The role of the RS1-NEDD4-YAP1 axis orchestrates epithelial-macrophage interplay in COPD

Abstract

Background

Chronic obstructive pulmonary disease (COPD) is a heterogeneous disorder characterized by persistent airflow limitation and is a leading cause of morbidity and mortality worldwide. While cigarette smoking is the primary environmental risk factor, only a subset of smokers develops clinically significant COPD, indicating roles for genetic susceptibility and dysregulated cellular responses. Although traditionally viewed as an immune-driven inflammatory disease, recent evidence underscores the active contribution of structural cells—particularly alveolar type II epithelial cells (AT2 cells)—to immune regulation and tissue remodeling.

Methods

This study integrated single-cell transcriptomics, conditional gene knockout mouse models, organoid culture, multiplex immunofluorescence validation, and flow cytometry to investigate the role of the RS1 gene in chronic obstructive pulmonary disease (COPD). Single-cell sequencing technology revealed RS1 as a signature gene of AT2 cells. Using cigarette smoke (CS)-exposed AT2 cell-specific RS1 knockout mice and organoids as models, we elucidated the mechanisms by which RS1 regulates AT2 cells and its associated role in COPD pathogenesis through molecular experiments.

Results

Single-cell RNA sequencing analysis revealed that RS1 is specifically expressed in AT2 cells in COPD models, and its expression exhibits interactions with immune cells. By constructing a COPD mouse model with AT2 cell-specific RS1 knockout (Sftpc^cre;RS1^fl/fl) and organoid models, we demonstrated that RS1-deficient AT2 cells display enhanced proliferative capacity and suppress macrophage infiltration. Mechanistically, RS1 mediates the interaction between NEDD4 and YAP1 through serine 105, while the RS1-NEDD4 interaction facilitates YAP1 dephosphorylation and nuclear translocation. Treatment with the selective NEDD4 inhibitor I3C further suppressed TGFB1 expression and macrophage infiltration.

Conclusion

RS1 suppresses AT2 cell proliferative capacity and promotes macrophage infiltration. RS1 interacts with NEDD4 to mediate YAP1 dephosphorylation and nuclear translocation, and treatment with the selective NEDD4 inhibitor I3C further suppresses TGFB1 expression and macrophage infiltration.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12931-026-03735-x.

Introduction

Chronic Obstructive Pulmonary Disease (COPD) is a heterogeneous disorder characterized by persistent airflow limitation, primarily driven by small airway pathology and parenchymal destruction [1]. It remains a leading cause of global morbidity and mortality. Although tobacco smoke is the principal environmental risk factor, only a subset of smokers develops clinically significant COPD, indicating that genetic susceptibility and altered molecular and cellular responses play crucial roles in disease pathogenesis [2, 3]. While traditionally viewed as a chronic inflammatory disease driven by immune cells, emerging evidence highlights the active contribution of lung structural cells—particularly alveolar type II epithelial cells (AT2 cells)—to immune regulation, tissue repair, and remodeling [4]. As the progenitor cells of the alveolar epithelium, AT2 cells are responsible for surfactant production and the maintenance of alveolar homeostasis [5]. In COPD, these cells exhibit dysfunction, abnormal proliferation, impaired differentiation, and senescence, collectively compromising repair capacity and perpetuating chronic inflammation [6].

Advances in single-cell RNA sequencing (scRNA-seq) [7] and spatial transcriptomics [8] have enabled high-resolution characterization of the dynamic cellular landscape and intercellular communication in the COPD lung. These techniques have revealed a distinct “disease-associated” AT2 cell subpopulation with a pro-inflammatory signature, whose specific molecular profile may drive disease progression. Notably, the Hippo signaling pathway, a key conserved pathway regulating organ size, tissue homeostasis, and regeneration, features the core effector molecules YAP/TAZ, which are pivotal for the proliferation, differentiation, and injury repair of alveolar epithelial cells [9–12]. Studies indicate that dysregulation of the Hippo/YAP pathway is closely associated with aberrant epithelial regeneration, senescence, and epithelial-mesenchymal transition, representing a crucial molecular mechanism underlying alveolar structural destruction and pathological remodeling [9–12].

The Retinoschisin 1 (RS1) gene encodes a protein primarily expressed in the retina, with mutations causing X-linked juvenile retinoschisis. Recent research suggests aberrant RS1 expression in various cancers and its correlation with immune cell infiltration, implying a potential role in cell proliferation and immune modulation [13–15]. However, its expression pattern, cellular source, and functional significance in the lung, particularly in COPD, remain unexplored. Given the regulatory functions of RS1 in other systems and the pivotal role of the Hippo pathway in COPD-related epithelial remodeling, we hypothesize that RS1 may be involved in regulating alveolar epithelial dysfunction and immune microenvironment dysregulation in COPD, potentially by influencing signaling pathways such as Hippo/YAP within AT2 cells. This study aims to systematically elucidate the specific role of RS1 in COPD pathogenesis and its underlying molecular mechanisms.

Methods

Single-cell data processing and analytical workflow

Public single-cell transcriptomic datasets (GSE136831, GSE168299) were systematically re-analyzed. The analytical pipeline strictly adhered to the methodological framework established by Sauler et al. [7]. Data processing was performed following the reported toolchain: raw sequencing data were processed using the mkfastq pipeline from Cell Ranger (v3.1); sequence trimming was performed using cutadapt (v3.4) and TrimGalore! (v0.6.7); and alignment to the reference genome was executed using STAR (v2.6.0c, v2.7.3a, v2.7.10b).

Following the generation of the single-cell transcriptome count matrix, core analysis was conducted using the Seurat R package (v4.0.4) and its recommended workflow. Specifically, a donor-level integration strategy was employed. Stringent quality control (QC) was first applied to each individual sample: cells with fewer than 200 or more than 6000 detected genes, or with a mitochondrial gene content exceeding 20%, were filtered out. Doublets were identified and removed using scDblFinder (v1.24.0). Subsequently, each sample underwent normalization and variance stabilization using SCTransform (v2), and highly variable genes were selected for downstream analysis. The key integration step used individual donors or mouse IDs as integration anchors, performing cross-sample batch correction based on 3000 integration features. UMAP visualizations before and after integration confirmed the effective removal of technical batch effects while preserving disease-associated biological heterogeneity.

Dimensionality reduction of the integrated dataset was performed using the first 14 principal components. Clustering was conducted on a shared nearest neighbor (SNN) graph using the Leiden algorithm (resolution = 0.8). Cell types were annotated based on canonical marker gene expression (AT2 cells: SFTPC⁺, RS1⁺). To investigate cell type-specific genetic contributions, gene expression specificity scores were computed using CELLEX (v1.2.1). CELLECT (v1.1.0) was then run with its recommended workflow (CELLECT-LDSC, using default parameters such as a 100 kb window upstream/downstream of genes) to integrate genome-wide association study (GWAS) summary statistics.

Differential expression analysis employed a donor-level pseudo-bulk approach: expression values for cells of the same type within a single donor were aggregated to form a “pseudo-sample”. Differential analysis was performed using DESeq2 (v1.48.1), with a model that included disease status and was adjusted for sex as a covariate. Differentially expressed genes were further subjected to KEGG and Gene Ontology (GO) pathway enrichment analysis using clusterProfiler (v4.19.2). Cell-cell communication network analysis was conducted using the Connectome R package (v1.0.0). Global interaction networks were constructed by calculating the average expression values of ligand-receptor pairs curated in the FANTOM5 database.

Additional tools used included featureCounts (v2.0.1) for gene quantification and UpsetR (v1.4) for set visualization. To validate the robustness of our analysis, sensitivity analyses were performed before and after integration, confirming the consistency of core findings. Furthermore, image analysis was performed using Fiji (v1.0) and CellProfiler (v4.2.8), and statistical graphs were generated using GraphPad Prism (v9.3.0).

Generation of AT2 cell-specific RS1 knockout mice (Sftpc^cre;RS1^fl/fl)

AT2 cell-specific RS1 knockout mice (Sftpc^cre;RS1^fl/fl) were generated using the Cre-loxP recombination system. The conditional RS1 allele (RS1^fl/fl) was designed by introducing loxP sites flanking critical exons, ensuring normal RS1 expression in the absence of Cre recombinase activity. Sftpccre transgenic mice, which express Cre recombinase under the control of the surfactant protein C (Sftpc) promoter, were used to achieve specific deletion in AT2 cells. To generate the experimental model, RS1^fl/fl mice were crossed with Sftpccre mice. The resulting F1 offspring (RS1^fl/+; Sftpc^cre/+) were then backcrossed to homozygous RS1^fl/fl mice. From the F2 generation, mice homozygous for the floxed RS1 allele and carrying the Sftpc^cre transgene (Sftpc^cre;RS1^fl/fl) were identified through genotyping at weaning (approximately 21 days postnatal).

Establishment of the Chronic Obstructive Pulmonary Disease (COPD) mouse model

Genotyped 8-week-old male [16, 17] Sftpc^cre; RS1^fl/fl mice and their male littermate RS1^fl/fl control mice (n = 5 per group) were assigned to four experimental groups: RS1^fl/fl + PBS, Sftpc^cre; RS1^fl/fl + PBS, RS1^fl/fl + cigarette smoke (CS) exposure, and Sftpc^cre; RS1^fl/fl + CS exposure. In two independent experiments, each group consisted of five mice, and the RS1^fl/fl + PBS group also consisted of five mice.

CS exposure was performed using a whole-body inhalation exposure system, which consisted of a smoke generator, a concentration-controlled exposure chamber, a real-time particulate matter monitor, and an exhaust system. Commercial research-grade cigarettes (3R4F) were used. Smoke was generated at fixed intervals and diluted with HEPA-filtered air to maintain a total particulate matter concentration of 100 mg/m³ (range: 80–150 mg/m³). The exposure protocol was as follows: 60 min per session, twice daily with an interval of at least 4 h between sessions, 5 days per week, for a duration of 24 weeks.

Animal Ethics and Compliance Statement: The animal experimental protocol involved in this study was reviewed and approved by the Experimental Animal Ethics Committee of Shenyang Medical College. All procedures were strictly conducted in accordance with the ARRIVE (Animal Research: Reporting of In Vivo Experiments) guidelines and relevant regulations. The entire experimental process adhered to the principles of the UK Animals (Scientific Procedures) Act 1986 and its associated guidelines, the European Union Directive 2010/63/EU on the protection of animals used for scientific purposes, and the National Institutes of Health Guide for the Care and Use of Laboratory Animals (NIH Publication No. 8023, revised 1978) to ensure animal welfare and standardize experimental conduct.

Bronchoalveolar Lavage Fluid (BALF) collection

For bronchoalveolar lavage fluid (BALF) collection, and the lung was lavaged three times with 0.5 ml of sterile ice-cold PBS containing protease inhibitors. BALF samples were centrifuged at 230×g for 10 min at 4 °C. Total cell counts in the BALF were obtained using an automated cell counter (Countstar, Shanghai, China), while differential cell counts were performed on cytospin preparations stained with Wright-Giemsa.

Hematoxylin and Eosin (H&E) staining

Tissue Processing and Sectioning: Following the completion of experiments, mice were anesthetized, and lung tissue was fixed in situ by intratracheal perfusion with 4% paraformaldehyde (prepared in PBS) at a constant pressure of 20 cm H₂O. Subsequently, the intact lungs were excised and immersed in 4% paraformaldehyde for post-fixation at 4 °C for 24 h. After fixation, tissues were rinsed three times with PBS for 5 min each. For processing, the right middle lobe was uniformly collected. Tissues were dehydrated through a graded ethanol series (70%, 80%, 90%, 95%, 30 min each), followed by two changes of absolute ethanol (30 min each) for complete dehydration. Tissues were then cleared in xylene twice (30 min each) until fully transparent. This was followed by infiltration in molten paraffin wax twice (2 h each) and embedding. After the blocks solidified at room temperature, consecutive sections of 4 μm thickness were cut using a microtome at 4 °C. Sections were floated in a 40 °C water bath, mounted onto glass slides, and baked at 60 °C for 4 h before being stored at room temperature for later use.

Staining Procedure: Prior to staining, sections were deparaffinized in eco-friendly dewaxing solutions I and II for 20 min each. Rehydration was performed on a horizontal shaker sequentially through absolute ethanol I (5 min), absolute ethanol II (5 min), and 75% ethanol (5 min), followed by a 20-minute rinse in ddH₂O. Sections were stained with hematoxylin for 3–5 min, rinsed under running water, differentiated, blued, and rinsed again. Dehydration was then carried out through a graded ethanol series (5 min each). Counterstaining was performed with eosin for 5 min. Finally, stained sections were dehydrated in absolute ethanol (three changes), cleared in xylene, and mounted with neutral resin.

Quantification of emphysema severity (measurement of mean alveolar septal distance)

To assess the degree of emphysema, we performed Mean Linear Intercept (MLI) analysis on H&E-stained sections as described by Ghosh et al. [18]. For each section, 10 non-overlapping fields of view, devoid of large airways or blood vessels, were randomly selected at 200x magnification. A grid containing evenly spaced (50 μm) test lines was superimposed onto each field. The total number of intersections between alveolar septa and the test lines within each field was counted. The MLI for that field was calculated using the formula: MLI = (Total length of test lines) / (Total number of intersections). The final MLI value for each animal was the average of the MLI values from all its fields. This method serves as the gold standard for quantitatively evaluating alveolar space enlargement and lung tissue destruction.

Establishment of AT2 cell-derived lung organoids

To model and investigate RS1-mediated mechanisms of COPD in a three-dimensional in vitro system, primary AT2 cells were isolated from the lung tissue of mice from the in vivo models to generate lung organoids. The specific protocol is as follows: Experimental mice from the four groups (RS1^fl/fl+CS, Sftpc^cre; RS1^fl/fl+CS, RS1^fl/fl+PBS, Sftpc^cre; RS1^fl/fl+PBS) that had been exposed to cigarette smoke (CS) or PBS for six months were euthanized. Lung tissue was harvested immediately and placed in sterile, ice-cold PBS.

A single-cell suspension was obtained via tissue digestion: a digestion enzyme mixture containing Collagenase II and Dispase was infused into the lungs via the trachea, followed by incubation at 37 °C for 30–45 min with gentle mechanical trituration. The cell suspension was filtered through a 70 μm cell strainer and centrifuged at 300–400 × g for 5 min.

To isolate highly pure primary AT2 cells, fluorescence-activated cell sorting (FACS) was performed. Cells were resuspended in PBS containing 2% FBS and stained with antibodies against AT2-specific surface markers (EpCAM, PDGFRa, SFTPC), while excluding CD45 + and CD31 + cells. The sorted AT2 cells were immediately resuspended in pre-cooled complete organoid medium.

Sorted cells were mixed on ice with pre-cooled Matrigel (or Cultrex RGF BME) at a density of 5,000–10,000 cells per 50 µL droplet. The mixture was plated as 50 µL droplets (2–3 per well) in a pre-warmed 24-well plate. After solidification at 37 °C for 15–20 min, 500–700 µL of pre-warmed complete organoid medium was added to each well. The medium was based on Advanced DMEM/F-12, supplemented with N21-MAX additive, GlutaMAX, HEPES, penicillin/streptomycin, nicotinamide, SB 202,190, Y-27,632, and N-acetylcysteine. For this study, EGF and Wnt3a-conditioned medium were additionally supplemented to support AT2 cell growth and maintenance. The medium was changed every 2–3 days. Within 3–5 days, AT2 cells from all groups formed three-dimensional spherical organoids with characteristic luminal structures.

EdU staining protocol

At the experimental endpoint, organoids were incubated with complete medium containing EdU at a final concentration of 10 µM for 4–6 h at 37 °C in a 5% CO₂ atmosphere. Subsequently, the medium was carefully discarded, and the organoid-Matrigel composites were gently washed with pre-cooled phosphate-buffered saline (PBS). Fixation was performed using 4% paraformaldehyde at room temperature for 15–30 min. Following fixation, samples were permeabilized with PBS containing 0.5% Triton X-100 and blocked with PBS containing 3% bovine serum albumin (BSA). A click chemistry reaction was then performed by incubating the samples with the reaction mixture from a commercial Click-iT EdU detection kit (Alexa Fluor 488 labeled) for 30 min in the dark. After thorough washing, nuclei were counterstained with DAPI (1 µg/mL). Samples were finally examined under a fluorescence or confocal microscope. EdU-positive nuclei exhibited bright green fluorescence (Alexa Fluor 488), and the EdU incorporation rate, which accurately reflects the proliferative activity of organoid cells, was calculated by comparing the number of EdU-positive nuclei with the total number of nuclei stained blue (DAPI). In the negative control group, PBS was used in place of EdU staining, confirming the specificity of the immunostaining.

Ki67 staining protocol

Organoid-Matrigel complexes cultured to the desired time points were fixed with 4% paraformaldehyde. After washing with PBS, samples were permeabilized with PBS containing 0.5% Triton X-100 to increase membrane permeability, followed by blocking of nonspecific binding sites with PBS containing 3% BSA. Subsequently, samples were incubated with a validated primary antibody against Ki67 (selected based on species, e.g., rabbit anti-mouse/human Ki67 monoclonal antibody) overnight at 4 °C. The following day, after thorough washing with PBS, the corresponding fluorescently labeled secondary antibody (e.g., Alexa Fluor 594-conjugated donkey anti-rabbit IgG) was applied, and the samples were incubated at room temperature for 1 h in the dark. Following another wash, nuclei were counterstained with DAPI. After mounting, samples were observed and images were acquired using a fluorescence microscope. For result interpretation, Ki67-positive signals (red fluorescence, Alexa Fluor 594) localized to the nuclei, overlapping with the blue nuclear outlines (DAPI). The proliferation index of the organoid cells was quantified by calculating the percentage of Ki67-positive nuclei relative to the total number of nuclei.

Tyramide Signal Amplification (TSA) multiplex immunofluorescence staining and image analysis

This experiment employed the Servicebio^® TSAPLus Six-Plex Seven-Color Fluorescence Staining Kit (G1257) to perform multiplex immunofluorescence on mouse lung paraffin sections, based on the Tyramide Signal Amplification (TSA) technology. The TSA technique utilizes HRP to catalyze the covalent deposition of fluorophore-labeled tyramides onto protein residues surrounding the antigen site, enabling significant signal amplification and stable deposition. Sequential staining of multiple targets is achieved by stripping the primary/secondary antibody complexes via microwave heat-induced epitope retrieval after each staining round, effectively preventing cross-reactivity even when primary antibodies from the same host species are used. Sample Pretreatment: After deparaffinization and rehydration, tissue sections underwent microwave heat-induced antigen retrieval using EDTA antigen retrieval buffer (pH 8.0). Endogenous Enzyme Quenching and Blocking: Endogenous peroxidase activity was quenched using 3% H₂O₂, followed by serum blocking with 3% BSA. Cyclic Staining: The following steps were performed sequentially and cyclically for each target: Primary Antibody Incubation: Sections were incubated with primary antibodies diluted in PBS overnight at 4 °C. HRP-Conjugated Secondary Antibody Incubation: After washing, HRP-conjugated secondary antibodies corresponding to the host species of the primary antibodies were applied and incubated for 50 min at room temperature. TSA Fluorescent Signal Amplification: Following washes, freshly prepared fluorophore-conjugated tyramide working solution (iF488, iF546, iF594, iF647, etc.) was applied according to the manufacturer’s instructions and incubated for 10 min at room temperature in the dark. Antibody Stripping: To prepare for the next staining cycle, sections were placed in EDTA retrieval buffer and subjected to microwave heating to remove the bound primary/secondary antibody complexes. Serum blocking was repeated after each stripping step. Counterstaining and Mounting: After completing all target stainings, cell nuclei were counterstained using the ready-to-use DAPI solution provided in the kit. An optional step of quenching tissue autofluorescence with Sudan Black was performed. Finally, sections were mounted with an anti-fade mounting medium. Image Acquisition: Images were acquired using a fluorescence microscope or slide scanner by capturing multi-channel images based on the excitation/emission wavelengths of the respective fluorophores. Negative control staining was performed with PBS substituted for the primary antibody, followed by incubation with the secondary antibody, confirming the specificity of the immunostaining.

Image J Quantitative Analysis: Image Preprocessing: Multi-channel fluorescence images were opened in ImageJ. The “Subtract Background” function was used to reduce background noise, and channels were aligned if necessary. Thresholding and Binarization: For each target channel, the “Image > Adjust > Threshold” function was used to manually or automatically set thresholds, separating specific signals from the background to create a binary mask. Cell Counting and Area Measurement: For punctate or granular signals (e.g., DAPI-stained nuclei), the “Analyze Particles” function was used with appropriate size and circularity parameters to automatically count objects and measure their area. For cytoplasmic or membrane signals (e.g., Sftpc, F4/80), the DAPI mask was first used to identify nuclei. Using the “Analyze > Analyze Particles” or “Measure” tools, parameters such as the Integrated Density of fluorescence or the positive area percentage (%Area) within a region expanded from the nuclei (using the “Dilate” function) were measured. Colocalization Analysis: Visual assessment was performed by creating composite images using the “Image > Color > Merge Channels” function. For quantitative analysis, the “Coloc 2” plugin was used to calculate coefficients such as Pearson’s Correlation Coefficient (Rr) or Manders’ Overlap Coefficients (M1/M2) to quantify the spatial overlap between signals from two channels. Data Export and Statistics: All measurement results were exported to spreadsheet software. Mean values and standard deviations were calculated for each group, and appropriate statistical tests (t-test, ANOVA) were applied. Results were presented as the number of positive cells, percentage of positive area, colocalization coefficients, etc.

Flow cytometric analysis of AT2 cells, CD8⁺ T cells, macrophages, and DC cells

Following the culture period, cells from the COPD mouse model were harvested and phenotypically characterized via flow cytometry. Analysis was performed using a FACS Calibur flow cytometer (BD Biosciences, USA), with data processed by Cell Quest software (BD Biosciences, USA). The following antibodies were employed: AT2 Cell Marker: SFTPC antibody (Thermo Fisher Scientific, #PA5-71680);CD8⁺ T Cell Marker: CD8 alpha (2.43) Rat Monoclonal Antibody (PerCP-Cy5.5^® Conjugate) (CST, #72461)༛Macrophage Marker: F4/80 Monoclonal Antibody (Cl: A3-1), FITC-labeled (Thermo Fisher Scientific, #MA5-16628)༛M1 Macrophage Co-stimulatory Molecule: CD86 Monoclonal Antibody, APC-labeled (Proteintech, #APC-65068)༛Dendritic Cell (DC) Markers: CD11c (N418) Hamster Monoclonal Antibody (FITC Conjugate) (CST, #51419); MHC Class II (I-A/I-E) (M5/114.15.2) Rat Monoclonal Antibody (APC Conjugate) (CST, #64776).

Cell isolation and cell lines

Lung tissues from wild-type mice (RS1^fl/fl+CS group) and alveolar type II epithelial cell-specific RS1-deficient mice (Sftpc^cre;RS1^fl/fl+CS group) were dissociated using a solution containing 25 mg/mL deoxyribonuclease (Sigma-Aldrich) and Liberase (0.5 mg/mL, Sigma-Aldrich). AT2 cells were isolated and purified from the lung tissues by flow cytometry. Subsequently, the AT2 cells were stained with antibodies and sorted by fluorescence-activated cell sorting (FACS) to achieve a purity ≥ 98%. The gating strategy for FACS is detailed in Supplementary Fig. 2. The activation of NEDD4 was selectively inhibited using Pevonedistat (I3C, MCE).

Plasmid construction, cell transfection, and establishment of stable overexpression systems in alveolar type II epithelial cells

To investigate the functions of RS1 and NEDD4 in the alveolar epithelium, genetic manipulations were performed using a lentiviral system. First, short hairpin RNA (shRNA) expression vectors targeting RS1 were constructed and transduced into a human alveolar type II epithelial cell model via lentivirus to knock down endogenous RS1 expression.

Concurrently, to establish alveolar type II epithelial cell lines stably overexpressing RS1 or NEDD4, the following protocol was employed: The full-length coding sequences of human RS1 or NEDD4 were cloned into a lentiviral overexpression vector (pCDH-CMV-MCS-EF1-Puro). This vector was then co-transfected with packaging plasmids into HEK293T cells to produce viral particles. The viral supernatant was collected and used to infect the target alveolar type II epithelial cells, with polybrene added during infection to enhance transduction efficiency. Following infection, cells were continuously selected using 2 µg/mL puromycin to obtain stable overexpression cell lines. Efficient overexpression of the target genes in the alveolar type II epithelial cells was confirmed at both the mRNA and protein levels by qRT-PCR and Western Blot, respectively, providing reliable cellular models for subsequent mechanistic studies.

Plasmid construction

To elucidate the interactions and molecular mechanisms between RS1, NEDD4, and YAP1, a series of expression vectors were constructed. First, expression vectors for MYC-tagged wild-type RS1 (MYC-RS1), Flag-tagged NEDD4 (Flag-NEDD4), and HA-tagged YAP1 (HA-YAP1) were generated. The cDNA sequences for all genes were inserted into the vectors via EcoRI/NotI multiple cloning sites. To further identify critical functional domains of RS1, based on molecular docking simulations, RS1 mutants were constructed: a deletion mutant lacking amino acids 100–110 (MYC-RS1-Δ100–110) and a point mutant with serine 105 (S105) mutated to alanine (MYC-RS1-S105A). The sequences of all constructs were verified by sequencing, and the expected molecular weights of the expressed fusion proteins were confirmed by Western Blot to ensure correct expression.

Western blot

Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was used to extract total cellular protein, which was then transferred onto a polyvinylidene fluoride (PVDF) membrane (Millipore, Billerica, MA, USA). Thereafter, BSA was added for membrane blocking and antibodies were added to incubate membrane under 4 °C overnight.

Antibodies against: RS1 Rabbit mAb (Proteintech, #24430-1-AP); YAP (D8H1X) Rabbit Monoclonal Antibody (CST, #14074); Phospho-YAP (Ser127) Antibody (CST, #4911); GAPDH (14C10) Rabbit mAb (CST, #2118); Anti-rabbit IgG, HRP-linked Antibody (CST, #7074); Anti-mouse IgG, HRP-linked Antibody (CST, #7076).

Co-immunoprecipitation

A previously described method was used to study protein-protein interactions through co-immunoprecipitation (Co-IP). Lipofectamine 3000 was used to transfect plasmids into lung cancer cells for 48 h. The cells that were transfected were broken down using RIPA buffer, which was enhanced with a blend of protease and phosphatase inhibitors. Primary antibodies were added to the lysate and left to incubate overnight at 4 °C. The mixture of cleavage products and antibodies was then combined with magnetic beads and rotated for 50 min at 25 °C. Subsequently, the plates were rinsed three times with 200 µL of PBS at room temperature, with each rinse taking 10 min. To conduct western blot analysis, a sample was prepared by heating 35 µL of SDS sample buffer at 95 °C for 10 min, then briefly centrifuged at 4500 r.p.m. for 30 s.

Antibodies against: RS1 Rabbit mAb [EPR28513-78] (abcam, ab314231); YAP1 Rabbit mAb (Proteintech, #13584-1-AP); NEDD4 (Y297) Rabbit mAb (abcam, #ab81264); Secondary antibodies: Mouse Anti-rabbit IgG (Conformation Specific) (L27A9) monoclonal antibody (HRP Conjugate) #5127.

Real-time PCR

To quantitatively analyze the mRNA expression levels of TGFB1, CCL-2, IL-3, and IL-4, quantitative real-time PCR (qPCR) was performed. Total RNA was extracted from cells using TRIzol Reagent and reverse-transcribed into cDNA using a StepOne Real-Time PCR system. PCR reactions were carried out in 20–25 µL volumes containing cDNA template, specific primers TGFB1 F: CTCCCGTGGCTTAGTGCR, R: GCCTTAGTTGGACAGATGTC;

CCL-2 F: TTAAAAACCTGGATCGGAACCAA, R: GCATTAGCTTCAGATTTACGGGT;

IL-3 F: TGTGCTGAATGTGGACTCAATCC, R: ACCCTAAGGCAGGCAGTTTG;

IL-4 F: CCATATCCAGGATGGCACA, R: AAAGCCCGAAAGAGTCTCTGC;

GAPDH F: CATCACTGCCACCCAGAAGACTG, R: ATGCAGGTGAGCTTCCGGTCAG, SYBR Green dye, and the corresponding enzymes and substrates. The thermal cycling protocol consisted of an initial denaturation at 95 °C for 5–10 min, followed by 40–45 cycles of amplification (95 °C for 15–30 s, 60 °C for 30–60 s), and a final melt curve analysis (from 60 °C to 95 °C, with signal acquisition at 0.5 °C increments) to verify product specificity. Each sample was run in triplicate. Relative gene expression was calculated using the 2^-ΔΔCt method and normalized to GAPDH. Data analysis was performed using StepOne Software v2.2.1.

RNA sequencing analysis

AT2 cells were later sorted from from wild type mice (RS1^fl/fl mice + CS group) and alveolar type II epithelial (AT2) specific RS1 deficient mice (Sftpc^creRS1^fl/fl+CS group) through flow cytometry. AT2 cells from the above mice were sequenced using the Illumina HiSeq 2500 system. Using the hypergeometric distribution, DEGs were subsequently enriched into KEGG pathways with the help of R software. Furthermore, gene functions were annotated using gene set enrichment analysis (GSEA) based on the Molecular Signature Database (MSigDB).

Statistics

GraphPad Prism 9.3 and SPSS 22.0 were used for drawing and statistical analysis, with each experiment being performed three times. The T-test or Mann-Whitney U test was used to compare measurement data between the two groups, depending on the normality test results. Depending on whether the data variance was uniform, either analysis of variance or the Kruskal-Wallis test was used to compare and analyze measurement data across different groups. When a statistically significant difference was found between groups, pairwise comparisons were conducted using the Least-Significant Difference (LSD) method. Fisher’s exact test or the chi-square test was used for comparing count data. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 stood for statistical significance.

Results

The interaction network of RS1⁺ SFTPC⁺ AT2 cells in COPD

To elucidate the interaction landscape between SFTPC⁺ AT2 cells and other cell types in Chronic Obstructive Pulmonary Disease (COPD), cell annotation was performed based on the human COPD single-cell transcriptomic dataset GSE136831. This encompassed populations including SFTPC⁻ AT2 cells, SFTPC⁺ AT2 cells, CD8⁺ T cells, CD4⁺ T cells, ciliated cells, dendritic cells (DCs), endothelial cells, fibroblasts, AT1 cells, mast cells, monocyte/macrophages, B cells, and plasma cells. Construction of a cell-cell interaction network (Fig. 1A) revealed distinct interaction patterns. SFTPC⁻ AT2 cells primarily interacted with CD8⁺ T cells, fibroblasts, and AT1 cells (Fig. 1B). In contrast, SFTPC⁺ AT2 cells exhibited extensive interactions with a variety of cell types, including CD8⁺ T cells, DCs, monocyte/macrophages, fibroblasts, AT1 cells, endothelial cells, and B cells (Fig. 1C). This indicates that SFTPC⁺ AT2 cells are associated with multiple immune cell types. Further analysis identified that RS1 expression was predominantly localized to AT1 and AT2 cells (Fig. 1D). UMAP dimensionality reduction and expression analysis confirmed that RS1 expression was significantly enriched in alveolar epithelial cells, mainly in AT1 and AT2 cells, with lower expression levels in other lung cell types (Supplementary Fig. 1A-C). Network analysis revealed that RS1⁻SFTPC⁺ AT2 cells were associated with CD4⁺ T cells, fibroblasts, and AT1 cells (Fig. 1E), whereas RS1⁺SFTPC⁺ AT2 cells were associated with various immune cells, most notably exhibiting interactions with monocytes/macrophages (Fig. 1F). This suggests that RS1⁺SFTPC⁺ AT2 cells may participate in COPD progression together with monocytes/macrophages.

Fig. 1.

Analysis of cellular interaction networks involving SFTPC⁺ AT2 cells and their RS1-expressing subpopulations in COPD. (A) Interaction network of major lung cell types constructed based on the human COPD single-cell dataset (GSE136831). (B, C) Specific interaction profiles of SFTPC⁻ AT2 cells (B) and SFTPC⁺ AT2 cells (C). (D) Expression distribution of the RS1 gene across lung cell types. (E, F) Interaction networks of RS1⁻ (E) and RS1⁺ (F) SFTPC⁺ AT2 cell subpopulations. (G) Expression localization of RS1 in the mouse COPD dataset (GSE168299). (H, I) Panoramic and detailed interaction views of RS1⁻SFTPC⁺ AT2 cells (H) and RS1⁺SFTPC⁺ AT2 cells (I) in mouse data. (J) KEGG signaling pathways significantly enriched for RS1 in AT2 cells. The analysis indicates that SFTPC⁺ AT2 cells, particularly the RS1⁺ subpopulation, are involved in COPD pathogenesis through interactions with immune cells and engagement in inflammation-related pathways

To validate these findings, a cross-species analysis was conducted using a mouse COPD single-cell dataset (GSE168299). Consistent with human data, RS1 was also primarily localized to AT1 and AT2 cells in mice (Fig. 1G). Furthermore, mouse RS1⁺SFTPC⁺ AT2 cells were associated with multiple immune cell types, especially monocytes/macrophages (Fig. 1H, I). AT2 cells were sorted by flow cytometry from wild-type mice (RS1^fl/fl mice + CS group) and AT2 cell-specific RS1-deficient mice (Sftpc^creRS1^fl/fl + CS group). Subsequent sequencing and KEGG pathway enrichment analysis demonstrated that RS1 in AT2 cells was significantly associated with inflammatory response and the WNT signaling pathway (Fig. 1J). In summary, these results indicate that RS1⁺SFTPC⁺ AT2 cells play a crucial role in COPD inflammation.

RS1 suppresses AT2 cell proliferation and RS1⁺ AT2 cells promote inflammatory infiltration in COPD

To investigate the regulatory role of RS1 in AT2 cells during Chronic Obstructive Pulmonary Disease (COPD) progression, we first generated RS1^fl/fl mice and AT2 cell-specific RS1 knockout mice (Sftpc^cre; RS1^fl/fl). Subsequently, COPD models were established using cigarette smoke (CS) exposure (RS1^fl/fl+CS and Sftpc^cre; RS1^fl/fl+CS groups), while PBS-treated groups served as controls (RS1^fl/fl+PBS and Sftpc^cre; RS1^fl/fl+PBS groups) (Fig. 2A). We first performed cell counting using bronchoalveolar lavage fluid (BALF). Compared with the RS1^fl/fl+PBS group, the Sftpc^cre; RS1^fl/fl+PBS group showed decreased total cell counts, macrophages, and lymphocytes, while neutrophils remained unchanged. Compared with the RS1^fl/fl+CS group, the Sftpc^cre; RS1^fl/fl+CS group, with specific RS1 knockout in AT2 cells, exhibited decreased total cell counts, macrophages, and lymphocytes, whereas neutrophils showed no significant change (Supplement Fig. 2A-D). Subsequently, H&E staining of lung tissues revealed prominent inflammatory cell infiltration in the lungs of RS1^fl/fl+CS mice, which was significantly attenuated in the Sftpc^cre; RS1^fl/fl+CS group with AT2 cell-specific RS1 knockout (Fig. 2B).

Fig. 2.

RS1 regulates AT2 cell function and disease pathology in a COPD mouse model. (A) Schematic of the experimental design: Wild-type (RS1^fl/fl) and AT2 cell-specific RS1 knockout (Sftpc^cre; RS1^fl/fl) mice were exposed to cigarette smoke (CS) or PBS. (B) Representative H&E-stained lung tissue images show that RS1 deficiency attenuates CS-induced inflammatory infiltration (n = 5); scale bar: 2 mm; inset scale bar: 200 μm. (C) Quantitative analysis of the mean alveolar intercept indicates that RS1 deficiency alleviates CS-induced emphysema (n = 5). (D) AT2 cells sorted from mouse lung tissue for organoid culture, scale bar: 50 μm. (E) Quantitative comparison of organoid diameters across groups (n = 5). (F, G) Assessment of organoid cell proliferation by EdU staining (F) and Ki67 staining (G) (n = 5); EdU (green), DAPI (blue), scale bar: 50 μm. In this experiment, each group consisted of five mice, and the RS1^fl/fl + PBS group also consisted of five mice. Data are presented as mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (two-tailed t-test)

Further assessment of emphysema severity by measuring the mean alveolar intercept showed no significant difference between the RS1^fl/fl+PBS and Sftpc^cre; RS1^fl/fl+PBS groups. However, compared with the RS1^fl/fl+CS group, the Sftpc^cre; RS1^fl/fl+CS group exhibited a significantly reduced mean alveolar intercept (Fig. 2C), indicating that RS1 deficiency attenuates COPD-associated lung structural destruction. To investigate the regulatory effect of RS1 on AT2 cell proliferative capacity, AT2 cells were sorted by flow cytometry and subjected to three-dimensional lung organoid culture (Fig. 2D). Organoid formation assays demonstrated that organoids derived from the RS1^fl/fl+PBS group had a significantly smaller diameter than those from the Sftpc^cre; RS1^fl/fl+PBS group; similarly, organoids from the RS1^fl/fl+CS group were smaller than those from the Sftpc^cre; RS1^fl/fl+CS group (Fig. 2E), suggesting that RS1 inhibits AT2 cell proliferation. To directly test this hypothesis, we employed EdU incorporation assay (Fig. 2F, negative control shown in Supplement Fig. 3A) and Ki67 immunofluorescence staining (Fig. 2G) to assess the proliferative activity of organoid cells. The results showed that, compared with the RS1^fl/fl+PBS group, the proportion of proliferating cells was elevated in organoids from the Sftpc^cre; RS1^fl/fl+PBS group; compared with the RS1^fl/fl+CS group, the proportion of proliferating cells was significantly increased in organoids from the Sftpc^cre; RS1^fl/fl+CS group. Collectively, these results demonstrate that RS1 suppresses AT2 cell proliferation and that RS1⁺ AT2 cells promote inflammatory infiltration in COPD.

RS1 suppresses AT2 cell proliferation and RS1⁺ AT2 cells promote macrophage infiltration

To elucidate the specific cell types through which RS1 exerts its regulatory effects in COPD, we first generated RS1^fl/fl mice and AT2 cell-specific RS1 knockout mice (Sftpc^cre; RS1^fl/fl). Subsequently, COPD models were established using cigarette smoke (CS) exposure (RS1^fl/fl+CS and Sftpc^cre; RS1^fl/fl+CS groups), while PBS-treated groups served as controls (RS1^fl/fl+PBS and Sftpc^cre; RS1^fl/fl+PBS groups) (Fig. 2A). Flow cytometry was performed to analyze the proportions of lung immune cells and AT2 cells (gating strategy detailed in Supplement Fig. 4). Flow cytometry results showed that the proportion of AT2 cells was higher in the Sftpc^cre; RS1^fl/fl+PBS group compared with the RS1^fl/fl+PBS group; similarly, the proportion of AT2 cells was significantly higher in the Sftpc^cre; RS1^fl/fl+CS group compared with the RS1^fl/fl+CS group (Fig. 3A). Furthermore, compared with the RS1^fl/fl+PBS group, the proportions of CD8⁺ T cells and macrophages were reduced in the Sftpc^cre; RS1^fl/fl+PBS group; compared with the RS1^fl/fl+CS group, the proportions of CD8⁺ T cells and macrophages were significantly reduced in the Sftpc^cre; RS1^fl/fl+CS group (Fig. 3B, C). However, the proportion of dendritic cells (DCs) showed no significant change between the RS1^fl/fl+PBS and Sftpc^cre; RS1^fl/fl+PBS groups, nor between the RS1^fl/fl+CS and Sftpc^cre; RS1^fl/fl+CS groups. These results indicate that RS1 suppresses AT2 cell proliferation, and RS1⁺ AT2 cells promote infiltration of CD8⁺ T cells and macrophages.

Fig. 3.

Conditional knockout of RS1 in AT2 cells alters lung cellular composition in a COPD model. (A-D) Flow cytometric quantification of the proportions of (A) AT2 cells (PDGFRα⁺EpCAM⁺SFTPC⁺; % of PDGFRα⁺EpCAM⁺ cells), (B) CD8⁺ T cells (CD3⁺CD8⁺; % of CD31⁻CD45⁺ cells), (C) M1 macrophages (CD11b⁺F4/80⁺CD86⁺; % of CD31⁻CD45⁺ cells), and (D) dendritic cells (CD11c⁺MHC II⁺; % of CD31⁻CD45⁺ cells) in the lungs of mice from each group (n = 5). (E) Representative multiplex immunofluorescence images of lung tissue showing AT2 cells (Sftpc, blue), M1 macrophages (F4/80, green; CD86, red), and nuclei (DAPI, grey). Scale bar: 2 mm; inset scale bar: 200 μm. (F, G) Quantitative image analysis showing the number of (F) AT2 cells and (G) M1 macrophages per 100 total nuclei (n = 5). In this experiment, each group consisted of five mice, and the RS1^fl/fl+PBS group also consisted of five mice. Data are presented as mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (two-tailed t-test)

Based on prior single-cell interaction analysis suggesting a close association between RS1⁺SFTPC⁺ AT2 cells and macrophages, we further employed tissue-based multiplex immunofluorescence for spatial validation. Representative images show the distribution of AT2 cells (SFTPC⁺) and M1 macrophages (CD86⁺) (Fig. 3E, negative control shown in Supplement Fig. 3B). Quantitative analysis confirmed that, compared with the RS1^fl/fl+PBS group, the number of AT2 cells was increased in the Sftpc^cre; RS1^fl/fl+PBS group; compared with the RS1^fl/fl+CS group, the number of AT2 cells was also increased in the Sftpc^cre; RS1^fl/fl+CS group (Fig. 3F). Additionally, compared with the RS1^fl/fl+PBS group, the number of macrophages was decreased in the Sftpc^cre; RS1^fl/fl+PBS group; compared with the RS1^fl/fl+CS group, the number of macrophages was significantly decreased in the Sftpc^cre; RS1^fl/fl+CS group (Fig. 3G). In summary, RS1 suppresses AT2 cell proliferation, and RS1⁺ AT2 cells promote macrophage infiltration.

RS1 inhibits YAP1 phosphorylation and promotes its nuclear translocation via NEDD4

To elucidate the molecular mechanism by which RS1 regulates AT2 cells, we performed transcriptome sequencing on AT2 cells sorted from the lungs of Sftpc^cre; RS1^fl/fl+CS and RS1^fl/fl+CS mice. Gene Set Enrichment Analysis (GSEA) revealed that RS1 was significantly associated with gene sets related to the Hippo signaling pathway and the HECT domain (Fig. 4A). To validate the association between RS1 and the Hippo signaling pathway, we extracted AT2 cells from RS1^fl/fl+CS mice and examined the effect of RS1 on YAP1 protein expression and phosphorylation. Western blot analysis showed that overexpression of RS1 in AT2 cells significantly reduced the level of phosphorylated YAP1 (p-YAP1) while increasing total YAP1 protein expression (Fig. 4B). Conversely, knockdown of RS1 in AT2 cells significantly increased p-YAP1 levels and decreased total YAP1 expression (Fig. 4C). These results indicate that RS1 inhibits YAP1 phosphorylation.

Fig. 4.

Mechanistic investigation of RS1 regulating the Hippo-YAP signaling pathway via NEDD4. (A) GSEA of the AT2 cell transcriptome showing gene sets significantly enriched upon RS1 deletion, related to the Hippo signaling pathway and the HECT domain. (B, C) Western blot analysis showing the effects of RS1 overexpression (B) or knockdown (C) on phosphorylated YAP1 (p-YAP1) and total YAP1 protein levels (n = 3). (D) Western blot verification that RS1 regulation of YAP1 phosphorylation is dependent on NEDD4 (n = 3). (E) Western blot results showing changes in p-YAP1 and total YAP1 protein levels in AT2 cells from RS1^fl/fl+CS mice treated with the NEDD4 inhibitor I3C (n = 3). (F) Nuclear Western blot analysis confirming that RS1 promotes YAP1 nuclear accumulation via NEDD4 (n = 3). Data are presented as mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (two-tailed t-test)

The HECT domain is the core functional module of HECT-type E3 ubiquitin ligases, responsible for substrate recognition and ubiquitination [19, 20]. NEDD4-mediated neddylation modification is crucial for regulating the stability and activity of YAP1, a key effector of the Hippo pathway [21, 22]. Therefore, we hypothesized that RS1 regulates YAP1 function by influencing NEDD4. Western blot experiments confirmed that knocking down RS1 promoted YAP1 phosphorylation, while subsequent overexpression of NEDD4 suppressed YAP1 phosphorylation (Fig. 4D). To elucidate the role of NEDD4 in the RS1-YAP1 axis, we determined a safe concentration of the selective NEDD4 inhibitor I3C via IC50 evaluation and treated AT2 cells isolated from RS1^fl/fl+CS mice with I3C. The results showed that inhibiting NEDD4 increased YAP1 phosphorylation levels and decreased total YAP1 protein expression in AT2 cells (Fig. 4E). Nuclear-cytoplasmic fractionation experiments further demonstrated that overexpressing NEDD4 on a background of RS1 knockdown promoted YAP1 accumulation in the nucleus, whereas inhibiting NEDD4 on a background of RS1 overexpression reduced YAP1 nuclear accumulation (Fig. 4F). In summary, these results demonstrate that RS1 inhibits YAP1 phosphorylation and promotes its nuclear translocation through NEDD4.

RS1 mediates the interaction between NEDD4 and YAP1 via serine 105

To further elucidate the molecular mechanism by which RS1 regulates YAP1 through NEDD4, we conducted mechanistic studies using lung tissues from RS1^fl/fl+CS mice. Immunofluorescence staining of lung tissue confirmed the co-localization of RS1 and NEDD4 within AT2 cells (Fig. 5A, negative control shown in Supplement Fig. 3C). Retinoschisin (RS1) is a soluble secreted protein typically localized to the extracellular side of the plasma membrane in an octameric form, while a monomeric form exists in the cytoplasm. Given that its intracellular monomeric form may participate in signal regulation, we extracted AT2 cells from RS1^fl/fl+CS mice for mechanistic studies. Co-immunoprecipitation (Co-IP) experiments verified the interaction between RS1 and NEDD4 at the protein level (Fig. 5B, C). Structure prediction via AlphaFold and molecular docking simulations further indicated the potential for interaction among RS1, NEDD4, and YAP1 (Fig. 5D). Subsequent Co-IP experiments confirmed a direct interaction between RS1 and YAP1 (Fig. 5E, F). Molecular docking simulations further suggested that the amino acid region 100–110 of the RS1 protein might constitute a key interface for its binding to NEDD4 and YAP1 (Fig. 5G). To test this prediction, we constructed plasmids for wild-type Myc-RS1 and a mutant with a deletion of amino acids 100–110, and co-transfected them with HA-YAP1 and Flag-NEDD4 into AT2 cells. Co-IP results showed that wild-type RS1 significantly enhanced the interaction between NEDD4 and YAP1, whereas the RS1 mutant lacking amino acids 100–110 weakened this interaction (Fig. 5H). Further studies identified serine 105 (S105) within this region as crucial: Co-IP results demonstrated that wild-type RS1 significantly enhanced the NEDD4-YAP1 interaction, while the RS1 mutant with S105 deletion diminished this interaction (Fig. 5I). In summary, these results indicate that RS1 mediates the interaction between NEDD4 and YAP1 via serine 105.

Fig. 5.

Mechanistic study of RS1 as a molecular scaffold mediating the interaction between NEDD4 and YAP1. (A) Immunofluorescence staining of lung tissue from a C57BL/6 mouse COPD model shows co-localization of RS1 (green) and NEDD4 (red) in AT2 cells (SFTPC⁺, purple). Nuclei are stained with DAPI (blue). Scale bar: 50 μm; inset scale bar: 20 μm. (B, C) Co-IP experiments validate the binding between endogenous RS1 and NEDD4. (D) AlphaFold-predicted structural model of the ternary complex involving RS1, NEDD4, and YAP1. (E, F) Co-IP experiments validate the binding between endogenous RS1 and YAP1. (G) Molecular docking simulation reveals the amino acid region 100–110 of RS1 (green) as a key interface for interaction with NEDD4 (purple) and YAP1 (red). (H, I) Co-IP experiments demonstrate that wild-type RS1 (H) enhances the binding between Flag-NEDD4 and HA-YAP1, whereas RS1 lacking amino acids 100–110 (H) or carrying a mutation at serine 105 (I) impairs this effect

RS1⁺SFTPC⁺ AT2 cells mediate macrophage interactions via the TGF-β signaling pathway

To investigate the downstream signaling network regulated by RS1 in AT2 cells, we performed transcriptome sequencing on AT2 cells isolated from the lungs of Sftpc^cre; RS1^fl/fl+CS and RS1^fl/fl+CS mice. Gene Set Enrichment Analysis (GSEA) revealed that RS1 was significantly enriched in the “cytokine-cytokine receptor interaction” pathway (Fig. 6A) and the “TGF-β signaling pathway” (Fig. 6B), indicating that RS1 is involved in regulating inflammation-related pathways in COPD. To further validate this association in the context of human disease, we analyzed a human COPD single-cell transcriptomic dataset (GSE136831). Cell-cell communication analysis demonstrated that RS1⁺SFTPC⁺ AT2 cells established interactions via the TGF-β signaling pathway with macrophages, fibroblasts, CD8⁺ T cells, and endothelial cells (Fig. 6C, D). Cellular interaction network analysis further revealed that the TGF-β signaling pathway mediates the interaction between SFTPC⁺ AT2 cells and macrophages (Fig. 6E, F). In summary, RS1⁺SFTPC⁺ AT2 cells interact with macrophages through the TGF-β signaling pathway.

Fig. 6.

RS1⁺SFTPC⁺ AT2 cells regulate the cellular interaction network in the COPD microenvironment via the TGF-β Signaling Pathway. (A, B) GSEA of the transcriptome from RS1-deficient AT2 cells showing significant enrichment in the “Cytokine-cytokine receptor interaction” (A) and “TGF-β signaling” (B) pathways. (C, D) Cell-cell communication analysis based on the human COPD single-cell dataset (GSE136831), illustrating specific interactions between RS1⁺SFTPC⁺ AT2 cells and macrophages, fibroblasts, CD8⁺ T cells, and endothelial cells via the TGF-β pathway. (E, F) The analysis reveals that the TGF-β signaling pathway is the core pathway mediating a complex network involving SFTPC⁺ AT2 cells, macrophages, fibroblasts, CD4⁺ T cells, CD8⁺ T cells, and endothelial cells

RS1 regulates TGFB1 expression and macrophage infiltration via NEDD4

To elucidate whether RS1 regulates downstream cytokine expression through its interaction with NEDD4, we first generated RS1^fl/fl mice and AT2 cell-specific RS1 knockout mice (Sftpc^cre; RS1^fl/fl). Subsequently, control groups (RS1^fl/fl+PBS and Sftpc^cre; RS1^fl/fl+PBS groups) were established using PBS treatment, while COPD model groups (Sftpc^cre; RS1^fl/fl+CS group) were established using cigarette smoke (CS) exposure. To validate the role of NEDD4 in this process, the Sftpc^cre; RS1^fl/fl+CS group was additionally treated with I3C, a specific inhibitor of the NEDD4-activating enzyme, constituting the Sftpc^cre; RS1^fl/fl+CS+I3C group. RT-PCR was performed to detect the mRNA levels of key cytokines in AT2 cells from each group. The results showed that, compared with the Sftpc^cre; RS1^fl/fl+CS group, the expression of TGF-β1 was significantly decreased in the Sftpc^cre; RS1^fl/fl+CS+I3C group, while the expression of CCL-2, IL-13, and IL-4 showed no significant changes (Fig. 7A-D). These results indicate that RS1 suppresses TGF-β1 expression through NEDD4. We further analyzed macrophage infiltration by multiplex immunofluorescence staining of lung tissue and flow cytometry. The results showed that, compared with the RS1^fl/fl+PBS group, the proportion of macrophages was significantly increased in the RS1^fl/fl+CS group; compared with the Sftpc^cre; RS1^fl/fl+CS group, the proportion of macrophages was decreased in the Sftpc^cre; RS1^fl/fl+CS+I3C group (Fig. 7E, F, negative control shown in Supplement Fig. 4D). Flow cytometry analysis further confirmed that, compared with the RS1^fl/fl+PBS group, the proportion of macrophages was significantly increased in the RS1^fl/fl+CS group; compared with the Sftpc^cre; RS1^fl/fl+CS group, the proportion of macrophages was decreased in the Sftpc^cre; RS1^fl/fl+CS+I3C group (Fig. 7G). These results demonstrate that RS1 suppresses TGF-β1 expression via NEDD4 and influences macrophage infiltration.

Fig. 7.

RS1 regulates downstream cytokine expression and macrophage infiltration via NEDD4. (A-D) Relative mRNA expression levels of TGF-β1, CCL-2, IL-13, and IL-4 in AT2 cells from each group, measured by RT-PCR (n = 5). Expression of these cytokines was significantly lower in the Sftpc^cre; RS1^fl/fl+CS group compared to the RS1^fl/fl+CS group. (E) Representative multiplex immunofluorescence images of lung tissue showing alveolar type II epithelial cells (SFTPC, blue) and M1 macrophage (F4/80, green; CD86, red) infiltration. Nuclei are stained with DAPI (grey) (n = 3). Scale bar: 200 μm; inset scale bar: 50 μm. (F) Quantification of M1 macrophages per 100 total nuclei (n = 5). (G) Flow cytometric quantification of the proportion of M1 macrophages (% of CD31⁺CD45⁺ cells) in the lung tissue of each group (n = 5). In this experiment, each group consisted of five mice, and the RS1^fl/fl + PBS group also consisted of five mice. Data are presented as mean ± SEM; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 (two-tailed t-test)

Discussion

This study established AT2 cell-specific RS1 knockout (Sftpc^cre;RS1^fl/fl) COPD mouse models and organoid models to elucidate the specific expression of retinoschisin 1 (RS1) in alveolar type II epithelial cells during chronic obstructive pulmonary disease (COPD). Phenotypically, it was found that RS1 deficiency enhanced the proliferative capacity of AT2 cells and inhibited macrophage infiltration. Mechanistically, RS1 mediates the interaction between NEDD4 and YAP1 through serine at position 105. Concurrently, the interaction between RS1 and NEDD4 facilitates YAP1 dephosphorylation and nuclear translocation. Finally, we demonstrated that the selective NEDD4 inhibitor I3C can further suppress TGFB1 expression and macrophage infiltration.

This study reveals the critical regulatory role of RS1 in alveolar type II epithelial (AT2) cells and its modulation of macrophage infiltration in chronic obstructive pulmonary disease (COPD). While previous research on RS1 has primarily focused on the pathogenesis, diagnosis, and targeted therapy of retinoschisis, with some studies reporting its role in lung cancer [13, 15], our work integrates single-cell transcriptomic analysis, conditional gene knockout mouse models, organoid cultures, and multiplex immunofluorescence validation. We discovered that RS1 is specifically expressed in AT2 cells of COPD patients and correlates with immune cell populations. Notably, the interaction network between the RS1⁺SFTPC⁺ AT2 cell subpopulation and monocytes/macrophages suggests this subset serves as a hub connecting AT2 cells with immune cells. Functionally, RS1 deficiency in AT2 cells significantly alleviated lung tissue damage, alveolar structure destruction, and inflammatory cell infiltration in COPD mice, while enhancing AT2 cell proliferative capacity. Further analysis demonstrated that RS1 deletion in AT2 cells not only increased the proportion of AT2 cells but also markedly reduced macrophage presence in the COPD mouse model.

Regarding the molecular mechanism, this study elucidates how RS1 regulates YAP1 activity through NEDD4 in alveolar type II epithelial (AT2) cells during chronic obstructive pulmonary disease (COPD). Transcriptomic sequencing and subsequent validation revealed that RS1 is closely associated with the Hippo signaling pathway, enhancing YAP1 transcriptional activity by inhibiting YAP1 phosphorylation and promoting total YAP1 protein expression. Within the Hippo signaling pathway, YAP1 plays a pivotal role in nuclear transcriptional regulation of inflammatory factors. When the Hippo pathway is inactivated, unphosphorylated YAP1 translocates to the nucleus, where it forms transcriptional complexes with transcription factors such as TEAD, directly activating the expression of various inflammation-related genes [23–25]. Mechanistically, RS1 utilizes its intracellular monomeric form to directly mediate the interaction with the E3 ubiquitin ligase NEDD4 and YAP1 through serine 105 (S105). Functional experiments demonstrated that RS1 relies on NEDD4 to exert its regulatory effect on YAP1: on one hand, RS1 promotes YAP1 nuclear translocation via NEDD4; on the other hand, treatment with the specific NEDD4 inhibitor I3C completely reverses RS1-mediated YAP1 dephosphorylation and nuclear translocation. These findings indicate that RS1 facilitates YAP1 dephosphorylation and nuclear translocation through NEDD4.

This study further uncovers the molecular mechanism by which RS1⁺SFTPC⁺ AT2 cells regulate macrophage infiltration through the TGF-β signaling pathway. Transcriptomic sequencing and signaling pathway enrichment analysis revealed significant enrichment of RS1 in the TGF-β signaling pathway and cytokine-related pathways; integrated with public single-cell dataset validation, RS1⁺SFTPC⁺ AT2 cells indeed exhibit an interaction network with macrophages through the TGF-β signaling pathway, suggesting that this cell subpopulation may reshape the local immune microenvironment via paracrine secretion of TGF-β family cytokines. Pulmonary epithelial cells not only serve as physical barriers but also actively participate in immune responses, maintaining pulmonary immune homeostasis through bidirectional communication with immune cells [26–28]. Functionally, experimental systems utilizing AT2 cell-specific RS1 knockout mice combined with NEDD4 inhibitor I3C intervention confirmed that RS1 upregulates TGF-β1 expression through a NEDD4-dependent mechanism; inhibiting NEDD4 activity not only significantly reduced TGF-β1 transcriptional levels but also effectively decreased the number of infiltrating macrophages in lung tissue, demonstrating that TGF-β1 is the primary molecule mediating macrophage chemotaxis through the RS1-NEDD4 axis.

This study has several limitations: First, although the role of RS1 in AT2 cells has been systematically confirmed through single-cell sequencing, conditional gene knockout mice, and organoid models, direct validation using clinical patient samples is lacking. Second, while we have demonstrated that RS1 mediates the interaction between NEDD4 and YAP1 through serine 105, the specific structural domains and post-translational modifications involved in RS1 binding to NEDD4 have not been fully elucidated. Third, although the mechanistic studies proposed that RS1 promotes YAP1 dephosphorylation and nuclear translocation via NEDD4, the precise molecular mechanism by which YAP1 regulates TGF-β1 expression following nuclear entry remains unresolved. Fourth, the mechanism by which TGF-β1 influences macrophage polarization status has not been thoroughly investigated.

Based on these findings, this study proposes the following conclusions: RS1 suppresses the proliferative capacity of AT2 cells while promoting macrophage infiltration. RS1 interacts with NEDD4 to mediate YAP1 dephosphorylation and nuclear translocation, and treatment with the selective NEDD4 inhibitor I3C can further inhibit TGFB1 expression and macrophage infiltration. This study elucidates at the molecular level the mechanism by which RS1 drives macrophage infiltration through NEDD4, providing new insights into how AT2 cells shape the immune microenvironment in COPD. These findings suggest that targeting the RS1-NEDD4 axis may represent a potential strategy for intervening in the inflammatory progression of COPD.

Authors' contributions

ZS-Z, WH and SY-X offered direction and guidance of the manuscript. ZS-Z, CF-C and HJ drafted the initial manuscript. ZS-Z, WH and SY-X revised the manuscript. ZS-Z, CF-C, HJ and WH illustrated the results for the manuscript. All authors approved the final manuscript.

Funding

This work was supported by the 2023 Annual Basic Research Projects of the Provincial Department of Education (JYTMS20231385), Shenyang Municipal Science and Technology Bureau Projects (241016, 2022062, 241020, 22-321-33-96), Autonomous Research Projects of the Liaoning Provincial Department of Education in Science and Technology (LJ212410164020), Research Projects of the Shenyang Municipal Health Commission (202316), Joint Fund of the Liaoning Provincial Natural Science Foundation (2023-MSLH-301).

Data availability

Not applicable.

Declarations

Ethics approval and consent to participate

The study was conducted in accordance with the Declaration of Helsinki. All procedures both on humans and mice were approved by the Ethics Committee of Shenyang Medical College.

All animal experiments in our study comply with the ARRIVE guidelines and should be conducted in accordance with the UK Animals (Scientific Procedures) Act 1986 and its associated guidelines, the EU Directive 2010/63/EU on animal experiments, and the NIH Guide for the Care and Use of Laboratory Animals (NIH Publication No. 8023, Revised 1978).

Consent for publication

All participants have signed informed consent forms according to the protocols approved by the Ethics Committee of Shenyang Medical College.

Competing interests

The authors declare no competing interests.