Primary SOX9⁺ airway basal cells, accounting for ~1% of the basal cells, were isolated from patients with IFP as reported in a previous study (15) and cultured in DMEM/F12 medium (Gibco; Thermo Fisher Scientific, Inc.) supplemented with 10% fetal bovine serum (FBS; Cytiva), 1% amphotericin, antibiotics and a growth factor cocktail, as previously described (19). Human bronchial epithelial cell line BEAS-2B (cat. no. SCSP-5067) and human embryonic fibroblasts MRC-5 (cat. no. GNHu41) were purchased from the Cell Bank of Chinese Academy of Sciences. BEAS-2B and MRC-5 were cultured with DEME medium (Gibco; Thermo Fisher Scientific, Inc.) and MEM medium (Gibco; Thermo Fisher Scientific, Inc.), respectively, both of which were supplemented with 10% FBS and 1% amphotericin. All cells were maintained in an humidified incubator with 5% CO₂ at 37˚C.

SOX9⁺ airway basal cells (within 10 passages) were cultured with exosome-free FBS for 48 h. Then, the resulting supernatant was collected and filtered using a 0.45-µm film to remove cell fragments. Next, Qiagen exoEasy Maxi Kit (Qiagen GmbH) was used to extract exosomes, according to the manufacturer's instructions. The number and size distribution of the resulting exosomes was analyzed using NanoSight NS300 (Malvern Instruments, Inc.). Transmission electron microscopy (TEM) was used to observe the morphology of the exosomes. Briefly, a drop of the diluted exosomes were loaded on a cooper mesh and fixed with 2.5% glutaraldehyde for 5 min at room temperature. After an incubated with 1% phosphotungstic acid for 10 min at temperature, the sample was used for TEM imaging. The exosomal positive protein markers [CD63 and tumor susceptibility gene 101 protein (TSG101)] and negative marker (β-Tubulin) were analyzed using western blotting.

BEAS-2B and MRC-5 were treated with 1 ng/ml TGF-β1 for 24 h before culturing with/without exosomes for an additional 24 h. Cells without any treatment were used as the control group. Next, the cells were lysed with RIPA buffer (Beyotime Institute of Biotechnology) and the proteins in the supernatant were quantified using a BCA kit (Beijing Solarbio Science & Technology Co., Ltd.) after the lysates had been centrifuged at 12,000 x g for 5 min at 4˚C. Then, 10 µg of protein was added and separated in 10% SDS-PAGE and the proteins in the gels were transferred onto PVDF membranes. Subsequently, membranes were blocked with 5% fat-free milk for 1 h at room temperature, which were then incubated with primary antibodies at 4˚C overnight, followed by the secondary antibody for 2 h at room temperature. Finally, the signals were enhanced using ECL Plus Western blotting system kit (Beyotime Institute of Biotechnology). The primary antibodies, namely ANO1 (cat. no. 14476; 1:1,000), vimentin (cat. no. 5741; 1:1,000), E-cadherin (cat. no. 3195; 1:1,000), CD63 (cat. no. 52090; 1:1,000), TSG101 (cat. no. 72312; 1:1,000), β-Tubulin (cat. no. 2146; 1:1,000), Fibronectin (FN1; cat. no. 26836; 1:1,000), Collagen I A1 (COL1A1; cat. no. 72026; 1:1,000), GAPDH (cat. no. 5174; 1:1,000) and secondary antibody HRP-labeled IgG (cat. no. 7074; 1:3,000), were obtained from Cell Signaling Technology, Inc. Gray values of the blot bands were analyzed using ImageJ software (version 1.8.0; National Institutes of Health).

After treating the cells (BEAS-2B and MRC-5) with 1 ng/ml TGF-β1 for 24 h or further treatment with 5x10⁸ particles/ml exosomes for 24 h at 37˚C, the total RNA of the four samples was extracted using RNeasy mini kit (Qiagen GmbH). Integrity of total RNA was assessed using the Agilent 2100 Bioanalyzer (Agilent Technologies Inc.). Paired-end libraries were synthesized by using the Stranded mRNA-seq Lib Prep Kit (cat. no. RK20301; ABclonal Biotech Co., Ltd.) in accordance with the manufacturer's protocols. Purified libraries were quantified by Qubit 3.0 Fluorometer (Thermo Fisher Scientific, Inc.) and validated by Agilent 2100 bioanalyzer (Agilent Technologies Inc.) to confirm the insert size and calculate the mole concentration. Cluster was generated by cBot (Illumina, Inc.) with the library diluted to 10 pM and then were sequenced on the Illumina NovaSeq 6000 (Illumina, Inc.). cDNA library construction and RNA sequencing were performed by Shanghai Biotechnology Corporation. Nucleotide length and the direction of sequencing were 150 bp and paired-end respectively. The raw data was deposited in the SRA database and accessible at https://www.ncbi.nlm.nih.gov/sra/PRJNA1051029.

Stringtie software (v1.3.3b; The Center for Computational Biology, Johns Hopkins University) was used to count the fragment within each gene and TMM algorithm was used for normalization. Differential expression analysis for mRNA was performed using the edgeR package in R (version 3.4.3; http://www.R-project.org/) (20). Fragments Per Kilobase Million (FPKM) was used to indicate the expression level of gene. Fold-change (FC) was calculated using the ratio of the mean FPKM of TGF-β1 + exosome group to that of TGF-β1 group. The dysregulated genes in TGF-β1 + exosome group were screened using the threshold of |log2(FC)| value >1(21). The P-value was not taken as the threshold, as the sample number in each group was less than three.

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis were conducted to determine the biological role of the dysregulated genes via the enrich package in R (version 3.4.3; http://www.R-project.org/) (20). Rich factor was calculated using the ratio of dysregulated_gene_in_this_pathway/dysregulated_gene_in_all_pathway to all_gene_in_this_pathway/all_gene_in_all_pathway. The P-value was used to screen the significant GO terms or KEGG pathways and the top 30 terms or pathways were further screened by rich factor to draw the corresponding bubble chart.

The total RNA of the 5x10⁵ cells was extracted using a RNeasy mini kit (Qiagen GmbH) according to manufacturer's protocols and quantified on a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies; Thermo Fisher Scientific, Inc.). Next, 500 ng of RNA was transcribed into cDNA using a ReverTra Ace qPCR RT kit (Toyobo Life Science) following manufacturer's instructions. Then, 2 µl of cDNA (n=3) was mixed with 10 µl of Master SYBR Green I mix (Roche Diagnostics), 1 µl of primers and water to obtain a mixture of 20 µl, which was further reacted on an ABI StepOnePlus Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). The reaction was performed at 95˚C for 5 min, followed by 40 cycles of 95˚C for 10 sec and 60˚C for 60 sec according to manufacturer's protocols. The relative expression levels of mRNA were calculated using the 2^-ΔΔCq method (22) and GAPDH was used as the internal control. The sequences of the primers use are listed in Table I.

The overexpression vector of ANO1 (pcDNA3.1-ANO1) and its short interfering (si)RNA (si-ANO1) were both synthetized by Sangon Biotech Co., Ltd. Next, the overexpression vector (2.5 µg) and si-ANO1 (2 µg) were transfected into 5x10⁵ cells (BEAS-2B and MRC-5) using Lipofectamine^® 3000 (Invitrogen; Thermo Fisher Scientific, Inc.) at 37˚C for 48 h, according to the manufacturer's instructions. The control vector and the si-RNA were also transfected into 5x10⁵ cells (BEAS-2B and MRC-5) as negative control (NC) and si-NC, respectively. Subsequently, cells transfected with vectors were screened by G418 for 10 days, followed by qPCR validation. Cells transfected with siRNAs were directly detected by qPCR to confirm the interference efficiency. si-NC: 5'-UUCUCCGAACGUGUCACGUTT-3', si-ANO1: 5'-CGTGTACAAAGGCCAAGTA-3'.

Cells were collected and resuspended in FBS-free medium. Then, 100 µl of cells at a density of 4x10⁵ cells/ml were seeded into the upper chamber and 600 µl of the complete medium was added into the lower chamber. After 24 h of migration, the residual cells on the Inner layer of the membrane were removed and the migrated cells were fixed with methanol for 20 min at room temperature, followed by staining with 0.1% crystal violet for 15 min at room temperature. Finally, the migrated cells were observed and counted under an inverted microscope.

Male C57BL/6 mice (6-8 weeks; 18-22 g) were purchased from Shanghai Laboratory Animal Research Center. All mice were maintained in 12-h light/dark cycle and with free access to food and water. The temperature and relative humidity were maintain at 22±2˚C and 50-60%, respectively. The 18 mice were randomly divided into 3 groups (n=6): Control group, model (bleomycin treatment) group and exosome group (bleomycin as well as exosome treatments). Pulmonary fibrosis was induced with a single intratracheal injection of 2 U/kg bleomycin (Nippon Kayaku Co., Ltd.; cat. no. H20090885) in 30 µl saline as reported (23). Mice administered with same volume of saline were served as controls. At 10 days post bleomycin treatment, exosomes (10x10⁹ particles/kg) were given for ~30 min/day for 7 consecutive days using a nebulizer (Trek S Portable Compressor Nebulizer Aerosol System; PARI GmbH). At 21 days post bleomycin treatment, all mice were anesthetized by intraperitoneal injection of chloral hydrate (350 mg/kg), then sacrificed by cervical dislocation. Lung tissues were collected for protein detection and histological examination. All experimental procedures were approved by the Animal Welfare Ethics Committee of Shanghai Sixth People's Hospital (approval no. 2023-0374).

Pulmonary tissue samples were fixed with 4% paraformaldehyde for 48 h at room temperature, followed by dehydration in graded alcohol and embedding in paraffin. Embedded tissue samples were then cut into 5 µm thick sections and stained with hematoxylin and eosin (H&E) for 5 min or Masson's trichrome reagents for 30 min at room temperature. Images of the histologic sections were captured using light microscopy.

Data are presented as the mean ± standard deviation from three independent experiments unless otherwise noted. All statistical analyses were performed using GraphPad Prism 7.0 (Dotmatics). Statistical significance was analyzed by unpaired two-tailed Student's t-test, and one-way ANOVA with Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference.

First, the exosomes were extracted from the medium of airway basal cells and the nanoparticle tracking analysis indicated that the majority of the particles had an average diameter of 145 nm, with some showing a larger size (Fig. 1A). The TEM image (Fig. 1B) showed that the obtained particles were round-like and had a size of ~100 nm. Furthermore, exosome makers CD63 and TSG101 were both expressed in these particles, while cellular maker β-Tubulin could not be detected (Fig. 1A, right). These results indicate that the extracellular vesicles were nanoparticles, confirming that they were exosomes.

The resulting exosomes were used to reverse the EMT of activated lung cells. As shown in Fig. 2A, most of the BEAS-2B cells were long shuttle after 1 ng/ml TGF-β1 treatment, then became short shuttle or polygon after further treated with exsomes. Similarly, the morphology of the MRC-5 in TGF-β1 group became more slender compared with the control cells and further treatment of exosomes could reverse this morphological change. Western blot analysis indicated that TGF-β1 stimulation resulted in the significant upregulation of vimentin and the downregulation of E-cadherin in the two types of cells (Fig. 2B and C), reflecting the occurrence of EMT. Notably, the morphological alterations of the two lung cells induced by TGF-β1 were both reversed after further treatment with exosomes. Consistently, the EMT of the two types of cells was also significantly inhibited in the TGF-β1 + exosomes group compared with the TGF-β1 group (Fig. 2B and C). These results suggest that airway basal cell-derived exosomes could relieve the EMT of lung epithelial cell and fibroblasts activated by TGF-β1.

To determine the underlying mechanism of the exosomes, the cells in the TGF-β1 and TGF-β1 + exosomes groups were collected and analyzed by RNA sequencing. As shown in Fig. 3A, although the expression profile of the four samples (without any parallels) varied, the fold-change of most genes was consistent in the TGF-β1 + exosomes treated cells compared with the corresponding cells treated with TGF-β1. A total of 4,158 dysregulated genes, including 1,819 upregulated and 2,339 downregulated genes, were screened under the threshold of |log2FC| value >1 (Fig. 3B). The P-value was not taken as the threshold, as the sample number in each group was <3.

As shown in Table II, among the top 20 genes with the highest FC (abs) values, AP005263.1 (unknown gene) had the highest FC (abs) value (1,442), followed by KRT81, PCDHGA12, ANO1 and CXCL8, sharing the FC (abs) values varying from 100-370. The P-value of the 20 genes were also presented and were all <0.05, indicating that the FC (abs) values were consistent in the two independent cells in the TGF-β1 + exosomes group (compared with the cells in TGF-β1 group).

GO enrichment analysis indicated that the dysregulated genes were enriched in various metabolism processes of quinolinate, glycine and tryptophan (Fig. 4), which were regulated by HAAO, KYNU, ACMSD and KMO (Table SI). Some genes (WNK2, WNK3, ATP1B2 and WNK3) were involved in the biological processes of ion import (Fig. 4, Table SI). The KEGG enrichment results also showed that a number of the pathways were involved in the metabolism of various amino acids and some other molecules, including thiamine, butanoate and arachidonic acid (Fig. 5). In addition, 51 genes were enriched in cytokine-cytokine receptor interaction and 18 genes were involved in TGF-β signaling pathway, including downregulated NOG, FST, GDF5, BAMBI and BMPR1B (Fig. 5; Table SII).

Another two migration-related GO terms and KEGG pathways were also extracted, since little evidence could be found associated with EMT in the top 30 terms and 30 pathways. As shown in Table III, 110 of the genes were enriched in the regulation of cell motility and 22 genes were involved in the pathway responsible for the regulation of the actin cytoskeleton. Next, nine overlapped genes (in red, Table III) were obtained through comparing the genes in the two collections, including three upregulated genes (CXCL12, ITGA2 and BDKRB1) and six downregulated ones (FGF5, FGF22, FGF17, PDGFD, FGF1 and FGF18). These may might be involved in the EMT process reversed by exosomes.

Since the top 20 dysregulated genes had much higher FC (abs) values, the first five known genes, excluding the unknown gene AP005263.1, were selected for qPCR validation. As a result, the log2FC values of four genes were consistent in the two lung cells. The FC (abs) values of ANO1 in the two cells were both higher than 4, whereas the value of the other three genes in BEAS-2B or MRC-5 was always lower than 4 (Fig. 6A). Therefore, ANO1 was selected for further analysis. The qPCR and western blotting results showed that ANO1 was successfully overexpressed in the two types of cells. As shown in Fig. 6B-D, the mRNA expression level was enhanced by 2-3 fold and the ANO1 protein level was also significantly increased. Further assays indicated that the migration abilities were both significantly enhanced after the overexpression of ANO1, along with increased levels of vimentin and downregulated levels of E-cadherin (Fig. 6E-G). These results suggest that the overexpression of ANO1 could induce the EMT and enhance the migration ability of lung cells.

To further confirm the role of ANO1, its expression was evaluated in lung cells. The qPCR results indicated that the expression level of ANO1 was reduced by ~60% (Fig. 7A). Consistently, its protein level was also significantly decreased (Fig. 7B and C). Next, ANO1 knockdown cells were further treated with TGF-β1 for 24 h. The Transwell results indicated that the migration ability of BEAS-2B in the si-NC + TGF-β1 group showed a significant enhancement compared with si-NC group, whereas the knockdown of ANO1 caused a weakness of migration ability compared with those in the si-NC + TGF-β1 group (Fig. 7D). The migration ability of MRC-5 also showed a similar trend (Fig. 7D). In addition, the knockdown of ANO1 in two lung cells also inhibited the EMT induced by TGF-β1 (Fig. 7E and F). These results suggest that the downregulation of ANO1 induced by exosomes is an important mechanism for the suppression of the TGF-β1-induced EMT by exosomes.

H&E and Masson's trichrome staining results indicated that, bleomycin treatment resulted in interstitial thickening, infiltration of inflammatory cells and collagen deposition (indicated by arrows; Fig. 8A and B). Typical fibrosis progression markers COL1A1 and FN1 were both significantly enhanced in model group, compared with control group (Fig. 8C and D). In addition, a significant increase of vimentin expression level as well as the reduction of E-cadherin was also found following bleomycin treatment (Fig. 8C and D), which indicated the EMT of lung cells and the progression of fibrosis. Notably, bleomycin treatment also caused a significant upregulation of ANO1 (Fig. 8C and D).

Following exosome treatment for 7 consecutive days, the fibrosis level was reduced, with fewer interstitial thickening and collagen deposition were observed in exosome group (indicated by arrows; Fig. 8A and B). Consistently, the protein levels of fibrosis makers were both decreased and the EMT level was also alleviated (Fig. 8C and D). Furthermore, exosome treatment also reduced the expression level of ANO1. These results indicated that airway basal cell-derived exosomes could ameliorate pulmonary fibrosis and inhibited the upregulation of ANO1, basically consistent with the in vitro results.