A serum-free culture medium for long-term expansion of human airway basal stem cells

Abstract

Background

Human basal stem cells, with their self-renewal and multilineage differentiation capacity, are essential tools for modeling airway diseases and advancing regenerative medicine. However, existing culture systems often rely on undefined components like serum, bovine pituitary extract, or feeder cells, limiting reproducibility and clinical translation. To address this limitation, we developed a well-defined culture medium that enables the long-term expansion of human airway basal stem cells while preserving their proliferative and differentiation potential.

Methods

We formulated a novel medium (sfBSC) comprising 14 defined components, including basal media, supplements, growth factors (EGF, FGF10), and signaling inhibitors (Y-27632, A-83-01, DAPT, DMH1). Human bronchial and small airway epithelial cells were cultured over multiple passages in sfBSC and assessed for morphology, population doublings, marker expression, and differentiation capacity via air-liquid interface and organoid cultures. RNA-seq was performed to explore molecular changes across passages.

Results

Bronchial and small airway epithelial cells were expanded up to 17 and 24 passages, respectively, with stable morphology and consistent cell size (10–15 μm). Cells maintained expression of canonical BSC markers (TP63, KRT5, NGFR) throughout long-term culture. Differentiation assays confirmed the ability to generate ciliated, goblet, and club cells. Optimized concentrations of EGF (1 ng/mL) and FGF10 (0.4 ng/mL) were critical for sustained proliferation. RNA-seq revealed stable marker expression and metabolic changes over time.

Conclusions

sfBSC medium offers a defined, reproducible, and scalable platform for basal stem cell culture, enabling applications in disease modeling, regenerative medicine, and clinical-grade cell production.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13287-026-04910-z.

Background

Within the complex architecture of the respiratory system, basal stem cells (BSCs), a specialized type of epithelial cell, reside in the basal layer of the airway epithelium [1, 2]. These cuboidal cells are distributed throughout the human airways, with their abundance varying along the proximal-to-distal axis. In larger airways, with diameters exceeding 4 millimeters, BSCs account for approximately 31% of the epithelial cell population [3]. In contrast, in airways with diameters less than 0.5 millimeters, their prevalence decreases to approximately 6% [4]. BSCs are multipotent, undifferentiated cells capable of self-renewal and proliferation. They can also differentiate into various specialized cell types, including ciliated cells, goblet cells, and club cells, thereby playing a critical role in maintaining epithelial homeostasis. This regenerative capacity is essential for preserving the integrity and function of the airway epithelium, particularly following injury [5, 6]. Consequently, BSCs are indispensable for epithelial repair and are central to the pathogenesis and progression of many airway diseases, including asthma [7, 8], chronic obstructive pulmonary disease (COPD) [9, 10], and lung cancer [11]. Given their pivotal role in airway homeostasis and disease progression, BSCs have become an invaluable tool for studying airway development and disease pathophysiology.

The growing demand for high-quality BSCs in both basic research and clinical applications has spurred ongoing efforts to refine culture systems. However, conventional BSC culture methods often rely on complex, undefined components, such as serum and bovine pituitary extract (BPE), to sustain cellular growth and composition [12–16]. Additionally, these methods typically require the use of feeder layer cells or conditioned medium to maintain BSC viability and support expansion [17, 18]. Despite these strategies, prolonged passaging significantly diminishes the ability of BSCs to differentiate into ciliated epithelial cells [19]. Furthermore, the complexity of these culture systems extends experimental timelines and substantially increases costs. The reliance on undefined components also hampers efforts to elucidate the molecular mechanisms underlying BSC stemness, thereby limiting both fundamental research and translational applications. These challenges highlight the urgent need for a more advanced, simplified, and compositionally defined culture system that enhances the efficiency, reproducibility, and clinical utility of BSC cultures.

To address these limitations, we developed a serum- and feeder-free culture system with a well-defined composition that supports the long-term expansion of human airway BSCs while maintaining their differentiation potential and morphological integrity. This optimized culture method not only increases the availability of primary cells but also facilitates large-scale experimentation and provides a more robust platform for advancing the understanding of BSC biology.

Methods

HBEC and HSAEC culture

Two healthy BSC lines were used in this study: human bronchial epithelial cells (HBECs, Lifeline, FC-0035) and human small airway epithelial cells (HSAECs, ATCC, PCS-301-010). Both cell types were cultured in T25 flasks with sfBSC medium. For comparison, cells were also cultured in the commercial PneumaCult™-Ex Plus Medium (STEMCELL Technologies, 05040). No feeder cells or coating treatments were necessary. The medium was changed every two days, and the cells were passaged when they reached 70% − 80% confluence. The cell morphology was captured via brightfield microscopy before each passage. The cells were digested with accutase (Sigma-Aldrich, A6964) at 37 °C for 10 min. After neutralization with an equal volume of medium, the cells were collected and centrifuged at 200 × g for 3 min. Cell counting and size measurement were performed via an Automatic Cell Fluorescence Analyzer (Countstar).

Patient-specific BSC isolation and culture

A disease-specific BSC line was derived from a patient with COPD via bronchoscopic brush biopsy of the airway epithelium in the main bronchus. All procedures involving cell acquisition were conducted in accordance with protocols approved by the Institutional Review Board of The First Affiliated Hospital of Guangzhou Medical University. The subject provided full informed consent prior to bronchoscopy. Following this procedure, the brush was retrieved in a 50 mL centrifuge tube, and the collected epithelial cells were washed with sfBSC medium under sterile conditions. The wash mixture was centrifuged, and the cell pellet was resuspended in culture medium. The suspension was then transferred to a T25 cell culture flask. The cells were cultured at 37 °C and 5% CO₂. The medium was replaced daily to remove red blood cells and other impurities, ensuring the purification of the BSCs. The cells were maintained until they reached approximately 50% − 70% confluence before passage.

SfBSC medium

The newly developed culture medium, with clearly defined components, consisted of a basal medium, supplements, growth factors, and inhibitors. The base medium was composed of a 1:1 mixture of Iscove’s Modified Dulbecco Medium (IMDM, Gibco, 12440061) and Ham’s F12 (Gibco, 11765054). The complete sfBSC medium was supplemented with the following components: 1X B27 supplement (Gibco, 17504044), 0.1% recombinant human serum albumin (rHSA, Sigma-Aldrich, A9731), 2 mM Glutamax (Gibco, 35050061), 50 µg/mL L-ascorbic acid (Sigma-Aldrich, A4544), 0.4 µM monothioglycerol (MTG, Sigma-Aldrich, M6145), 1 µg/mL hydrocortisone (STEMCELL Technologies, 07926), 1 ng/mL recombinant human EGF (Sigma-Aldrich, SRP3027), 0.4 ng/mL recombinant human FGF10 (R&D systems, 345-FG), 10 µM Y27632 (Rho-associated protein kinase (ROCK) inhibitor, Selleck, S1049), 1 µM DAPT (γ-secretase/Notch signaling pathway inhibitor, Sigma-Aldrich, D5942), 1 µM A83-01 (ALK5 inhibitor, Sigma-Aldrich, SML0788), and 1 µM DMH-1 (BMP signaling inhibitor, Sigma-Aldrich, D8946). Inhibitor powders were dissolved in DMSO, aliquoted, and stored at -20 °C. Cytokines were diluted in PBS containing 5% trehalose, aliquoted, and stored at -80 °C. The complete sfBSC medium can be stored at 2–8 °C for up to four weeks. For comparison, other basal media evaluated in this study included Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F-12; Gibco, 11320033), DMEM (Gibco, 11965092), and Roswell Park Memorial Institute Medium 1640 (RPMI 1640; Gibco, 11875119).

Cumulative population doublings (cPDs)

The number of inoculated cells and the number of cells harvested during the expansion culture process were documented. The cumulative population doublings were calculated according to the ATCC recommendation via the following formula:

where:

• n = the final population doubling level number at the end of a given subculture.

• Xe = the cell yield at that point.

• Xb = the number of cells used as the inoculum to begin subculture.

• S = the starting population doubling level.

Flow cytometry

Cells from different passages of HBECs and HSAECs were harvested by enzymatic dissociation and resuspended in PBS. For surface marker detection, cells were incubated with APC-conjugated anti-CD271 (NGFR) (mouse, BioLegend, 345108) for 30 min at 4 °C in the dark. After surface staining, cells were fixed with 4% paraformaldehyde (PFA; Alfa Aesar, 043368), and permeabilized using an enhanced permeabilization solution (Beyotime, P0097) for 10 min. For intracellular staining, cells were subsequently incubated with anti-P63 (rabbit, Abcam, ab124762) for 30 min, followed by FITC-conjugated anti-rabbit IgG secondary antibody (Abcam, ab150077) for 30 min. Between each incubation step, cells were washed and centrifuged to remove unbound antibodies. Appropriate isotype controls were included: APC-conjugated IgG1,κ isotype control (mouse, BioLegend, 400122) and IgG isotype control (rabbit, Invitrogen, 31235). Samples were analyzed using a CytoFLEX S flow cytometer (Beckman), and data were processed with FlowJo software (version 10).

Cell proliferation assays

HBECs were enzymatically dissociated, centrifuged, and counted prior to seeding. For the CCK-8 assay, 2000 cells per well were seeded into 96-well plates, and cultured in different media formulations. Plates were incubated in a 37 °C, CO₂ incubator, and cell viability was assessed at 0, 1, 3, and 5 days. At each time point, 10 µL of CCK-8 reagent (Beyotime, C0042) was added to each well, followed by a 2-hour incubation. Absorbance was measured at 450 nm using a SpectraMax Paradigm microplate reader (Molecular Devices).

For the growth curve analysis, 6 × 10⁴ cells per well were seeded in 12-well plates, and cultured 1, 3, 5, and 7 days in different media. At each time point, cells were harvested and counted using an automatic cell fluorescence analyzer (Countstar).

For the colony formation assay, 200 cells were seeded into 6-well plates and cultured for 8 days, with medium changes every 2 days. Colonies were fixed with 4% PFA for 30 min, stained with crystal violet (Beyotime, C0121) for 10 min, and washed three times with PBS. Plates were air-dried, imaged, and colony numbers were quantified using ImageJ software [20].

Air‒liquid interface (ALI) culture

HBECs or HSAECs were seeded onto collagen IV-coated (Sigma‒Aldrich, 35533) transwell inserts (Corning, 3460, 12 mm with 0.4 μm pore) at a density of 3 × 10⁴ cells per well. The upper chamber contained 500 µL of medium, and the lower chamber contained 1 mL of medium. Cells were maintained under submerged conditions until reaching confluence, at which point the medium in the upper chamber medium was removed to initiate air-liquid (ALI) culture. Differentiation was induced by supplying PneumaCult™-ALI Medium (STEMCELL Technologies, 05001) for HBECs and PneumaCult™-ALI-S Medium (STEMCELL Technologies, 05050) for HSAECs to the lower chamber. The medium was changed every 2 days, and mucus was cleared from the upper chamber with PBS weekly. Cells were maintained under ALI culture conditions for 21 days and subsequently collected for analysis. The dynamic GIF images of cilia presented were captured using the Eclipse Ti2 inverted microscope platform (Nikon).

Hematoxylin and eosin (H&E) staining

Differentiated ALI cultures were fixed with 4% PFA at 4 °C for 1 h, followed by three washes with PBS. The polycarbonate membrane was then carefully excised and embedded in molten 4% (w/v) agarose (Biowest, BY-R0100), which was allowed to solidify on ice for 30 min. The resulting agarose block was placed in an embedding cassette and dehydrated through a graded ethanol series, followed by two clearing steps in xylene (Sigma-Aldrich, 534056). After clearing, the samples were infiltrated with paraffin, embedded, and sectioned at 5 μm thickness using a microtome (LEICA, HistoCore AUTOCUT). Sections were floated on a 40 °C water batch, mounted on glass slides, and baked at 65 °C in a slide dryer (LEICA, HI1220). For H&E staining, paraffin sections were deparaffinized in xylene, rehydrated through a graded ethanol series, and stained sequentially with Harris hematoxylin (Beyotime, C0107) for 5 min and eosin (Beyotime, C0109) for 3 min. After dehydration and clearing, the slides were mounted with coverslips and imaged using an Axio Vert A1 inverted microscope (Zeiss).

Western blotting

Total protein was extracted from ALI monolayers using RIPA lysis buffer (Beyotime, P0013B). Following the guidelines of the BCA protein concentration assay kit (Beyotime, P0009), total protein concentrations were measured and adjusted to 2 µg/µL. The SDS-PAGE gels with varying acrylamide concentrations were prepared according to the molecular weights of target proteins. Protein markers and samples were loaded onto the gels, followed by electrophoresis, membrane transfer, blocking, primary antibody incubation, and secondary antibody incubation. The primary antibodies used included anti-MUC5AC (rabbit; Abways, CY6826), anti-Ac-Tub (mouse; Sigma-Aldrich, T7451), anti-KRT5 (mouse; Invitrogen, MA5-17057), anti-NGFR (mouse; R&D Systems, MAB367) and anti-GAPDH (mouse; Proteintech, 60004-1-Ig). Secondary antibodies were anti-rabbit IgG H&L (HRP) (goat, Abcam, ab6721) or anti-mouse IgG H&L (HRP) (goat, Abcam, ab6789). Protein signals were detected by chemiluminescence using ECL substrate (Abcam, ab133406) and visualized with a FUSION Solo S imaging system (Tanon, 5200 Multi). The original full-length, uncropped Western blotting images have been provided as part of the supplementary material.

Scanning electron microscopy (SEM)

Samples were initially fixed in 2.5% glutaraldehyde (Macklin, G916054) solution at room temperature for 1 h, followed by treatment with 1% osmium tetroxide solution (Sigma, 208868) for 1 h. Subsequently, dehydration was performed using a graded ethanol series, after which the samples were treated with anhydrous tert-butanol (Aladdin, T119717) for 30 min at room temperature. The specimens were then vacuum-dried and mounted onto aluminum stubs. To enable electron conductivity, samples were sputter-coated with gold using a high-vacuum ion sputter coater (Quorum) and subsequently imaged using a GeminiSEM 500 scanning electron microscope (Zeiss).

Apical-out airway organoid differentiation

HBECs and HSAECs were cultured in sfBSC medium and seeded onto AggreWell™400 24-well plates (STEMCELL Technologies, 34415) pre-treated with anti-adherence rinsing solution (STEMCELL Technologies, 07010) at a density of 1.2 × 10⁵ cells. The cells were cultured for 6 days in PneumaCult™ Apical-Out Airway Organoid Medium (STEMCELL Technologies, 100–0620). Afterward, the cells were transferred to flat-bottom 24-well plates treated with anti-adherence rinsing solution and maintained for an additional 9 days with continued culture in PneumaCult™ Apical-Out Organoid Medium.

Immunofluorescence staining

Cells cultured on confocal dishes or in the ALI culture system were fixed with 4% PFA for 30 min, washed with PBS, and permeabilized with 0.1% Triton X-100 (Sigma‒Aldrich, X100) for 10 min. After being blocked with 1% goat serum (Beyotime, C0265) and 0.1% Tween-20 (Sigma‒Aldrich, P1379) in PBS for 1 h, the cells were incubated overnight at 4 °C with the following primary antibodies: anti-P63 (rabbit, Abcam, ab124762), anti-KRT5 (mouse, Abcam, ab17130), anti-NGFR (mouse, R&D Systems, MAB367), anti-MUC5AC (rabbit, Abways, CY6826), anti-Ac-Tub (mouse, Sigma-Aldrich, T7451), and anti-SCGB1A1 (rat, R&D Systems, MAB4218). After washing, the samples were incubated with secondary antibodies conjugated to Alexa Fluor 488-conjugated anti-rabbit IgG (goat, Abcam, ab150077), Alexa Fluor 647-conjugated anti-mouse IgG (donkey, Abcam, ab150107), or Alexa Fluor 488-conjugated anti-rat IgG (donkey, Invitrogen, A-21208). Nuclei were stained with DAPI (Sigma-Aldrich, D9542) for 15 min. Next, the samples were washed with PBS containing 0.1% Tween 20 and stored in antifade mounting medium (Invitrogen, P36935). Images were captured via confocal microscopy.

For 3D apical-out airway organoids, the organoids were collected in 1.5 mL centrifugation tubes and treated with anti-adherence rinsing solution before fixation. The staining process was carried out in tubes, and the organoids and mountant suspension were dispensed onto the center of the confocal dish prior to imaging.

Images were captured via Z-stack imaging via LSM900 (Zeiss) and FV3000 (Olympus) confocal microscopes.

Forskolin-induced swelling of organoids

On day 15 of cell differentiation, forskolin-induced organoid swelling was analyzed. One day prior to analysis, organoids were transferred to 48-well plates in 3 µL Matrigel (Corning, 354277) droplets per well. For swelling induction, organoids were incubated with 10 µM forskolin (Selleck, S2449) for 24 h at 37 °C and 5% CO₂. Images were captured using a Axio Vert A1 inverted microscope (ZEISS) and analyzed with ilastik and ImageJ software [21]. Organoid areas were normalized to the initial well area (time = 0 h set as 1) using GraphPad Prism. Statistics were calculated from separate wells of organoids, with each value comprising 10–20 individual organoids.

RNA sequencing (RNA-seq) and data analysis

Total RNA was extracted from HBECs (P3, P6, P9, and P13) and HSAECs (P2, P7, P14, and P21) using TRIzol reagent (Invitrogen, 15596026CN) according to the manufacturer’s instructions. Briefly, approximately 1 × 10⁶ cells were collected into a 1.5 mL microcentrifuge tube and lysed by adding 0.5 mL of TRIzol reagent, followed by incubation at room temperature for 5 min. Subsequently, 0.1 mL of chloroform was added, the mixture was vortexed vigorously, and incubated at room temperature for 3 min. After centrifugation at 12,000 x g for 15 min at 4 °C, the upper colorless aqueous phase (supernatant) was carefully transferred to a new microcentrifuge tube. Then, 0.25 mL of isopropanol was added, mixed thoroughly, and incubated at 4 °C for 10 min, followed by centrifugation at 12,000 x g for 10 min at 4 °C. The supernatant was discarded, and the resulting white RNA pellet was washed by resuspending it in 0.5 mL of 75% ethanol; centrifugation was then performed at 7,500 x g for 5 min at 4 °C. This washing step was repeated twice. After discarding the ethanol, the pellet was air-dried for 5 min in a biosafety cabinet. Finally, the RNA pellet was dissolved in 25–50 µL of DEPC-treated water. The quality and concentration of the extracted total RNA were assessed using a NanoDrop spectrophotometer (Thermo Fisher Scientific) and a 2100 Bioanalyzer (Agilent).

mRNA was enriched using Oligo (dT) magnetic beads and subsequently fragmented. First-strand cDNA was generated using random hexamer-primed reverse transcription, followed by a second-strand cDNA synthesis. Afterwards, A-Tailing Mix and RNA Index Adapters were added by incubating to end repair. The adapter-ligated cDNA fragments were amplified by PCR. The PCR products were purified using Ampure XP beads, eluted in EB buffer, and quality-controlled using a 2100 Bioanalyzer. The purified double-stranded PCR products were then heat-denatured. Single-stranded circular DNA (ssCir DNA) libraries were constructed via adapter-mediated circularization. DNA Nanoballs (DNBs) were generated from the circularized libraries using phi29 DNA polymerase-based amplification, yielding > 300 copies per template molecule. Finally, the DNBs were loaded onto a patterned nanoarray and subjected to 150-bp paired-end sequencing on the MGIseq-2000 platform (BGI-Wuhan, China).

The raw data were filtered via SOAPnuke (v1.4.0) to obtain clean data, which were aligned to the reference genome (GCF_000001405.39_GRCh38.p13) via HISAT (v2.2.1) and Bowtie2 (v2.4.5). Gene expression levels in each sample were quantified via the RESM (v1.3.1) software package. The Pearson correlation coefficients for each sample of HBECs or HSAECs were computed using the FPKM values of all genes present in each respective sample. DESeq2 was used for intergroup differential gene analysis, whereas the Poisson distribution was used for intersample difference analysis with a threshold of |log2(foldchange)| ≥ 2 and a Q value ≤ 0.05. The heatmap function was applied to generate a heatmap illustrating the clusters of DEGs, and TBtools-II was used for visualization [22]. According to the GO annotation results and official classifications, the DEGs were functionally classified, and the TermFinder package was used to perform GO enrichment analysis.

Seahorse metabolic flux measurements

Seahorse XF analysis was performed using the Agilent Seahorse XF Cell Mito Stress Test Kit (Agilent Technologies) according to the manufacturer’s instructions. Briefly, HBECs (P3 and P13) and HSAECs (P2 and P21) were enzymatically digested and seeded onto XF96 plates at a density of 2 × 10⁴ cells/well in Seahorse assay medium (phenol red-free DMEM supplemented with 10 mM glucose, 1 mM sodium pyruvate, and 2 mM L-glutamine; pH 7.4). Plates were incubated for 45 min at 37 °C in a to a non-CO₂ environment. Cells were sequentially treated with 1.5 µM oligomycin, 1.0 µM FCCP, and 0.5 µM rotenone/antimycin A for oxygen consumption rate (OCR) measurements. All OCR values were normalized to cell numbers, which were quantified post-assay via nuclear and viability staining with Hoechst 33342 and propidium iodide using the BioTek Cytation 1 imaging reader (Agilent Technologies). Data were analyzed using Agilent Wave Desktop Software (Agilent Technologies).

Statistical analysis

Statistical analyses were conducted using GraphPad Prism (version 10). Unless otherwise indicated, data are either representative of three independent experiments or pooled from three independent experiments, as specified. Results are presented as mean ± SEM. Statistical comparisons were primarily performed using unpaired two-tailed Student’s t-test, unless stated otherwise. A p value < 0.05 was considered statistically significant.

Results

Serum-free medium for the sustained long-term expansion of BSCs

Previous studies have identified specific growth factors and signaling pathways critical for the survival and function of basal stem cells (BSCs). Epidermal growth factor (EGF), for instance, is indispensable for the proliferation of mouse BSCs [23, 24], while signaling pathways such as Wnt, ROCK, TGF-β, BMP, and FGF have been shown to promote proliferation or suppress differentiation in human or mouse BSCs [6, 25–28]. However, current culture systems for human BSCs remain suboptimal, often lacking standardization and relying on complex, undefined components, such as serum or bovine pituitary extract (BPE), which compromise reproducibility and translational potential.

To address these limitations, we systematically tested multiple combinations and developed a serum-free, fully defined BSC culture medium (sfBSC) consisting of 14 characterized components. The formulation includes a basal medium (IMDM and Ham’s F12) supplemented with B27 supplement, recombinant human serum albumin (rHSA), GlutaMAX, L-ascorbic acid, monothioglycerol (MTG) and hydrocortisone. Additionally, it contains specific growth factors (EGF and fibroblast growth factor 10 [FGF10]) and signaling inhibitors targeting ROCK (Y-27632), Notch (DAPT), and SMAD (DMH1 and A-83-01) pathways (Table 1). To assess the medium’s applicability, we cultured human bronchial epithelial cells (HBECs) and human small airway epithelial cells (HSAECs) in sfBSC medium. Both cell types were successfully seeded on uncoated culture flasks without the need for feeder cells, demonstrating the ability of the medium to support cell attachment and proliferation under simplified conditions.

Table 1.

Complete formulation for SfBSC medium

Component		Final concentration	500mL medium
Basal medium	IMDM	50%	250 mL
	Ham’s F12	50%	250 mL
Supplementary components	B27 Supplement	1X	10 mL
	rHSA	0.1%	0.5 mL
	GlutaMAX (mM)	2	1 mmol
	L-ascorbic acid (µg/mL)	50	25 mg
	Monothioglycerol (µM)	0.4	0.2 µmol
	Hydrocortisone (µg/mL)	1	500 µg
Inhibitors	Y-27632(µM)	10	5 µmol
	DAPT (µM)	1	0.5 µmol
	DMH-1 (µM)	1	0.5 µmol
	A-83-01 (µM)	1	0.5 µmol
Growth factors	EGF(ng/mL)	1	500 ng
	FGF10(ng/mL)	0.4	200 ng

Inhibitor powders were dissolved in DMSO, aliquoted, and stored at -20 °C. Cytokines were diluted in PBS containing 5% trehalose, aliquoted, and stored at -80 °C

The prepared complete sfBSC medium can be stored at 2–8 °C for up to four weeks

Continuous passaging of both cell lines in sfBSC medium yielded promising results. HBECs were successfully expanded to 17 passages (Fig. 1A), whereas HSAECs were expanded to 24 passages (Fig. 1B). In contrast, the expansion capacity of the cells cultured in a serum-free commercial medium was more limited, with that of the HBECs and HSAECs reaching only 8 and 11 passages, respectively (see Figure S1A, S1B). Notably, extended culture in commercial media often led to pronounced morphological heterogeneity (see Figure S1A, S1B), whereas cells maintained in sfBSC media preserved a stable and uniform cobblestone-like morphology throughout long-term expansion (Fig. 1A and B). The cell size also remained consistent, ranging from 10 to 15 μm across passages (Fig. 1C). Furthermore, cumulative population doublings (cPDs) were calculated for both cell lines, confirming sustained proliferation and growth potential throughout long-term culture (Fig. 1D and E).

Fig. 1.

HBECs and HSAECs were cultured in sfBSC medium. A Brightfield images of HBECs from passage 3 (P3) to passage 17 (P17). Scale bar = 200 μm. White boxes delineate representative cellular areas. B Brightfield images of HSAECs from P2 to P24. Scale bar = 200 μm. White boxes delineate representative cellular areas. C Cellular diameter of HBECs and HSAECs at different passages. D cPD curves for HBECs at different passages. E cPD curves for HSAECs at different passages

To further assess the long-term maintenance of BSC properties, we performed immunofluorescence staining on HBECs and HSAECs at early, middle and late passages. Even after prolonged culture, the majority of cells retained the expression of canonical BSC markers, including the transcription factor TP63 (P63), cytokeratin 5 (KRT5), and nerve growth factor receptor (NGFR) (Fig. 2A and B). Flow cytometric quantification revealed relatively stable P63/NGFR expression in HBECs and HSAECs, with minor variations observed between early and late passages (Fig. 2C). In contrast, cells maintained in the commercial medium exhibited a marked decline in these marker expressions during prolonged passaging (see Figure S1C, S1D).

Fig. 2.

Identification of canonical BSC markers for HBECs and HSAECs cultured in sfBSC medium. A Immunofluorescence images showing the expression of P63, KRT5, and NGFR in expanded HBECs at P4, P7, and P17. Scale bar = 50 μm. B Immunofluorescence images showing the expression of P63, KRT5, and NGFR in expanded HSAECs at P4, P7, and P24. Scale bar = 50 μm. C Flow cytometric analysis of P63 and NGFR expression in HBECs (P4 and P17) and HSAECs (P4 and P24)

Initially, HBECs and HSAECs were purchased from a commercial source and cultured in the commercial medium for the first two (HBECs) or one (HSAECs) passage before transitioning to sfBSC medium. To confirm that sfBSC medium is also suitable for primary HBEC isolation and expansion, BSCs derived from a COPD patient were directly cultured from bronchial brush biopsies in sfBSC medium. By day 7, P0 cells had formed well-defined, stem cell-like clones with distinct boundaries (see Figure S2A). These clones consisted of cube- or pebble-shaped cells that formed tight intercellular connections. After an additional 4 days of continuous culture, the cell density reached approximately 70% (see Figure S2B). The isolated patient-derived BSCs were successfully passaged and expanded (see Figure S2C).

These findings demonstrate that the sfBSC medium enables the long-term expansion of BSCs while preserving their morphology and the expression of key BSC markers, making it an effective and reproducible system.

Analysis of the impact of each SfBSC medium component on BSC growth

To determine an optimal basal medium for BSC culture, we first compared several commonly used formulations, including Dulbecco’s Modified Eagle Medium/Nutrient Mixture F-12 (DMEM/F12), DMEM, and Roswell Park Memorial Institute Medium 1640 (RPMI 1640). Cells cultured in DMEM/F12, DMEM, or RPMI 1640 exhibited markedly slower proliferation and reduced viability compared with those cultured in IMDM and Ham’s F12 (see Figure S3). Therefore, IMDM and Ham’s F12 were selected as the basal components of the sfBSC medium to ensure optimal growth and maintenance of BSCs.

We next performed a systematic evaluation of the effects of 12 individual additives. Our analysis revealed that these components played distinct roles in supporting BSC growth and maintenance. B27 supplement, GlutaMAX, L-ascorbic acid, EGF and Y-27632 were identified as core components essential for maintaining cell survival and promoting growth. Removing any one of these five components severely impaired growth, with a marked reduction in adherent cells at passage 3 (P3) and a near-complete loss of adhesion by passage 4 (P4) (Fig. 3A), as further confirmed by quantitative CCK-8 assays (Fig. 3B).

Fig. 3.

Evaluation of the effects of specific components in sfBSC medium on the growth of HBECs. A Brightfield images showing the morphology of HBECs at P3 and P4 after the removal of B27 supplement, GlutaMAX, L-ascorbic acid, Y27632, or EGF from the sfBSC medium. Complete sfBSC medium served as a control. Scale bar = 200 μm. B Quantitative assessment of HBEC proliferation using the CCK-8 assay following the removal of individual components (B27 supplement, GlutaMAX, L-ascorbic acid, Y27632, or EGF) from the sfBSC medium (n = 3 per group). C Growth curves showing changes in cell numbers over time following the removal of rHSA, hydrocortisone or FGF10 from the sfBSC medium (n = 3 per group). D Quantification of colony formation from single-cell seeding follwoing the removal of rHSA, hydrocortisone, or FGF10 (n = 3 per group). E Brightfield images showing the morphology of HBECs at P3 to P6 after the removal of A-83-01, DMH1, or DAPT from the sfBSC medium. Complete sfBSC medium served as a control. Scale bar = 200 μm. White boxes highlight regions containing differentiated cells. F cPD curves for HBECs at different passages following removal of A-83-01, DMH1, or DAPT in sfBSC medium. Statistical significance at last day was determined by one-way ANOVA followed by Dunnett’s multiple comparisons test vs. complete medium; ***p < 0.001

In contrast, removal of hydrocortisone, rHSA, or FGF10 did not cause immediate cell loss but significantly reduced expansion capacity, as evidenced by slower growth curves (Fig. 3C) and a marked decrease in colony formation (Fig. 3D, Figure S4). While Y-27632 played a critical role in maintaining cell survival, the other three inhibitors (A-83-01, DMH-1 and DAPT) did not directly affect viability (Fig. 3E and F). Instead, they primarily prevented spontaneous differentiation of BSCs during continuous passaging. The removal of any of these inhibitors led to noticeable morphological changes, indicating differentiation (Fig. 3E).

Although removal of MTG did not immediately impact BSC viability; its prolonged absence compromised cellular fitness during extended culture. By P11, cells displayed notable morphological changes—including irregular contours, loss of uniform cobblestone morphology, and disrupted monolayer organization—highlighting the essential role of MTG in maintaining BSC stability, defined as sustained proliferative capacity and phenotypic integrity over time (see Figure S5).

Optimization of EGF and FGF10 concentrations for sustained BSC growth and phenotypic stability

The concentration of growth factors in culture media is crucial for maintaining optimal cellular behavior, including cell proliferation, differentiation, and survival. We optimized the concentrations of EGF and FGF10 to support sustained BSC expansion. Consistent with prior findings, cultures lacking EGF failed to progress beyond the first passage, underscoring its indispensability (Figs. 3A and 4A). Interestingly, although low EGF concentrations (e.g. 0.1 ng/mL) reduced the proliferation rate, higher concentrations (≥ 10 ng/mL) induced an elongated cell morphology, suggesting that excessive EGF may adversely affect cellular phenotype (Fig. 4A).

Fig. 4.

Optimization of EGF and FGF10 concentrations for maintaining BSC morphology. A Brightfield images showing the morphology of HBECs cultured in media supplemented with different concentrations of EGF. Scale bar = 200 μm. White boxes demarcate cells exhibiting morphological alterations. B Brightfield images showing the morphology of HBECs cultured in media supplemented with different concentrations of FGF10. Scale bar = 200 μm. White boxes demarcate cells exhibiting morphological alterations. C Immunofluorescence staining of P63 and KRT5 in HBECs grown in medium supplemented with 40 ng/mL FGF10. Scale bar = 50 μm. D Immunofluorescence staining of NGFR in HBECs grown in media supplemented with 40 ng/mL FGF10. White borders demarcate regions with altered NGFR expression patterns. Scale bar = 50 μm

Similarly, insufficient FGF10 (0.04 ng/mL) reduced the proliferation rate of BSCs, whereas elevated levels (≥ 4 ng/mL) resulted in morphological alterations and disrupted intercellular junctions (Fig. 4B). Notably, at higher FGF10 concentrations, two morphologically distinct populations emerged: tightly-connected and loosely-connected cells. When the FGF10 concentration reached 40 ng/mL, the loosely-connected population predominated. To further investigate these phenotypes, we performed immunofluorescence staining on BSCs cultured with 40 ng/mL FGF10. The results revealed that, although all BSCs expressed canonical BSC markers (KRT5 and P63), only the tightly connected population retained membrane-localized NGFR; in contrast, the loosely connected cells showed cytoplasmic NGFR localization (Fig. 4C and D). Notably, under low confluence conditions, BSCs maintained in 0.4 ng/mL FGF10 still retained tight morphology and membrane NGFR localization (see Figure S6), further supporting that the effect is not driven by density alone. Based on these observations, we determined that 1 ng/mL EGF and 0.4 ng/mL FGF10 are the optimal concentrations for maintaining BSC morphology, proliferation, and marker expression in sfBSC medium.

Overall, our analysis highlights the essential role of each component in the sfBSC medium, demonstrating its effectiveness as a well-defined culture system for the reproducible expansion of BSCs. This optimized medium not only supports long-term BSC culture but also provides valuable insights into the molecular mechanisms regulating BSC stemness and homeostasis.

Differentiation potential of BSCs cultured in SfBSC medium

Long-term in vitro expansion of BSCs often leads to loss of stemness, resulting in reduced proliferation and diminished differentiation capacity. To evaluate whether BSCs maintain their differentiation potential in sfBSC medium, early- and late-passage HBECs and HSAECs were differentiated into airway epithelial cells via air-liquid interface (ALI) culture. After three weeks of differentiation, hematoxylin and eosin (H&E) staining showed that HBECs formed a pseudostratified epithelium resembling the in vivo bronchial epithelium, whereas HSAECs formed a thin, cuboidal epithelial layer representative of the in vivo small airway epithelium (Fig. 5A). Immunostaining revealed abundant ciliated and goblet cells in the HBEC-derived epithelium (Fig. 5B), while the HSAEC-derived epithelium additionally exhibited club cells, indicating multilineage differentiation capacity (Fig. 5C). Western blotting (WB) analysis further confirmed increased expression of MUC5AC and acetylated tubulin (Ac-Tub) in differentiated versus undifferentiated HBECs or HSAECs, at both early and late passages (Fig. 5D), accompanied by a concomitant reduction in NGFR expression (see Figure S7). To better visualize ciliary differentiation in ALI cultures, we employed scanning electron microscopy (SEM) for static cilia imaging (Fig. 5E) and Eclipse Ti2 inverted microscopy for dynamic ciliary imaging (see Video S1-4). Both approaches consistently demonstrated well-preserved ciliogenesis across passages.

Fig. 5.

Differentiation of expanded HBECs and HSAECs into airway epithelium via sfBSC medium. A H&E staining of HBECs (P7 and P13) and HSAECs (P7 and P17) following 21-day ALI differentiation. Scale bar = 100 μm. B Immunofluorescence staining of MUC5AC and Ac-Tub after ALI differentiation of P7 and P13 HBECs. Scale bar in the left panel = 50 μm; scale bar in the right panel = 20 μm. C Immunofluorescence staining of MUC5AC, Ac-Tub, and CC10 after ALI differentiation of P7 and P17 HSAECs. Scale bar in the left panel = 50 μm; scale bar in the right panel = 20 μm. D Western blotting was performed to assess expression of MUC5AC and Ac-Tub in HBECs (P7 and P13) and HSAECs (P7 and P17) at days 0 and 21 of ALI differentiation. F SEM was utilized to assess ciliary distribution in HBECs (P7 and P13) and HSAECs (P7 and P17) at day 21 post-ALI differentiation. Scale bar = 20 μm

Multilineage differentiation potential was further evaluated using a 3D organoid culture system. HBECs and HSAECs from early and late passages successfully formed airway organoids (see Figure S8). The derived organoids exhibited ciliated cells, as confirmed by the detection of Ac-Tub on their outward-facing apical surface via immunofluorescence (Fig. 6A and B). Consistent with the ALI differentiation results, MUC5AC was detected in both the HBEC- and HSAEC-derived organoids, whereas CC10 was expressed predominantly in the HSAEC-derived organoids. To assess functional maintenance of epithelial ion transport, we evaluated cystic fibrosis transmembrane conductance regulator (CFTR) activity using a forskolin-induced swelling assay. Organoids derived from early- (HBEC P7 and HSAEC P7) and late-passage (HBEC P13 and HSAEC P17) cells demonstrated comparable swelling in response to 10 µM forskolin over 24 h (Fig. 6C and D), indicating preserved CFTR-mediated ion channel function over extended passaging.

Fig. 6.

Differentiation of expanded HBECs and HSAECs into organoids via sfBSC medium. A Immunofluorescence staining of MUC5AC and Ac-Tub in P7 and P13 HBEC-derived airway organoids. Scale bar = 20 μm. B Immunofluorescence staining of MUC5AC, Ac-Tub, and CC10 in P7 and P17 HSAEC-derived airway organoids. Scale bar = 20 μm. C Bright-field microscopy images showing organoid swelling of HBECs (P3 and P7) and HSAECs (P7 and P17) following 24-hour treatment with 10 µM forskolin. Scale bar = 100 μm. D Quantification of forskolin-induced organoid swelling based on normalized area change at 24 h relative to 0 h (n = 3 per group). Statistical significance was determined using a two-tailed unpaired t test. *p < 0.05, **p < 0.01

Collectively, these results demonstrate that BSCs cultured in sfBSC medium retain robust multilineage differentiation capacity and functional epithelial characteristics after long-term expansion.

RNA-seq analysis of BSCs cultured in SfBSC medium

Although HBECs and HSAECs could be expanded for 17 and 24 passages, respectively, in sfBSC medium, providing sufficient cells for most basic, translational, and potential clinical applications, proliferation rates eventually declined after prolonged culture. To investigate the molecular changes associated with long-term passage, RNA transcriptomic sequencing (RNA-seq) was performed on HBECs at passages P3, P6, P9, and P13 and on HSAECs at passages P2, P7, P14, and P21. Notably, the expression of specific BSC markers for both cell types remained stable over multiple passages, with no significant differences observed, corroborating our previous immunostaining results (Fig. 7A and B). To assess the stability of transcriptomic profiles during serial passaging in both cell lines, sample correlation analysis demonstrated high concordance across different passages, with all sample pairs exhibiting Pearson correlation coefficients > 0.93 (range: 0.93–0.99) (Fig. 7C and D).

Fig. 7.

Transcriptomic analysis of HBECs and HSAECs. A Heatmap showing the expression patterns of BSC-specific markers in HBECs across different passages. B Heatmap showing the expression patterns of BSC-specific markers in HSAECs across different passages. C Sample correlation heatmap illustrating transcriptomic similarity across different passages of HBECs. D Sample correlation heatmap illustrating transcriptomic similarity across different passages of HSAECs. E GO enrichment analysis of biological processes associated with DEGs in HBECs (P3 vs. P13). F GO enrichment analysis of biological processes associated with DEGs in HSAECs (P2 vs. P21). G Venn diagram depicting the overlap of DEGs between HBECs (P3 vs. P13) and HSAECs (P2 vs. P21) during passaging. H Heatmap displaying 10 upregulated genes and 10 downregulated genes among commonly identified DEGs between HBECs and HSAECs during long-time culture

A threshold of |log2|≥2 was applied to identify differentially expressed genes (DEGs) between early- and late-passage HBECs and HSAECs. Gene Ontology (GO) analysis revealed that the biological processes enriched with the DEGs were highly consistent between the HBECs and HSAECs. Notably, these DEGs were predominantly associated with cell proliferation, cell cycle regulation, and cell division, which aligns with the observed decrease in proliferation capacity after long-term passaging. Additionally, pathways related to lipid metabolism and the tricarboxylic acid (TCA) cycle were significantly enriched, suggesting that metabolic adaptations occurred during extended culture. Importantly, the EGF receptor signaling pathway was significantly enriched, reinforcing the critical role of EGF in BSC maintenance. This finding is consistent with our previous results, which demonstrated that EGF withdrawal severely compromised BSC survival (Fig. 7E and F).

To further investigate genes essential for maintaining the stability of both HBECs and HSAECs, we performed integrated analysis of sequencing data from both cell lines. A total of 142 common DEGs were identified (Fig. 7G). Among these, we focused on genes whose expression gradually and consistently changed from early to late passages (Fig. 7H). These genes were categorized into various functional groups, including those encoding transcription factors, enzymes, cell adhesion molecules, and extracellular matrix proteins. Notably, many of these genes were linked to metabolism regulation, particularly amino acid metabolism and energy homeostasis. For example, two upregulated genes were identified: phosphodiesterase 3B (PDE3B), which regulates cyclic AMP (cAMP) signaling, a crucial pathway in cellular metabolism and signaling, and aspartoacylase (ASPA), which facilitates the breakdown of N-acylated aspartate and plays a role in fatty acid metabolism. Conversely, three downregulated metabolism-related genes were identified: Cystathionine Beta-Synthase (CBS), a key enzyme in sulfur amino acids metabolism and homocysteine regulation; branched-chain aminotransferase 1 (BCAT1), which is involved in the metabolism of branched-chain amino acids (valine, leucine, isoleucine), which are critical for protein synthesis and energy metabolism; and glutathione peroxidase 3 (GPX3), an antioxidant enzyme that protects cells from oxidative stress and regulates cellular metabolism.

Overall, these findings suggest that while key BSC markers remain stably expressed during long-term culture, metabolic reprogramming occurs over successive passages.

BSCs exhibit dynamic metabolic changes during serial passaging

To further investigate the metabolic changes associated with prolonged in vitro expansion, we performed real-time metabolic flux analysis using Seahorse XF technology. HBECs (P3 and P13) and HSAECs (P2 and P21) cultured in sfBSC medium were analyzed to assess oxgen comsumption rate (OCR) and mitochondrial function. Early-passage cells (HBEC P3 and HSAEC P2) exhibited significantly higher basal respiration, increased maximal respiratory capacity, and elevated mitochondrial ATP production compared to their late-passage counterparts (HBEC P13 and HSAEC P21) (Fig. 8). These findings suggest that early-passage BSCs maintain more oxidative phosphorylation activity and a more robust tricarboxylic acid (TCA) cycle.

Fig. 8.

BSCs exhibited enhanced mitochondrial metabolism during early stages. A Real-time OCRs of HBEC at P3 and P13, measured using the Seahorse XF96 Analyzer and normalized to cell number. B Quantification of basal respiration, maximal respiration, and ATP production in HBECs (P3 vs. HBEC P13) (n = 3 per group). C Real-time OCRs of HSAECs at P2 and P21 measured using the Seahorse XF96 Analyzer and normalized to cell number. D Quantification of basal respiration, maximal respiration, and ATP production of HSAEC (P2 vs. P21) (n = 3 per group). Statistical significance was determined using a two-tailed unpaired Student’s t test. *p < 0.05, **p < 0.01

This shift in metabolic profile during serial passaging highlights the importance of cellular metabolism in maintaining BSC function and supports the need for further optimization of culture conditions to preserve stemness and proliferative capacity over extended culture periods.

Discussion

Previous studies have employed various strategies to achieve long-term expansion of BSCs, including co-culturing with feeder cells, using conditioned media, or incorporating undefined components such as serum or BPE. For example, BSCs were cultured via the well-established conditional reprogramming culture (CRC) technique. For this method, BSCs were co-cultured with irradiated mouse embryonic fibroblasts (3T3-J2 cells) in medium supplemented with Y-27632. Compared with conventional bronchial epithelial growth medium (BEGM) alone, the CRC approach facilitates more efficient cell expansion, enabling BSCs to maintain long-term growth for more than three months in vitro [19, 29–31].

Despite its efficacy, the complex handling of feeder cells, as well as concerns regarding genetic stability and safety, has limited its broader clinical application. As an alternative, researchers have explored the use of conditioned media to maintain BSC stability. For example, pre-coating cell culture containers with conditioned medium from 804G rat bladder epithelial cells, which secrete laminin- and collagen-rich extracellular matrix proteins, has been shown to support BSC adhesion and prolonged culture [32–34]. Similarly, 3T3-conditioned medium in combination with Matrigel-coated culture dishes has enabled the stable proliferation of mouse BSCs for up to 15 passages [18].

Another widely adopted strategy for BSC culture involves supplementing media with serum or BPE, which contain various hormones and growth factors that support cellular proliferation. When combined with additional growth factors or signaling inhibitors, these components can effectively sustain BSC culture. For example, the addition of Y-27632 to media containing serum or BPE has been reported to increase the proliferation of primary airway epithelial cells without compromising their differentiation potential [25]. Similarly, SMAD signaling inhibitors have been used to prevent spontaneous differentiation and prolong BSC culture time [26]. Although both serum and BPE have been shown to promote cell growth, their practical application is constrained by issues such as batch-to-batch variability, complex compositions, and challenges in achieving standardization.

To overcome these limitations, we developed a simplified, well-defined, serum- and BPE-free BSC culture medium that supports robust long-term expansion. Our sfBSC medium does not require feeder layers, conditioned media, or specialized substrate coatings but enables BSCs to maintain viability, self-renewal, and differentiation capacity over extended culture periods. Notably, our medium is compatible with BSCs derived from different airway locations, including bronchial and small airway epithelial cells, and is suitable for both basal cell lines and primary BSC isolation from bronchial brush biopsies.

Through a systematic evaluation of individual medium components, we identified essential factors that contribute to BSC survival, proliferation, and differentiation maintenance. In particular, we demonstrated the specific roles of EGF and FGF10 in BSC culture and determined their optimal concentrations for long-term expansion. Our findings indicate that while low EGF concentrations lead to cell death, excessively high concentrations induce abnormal cellular morphology, highlighting the need for careful concentration regulation.

FGF10 is known to play a pivotal role in lung homeostasis and epithelial stem cell expansion [35, 36]. Our results confirmed that adding FGF10 at an optimal concentration significantly enhances BSC proliferative capacity. Interestingly, we also observed that high FGF10 concentrations altered NGFR expression patterns in BSCs, affecting their membrane localization in a subset of cells. NGFR is a low-affinity receptor for nerve growth factor (NGF) that plays key roles in neural, embryonic, and intestinal stem cells [37, 38] and is commonly used as a mature BSC marker. Under high-FGF10 conditions, we identified two morphologically distinct populations: one consisting tightly-connected cells with membrane-localized NGFR, and another composed of loosely-connected cells exhibiting abnormal morphology and predominantly cytoplasmic NGFR expression. These findings suggest that changes in NGFR localization may be associated with alterations in its functional activity. However, further in-depth investigations are needed to determine how these changes influence the biological behavior of BSCs.

Finally, to further elucidate the molecular changes associated with long-term BSC culture, we conducted RNA-seq analysis across different passages of HBECs and HSAECs. Our analysis aimed to identify key factors that could improve BSC stability in prolonged culture. We specifically focused on 142 common DEGs shared between both cell lines, many of which were associated with metabolism regulation. These transcriptional changes indicate alterations in metabolic states that may underlie the gradual loss of stemness during prolonged culture, ultimately impairing both proliferative capacity and differentiation potential.

It is well established that cell differentiation is an energy-intensive process, typically accompanied by increased ATP demand and enhanced mitochondrial activity during early differentiation [39]. Given the observed metabolic attenuation in later passages, it is reasonable to speculate that further extending culture may exacerbate metabolic insufficiency and compromise differentiation capacity. Accordingly, we used HBEC P13 and HSAEC P17 for differentiation assays instead of higher passages (HBEC P17 and HSAEC P24). These cells retained the ability to form organized epithelial structures; however, experimentally validating this hypothesis using HBEC P17 and HSAEC P24 would be ideal. Additionally, future in-depth investigation into these transcriptional changes and the metabolic status of BSCs will help improve strategies for extending the longevity and enhancing the stability of basal cells in vitro.

Conclusions

This study established a well-defined culture system that enables long-term stable in vitro expansion of human BSCs while preserving their self-renewal capacity and multipotent differentiation potential.

Unlike previous published BSC culture media, which often contain undefined serum, BPE, or undisclosed formulations, our custom-prepared culture system consists entirely of defined and disclosed components. This transparency not only serves as a foundation for further optimization but also supports three promising applications. First, this well-defined system enables a detailed investigation of the molecular mechanisms regulating BSC homeostasis. Second, given the demonstrated therapeutic potential of BSC transplantation for COPD and IPF patients, this serum-free medium provides a feasible approach for generating clinical-grade BSCs. Third, owing to its simplified and reproducible culture protocol, this system is well suited for large-scale, automated expansion of BSCs, facilitating future translational and industrial applications.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

BSCs

Basal Stem Cells

COPD

Chronic Obstructive Pulmonary Disease

BPE

Bovine Pituitary Extract

HBECs

Human Bronchial Epithelial Cells

HSAECs

Human Small Airway Epithelial Cells

rHSA

Recombinant Human Serum Albumin

MTG

Monothioglycerol

EGF

Epidermal Growth Factor

FGF10

Fibroblast Growth Factor 10

DMH1

Dorsomorphin Homolog 1

cPDs

Cumulative Population Doublings

KRT5

Cytokeratin 5

NGFR

Nerve Growth Factor Receptor

PFA

Paraformaldehyde

ALI

Air-Liquid Interface

Ac-Tub

Acetylated Tubulin

H&E

Hematoxylin and eosin

SEM

Scanning Electron Microscopy

CFTR

Cystic fibrosis transmembrane conductance regulator

RNA-seq

RNA Transcriptomic Sequencing

DEGs

Differentially Expressed Genes

GO

Gene Ontology

OCRs

Oxygen consumption rates

TCA

Tricarboxylic Acid

CBS

Cystathionine Beta-Synthase

BCAT1

Branched-Chain Aminotransferase 1

GPX3

Glutathione Peroxidase 3

PDE3B

Phosphodiesterase 3B

cAMP

Cyclic AMP

ASPA

Aspartoacylase

CRC

Conditional Reprogramming Culture

BEGM

Bronchial Epithelial Growth Medium

Author contributions

N.M. conceived the idea of this work. N.M. and M.J. designed the experiments and managed the project. M.J. and Y.Z. performed the experiments. H.H.Z. performed the analysis of bulk RNA-seq data. S.L., H.C., and Z.S. collected the patient-derived cells. M.J. wrote the manuscript draft. N.M., B.H.Z., S.L., W.L., H.H.Z., M.J., and Y.Z. revised the manuscript. All the authors read and approved the final manuscript.

Funding

This study is supported by the Major Project of Guangzhou National Laboratory (Grant Nos. GZNL2023A02002, GZNL2023A02006 and GZNL2025C02020), the National Natural Science Foundation of China (Grant Nos. 82470243 and 82200355), the Guangzhou Basic and Applied Basic Research Foundation (Grant No. 2023A04J1204), the Young Scientists Program of Guangzhou National Laboratory (Grant No. QNPG23-19), and the Pearl River Talent Recruitment Program (Grant No. 2021QN02Y016).

Data availability

The sequencing data generated and analyzed in this study have been deposited in the Gene Expression Omnibus (GEO) repository under accession number GSE304935. All other datasets supporting the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the protocol approved by the Institutional Review Board of The First Affiliated Hospital of Guangzhou Medical University (Project Title: Construction and Application Evaluation of a Standardized Animal Model for COPD; Approval No.: ES-2024-085-01; Date of Approval: June 13, 2024). Written informed consent was obtained from all subjects prior to bronchoscopy. The human bronchial epithelial cells (HBECs, Lifeline, FC-0035) and human small airway epithelial cells (HSAECs, ATCC, PCS-301-010) used in this study were purchased from commercial vendors. According to the information provided by the respective manufacturers, ethical approval for donor tissue collection was obtained, and informed consent was secured from all donors. Relevant details can be found on the ATCC website (https://www.atcc.org/products/pcs-301-010) and the Lifeline Cell Technology website (https://www.lifelinecelltech.com/knowledge-base/ethical-and-legal-standards/).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.