activated protein kinase (MAPK) cascades, ultimately inducing the secretion of Th2-associated cytokines, including IL-4, IL-5 and IL-13 [13]. IL-33 is expressed in fibroblasts, endothelial, epithelial, and other cells [14]. Studies have indicated that increased IL-33 production is closely associated with the progression [15] and exacerbation of asthma [16]. IL-33 can promote airway inflammation, primarily the type 2 phenotype. Furthermore, IL-33 is involved in collagen deposition in asthmatic airways, suggesting that IL-33 may participate in the epithelial-mesenchymal transition (EMT) process in lungs [17]. Hence, targeting the suppression of IL-33 might play a vital role in curbing the progression of asthma.

Suhuang antitussive capsule (SH) is currently the only traditional Chinese patent medicine approved for the clinical use to treat CVA. Arctigenin (AG) represents a bioactive lignan obtained through extraction of Arctium lappa roots [18], which has a high content in SH and was selected based on activity screening in three in vitro assays: anti-inflammatory, antitussive, and anti-asthmatic (unpublished data). Recent studies have shown that AG has many pharmacological activities such as anti-oxidative stress [19], anti-viral [20], anti-tumor [21], and anti-inflammatory [22]. AG exhibits poor oral bioavailability and is minimally absorbed into the circulation [20]. Additionally, it has been reported that AG can prevent obesity by regulating the disorder of the gut microbiota [23]. However, the molecular mechanism underlying its treatment for CVA remains to be further elucidated.

Herein, we demonstrated that AG and SH significantly improved airway inflammation and remodeling in ovalbumin (OVA)-induced CVA rats and that the underlying mechanism partly depended on the gut microbiota. 16S rDNA sequencing revealed that AG regulated the diversity and composition of intestinal mucosa, promoting probiotic colonization and inhibiting potential pathogens. After screening among the differential microbial species altered by AG, we found Bifidobacterium pseudolongum (B. pseudolongum), which is significantly depleted in CVA, plays a role through short-chain fatty acid metabolism. Subsequently, we demonstrated that the mono-colonization with B. pseudolongum significantly prevented CVA formation. The CVA-suppressing effect of B. pseudolongum was mediated by acetate, a metabolic product generated by B. pseudolongum. Further, acetate could target and activate aryl hydrocarbon receptor (AHR) to decrease the secretion of IL-33, leading to inhibit CVA formation. Meanwhile, SH administration also increased the abundance of B. pseudolongum, activating the AHR pathway to suppress IL-33 release, thereby ameliorating CVA. Together, our current findings not only propose a novel “prebiotics-probiotics” therapeutic approach for the therapy of CVA, but also further confirm the argument that AG is one of the effective components in SH.

2. Materials and methods

2.1. Chemicals and reagents

AG (CAS No. 7770-78-7, purity≥98%) was provided by Chengdu Lemeitian Pharmaceutical Technology Co., Ltd. (Chengdu, Sichuan, China). Dexamethasone (Dex; Lot No. 2304071; pharmaceutical grade) was purchased by Zhejiang Xianju Pharmaceutical Co., Ltd. (Taizhou, Zhejiang, China). Ampicillin (CAS No. 69-52-3, purity≥90%), gentamicin (CAS No. 1405-41-0, purity≥95%), neomycin (CAS No. 1405-10-3, purity≥90%), metronidazole (CAS No. 443-48-1, purity≥98%), and vancomycin (CAS No.1404-93-9, purity≥96%) were acquired from Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China). SH (National Drug Standards of China, YBZ00172008) was provided by Beijing Haiyan Pharmaceutical Co., Ltd., Yangtze River Pharmaceutical Group (Beijing, China). OVA (CAS No. 9006-59-1, purity≥98%) and sodium acetate (SA; CAS No. 127-09-3, purity≥99%) were purchased from Sigma-Aldrich (St. Louis, MO, USA). AHR inhibitor CH-223191 (CAS No. 301326-22-7, purity≥99.96%), GPR43 agonist 4-CMTB (CAS No. 300851-67-6, purity≥99.66%), GPR109A agonist niacin (CAS No. 59-67-6, purity≥99.96%) and GPR41 agonist AR420626 (CAS No. 1798310-55-0, purity≥98.29%) were purchased from MedChem Express (Monmouth Junction, NJ, USA). Authentic standards of acetic acid (CAS No. 64-19-7, purity≥99.8%), propionic acid (CAS No. 79-09-4, purity≥99.5%), and butyric acid (CAS No. 107-92-6, purity≥99.5%) were purchased from Shanghai Macklin Biochemical Co., Ltd. (Shanghai, China). Enzyme-linked immunosorbent assay (ELISA) kits for rat IL-4, IL-5, IL-13, and lipopolysaccharide (LPS) were purchased from Wuhan Mosak Biotechnology Co., Ltd. (Wuhan, Hubei, China). Human transforming growth factor-β (TGF-β) recombinant protein was purchased from Proteintech (Rosemont, IL, USA). Antibodies for epithelial cadherin (E-cadherin), AHR, IL-33, occludin, claudin-1, and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were supplied by Proteintech (Rosemont, IL, USA). Antibodies for vimentin, alpha-smooth muscle actin (α-SMA), and snail family transcriptional repressor 1 (Snail) were obtained from Cell Signaling Technology (Danvers, MA, USA).

2.2. Animal studies

Experiments used specific pathogen-free (SPF)-grade male Wistar rats (220–260 g, 8–10 weeks) sourced from Nanjing Qinglongshan Laboratory Animal Center (Nanjing, Jiangsu, China). Animals were housed in controlled conditions with ad libitum feeding. The CVA model was induced as previously described [24]. Briefly, on days 1 and 7, the rats were sensitized with 1 mg of OVA and 100 mg of aluminum hydroxide (Al (OH)₃). Starting from day 14, the rats were stimulated with 1% OVA in saline for 13 days.

For evaluating the effect of AG on OVA-induced CVA in rats, the rats were randomly divided into 6 groups (n = 6 per group): control (CON) group, OVA group, OVA + SH group (intragastrically (i.g.), 3.6 g/kg), OVA + Dex group (i.g., 0.5 mg/kg), OVA + arctigenin-low (AG-L) group (i.g., 30 mg/kg), and OVA + arctigenin-high (AG-H) group (i.g., 60 mg/kg). Each gavage group was treated once a day for 13 consecutive days.

For gut microbiota depletion treatment (n = 6 per group), an antibiotic cocktail (ABX) was administered in vivo to determine whether the anti-CVA effect of AG was associated with the gut microbiota. The ABX, consisting of ampicillin (1 mg/mL), gentamicin (1 mg/mL), neomycin (1 mg/mL), metronidazole (1 mg/mL) and vancomycin (0.2 mg/mL), was given to the rats in their drinking water for 26 days, and followed by either AG (30 mg/kg) or 0.9% saline from day 14 to 26.

For fecal microbiota transplantations (FMT) (n = 6 per group), feces from the donor rats of each group were collected aseptically every day and homogenized in 0.125 g/mL phosphate-buffered saline (PBS). After centrifugation at 2000 g at 4 °C for 1 min, the supernatant was then carefully filtered through a pre-sterilized 70 μm mesh membrane. The CVA rat recipients underwent a 7-day ABX pretreatment period, followed by daily oral gavage administration of 1 mL supernatant for 26 consecutive days.

For single-microbe treatment, the rats were assigned to the following four treatment groups (n = 6 per group): CON group, OVA group, OVA + live B. pseudolongum group (2 × 10⁸ CFU/rat, suspended in PBS), and OVA + Heat-inactivated B. pseudolongum (2 × 10⁸ CFU/rat, suspended in PBS). Each gavage group was treated once a day for 26 consecutive days.

For microbial metabolite supplement and inhibitor treatment, the rats were assigned to the following four treatment groups (n = 6 per group): CON group, OVA group, OVA + SA group (200 mM in drinking water), and OVA + SA + CH-223191 group (i.g., 10 mg/kg). Each gavage group was treated once a day for 13 consecutive days.

On the last day for all treatment mentioned above, the number of coughs and cough latencies were determined using standardized procedures [25]. Feces were collected for subsequent analysis. Subsequently, the rats were euthanized after overnight fasting, and the plasma, bronchoalveolar lavage fluid (BALF), lungs, and intestinal tissues were collected for analysis. The Ethics Committee of China Pharmaceutical University approved all animal experiments (Approval No.: 2024-07-050), which followed the NIH Guide for the Care and Use of Laboratory Animals.

2.3. Cell lines and treatment

Human bronchial epithelial cell line (16HBE HBE135-E6E7 (ATCC® CRL-2741), Cat No. FH1013, RRID: CVCL_3695) was provided by Shanghai Fuheng Biotechnology Co., Ltd. (Shanghai, China). 16HBE were cultured in Roswell Park Memorial Institute 1640 medium (RPMI 1640) medium supplemented with 10% fetal bovine serum (FBS; CAS No. BC-SE-FBS01, Bio-Channel Biological Technology Co., Ltd., Nanjing, China) and 1% penicillin-streptomycin. During the experiments, the cell line exhibited robust viability and optimal growth, with routine screenings confirming the absence of mycoplasma and other contaminants. For transfection, 16HBE cells were transfected with the specified plasmid or AHR siRNA using Lipofectamine 2000 (Invitrogen, Thermo Fisher Scientific, Waltham, MA, USA). The siRNA sequences of AHR are listed in Table S1. For the airway remodeling model, TGF-β (10 ng/mL) was used to stimulate 16HBE cells for 24 h for the correlation assays.

2.4. Microbial strain and cultivation

Bifidobacterium pseudolongum (JCM 11662; BMZ010547; Mingzhou Biotech, Ningbo, China) was grown at 37 °C under anaerobic conditions (80% N₂, 10% CO₂, 10% H₂) in MRS broth (Hope Bio-Technology Co., Ltd., Qingdao, Shandong, China). The optical density (OD₆₀₀) of the medium containing AG (50, 100, and 200 μM) for in vitro culture of B. pseudolongum was measured to plot the growth curve, reflecting microbial growth. For animal administration experiments, the bacteria were enriched by centrifugation at 8000 g for 10 min at 4 °C and resuspended in anaerobically sterilized PBS to a final concentration of 2 × 10⁸ CFU/mL for animal administration. For the discovery of metabolites of B. pseudolongum, the supernatant was collected for subsequent metabolomic analysis.

2.5. Histopathological examination

Lung tissues were paraformaldehyde (PFA)-fixed and paraffin-embedded following standard procedures, sectioned, and further stained with haematoxylin and eosin (H&E) and Masson's trichrome (Masson) staining. Optical microscopy images were acquired using a Leica DMi8 (Wetzlar, Hesse, Germany).

2.6. Immunofluorescence analysis

The 16HBE cells and lung tissues were used for immunofluorescence as previously described [26]. 16HBE were incubated with anti-AHR (1:200 dilution) at 4 °C overnight. After incubation with fluorescent secondary antibody (1:200 dilution), the nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI; Beyotime Biotechnology, Shanghai, China). For immunofluorescence staining of AHR, IL-33, and E-cadherin in lung tissues, the tissue sections were incubated with antibodies against AHR, IL-33, and E-cadherin. The images were examined with a microscope (Leica DMi8, Wetzlar, Hesse, Germany).

2.7. Quantitative real-time PCR (qPCR) analysis

The stool DNA isolation was performed with the TIANamp Stool DNA Kit (TIANGEN Biotech, Beijing, China) following the manufacturer's recommended procedure. Total RNA was isolated from the tissue homogenates and 16HBE cells using RNAiso Plus (Takara Bio Inc., Kusatsu, Shiga, Japan) and RNA was reverse transcribed to cDNA by Hifair III 1st Strand cDNA Synthesis SuperMix (Cat No. 11141ES60; Yeasen, Shanghai, China) for qPCR. Statistical analysis was performed to evaluate the expression levels of key genes by normalization to GAPDH or 16S rDNA. The primer sequences were listed in Table S2.

2.8. ELISA assay

The protein levels of IL-4 (KT30491), IL-5 (KT30409), IL-13 (KT30399), and IL-33 (KT20440) in BALF, as well as LPS (KT22085) concentrations in rat serum and feces, were determined using ELISA kits (MSKBIO, Wuhan, Hubei, China) according to the manufacturer's instructions. The absorbance was measured with a microplate reader (BioTek Instruments, Inc., Winooski, VT, USA) at 450 nm.

2.9. Western blotting analysis

Tissue and cellular proteins were extracted using ice-cold radioimmunoprecipitation assay (RIPA) lysis buffer supplemented with 1 mM phenylmethylsulfonyl fluoride (PMSF) protease inhibitor, followed by protein concentration determination via the bicinchoninic acid (BCA) method (Beyotime Biotechnology, Shanghai, China). Following primary and secondary antibody incubation, protein bands were visualized with the Tanon 5200 Multi chemiluminescence system (Tanon Science & Technology Co., Ltd., Shanghai, China).

2.10. Luciferase reporter assay

For evaluating the transcriptional regulation of AHR, 16HBE cells were seeded in 24-well plates, and after transfection with AHR siRNA, either the pGL3 vector or pGL3-IL-33 full-length promoter reporter plasmid, along with the renilla luciferase plasmid pRL-TK (GenePharma, Shanghai, China), was transfected per well for 12 h, respectively. Then, the cells were incubated with SA (4.8 mM) for 24 h at 37 °C. Luciferase activities were detected using the Dualucif™ Firefly & Renilla Assay Kit System (F6075S; UElandy Inc., Suzhou, Jiangsu, China).

2.11. Chromatin immunoprecipitation (ChIP) assay

ChIP assays were performed using the SimpleChIP® Enzymatic Chromatin IP Kit (Cat. No. 9003, Cell Signaling Technology, Danvers, MA, USA) strictly following the manufacturer's instructions. Briefly, 16HBE cells were treated with SA for 24 h and then cross-linked with 1% formaldehyde to preserve protein-DNA interactions. The chromatin was subsequently fragmented via sonication to generate DNA fragments ranging from 200 to 1000 bp in length. Immunoprecipitation was conducted using a specific anti-AHR antibody (ab19079, Abcam, Cambridge, UK), with normal IgG serving as a negative control. Following extensive washing, the protein-DNA cross-links were reversed, and the samples were subjected to proteinase K and RNase A treatment. The DNA was then purified, and the enrichment of the IL-33 promoter region was analyzed via quantitative real-time PCR (qPCR). ChIP signals were normalized to total input chromatin and expressed as the percentage of input (% Input). Specific primers targeting the IL33 promoter were designed, and the sequences are as follows: human IL33 promoter, forward: 5′-TCCAAAGGGCAGACATAGCC-3′, reverse: 5′-TCATCCCCAAACCTGAGCAA-3′.

2.12. AHR binding assay

For molecular docking simulation, the mol2 file of SA structure was downloaded from the TCMSP website, and the AHR crystal structure was retrieved from the Protein Data Bank (PDB) (https://www.rcsb.org/). The downloaded molecule and target structure were imported into AutoDock 4.2.6 software for simulation docking, and PyMOL software was used for visualization of the docking result.

For cellular thermal shift assay (CETSA), 16HBE cells were treated with SA (4.8 mM) or dimethyl sulfoxide (DMSO) for 2 h at 37 °C in 5% CO₂ incubators. The cell lysates were harvested containing a protease inhibitor and divided into six groups, and an 80 μL aliquot of this suspension was dispensed into 0.2 mL PCR tubes and subjected to heating in a PCR machine using a gradient protocol (42–67 °C, 3 min incubation). After cooling at room temperature for 3 min, the samples were centrifuged at 4 °C at 12000 rpm for 15 min. The supernatant was collected and quantified using the BCA assay for general Western blot experiments.

For drug affinity responsive target stability (DARTS), harvested 16HBE cells were lysed with 500 μL of NP-40 lysis buffer. The lysate was centrifuged to remove debris. Subsequently, the supernatant was diluted with Tris-NaCl-CaCl₂ (TNC) buffer to a final concentration of 50 mM Tris-HCl (pH 8.0), 50 mM NaCl, and 10 mM CaCl₂. The lysate was divided into five aliquots: one treated with 0.1% DMSO and the others with SA at 0, 2.4, 4.8, and 9.6 mM, respectively. After incubation for 2 h at room temperature with gentle inversion, pronase E was added, and the samples were digested for 5 min at 37 °C. Finally, the samples were mixed with 5 × sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer at a specified ratio, followed by heat denaturation at high temperature for 10 min in a protein denaturing device before being collected for general Western blot experiments.

2.13. DNA isolation and 16S rDNA amplicon sequencing

Total genomic DNA was extracted from fecal samples using the TIANamp Stool DNA Kit (TIANGEN Biotech, Beijing, China) following the manufacturer's instructions, and quality evaluation involved gel electrophoresis on agarose, with quantification performed using UV absorbance measurements. The universal primers 341F (5′-CCTAYGGGRBGCASCAG-3′) and 806R (5′-GGACTACNNGGGTATCTAAT-3′) were used to amplify the V3-V4 regions of the 16S rDNA gene. The PCR mixture (25 μL) included Master Mix (12.5 μL, Phusion Hot Start Flex 2X; New England Biolabs, Ipswich, MA, USA), primers (2.5 μL each), and template DNA (50 ng), and nuclease-free water to adjust the volume. Thermocycling consisted of 98 °C for 30 s (initial denaturation), then 35 cycles of denaturation (98 °C, 10 s), annealing (54 °C, 30 s), and extension (72 °C, 45 s), concluding with a 10 min 72 °C step. Post-PCR analysis was conducted on an Agilent 2100 bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) with Illumina's quantification kit (Illumina, Inc., San Diego, CA, USA) for library assessment, and finally, a NovaSeq 6000 sequencer (Illumina, San Diego, CA, USA) was used to perform 2 × 250 bp paired-end sequencing. 16S rDNA gene sequencing analysis was performed using QIIME 2 (Quantitative Insights Into Microbial Ecology 2, https://qiime2.org/).

2.14. RNA-sequencing

RNA sequencing library preparation and sequencing were performed by LC-Bio Technology Co., Ltd. (Hangzhou, Zhejiang, China). Briefly, following RNA quality assessment, mRNA was enriched using oligo(dT) magnetic beads, fragmented, and used for cDNA synthesis. Double-stranded cDNA was subsequently synthesized, purified using AMPure XP beads (Beckman Coulter Life Sciences, Indianapolis, IN, USA), and subjected to end repair, A-tailing, and adapter ligation. After size selection using AMPure XP beads, the libraries were amplified by PCR and subjected to quality assessment. Qualified libraries were sequenced on an Illumina NovaSeq 6000 platform (Illumina, Inc., San Diego, CA, USA).

2.15. Short-chain fatty acids (SCFAs) analysis

SCFAs were determined with the reference to a previously established protocol [27]. Briefly, for bacterial supernatants and plasma samples, 200 μL of 0.005 M NaOH (containing 10 μg/mL internal standard 2-ethyl butyric acid) was added to 200 μL of the sample and vortexed for 1 min. The samples were centrifuged at 13000 rpm for 10 min at 4 °C. The supernatant was collected and mixed with 125 μL of a propanol/pyridine (v/v = 3:2) mixture and vortexed for 1 min. 25 μL of propyl chloroformate was added, and derivatization was performed at 60 °C for 60 min. Then, 300 μL of n-hexane was added to the derivatized sample, followed by centrifugation at 10000 rpm for 10 min. Subsequently, 200 μL of the upper solution was transferred into a 1.5 mL Eppendorf tube. Then, 200 μL of n-hexane was added again to the remaining sample, and the mixture was vortexed for 1 min and centrifuged at 10000 rpm for 10 min. The upper solution obtained from the second extraction was collected and combined with the first extract. A 100 μL aliquot was transferred to a 1.5 mL tube, bringing the total sample volume to 300 μL. The solution was vortexed and mixed twice, centrifuged at 10000 rpm for 10 min, and then stored at −20 °C. Before use, gas chromatography-mass spectrometry (GC-MS) injections were performed by filtering through a 0.22 μm filter membrane. For lung, fecal, and cecal content samples, 50 mg of each sample was homogenized with 400 μL of 0.005 M NaOH, and subsequent processing was similar to that of plasma samples. The samples were analyzed using an Agilent 7890B-5977B GC-MS system (Agilent Technologies, Santa Clara, CA, USA).

2.16. Quantification of AG levels by UPLC-QQQ-MS/MS

For plasma and tissue samples: plasma aliquots (100 μL) were deproteinized with 400 μL of ice-cold methanol/acetonitrile (1:1, v/v). Lung tissues (40 mg) were homogenized in 200 μL of ultrapure water, then mixed with 800 μL of methanol/acetonitrile (1:1, v/v) to precipitate proteins.

For fecal samples: lyophilized feces (60 mg) were homogenized in 600 μL of methanol/water (4:1, v/v) using a bead-beating homogenizer.

For extraction procedure: all mixtures were vortexed, incubated at −20 °C for 30 min, and centrifuged at 13000 rpm for 15 min at 4 °C. The supernatants were evaporated to dryness under a gentle nitrogen stream and reconstituted in acetonitrile/water (1:1, v/v). Following a final centrifugation (13000 rpm, 10 min, 4 °C), the supernatants were analyzed.

Chromatographic separation of AG was performed on a Waters ACQUITY UPLC® H-Class system (Waters Corporation, Milford, MA, USA) equipped with an ACQUITY UPLC HSS T3 column (100 × 2.1 mm, 1.8 μm) maintained at 28 °C. The mobile phase consisted of 0.1% aqueous formic acid (A) and acetonitrile (B), delivered at 0.3 mL/min. The gradient profile was: 0–5.0 min, 5%–100% B; 5.0-7.0 min, held at 100% B; 7.0–7.5 min, return to 5% B; and 7.5–10.0 min, equilibration at 5% B. The injection volume was 2 μL. MS detection was performed on a XEVO® TQD triple-quadrupole mass spectrometer with an electrospray ionization (ESI) source (Waters Corporation, Milford, MA, USA), controlled by MassLynx™ v4.2 software. Quantification was achieved in multiple reaction monitoring (MRM) mode using the transition m/z 373.15 → 137.15. UV absorbance at 254 nm was monitored for quantitative verification.

2.17. Statistical analysis

Experimental results are reported as mean ± standard error of the mean (SEM), analyzed through GraphPad Prism 9.0. To determine the appropriate statistical test, we firstly conduct a normality test for all data sets. For comparing two groups: If the data is normally distributed and the variances are equal, we apply the Student's t-test; If the data is normally distributed but the variances differ, we use Welch's t-test; If the data does not follow a normal distribution, we opt for the Mann-Whitney test. For comparing multiple groups: After confirming normality, we perform a one-way analysis of variance (ANOVA) to evaluate the statistical significance across the groups; Should the data not meet the normality assumption, we utilize the Kruskal-Wallis test instead.

3. Results

3.1. AG treatment alleviates the phenotypes of CVA by the effects consistent with those of SH

Herein, we provide the chromatograms of the quality control components of SH and the chromatogram of AG (Fig. S1), which proves that AG is one of the chemical components in SH. We initially assessed the potential protective effect of AG against OVA-associated CVA in rats by the procedure (Fig. 1A). To explore the effects of AG treatment on cough-relieving function, number of coughs and coughs latencies were examined. Compared with OVA-induced rats, AG and SH treatment significantly decreased the number of coughs and extended cough latencies (Fig. 1B). To clarify the role of AG in improving airway inflammation, inflammatory cells and cytokines in BALF were detected and the results revealed that AG and SH treatment could reduce the number of inflammatory cells (total cells, eosinophils, and macrophages) and the secretion of inflammatory cytokines (IL-4, IL-5 and IL-13) (Figs. 1C and D). In histological analysis, H&E staining showed that AG and SH treatment significantly reduced inflammatory cell infiltration in the lung tissues compared to CVA rats; Masson's trichrome staining indicated that a significant attenuation of collagen deposition following AG and SH administration; periodic acid-Schiff (PAS) staining showed that AG and SH treatment significantly decreased the abundance of PAS-positive goblet cells in the airway epithelium of CVA rats (Figs. 1E and F). To evaluate the effect of AG in OVA-induced airway remodeling, the protein expression (E-cadherin, vimentin, and α-SMA) revealed that AG and SH treatment could improve OVA-exposed airway remodeling (Figs. 1G and H).

Fig. 1.

Arctigenin (AG) and Suhuang antitussive capsule (SH) treatment ameliorates ovalbumin (OVA)-induced cough variant asthma (CVA) in rats. (A) Schematic diagram of the experimental procedure of AG, dexamethasone (Dex) and SH treatment. (B) Number of coughs and cough latencies were recorded (n = 5 per group). (C) Number of inflammatory cells (total cells, eosinophils, and macrophages) in the bronchoalveolar lavage fluid (BALF) were counted (n = 5 per group). (D) Pro-inflammatory type 2 cytokines expression of interleukin-4 (IL-4), IL-5 and IL-13 were analyzed by enzyme-linked immunosorbent assay (ELISA) (n = 5 per group). (E, F) Representative images of haematoxylin and eosin (H&E), Masson's trichrome (Masson) and Periodic Acid-Schiff (PAS) staining of the lung sections from rats (E), and pathological scores were measured (F) (n = 5 per group). (G, H) Expression of proteins epithelial cadherin (E-cadherin), vimentin, alpha-smooth muscle actin (α-SMA) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were analyzed by Western blot (G) and quantified by densitometry (H) (n = 5 per group). (I) AG levels in lung, plasma, and fecal samples were quantified by ultra-performance liquid chromatography-triple quadrupole tandem mass spectrometry (UPLC-QQQ-MS/MS). (n = 6). Data are expressed as mean ± standard error of the mean (SEM). i.p.: intraperitoneal; i.g.: intragastric; ns: not significant.

To further substantiate the functional significance of AG in CVA, we treated 16HBE cells with varying concentrations of AG (25, 50, and 100 μM) and observed that it could modulate EMT-related proteins expression by upregulating E-cadherin while downregulating α-SMA and Snail, thereby counteracting airway remodeling (Fig. S2). Notably, following oral administration of AG, significant accumulation of AG was primarily observed in fecal samples (Fig. 1I). This suggests that AG might partly exert its anti-CVA effects, potentially through the modulation of gut microbiota. Collectively, our findings suggest that AG exerts the effects consistent with those of SH, both contributing to the amelioration of CVA.

3.2. AG attenuates CVA in a gut microbiota-dependent manner

To determine whether the gut microbiota is involved in the anti-CVA effect of AG, we exposed rats to ABX through drinking water to deplete the gut microbiota (Fig. 2A). As expected, the therapeutic effects of AG on cough, airway inflammation, and remodeling were significantly diminished following the antibiotic treatment (Figs. 2B–H). This suggests that the gut microbiota may play a crucial role in mediating the anti-CVA effects of AG.

Fig. 2.

Arctigenin (AG) attenuates cough variant asthma (CVA) in a gut microbiota-dependent manner. (A) Schematic diagram of the experimental procedure of antibiotic cocktail (ABX) treatment. (B) Number of coughs and cough latencies were quantified (n = 5 per group). (C) Number of the total inflammatory cells, eosinophils and macrophages in the bronchoalveolar lavage fluid (BALF) were counted (n = 5 per group). (D, E) Representative images of haematoxylin and eosin (H&E) and Masson's trichrome (Masson) staining of the lung sections from rats (D) and histological scores were shown (E) (n = 5 per group). (F) Levels of the cytokines interleukin-4 (IL-4), IL-5, IL-13 in the BALF from each group were measured by enzyme-linked immunosorbent assay (ELISA) (n = 5 per group). (G, H) Levels of epithelial cadherin (E-cadherin), vimentin, alpha-smooth muscle actin (α-SMA) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were analyzed by Western blot analysis (G) and statistical analysis was performed by densitometry (H) (n = 5 per group). (I) Procedure for establishing a fecal microbiota transplantation (FMT) model. (J) Levels of the cytokines IL-4, IL-5, IL-13 in the lung tissues from each group were measured by ELISA (n = 5 per group). (K, L) Representative images of H&E and Masson staining of the lung sections from rats (K) and histological scores were shown (L) (n = 5 per group). Scale bar, 100 μm. (M, N) Levels of E-cadherin, vimentin, α-SMA and GAPDH were analyzed by Western blot (M) and quantified by densitometry (N) (n = 5 per group). Data are expressed as mean ± standard error of the mean (SEM). i.p.: intraperitoneal; i.g.: intragastric; OVA: ovalbumin; ns: not significant.

To further elucidate whether the ameliorative effects of AG in the CVA rats are mediated by the gut microbiota, we transplanted the microbiota from AG-treated CVA rats to recipient CVA rats (FMT-OVA + AG) by daily gavage (Fig. 2I). Compared with CVA recipient rats receiving the gut microbiota from CVA donor rats (FMT-OVA), CVA recipient rats receiving the gut microbiota from AG-treated CVA donor rats replicated the effects of AG treatment on OVA-exposed symptoms (Figs. 2J–N). These changes were associated with attenuated inflammatory cell infiltration and collagen deposition in lung tissues, along with decreased levels of pro-inflammatory cytokines (IL-4, IL-5, and IL-13) and EMT-related proteins (vimentin and α-SMA), whereas E-cadherin expression was significantly upregulated. These findings suggested that fecal microbiota from AG-treated CVA donor rats exerted ameliorative effects against CVA similar to those observed in the AG-treated group. Hence, both the ABX treatment and FMT experiments mentioned above suggest that the gut microbiota contributes to the protective effects of AG in CVA-induced rats. Taken together, these results indicate that the beneficial effects of AG on CVA are modulated by gut microbiota.

3.3. AG reverses gut dysbiosis in CVA rats and enriches B. pseudolongum

Given that AG improves CVA in a gut microbiota-dependent manner, we next investigated how AG affected the gut microbiota composition by performing 16S rDNA gene amplicon sequencing on stool samples. Chao1 and Shannon indices were analyzed to evaluate the microbiota α-diversity. Compared with the OVA group, AG treatment remarkably increased the Chao1 index and Shannon index (Fig. 3A), indicating that AG could increase the reduced alpha diversity. In the principal component analysis (PCA), we observed a distinct clustering of the microbiota composition for CON, OVA, and OVA + AG treatment groups, with the OVA + AG treatment group showing a tendency to converge towards the CON group (Fig. 3B). Microbiota at the phylum was further analyzed. The results revealed that the major Bacteroidetes and Firmicutes showed no significant difference, but significantly the increased Actinobacteria abundance after AG treatment (Figs. 3C and D). Microbiota at the genus level (top 20) was also analyzed. Bifidobacterium was markedly enhanced after AG treatment (Figs. 3E and F). We observed no significant changes in the relative abundance of B. adolescentis, B. animalis, B. breve, B. catenulatum group, and B. longum in the CON, OVA and OVA + AG groups, but the relative abundance of B. pseudolongum was decreased in OVA-challenged rats and was significantly enriched after AG and SH treatment (Figs. 3G and S3A). To investigate the effects of AG on B. pseudolongum growth, we performed a growth curve analysis supplemented with 50, 100, and 200 μM of AG. The results showed that AG accelerated the growth of B. pseudolongum, suggesting that AG creates a conducive microenvironment that promotes the proliferation of B. pseudolongum (Fig. S3B). Given that the gut barrier dysfunction enables extraintestinal dissemination of microbial products (e.g., whole bacteria, LPS, toxins, etc.) which play important roles in the pathogenesis of many diseases [28], we examined the effect of AG treatment on intestinal barrier function in CVA rats. The protein levels of colonic tight junction markers (occludin and claudin-1) were significantly increased by AG treatment (Figs. 3H and I). In addition, the serum and fecal LPS concentrations were significantly decreased in AG gavaged rats (Fig. 3J). These findings suggest that AG can significantly enrich B. pseudolongum and restore the intestinal barrier function.

Fig. 3.

Arctigenin (AG) treatment significantly ameliorates gut microbiota dysbiosis and enriches Bifidobacterium pseudolongum (B. pseudolongum). The fresh stool from rats was obtained for 16S rDNA analyses. Fig. 2, Fig. 3 were conducted in the same experiment condition and the AG group is AG-30 mg/kg (A) Chao1 and Shannon diversity of gut microbiota among different groups (n = 6 per group). (B) Weighted UniFrac principal component analysis (PCA) plot analysis (n = 6 per group). (C) Bar plot of the phylum taxonomic level in the control (CON), ovalbumin (OVA) and OVA + AG groups (n = 6 per group). (D) Relative abundance of Actinobacteriota in the three groups (n = 6 per group). (E) Relative abundance of Bifidobacterium in the three groups (n = 6 per group). (F) Circle heatmap of the 20 most differentially abundant taxa in the three groups at the genus level (n = 6 per group). (G) Relative abundance of Bifidobacterium species in the feces of three groups, as determined by quantitative polymerase chain reaction (qPCR) (n = 5 per group). (H, I) Colon tissue relative protein abundance of tight junction proteins (occludin, and claudin-1) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were determined by Western blot (H) and quantified by densitometry (I) (n = 5 per group). (J) Levels of the serum and fecal lipopolysaccharide (LPS) were analyzed in the three groups by enzyme-linked immunosorbent assay (ELISA) (n = 5 per group). Data are presented as mean ± standard error of the mean (SEM). B. adolescentis: Bifidobacterium adolescentis; B. animalis: Bifidobacterium animalis; B. breve: Bifidobacterium breve; B. catenulatum group: Bifidobacterium catenulatum group; B. longum: Bifidobacterium longum; ns: not significant.

3.4. B. pseudolongum protects against CVA formation

At present, our research has observed an increase in the presence of B. pseudolongum in rats that received AG treatment. This suggests that B. pseudolongum could play a significant role in modulating the host's response to CVA. To determine the role of B. pseudolongum in the development of CVA, we orally administered live or heat-inactivated B. pseudolongum to CVA model rats via gavage once daily for 26 days, followed by an examination of CVA-related indicators by the protocol (Fig. 4A). We verified that the CVA rats gavaged with live B. pseudolongum produced good colonization (Fig. S4). Compared with OVA-exposed rats, fewer cough times and longer cough latencies were observed in B. pseudolongum-gavaged rats (Fig. 4B). Furthermore, B. pseudolongum-gavaged rats displayed a notable decrease in inflammatory cell infiltration in the BALF (Fig. 4C). H&E and Masson's staining revealed that B. pseudolongum treatment successfully reversed the obvious inflammatory cell infiltration and airway collagen deposition induced by OVA (Figs. 4D and E). Consistent with these findings, this protective effect was further supported by a significant reduction in the levels of secreted inflammatory cytokines (IL-4, IL-5, and IL-13) (Fig. 4F). B. pseudolongum-gavaged rats accompanied with ameliorated CVA-induced airway remodeling, as demonstrated by downregulated EMT-related proteins (vimentin and α-SMA) and upregulated E-cadherin expression. Interestingly, heat-inactivated bacteria did not improve the effect of CVA (Figs. 4G and H). Oral B. pseudolongum could also improve the intestinal barrier function in CVA rats, as indicated by the reduced concentrations of LPS in serum and feces and increased expression of intestinal tight junction proteins in colon (E-cadherin and claudin-1) (Figs. 4I–K). These findings suggest that B. pseudolongum offers protective effects against the development of CVA.

Fig. 4.

Bifidobacterium pseudolongum (B. pseudolongum) ameliorates cough, airway inflammation and remodeling in cough variant asthma (CVA) rats. (A) Schematic protocol for live B. pseudolongum (B.p) or heated-inactivated B. pseudolongum (HB.p) treatment on ovalbumin (OVA)-induced CVA rats. (B) Number of coughs and cough latencies were recorded in the control (CON), OVA, OVA + B. p and OVA + HB. p groups (n = 5 per group). (C) Number of inflammatory cells in the bronchoalveolar lavage fluid (BALF) were counted in the four groups (n = 5 per group). (D, E) Representative drawings of haematoxylin and eosin (H&E) and Masson's trichrome (Masson) staining of the lung sections in the four groups (D) and histological scores were shown (E) (n = 5 per group). (F) Cytokines expression levels of interleukin-4 (IL-4), IL-5, IL-13 in the different groups were measured by enzyme-linked immunosorbent assay (ELISA) (n = 5 per group). (G, H) Protein expression levels of epithelial cadherin (E-cadherin), vimentin, alpha-smooth muscle actin (α-SMA) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were analyzed in the CON, OVA, OVA + B. p and OVA + HB. p groups by Western blot (G) and quantified by densitometry (H) (n = 5 per group). (I) Levels of serum and fecal lipopolysaccharide (LPS) in the CON, OVA and OVA + B. p groups were analyzed by ELISA (n = 5 per group). (J, K) Levels of E-cadherin, claudin-1 and GAPDH in the colon tissues in the CON, OVA and OVA + B. p groups were analyzed by Western blot (J) and quantified by densitometry (K) (n = 5 per group). Data are presented as mean ± standard error of the mean (SEM). i.p.: intraperitoneal; i.g.: intragastric; ns: not significant.

3.5. AG treatment increases acetate production associated with the acetate-producing bacterium B. pseudolongum

In addition to promoting the growth of B. pseudolongum, we hypothesized that AG might affect the metabolism of B. pseudolongum and produce beneficial metabolites, thereby playing a role in the treatment of CVA. Given that B. pseudolongum is an SCFA-producing bacterium [29], we then performed targeted SCFA analysis in the B. pseudolongum supernatant and MRS control (Fig. 5A). Acetic acid was the highest metabolite in the conditional medium of acetate-producing B. pseudolongum (B. p CM) compared to the MRS control, as determined by targeted SCFA analysis (Fig. 5B). To verify whether B. pseudolongum could modulate acetate levels in vivo, we analyzed samples from feces, serum, and lungs. Compared with the CON group, acetate levels were depleted in OVA-induced rats. However, administration of live B. pseudolongum significantly increased acetate levels in all three tissues compared to the OVA group, whereas the heat-inactivated strain failed to elicit a significant increase (Fig. 5C). To further identify metabolites that respond to treatment with AG, acetate was measured between OVA-induced rats and AG-gavaged rats in the feces, cecal content, serum, and lung samples. Notably, treating AG rats caused the marked elevations of the acetate levels (Fig. 5D). We further also investigated the correlations between key CVA parameters (number of coughs, cough latencies, IL-4, IL-5, and IL-13) and fecal acetate content after B. pseudolongum colonization using Pearson's correlation analysis. Acetate levels were upregulated in B. pseudolongum-treated rats and were positively associated with cough latencies and negatively correlated with CVA features (number of coughs, IL-4, IL-5, and IL-13) (Figs. 5E and F). These results indicate that the anti-CVA effects were enhanced concomitantly with the enrichment of acetate, a key potential metabolite. Therefore, we further focused on acetate to investigate its relationship with CVA and to clarify the underlying mechanisms.

Fig. 5.

Bifidobacterium pseudolongum (B. pseudolongum) produces acetate. (A) A schematic diagram illustrating the cultivation procedure and gas chromatography-mass spectrometry (GC-MS) analysis of B. pseudolongum. (B) Concentrations of short-chain fatty acids (SCFAs) from the B. pseudolongum conditioned supernatant were measured using GC-MS (n = 3 per group). (C) Acetate concentration from the feces, serum and lungs in the control (CON), ovalbumin (OVA), OVA + B. pseudolongum (B.p) and OVA + heated-inactivated B. pseudolongum (HB.p) groups were quantified by GC-MS (n = 5 per group). (D) Level of acetate from the feces, cecal content, serum and lungs in the CON, OVA and OVA + arctigenin (AG) groups, respectively (n = 5 per group). (E, F) Pearson's correlation was conducted to examine the relationships between the levels of fecal acetate in the CON, OVA and OVA + B. p groups and five cough variant asthma (CVA) markers: number of coughs, cough latencies (E), interleukin-4 (IL-4), IL-5 and IL-13 (F). Data are presented as mean ± standard error of the mean (SEM). ns: not significant.

3.6. Acetate attenuates CVA-related phenotypes

Our animal data have shown a significantly increased level of acetate in the feces, serum, and lung of B. pseudolongum-treated subjects. We further evaluated the therapeutic effect of SA on the CVA model rat. The OVA-induced CVA rats were administered with SA in the drinking water for 13 days (Fig. 6A). The results showed that SA markedly reduced the number of coughs and extended cough latencies (Fig. 6B). The expression of pro-inflammatory type 2 cytokines, IL-4, IL-5, and IL-13, was significantly decreased in the OVA + SA group (Fig. 6C). Histological analysis using H&E and Masson's staining showed that the lungs from the SA treatment group had fewer pathological changes than the OVA control group (Figs. 6D and E). The administration of SA significantly downregulated EMT-related proteins (α-SMA and Snail) while upregulating E-cadherin expression (Figs. 6F and G), and could also improve the intestinal barrier function in CVA rats, as evidenced by the increased expression of intestinal tight junction proteins in the colon (E-cadherin and claudin-1) (Figs. 6H and I).

Fig. 6.

Acetate attenuates cough variant asthma (CVA) and gut barrier dysfunction. (A) Protocol for acetate (sodium acetate (SA), 200 mM in drinking water) treatment on the ovalbumin (OVA)-induced CVA rats. (B) Number of coughs and cough latencies were recorded in the control (CON), OVA and OVA + SA groups (n = 5 per group). (C) Cytokines expression levels of interleukin-4 (IL-4), IL-5, IL-13 in the different groups were measured by enzyme-linked immunosorbent assay (ELISA) (n = 5 per group). (D, E) Representative drawings of haematoxylin and eosin (H&E) and Masson's trichrome (Masson) staining of the lung sections in the three groups (D) and histological scores were shown (E) (n = 5 per group). (F, G) Protein expression levels of epithelial cadherin (E-cadherin), alpha-smooth muscle actin (α-SMA), snail family transcriptional repressor 1 (Snail) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were analyzed by Western blot (F) and quantified by densitometry (G) (n = 5 per group). (H, I) Levels of E-cadherin, claudin-1 and GAPDH in the colon tissues in the CON, OVA and OVA + SA groups were analyzed by Western blot (H) and quantified by densitometry (I) (n = 5 per group). (J, K) Expression levels of E-cadherin, vimentin, α-SMA and GAPDH in 16HBE cells incubated with SA and B. pseudolongum-conditioned medium (B. p CM) were assessed by Western blot (J) and quantified by densitometry (K) (n = 5 per group). Data are presented as mean ± standard error of the mean (SEM). i.p.: intraperitoneal; ns: not significant; TGF-β: transforming growth factor-β.

To further prove the functional significance of acetate produced by B. pseudolongum in CVA, we treated 16HBE cells with the B. p CM and SA. Consistent with the effects of B. p CM (5%), SA reduced EMT-related proteins (vimentin, α-SMA) and increased E-cadherin, thereby alleviating airway remodeling (Figs. 6J and K). Taken together, these findings suggest that the treatment of SA exerts strong anti-CVA effects.

3.7. Acetate administration attenuates OVA-induced CVA in rats by inhibiting IL-33 production

To identify the potential molecular targets for the beneficial effects of AG, we conducted RNA sequencing (RNA-seq) analysis using the lung tissues from the OVA and OVA + AG groups. According to the Kyoto Encyclopedia of Genes and Genomes (KEGG) functional pathway data analysis, cytokine-cytokine receptor interaction pathway was significantly affected (Fig. 7A). Previous studies have shown that IL-33 blockade could shut down the established type 2 response and reduce the expression levels of pro-inflammatory factors (IL-4, IL-5 and IL-13) to improve airway inflammation [30]. Meanwhile, IL-33 is critical for airway remodeling in chronic asthma, IL-33 deficiency alleviated disease severity and decreased EMT in asthma [17]. Based on the above research, we then analyzed the expression of IL-33. Volcano plot on the gene abundance found that IL-33 was significantly downregulated after AG treatment (Fig. 7B). The ELISA analysis validated the results from RNA-seq that IL-33 was downregulated in the OVA + AG and OVA + B. p groups (Figs. 7C and D). Based on the results above, we assumed that B. pseudolongum-derived acetate regulated the secretion of IL-33 to play an anti-inflammatory and antifibrotic role. Western blotting and immunofluorescence analysis showed that IL-33 was downregulated along with a marked upregulation of E-cadherin in the OVA + SA group (Figs. 7E–G). We further designed the TGF-β-induced 16HBE cell model to confirm the potential role of IL-33 in the beneficial effects of SA. The qPCR analysis validated that IL33 and its receptor ST2 were downregulated in cells cultured with SA, a result also confirmed by Western blotting (Figs. 7H–J).

Fig. 7.

Acetate inhibits interleukin-33 (IL-33) production in vivo and in vitro. (A) Kyoto Encyclopedia of Genes and Genomes (KEGG)-based functional analysis of RNA sequencing data (ovalbumin (OVA) vs. OVA + arctigenin (AG)). (B) Volcano plot of gene abundance based on the RNA sequencing data (OVA vs. OVA + AG). (C) Bronchoalveolar lavage fluid (BALF) expression level of IL-33 in the control (CON), OVA, and AG-treated cough variant asthma (CVA) rats were determined by enzyme-linked immunosorbent assay (ELISA) (n = 5 per group). (D) BALF expression level of IL-33 in the CON, OVA, and B. pseudolongum (B.p)-treated CVA rats were determined by ELISA (n = 5 per group). (E, F) Lung expression level of IL-33 in the CON, OVA, and sodium acetate (SA)-treated CVA rats were assayed by Western blot (E) and quantified by densitometry (F) (n = 5 per group). (G) Immunofluorescence images of IL-33 and epithelial cadherin (E-cadherin) in the lungs from the CON, OVA, and SA-treated CVA rats. (n = 5 per group). (H) Relative mRNA levels of IL33 and ST2 in the transforming growth factor-β (TGF-β) induced human bronchial epithelial (16HBE) cells cultured with SA were measured (n = 5 per group). (I, J) Protein expression of IL-33 in the TGF-β induced 16HBE cells incubated with SA or B. pseudolongum-conditioned medium (B. p CM) were assessed by Western blot (I) and quantified by densitometry (J) (n = 4 per group). (K, L) Protein expression levels of E-cadherin, vimentin, alpha-smooth muscle actin (α-SMA), IL-33 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in the TGF-β induced 16HBE cells incubated with SA or anti-IL33 were analyzed by Western blot (K) and quantified by densitometry (L) (n = 3 per group). Data are presented as mean ± standard error of the mean (SEM). DAPI: 4′,6-diamidino-2-phenylindole; ns: not significant.

Furthermore, we used anti-IL-33 mAb treatment to verify the mechanism underlying the anti-CVA effect of SA in TGF-β-induced 16HBE cells. The results showed that anti-IL-33 mAb treatment significantly reduced levels of EMT-related proteins (vimentin, and α-SMA) and IL-33 while upregulating E-cadherin compared to the TGF-β control. However, combined SA and anti-IL-33 mAb treatment did not further enhance these effects beyond anti-IL-33 mAb alone (Figs. 7K and L). These results indicate that the anti-CVA effects of SA may be partially attributable to inhibiting IL-33 expression.

3.8. Acetate binds to AHR for suppressing IL-33 expression in 16HBE cells

Finally, we revealed the underlying mechanism whereby SA suppressed IL-33 expression in 16HBE cells. GPR43, GPR109A, and GPR41, belonging to the GPCR family, critically regulate metabolic processes and inflammatory responses, contributing to disease pathogenesis [31], which have been identified as receptors for SCFAs. Next, we used specific GPCR activators to investigate whether the regulatory effects of acetate on IL-33 production in 16HBE cells were mediated through GPCRs activation. The results showed that 4-CMTB (GPR43 activator), niacin (GPR109A activator) and AR420626 (GPR41 activator) could not regulate the production of IL-33 in 16HBE cells (Fig. S5). Based on our findings, GPCR signaling may not contribute to SA-mediated suppression of IL-33 expression. However, studies have shown that propionate inhibits IL-33 production in keratinocytes by inhibiting HDAC2/3 or enhancing AHR nuclear translocation [32], and acetate is a weak HDAC inhibitor or lacks this effect [33], so we speculated that SA might regulate IL-33 production by acting on AHR. Similarly, SH has been demonstrated to alleviate post-infectious cough by activating the AHR [34]. The Human Transcription Factor Database (Human TFDB) and the PROMO database (PROMO) were utilized to predict potential transcription factors that might regulate IL-33 expression. In silico analysis identified 12 shared candidate transcription factors (Fig. 8A). Further sequence analysis predicted a highly conserved AHR binding motif within the IL33 promoter (Fig. 8B). Subsequent qPCR analysis confirmed that SA treatment significantly upregulated AHR mRNA expression (Figs. 8C and D), as well as the expression of the AHR-regulated genes, AHRR and CYP1A1 (Fig. 8E). Correspondingly, Western blot analysis results demonstrated that SA also increased the protein expression of AHR (Figs. 8F and G). This prompted investigation into AHR's role in SA-mediated IL-33 downregulation. Following siRNA-mediated knockdown of AHR (Figs. 8H and I), we observed that depletion of AHR eliminated SA-mediated suppression of IL-33 in 16HBE cells, suggesting AHR's essential role in this regulatory pathway. Additionally, the modulatory effects on airway remodeling-related protein (E-cadherin, vimentin and α-SMA) expression induced by SA and 5% B. p CM were also reversed (Figs. 8J and K). Imaging analysis demonstrated SA-induced AHR nuclear accumulation (Fig. 8L), which was inhibited by the AHR antagonist CH-223191. Given the presence of putative dioxin response elements (DREs) within the IL33 promoter, we constructed a luciferase reporter to examine whether AHR mediates the inhibitory effect of SA. The results showed that, in the presence of SA, AHR knockdown significantly restored IL33 promoter activity (Fig. 8M). Furthermore, ChIP assays demonstrated that SA treatment significantly increased AHR recruitment to the IL33 promoter in 16HBE cells (Fig. 8N). Collectively, these data suggest that SA inhibits IL-33 expression by promoting the direct binding of AHR to the IL33 promoter. The proposed regulatory mechanism is shown in Fig. 8O. Additionally, we investigated the effects of AG, B. p, and SH treatment groups on the AHR-IL-33 pathway in vivo and discovered that they were all capable of activating AHR and inhibiting the expression of IL-33 (Fig. S6).

Fig. 8.

Acetate suppresses interleukin-33 (IL-33) expression in 16HBE cells through inducing an increase in aryl hydrocarbon receptor (AHR) expression. (A) Potential transcription factors that bind to IL-33 promoter were predicted through the PROMO database (PROMO) and the Human Transcription Factor Database (Human TFDB). (B) Consensus DNA-binding motifs of IL-33 according to the PROMO, and the prediction of the binding sites of AHR within IL33 promoter. (C) Quantitative polymerase chain reaction (qPCR) analysis of the mRNA expression of potential transcription factors in 16HBE cells subjected to sodium acetate (SA) for 24 h (n = 5 per group). (D) mRNA levels of AHR from the transforming growth factor-β (TGF-β) induced 16HBE cells cultured with SA (2.4, 4.8, 9.6 mM) were quantified by qPCR (n = 5 per group). (E) qPCR analysis of the mRNA expression of aryl hydrocarbon receptor repressor (AHRR) and cytochrome P450 1A1 (CYP1A1) in the TGF-β induced 16HBE cells subjected to SA for 24 h (n = 5 per group). (F, G) Western blotting analysis of the protein expression of AHR in the TGF-β induced 16HBE cells treated with SA (4.8 mM) and 5% B. pseudolongum-conditioned medium (B.p CM) (n = 5 per group). (H, I) Western blotting analysis of the protein expression of AHR in the 16HBE cells treated with AHR-specific siRNA (n = 3 per group). (J, K) Western blotting analysis of the protein expression of epithelial cadherin (E-cadherin), vimentin, alpha-smooth muscle actin (α-SMA), IL-33 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in the TGF-β induced 16HBE cells treated with SA (4.8 mM) and 5% B. p CM after transfecting with AHR-specific siRNA (n = 3 per group). (L) Representative confocal images showing the localization of AHR in 16HBE cells treated with SA (4.8 mM) for different times with or without CH-223191. (M) Dual luciferase reporter assay of IL33 promoter activity in si-NC or si-AHR-transfected 16HBE cells subjected to SA for 24 h. Luciferase activity was normalized with the internal Renilla luciferase control (n = 3 per group). (N) Detection of AHR enrichment in the IL33 promoter region using chromatin immunoprecipitation (ChIP)-qPCR (n = 3 per group). (O) A schematic figure showing regulatory mechanism through which SA regulates IL-33. Created by BioRender. https://BioRender.com/m04h973. Results are expressed as mean ± standard error of the mean (SEM). AR: androgen receptor; MAZ: MYC-associated zinc finger protein; POU2F1: POU class 2 homeobox 1; POU2F2: POU class 2 homeobox 2; STAT5A: signal transducer and activator of transcription 5A; SRY: sex determining region Y; VDR: vitamin D receptor; STAT4: signal transducer and activator of transcription 4; FOXP3: forkhead box P3; YY1: Yin Yang 1; ATF3: activating transcription factor 3; DAPI: 4′,6-diamidino-2-phenylindole; ns: not significant.

We next investigated whether SA and AHR could combine with each other through molecular docking technology. SA bound to the active pocket of AHR with a binding energy of −5.28 kcal/mol (Fig. S7A). To corroborate the docking predictions, we employed two biophysical methods: DARTS for evaluating drug-target interactions and CETSA for analyzing thermal stability shifts. These results demonstrated that the combination of SA with AHR enhanced the stability of AHR and prevented it from being hydrolyzed by pronase E and precipitated at high temperature (Figs. S7B and C). Collectively, our data indicate that SA can bind to AHR and inhibit the expression of IL-33.

3.9. Acetate protects CVA rats by modulating the AHR-IL-33 pathway

The above in vitro validation found that the inhibition of key target IL-33 by acetate was in an AHR-dependent manner. For in vivo validation, we further designed AHR inhibitor (CH-223191)-induced CVA rats to confirm the potential role of AHR in the beneficial effects of SA (Fig. 9A). AHR inhibition reversed the cough-suppressing effects observed in the intervention groups (Fig. 9B). The alleviating effects on the production of inflammatory cells and the release of inflammatory factors were eliminated after AHR inhibitor treatment (Figs. 9C and D). Lung histopathological analysis revealed the attenuated therapeutic efficacy of SA in AHR-inhibited CVA model rats (Figs. 9E and F). The elevated expression of AHR and E-cadherin, as well as the reduced expression of EMT-related proteins (vimentin, and α-SMA) and IL-33, was also blunted in the AHR-inhibited group (Figs. 9G and H). We further employed multi-immunohistochemistry techniques to analyze the lung tissues from the CON, OVA, OVA + SA and OVA + SA + CH-223191 groups (Fig. 9I). The results indicated that the OVA + SA group exhibited abundant expression of AHR and E-cadherin proteins, while the expression of IL-33 was notably low. In contrast, following CH-223191 administration, there was a significant decrease in both AHR and E-cadherin levels, suggesting that SA downregulates IL-33 protein and modulates CVA-related phenotypes through the regulation of AHR.

Fig. 9.

Sodium acetate (SA) attenuated cough variant asthma (CVA)-related phenotypes with aryl hydrocarbon receptor-interleukin-33 (AHR-IL-33) as the key axis. (A) Schematic diagram of the SA treatment (200 mM in drinking water) in combination with selective AHR antagonist-CH223191 (10 mg/kg). (B) Number of coughs and cough latencies were recorded in the control (CON), ovalbumin (OVA), OVA + SA and OVA + SA + CH223191 groups (n = 5 per group). (C) Number of inflammatory cells in the bronchoalveolar lavage fluid (BALF) were counted in the four groups (n = 5 per group). (D) Cytokines expression levels of interleukin-4 (IL-4), IL-5, IL-13 in the different groups were measured by enzyme-linked immunosorbent assay (ELISA) (n = 5 per group). (E, F) Representative images of haematoxylin and eosin (H&E) and Masson's trichrome (Masson) staining of the lung sections in the four groups (E) and histological scores are shown (F) (n = 5 per group). (G, H) Protein expression levels of AHR, epithelial cadherin (E-cadherin), vimentin, alpha-smooth muscle actin (α-SMA), IL-33 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in the four groups were assessed by Western blotting analysis (G) and quantified by densitometry (H) (n = 5 per group). (I) Representative immunofluorescence images of AHR, IL-33 and E-cadherin in the lung sections in the four groups (n = 5 per group). Data are presented as the mean ± standard error of the mean (SEM). DAPI: 4′,6-diamidino-2-phenylindole; i.p.: intraperitoneal; i.g.: intragastric; ns: not significant.

In vitro, we went on to identify the effects of the highly selective and specific AHR inhibitor CH-223191 on the protein expression of AHR, epithelial mesenchymal transformation (E-cadherin, Vimentin and α-SMA), and IL-33 in 16HBE cells. Consistent with the in vivo results, the AHR inhibitor CH-223191 distinctly reversed the protective effects (Fig. S8). Altogether, the above data indicate that SA alleviates CVA-associated cough, airway inflammation, and remodeling by regulating the AHR-IL-33 pathway.

4. Discussion

CVA is a distinct subtype of asthma, akin to typical asthma, which is caused by the combined effects of immune, genetic, and environmental factors, leading to chronic airway inflammation, airway hyperresponsiveness, allergen sensitization, and airway remodeling. There is growing evidence that gut-lung interactions are closely involved in the development of asthma, paving the way for novel therapeutic strategies [35]. AG is one of the components detected in SH, which exhibited favorable activity during our primary screening of antitussive efficacy components of SH (unpublished data). Despite previous research indicating that AG is an effective anti-inflammatory and antioxidant agent, and its synergistic treatment with β2 AR agonists for asthma [36], the mechanism by which it improves CVA is not yet fully understood, and the potential role of AG through the gut microbiota has not been investigated.

In this study, we identified a potential dual mechanism for AG. We validated the direct effects of AG using 16HBE cells; the results showed that AG can modulate the AHR pathway to suppress IL-33 and inhibit the expression of airway remodeling-related proteins (Figs. S2 and S9), indicating that AG itself possesses anti-inflammatory and anti-fibrotic properties. However, given the known poor oral bioavailability of AG, the concentration reaching pulmonary tissues is likely limited, suggesting that its in vivo therapeutic effects are primarily mediated through indirect regulation rather than direct pulmonary targeting. Our study showed that AG and SH could significantly alleviate the symptoms of CVA in rats by regulating the gut microbiota, especially by enhancing the abundance of B. pseudolongum. Oral administration of B. pseudolongum and its metabolite, acetate, also significantly inhibited CVA-related symptoms. This supports the conclusion that while AG possesses direct anti-inflammatory potential, its efficacy in CVA is largely driven by the gut-lung axis. To our knowledge, this is the first report to demonstrate that B. pseudolongum is a beneficial bacterium in protecting against CVA.

The review reported the common connection between the gut microbiota and the development of pulmonary diseases [10]. The disruption of gut-lung axis crosstalk is associated with an increased susceptibility to airway diseases and infections, including asthma [37]. The role of oral probiotics in limiting allergic diseases is increasingly being emphasized, as they can help prevent or mitigate the progression of asthma, normalize gut permeability, and balance immune responses [38]. Consistent with these findings, supplementation with bifidobacteria has been reported to modulate systemic inflammation and immune dysregulation, while also reducing the risk of allergies, asthma, and other inflammatory diseases [39]. We observed that antibiotic treatment partially blocked the protective effect of AG on CVA, and the transfer of AG-treated microbiota to CVA-sensitized rats demonstrated a transferable protective effect against CVA. These data provided functional evidence that AG acts, at least in part, through microbiota-dependent mechanisms in CVA.

Although there have been reports that the gut microbiota can influence the development of asthma, the mechanisms are not yet fully characterized. Growing evidence suggests that circulating gut microbial metabolites and soluble components serve as key mediators in the gut-lung axis communication [40]. Propionate oral gavage prevents allergic asthma in mice via bone marrow reshaping and DCs functional modulation [12]. Dietary SCFA supplementation attenuated asthma by regulating the activity of T cells and DCs [41]. The microbial-associated metabolite desaminotyrosine prevents influenza by enhancing type I IFN signaling and reducing pulmonary immunopathology [42]. We found that B. pseudolongum itself could not reach the lungs after oral gavage. B. pseudolongum supernatant significantly inhibited EMT and the expression of IL-33, suggesting that metabolites secreted by B. pseudolongum might play an important role in mediating its anti-CVA effects. Based on targeted metabolomics analysis, it was found that acetate was consistently enriched as the top metabolite in the B. pseudolongum supernatant, as well as in the feces, plasma, and lung samples of rats gavaged with B. pseudolongum. Functional studies have confirmed that acetate could improve CVA-related phenotypes both in vivo and in vitro. These findings collectively suggested that acetate was one of the key beneficial metabolites produced by B. pseudolongum, which helped to suppress the development of CVA.

However, it is worth noting that acetate acts as a crucial, but likely not the sole, bioactive mediator. Recent studies have demonstrated that B. pseudolongum also produces indole derivatives, such as indole-3-acetic acid, which can activate the AHR signaling pathway to exert neuroprotective and immunomodulatory effects [43]. This suggests that the therapeutic efficacy of B. pseudolongum may stem from the synergistic action of its multi-metabolic profile, including both SCFAs (such as acetate) and tryptophan metabolites. Furthermore, the complexity of the gut ecosystem implies that other functional strains may also contribute to alleviating airway inflammation. For instance, Bifidobacterium infantis has been reported to attenuate allergic asthma by regulating the Th1/Th2 balance [44], while butyrate producers such as Roseburia hominis and Ruminococcus callidus significantly reduce airway hyperresponsiveness and eosinophilia [45]. Therefore, future studies should explore the potential synergistic interactions between B. pseudolongum and other functional microbes or metabolites to develop more comprehensive intervention strategies.

Ligand-specific activation of the AHR differentially regulates immune homeostasis and inflammatory cascades in diverse pathological contexts [46]. Furthermore, endogenous ligands, including metabolites of tryptophan and ligands provided by the commensal microbiota, can control immune homeostasis or inflammation during sustained inflammatory responses, such as asthma [47]. Therefore, microbiota-derived AHR ligands could be utilized for the effective treatment of CVA. Parallel to the previous findings where propionate was shown to alleviate skin inflammation by modulating AHR signaling to suppress IL-33 in keratinocytes [32], our study uncovers a distinct but analogous mechanism in the lung. IL-33, a classic alarmin critical for type 2 immunity, acts as a rapid initiator of allergic inflammation and airway remodeling [48]. Herein, we elucidate a specific “gut-metabolite-lung” axis. AG treatment enriches B. pseudolongum, leading to enhancing the systemic supply of acetate. Crucially, we provide mechanistic evidence that upon reaching the lung epithelium, acetate functions beyond indirect metabolic support; it acts as a potent AHR agonist. As confirmed by our ChIP and luciferase assays (Figs. 8M and N), acetate induces AHR nuclear translocation and promotes its direct binding to putative DREs within the IL33 promoter. This direct transcriptional repression effectively curbs the IL-33-driven allergic cascade, thereby ameliorating CVA pathology.

While our study demonstrates that AG ameliorates CVA through modulation of the gut-lung axis, several limitations must be acknowledged. First, although the OVA-induced model effectively recapitulates Th2-driven eosinophilic inflammation, it may not fully capture the heterogeneity inherent to human CVA, particularly regarding immunological mechanisms; consequently, our findings primarily elucidate the underlying inflammatory pathways. Second, the specific causal contribution of B. pseudolongum should be further validated using germ-free mouse models. Third, the lack of comprehensive (untargeted) metabolic profiling limits the identification of other potential effector metabolites; it remains possible additional molecules produced by B. pseudolongum or induced by AG treatment act synergistically with acetate to mediate protective effects. Finally, our study was conducted exclusively in male rats to minimize the confounding influence of hormonal fluctuations. Asthma exhibits significant sexual dimorphism, with females often displaying distinct immune profiles, such as enhanced adaptive immunity (e.g., T cell activation) and eosinophilic infiltration, compared to the innate-dominant responses in males [49,50]. Consequently, relying solely on male animals may limit the full representation of clinical heterogeneity, particularly mechanisms relevant to female patients [51]. To address these gaps, future studies will incorporate both male and female animals, clinical samples, and integrated multi-omics approaches, including untargeted metabolomics, to enhance the mechanistic depth and clinical translation of our findings.

In conclusion, our results demonstrated that the lignan component AG could significantly alleviate airway inflammation and remodeling in CVA in a gut microbiota-dependent manner, with effects comparable to SH, confirming its role as one of the effective components of SH. AG was found to improve microbial imbalance and intestinal barrier dysfunction, significantly increasing the abundance of B. pseudolongum, which is markedly depleted in CVA and has been found to reduce CVA symptoms through mono-colonization. Additionally, B. pseudolongum offered protection against CVA by producing acetate. Mechanistically, supplementation of CVA rats with acetate could reduce CVA-associated inflammation and EMT by activating AHR to decrease the secretion of IL-33 (Fig. 10). The current work proposes a novel mechanism involving the gut-lung axis for the protective effects of AG on CVA, and our results suggest that targeting key metabolites or intestinal bacteria may provide a promising therapeutic approach for the treatment of CVA.

Fig. 10.

Proposed mechanisms of arctigenin (AG) in Suhuang antitussive capsule (SH) for treating cough variant asthma (CVA). Created by BioRender. https://BioRender.com/pnzke09). AHR: aryl hydrocarbon receptor; B. pseudolongum: Bifidobacterium pseudolongum; CVA: cough variant asthma; IL-33: interleukin-33; LPS: lipopolysaccharide; Suhuang: Suhuang antitussive capsule.

CRediT authorship contribution statement

Hong Jiang: Writing – review & editing, Writing – original draft, Investigation, Formal analysis, Conceptualization. Yongxiang Wang: Methodology, Conceptualization. Qimeng Tian: Investigation, Formal analysis. Yihang Sui: Investigation. Jingge Sun: Investigation. Ling Chen: Formal analysis. Xiaoqiong Liu: Investigation. Ninghua Tan: Writing – review & editing, Supervision, Resources, Project administration, Funding acquisition, Conceptualization.

Data availability

The experimental datasets supporting this work can be reasonably available from the corresponding author for legitimate research purposes.

Declaration of competing interest

Yongxiang Wang is an employee of Yangtze River Pharmaceutical Group Co., Ltd. and contributed to the conceptualization and methodology of this study as an author. The company provided the Suhuang antitussive capsules (Chinese National Drug Standard No. YBZ00172008) used in this study free of charge but had no role in data collection, data analysis, interpretation of the results, manuscript preparation, or the decision to submit the manuscript for publication. All other 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 is the Supplementary data to this article: