Polydatin mitigates cigarette smoke-induced pulmonary apoptosis by regulating Bim via FoxO3 post-translational modifications and modulating the gut microenvironment

Abstract

Background

Cigarette smoke-mediated chronic pulmonary damage constitutes a critical predisposing factor for chronic obstructive pulmonary disease and lung cancer pathogenesis. Polydatin, a naturally occurring stilbenoid characterized by its potent anti-inflammatory and antioxidant properties, has yet to be fully elucidated regarding its protective role and the underlying molecular mechanisms in CSE-induced pulmonary injury.

Methods

Cigarette smoke extract (CSE)-induced lung injury murine models and MLE12 cells were treated with polydatin or other conditions. The investigation encompassed both intravital and extracorporeal modalities. Evaluations included pulmonary index, histopathology, oxidative stress markers, transcriptomic sequencing, protein expression (AKT/FoxO3/Bim), apoptosis (TUNEL), intestinal metabolomics, 16S rRNA microbiota profiling, and immunohistochemical staining of colon tissue.

Results

The administration of polydatin dose-dependently in vivo alleviated CSE-induced systemic impairments including weight loss, pulmonary edema, and oxidative stress concurrently demonstrating high biocompatibility and low systemic toxicity. Furthermore, it modulates the intestinal microecology by depleting g__Pantoea and g__Enterococcus abundance, concurrently restoring mucosal barrier integrity via the enhancement of ZO-1 and Occludin levels. Mechanistic analysis demonstrated that polydatin directly interacts with AKT1 at the Thr308 residue, thereby reversing the CSE-induced attenuation of AKT1 and FoxO3 (Ser253) phosphorylation. This restoration facilitates the nuclear-to-cytoplasmic translocation of FoxO3 and subsequent downregulation of the pro-apoptotic protein Bim. This regulatory cascade effectively curtails apoptosis and DNA fragmentation, with a concomitant decrease in Bax and Cleaved-Caspase 3 expression.

Conclusions

In conclusion, polydatin exerts suitable protective effects against CSE-induced pulmonary injury by attenuating apoptosis through the AKT-FoxO3-Bim signaling cascade and simultaneously orchestrating the gut microenvironment. These findings highlight the potential of polydatin as a promising natural therapeutic agent for lung disorders.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s13062-026-00924-y.

Introduction

Smoking represents a major global health challenge, accounting for over 10% of total deaths worldwide and claiming more than 6 million lives annually. Smoking-attributable causes are also responsible for the loss of nearly 150 million disability-adjusted life-years [1]. The prevalence of smoking among Chinese men has shown an upward trend since the 1990s. Drawing on the experiences of developed countries, the detrimental consequences of this pattern are expected to emerge progressively after a lag period of several decades [2]. Notably, China represents the largest share of smoking-related male deaths worldwide [3]. Accordingly, it is currently a critical period for identifying robust interventions to counteract the systemic functional deficits attributable to smoking in the general populace [4]. As the major constituent of the alveolus, alveolar epithelial cell type 2 (AT2) is essential for maintaining airway patency via synthesizing and secreting surfactant. The airway epithelium serves as the initial site of contact for cigarette smoke. Chronic exposure results in AT2 dysregulation, ultimately predisposing individuals to chronic respiratory conditions such as COPD and pulmonary neoplasms [5]. Limited attention has been directed toward the consequences of cigarette smoke for the normal functionality of AT2 cells. Therefore, we sought to screen for active constituents within natural herbal plants that may serve as potential agents to attenuate smoking-induced cellular impairment.

Polydatin is chemically designated as resveratrol-3-O-β-mono-D-glucoside. It represents a primary bioactive constituent extracted from the botanical source Polygonum cuspidatum [6]. Polydatin appertains to the stilbenoid family, being typified by a trans-diphenylethylene nucleus with three hydroxyl groups and a glucopyranosyl substituent at C‑3 [7]. The conjugated π‑electron system and multiple phenolic hydrogens confer potent radical‑scavenging and metal‑chelating properties, forming the chemical foundation for its antioxidant and anti‑inflammatory activities [8]. Owing to its structural classification as a natural glucoside of resveratrol, early investigations primarily attributed its biological actions to its potent antioxidant properties [9]. The pharmacological exploration of polydatin in respiratory medicine has predominantly focused on its efficacy against acute lung injury, pulmonary fibrosis, and lung neoplasms. A study reveals that polydatin mitigates the inflammatory and fibrotic sequelae induced by mycoplasma through the targeted suppression of the NLRP3 inflammasome and NF-кB activation [10, 11]. Previous finding indicates that polydatin attenuates radiation-induced lung damage through Sirt family activation [12]. Intriguingly, evidence suggests that polydatin effectively inhibits the proliferation and metastasis of lung cancer cells via the suppression of canonical inflammatory pathways [13]. The prevailing studies simply focus on the modulation of various signaling cascades within end-target organs. Regrettably, the antecedent physiological actions in gut initiated by polydatin remain largely under-investigated.

As an extensively characterized constituent of the FoxO subfamily, FoxO3 (FKHRL1) is constitutively and widely distributed throughout diverse mammalian physiological compartments [14, 15]. FoxO3 functions as a human transcription factor consisting of 673 or 453 amino acid residues [16]. FoxO3 possesses various functional subdomains that enable its dynamic localization within both the cytoplasm and the nucleus [17]. The phosphorylation state serves as a pivotal determinant of its biological activity, effectively governing its functional compartmentalization between the nucleus and the cytoplasm [18]. Cumulative evidence consistently demonstrates that it exerts its regulatory effects on apoptosis, autophagy and oxidative stress predominantly through its direct interaction with the promoter regions of specific target genes [19].

In mammals, apoptotic processes play a critical role in preserving the functional stability of biological systems [20]. The intrinsic signaling cascade is frequently initiated by genotoxic effects or oxidative stress. The subsequent imbalance between pro- and anti-apoptotic mediators dictates the cellular decision-making process, ultimately leading to the execution of programmed cell death [21, 22]. Apoptotic signaling proceeds via either extrinsic pathways, initiated by cell surface receptors, or intrinsic pathways involving mitochondrial outer membrane permeabilization [23]. Both routes culminate in the activation of executioner caspases 3 or 7, which are responsible for the characteristic nuclear remodeling [24]. As a strictly pro-apoptotic constituent of the Bcl-2 protein family, Bim (Bcl2l11) exists in three major isomeric forms. Its apoptotic activity can be attenuated either by sequestration through interactions with anti-apoptotic Bcl-2 mediators or by the direct down-regulation of its gene transcription at the source [25].

This work endeavors to elucidate how polydatin exerts pulmonary protection against CSE-induced damage, leveraging multi-omics-derived insights. The research expounds from the standpoint of the gut microhabitat how polydatin modulates enteric metabolites and the commensal flora during its mitigative efficacy against lung damage. Additionally, the research specifically explores the Akt/FoxO3/Bim signaling axis as a pivotal mechanism governing the programmed cell death of ATII cells.

Materials and methods

Drugs and materials

Polydatin (Fig. 1A) and GDC-0068 were procured from MedChemExpress (Monmouth Junction, NJ, USA). SC-79 was acquired from Beyotime (Shanghai, China). Dexamethasonum was procured from Sigma-Aldrich (St. Louis, MO, USA). Above standards obtained from the provider demonstrated a purity greater than 98%. Supplies for in vitro test covering DMEM/Nutrient Mixture F-12 (DMEM/F12) were sourced from Procell (Wuhan, China). Fetal bovine serum (FBS) was sourced from HyClone Laboratories, Inc. (Logan, UT, USA).

Fig. 1.

Schematic representation of polydatin and the animal experimental workflow. (A) Molecular structure of polydatin. (B) Chronological overview of the in vivo study design and experimental procedure

Animals

A total of 30 male BALB/c mice (18–20 g) were procured from Jilin Yisi biotechnology Co., Ltd. Above mice allowed to acclimatize to standard environmental conditions for one week prior to all experimental procedures. Mice were maintained under a 12 h light/dark cycle at 22 ± 2 °C and 50 ± 10% relative humidity, and were fed standard chow. The animal study was approved by Animal Ethics Committee of College of Basic Medical Sciences, Jilin University (approval number: 2025 − 322).

Preparation of Cigarette Smoke Extract (CSE)

The methodology for CSE preparation was refined based on previously described protocols [26, 27]. Two distinct variants of CSE were generated for this study: an in vitro formulation utilizing basal medium as the solvent, and an in vivo formulation prepared in phosphate-buffered saline. Research-grade cigarettes (China Tobacco Co., Ltd., China) conforming to 3R4F criterion were employed. The extraction process utilized a glass T-type gas filter unit. The aforementioned device ensures complete retention of the smoke within the liquid without any subsequent dispersion. A burning nicotiana was coupled to the air access. Smoke was bubbled through the solvent by manual aspiration using a plunger connected to gas exit. Smoke was extracted in 50-mL increments, totaling ten cycles per nicotiana to ensure full combustion up to the tip. CSE was formulated at concentrations of one nicotiana per 10 mL culture medium for in vitro use or per 2 mL phosphate solution for in vivo use. Post-preparation, the pH was promptly stabilized at 7.40 under constant surveillance, followed by sterile filtration to yield the final working solutions.

Establishment of CSE-induced Lung injury model and pharmacological interventions

Mice were randomly allocated into five experimental groups (n = 6): Control, Model, Dex (dexamethasonum, 3 mg/kg), low-dose Polydatin (pol10, 10 mg/kg), and high-dose Polydatin (pol40, 40 mg/kg). To prepare the oral delivery vehicles, dexamethasone and polydatin standards were initially dissolved in DMSO and subsequently emulsified in corn oil to achieve a final working concentration of 3% (v/v) DMSO. Mice assigned to the pharmacological intervention groups were treated with their respective compounds via daily gavage over a period of 28 consecutive days. Systemic lung injury was induced through intraperitoneal (i.p.) injections of CSE (15 mL/kg, equivalent to 15 µL/1 g body weight) for three times. Control animals were administered volume-matched PBS injections and received gavage of 3% DMSO in corn oil on the same schedule. Following one-day fasting period, above animals were sacrificed on the 28th research-day. Excrement pellets were collected immediately prior to sacrifice. Under anesthesia induced by tribromoethanol (AbMole, USA), serum was harvested via cardio-puncture. The lungs or colon were subsequently excised following thoracolaparotomy. The comprehensive experimental timeline is schematically illustrated in Fig. 1B.

Histological stain

Hematoxylin and eosin (H&E) of mice tissues using the commercial Kit (R32933, Yuanye, China). Immunohistochemistry (IHC) was conducted on gut sections to detect the expression of ZO-1 and occludin. Slides were scanned using an Olympus VS120 virtual slide scanner (Olympus Corporation, Tokyo, Japan). Whole-slide images were viewed and analyzed using OlyVIA software (Olympus Corporation).

Quantification of serum malondialdehyde (MDA)

The concentration of MDA in serum was determined using a Lipid Peroxidation MDA Assay Kit (S0131S, Beyotime, China) according to the manufacturer’s instructions. Briefly, clarified serum samples were incubated with thiobarbituric acid (TBA) at 95 °C for 45 min to facilitate the formation of MDA-TBA adducts. After cooling the reaction mixture on ice, the absorbance was measured spectrophotometrically at a wavelength of 532 nm. All final concentrations were normalized as µM/mL serum.

Superoxide dismutase (SOD)

Activity was quantified using a commercial SOD Assay Kit (S0101, Beyotime, China) in accordance with the manufacturer’s protocol. A reaction mixture consisting of 20 µL supernatant was prepared in a 96-well format. SOD was calculated based on its capacity to inhibit the reduction of WST-8 by superoxide anions, as calibrated by a standard curve. The final values were normalized to protein content and reported as U/mg protein.

Real-time quantitative PCR (RT-qPCR)

Total RNA isolation was conducted using the RNA-Quick Purification Kit (400 − 100, Goonie, China) according to the spin-column protocol. For reverse transcription, 1 µg of the isolated RNA was converted into cDNA using the Hifair III 1st Strand cDNA Synthesis SuperMix (Yeasen, Shanghai, China). The resulting cDNA was then subjected to qPCR analysis utilizing Hieff qPCR SYBR Green Master Mix (Yeasen, Shanghai, China). All amplifications utilized a CFX96-detection instrument (Bio‑Rad, USA). Relative transcript abundances were determined through the 2^−ΔΔCt method. Forward/reverse primer specifics appear within supplementary Table S1.

Western blot (WB)

Total lysates were clarified by centrifugation, and the protein concentrations in the resulting supernatants were determined using a BCA Protein Assay Kit (P0010, Beyotime, China). Equal protein aliquots were mixed with loading buffer. Following wet-transfer, the imprinted-membranes were incubated with primary antibodies, followed by their respective enzyme-conjugated secondary antibodies. Immunoreactive protein bands were visualized using an enhanced chemiluminescence substrate on a Minichemi Imaging System (Sinsage, China). Comprehensive specifications for all antibodies utilized are detailed in supplementary Table S2.

Cell culture and in vitro experimental design

Murine lung epithelial (MLE12) cell line was employed for in vitro experimental procedures. To establish the cellular model, cultures at 50% confluence were incubated with serial dilutions of CSE for 6 h. Vehicle cells were maintained in basal medium only. The optimal CSE concentration for subsequent experiments was determined by evaluating cell viability via CCK-8 assays. The drug-operational menstrua directly osculating the cells of each therapeutic cohort uniformly contained DMSO at a volumetric proportion of 0.1%. The protective effects of these agents were then evaluated using the pre-determined, optimized CSE-induced modeling conditions.

Cell viability testing

Cellular viability was evaluated via the CCK-8 testing kit (Solarbio, China). Upon two-fifths substratum coverage, experimental cohorts experienced graded agent doses (spanning one diel) or CSE (lasting six hours). Supernatants were discarded post-challenging, then 0.1 mL of working fluid was admixed. After a 1.5-h incubation, the optical density at 450 nm was recorded.

Immunofluorescence (IF)

The experimental cells were seeded at a bottom area ratio close to 7.5% of the confocal dish. Upon reaching 15% confluence, the cells were subjected to various experimental treatments. Subsequently, the cultures were fixed with 4% paraformaldehyde, permeabilized using 0.5% Triton X-100, and blocked with normal goat serum to minimize non-specific binding. The cells were then incubated with primary antibody working fluid over night. Then, suitable dye-linked second-stage immunoglobulins plus DAPI (BL105A, Labgic) enabled ultimate staining. Finally, the samples were visualized using the Zeiss LSM 900 (Jena, Germany).

Molecular docking

Three-dimensional coordinates of murine AKT1 alongside polydatin’s SDF-encoded data were fed into CB-Dock2’s web portal for computational fit plus interaction locus forecast. Subsequently, derived interaction energies alongside orientations underwent evaluation. The foremost juxtaposition was prioritized by maximal Vina numeric magnitude.

Transcriptome sequencing

Pulmonary parenchyma fragments were lysed in TRIzol concoction (Invitrogen, USA) via rotor‑stator homogenization. Cytosolic polynucleotide integrities were verified electrophoretically on 1% agarose gel. Directional complementary DNA archives were fabricated employing VAHTS Universal V8 RNA‑seq Prep Kit (Vazyme, China), strictly adhering to commercial manuals. Amplified libraries underwent paired‑end 150‑base flowcell seeding aboard an Illumina NovaSeq 6000 sequencer. Primary signal outputs (FASTQ) were stripped of adaptor leftovers and low‑quality bases via fastp. Purified read cohorts were aligned onto GRCh38 coordinates with HISAT2. Transcript abundances, expressed as TPM, were tallied via featureCounts.

Untargeted metabolomics (UM) of intestinal metabolites

Intestinal content samples (approximately 50 mg) were thawed at 4℃ and subsequently extracted using a chilled solvent system consisting of methanol, acetonitrile, and water (2:2:1, v/v/v). To remove proteins, the mixtures were incubated at -20℃ followed by centrifugation. The resulting supernatants were concentrated to dryness under vacuum, reconstituted in 30% acetonitrile, and further clarified by centrifugation. The final supernatants were subjected to LC-MS analysis. Chromatographic separation was performed on a UPLC system (Thermo Fisher Scientific, USA) equipped with a Waters HSS T3 column, using a mobile phase gradient of water and acetonitrile, both supplemented with 0.1% formic acid. Mass spectrometry was conducted on a Q Exactive HFX mass spectrometer (Thermo Fisher Scientific, USA) using a full MS/dd-MS² acquisition strategy. Unrefined primary acquisitions subjected to Progenesis QI enabled chromatographic feature correspondence alongside small molecule annotation via proprietary and locally compiled libraries.

16S rRNA sequencing

The intestinal mucosal microbiota of the target group was analyzed using 16S rRNA gene sequencing. Prokaryotic genomic polynucleotides were procured from enteric mucosae employing a bead‑beating protocol (QIAamp PowerFecal Pro Kit, Qiagen, Hildesheim, FRG). Hypervariable V3–V4 segments within the 16S ribosomal cistron were propagated using barcoded amplimers 338 F/806R. Cleaned amplicon pools underwent paired‑end 300‑base flowcell seeding aboard an Illumina MiSeq sequencer. Nascent sequencing outputs (FASTQ) were purged of low‑quality bases and adapter vestiges via Trimmomatic, then concatenated with FLASH. Pseudo‑taxonomic units (OTUs) were delimited at 97% nucleotide homology utilizing QIIME2. Phylogenetic assignments were anchored to the SILVA 138.1 corpus.

Statistical analysis

All statistical calculations run on GraphPad Prism 9.5 (GraphPad Software, San Diego, CA, USA). Shapiro‑Wilk testing initially checked each dataset’s Gaussianity. Non‑Gaussian datasets handled through Kruskal‑Wallis plus Dunn’s pairwise contrasts. Gaussian datasets examined concerning dispersion uniformity employing Bartlett’s. Gaussian plus uniform dispersions underwent univariate ANOVA coupled with Dunnett’s pairwise contrasts (versus CSE). Non‑uniform dispersions underwent Welch’s ANOVA. p < 0.05 deemed meaningful difference.

Results

Polydatin ameliorates lung injury triggered by CSE in murine models

In vivo experiments were categorized into five distinct cohorts based on specific treatment protocols (Fig. 2A). Upon termination of the study, terminal corporeal mass evaluations demonstrated murine subjects exposed to CSE insult manifested one marked decline of mass. Conversely, administration of dexamethasone and polydatin resulted in a significant restoration of body weight (Fig. 2B). Lung index was utilized as a physiological indicator to evaluate the edemato‑inflammatory severities. Quantitative analysis indicated lung index was severely elevated following CSE exposure, whereas this increase was significantly attenuated by dexamethasone or polydatin treatments (Fig. 2C). Additionally, a significant dose-dependent manner was occurred in the attenuation of the index. Minimal plus maximal polydatin dosages potently abated serum malondialdehyde concentrations (Fig. 2D). Pneumonary parenchymal superoxide dismutase metrics revealed augmented antioxidase abundance following either dexamethasone or polydatin regimens (Fig. 2E). To assess the magnitude of the inflammatory or apoptosis response across experimental groups, mRNA expression levels of interleukins in lung tissues were quantified via RT-qPCR. The results demonstrated that CSE exposure markedly increased the transcriptional levels of Il1b, Il6 and bax. Nevertheless, dexamethasone and dose-escalation polydatin treatments significantly suppressed their expression (Fig. 2F). Furthermore, histopathological examination via H&E staining revealed that CSE induced diffuse pulmonary edema, tissue consolidation. These pathological damages were ameliorated by both dexamethasone and polydatin, with the high-dose group exhibiting the most pronounced anesis in pulmonary consolidation (Fig. 2G). Finally, representative WB confirmed that polydatin treatment substantially inhibited the protein expression of key inflammatory indicators (Figs. 2H-I).

Fig. 2.

Polydatin attenuates CSE-induced pulmonary pathology, oxidative stress, and the inflammatory response in a murine model. (A) Schematic illustration outlining the treatment regimens across the five experimental cohorts. (B) Terminal body weights of mice in each group recorded at the conclusion of the study (n = 6). (C) Pulmonary index assessment (n = 6). (D) Serum concentrations of malondialdehyde (n = 3). (E) Superoxide dismutase enzymatic activity within lung homogenates (n = 3). (F) Relative mRNA abundance of pro-inflammatory indicators (Il1b, Il6) and the pro-apoptotic factor (Bax) in lung tissues (n = 3). (G) Representative photomicrographs of H&E-stained pulmonary sections (n = 3). Scale bar, 100 μm. (H, I) Representative WB profiles and the associated densitometric quantification of pivotal inflammatory protein markers in lung tissues(n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 vs. CSE group. Data are expressed as mean ± standard deviation (SD)

Polydatin modulates the FoxO3 and apoptosis signaling axis in lung by high-throughput sequencing

To elucidate the molecular basis of polydatin-mediated alleviation of lung injury in vivo, lung tissues from three mice randomly selected per group (control, model, and pol40) were subjected to NGS. Heatmap visualization revealed distinct sample and gene clustering patterns. These patterns were characterized by high intra-group homogeneity and pronounced inter-group differentiation, thereby confirming the robustness of the clustering analysis (Fig. 3A). Comparison among the differentially expressed genes (DEGs) of three groups via Gene Set Enrichment Analysis (GSEA) highlighted significant enrichment in both the chemokine signaling and apoptosis pathways (Fig. 3B). Consistently, Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis reinforced the GSEA results and additionally revealed prominent enrichment within the FoxO signaling pathway (Fig. 3C). Gene Ontology (GO) enrichment analysis indicated that the DEGs were significantly enriched in the regulation of immune system processes and the positive regulation of response to stimulus (Fig. 3D). Given that protein phosphorylation represents a defining feature of the positive regulation of stimulus response, these findings provided a mechanistic link to the FoxO signaling axis transduction process being investigated. Based on the established crosstalk between the FoxO and chemokine pathways, we focused our follow-up study on the regulatory relationship between the FoxO signaling axis and apoptosis.

Fig. 3.

Transcriptomic profiling elucidates the FoxO and apoptosis cascades as pivotal regulatory targets of polydatin in pulmonary tissues. (A) Hierarchical clustering heatmap of differentially expressed genes (DEGs) in lung tissues across the Control, Model, and Pol40 (40 mg/kg polydatin) cohorts (n = 3 per group). (B) Gene Set Enrichment Analysis (GSEA) profiles demonstrating the statistically significant enrichment of chemokine signaling and apoptosis-related gene sets among the treatment cohorts. (C) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of DEGs, underscoring the prominent modulation of the FoxO signaling cascade. (D) Gene Ontology (GO) functional enrichment analysis

Polydatin stimulates AKT1 phosphorylation at Thr308 and FoxO3 phosphorylation at Ser253

As AKT1 is a well-documented kinase responsible for FoxO3 phosphorylation [19], we performed molecular docking to evaluate its interaction with polydatin. The results demonstrated a robust binding affinity (binding energy = -7.0 kcal/mol) between polydatin and activation loop of AKT1. Detailed binding mode analysis identified specific interactions with the T308 residue, an essential amino acid residue for the activation of AKT1 [28] (Fig. 4A). WB analysis and subsequent quantification demonstrated that CSE did not significantly alter the total protein abundance of AKT and FoxO3 in lung tissues. However, it exerted a marked inhibitory effect on the phosphorylation levels of AKT1 (Thr308) and FoxO3 (Ser253). Polydatin significantly reinstated the phosphorylation of both indicators, whereas dexamethasone proved ineffective (Figs. 4B-C). We performed an unprecedented co-staining analysis of P-AKT1 (Thr308) and FoxO3 within pulmonary alveolus. The findings revealed a robust colocalization of P-AKT1 (Thr308) and FoxO3 in normal lung specimens, which was significantly disrupted by CSE treatment. Notably, polydatin treatment successfully reinstated this protein interaction, showing statistically significant efficacy even at the lower dosage level. By contrast, dexamethasone demonstrated any no modulatory impact on the aforementioned indicators (Fig. 4D).

Fig. 4.

Polydatin activates AKT1/FoxO3 signaling via direct binding to AKT1. (A) Molecular docking analysis of polydatin binding to the AKT1 activation loop (T308). (B-C) WB (B) and quantification (C) of total and P-AKT1 (Thr308) and FoxO3 (Ser253) in lung tissues(n = 3). (D) Immunofluorescence co-staining of P-AKT1 (Thr308) and FoxO3 in pulmonary alveoli. Scale bar, 50 μm. *p < 0.05, **p < 0.01, ***p < 0.001 vs. CSE group. Data are expressed as mean ± standard deviation (SD)

Polydatin suppresses pro-apoptotic protein Bim and concomitantly mitigates apoptosis within lung tissues

FoxO3 serves as a pivotal transcription factor that governs the expression of Bim [29]. Immunofluorescence analysis revealed that the intensity of Bim-positive red fluorescence within the alveolar septa was substantially elevated in the CSE group relative to the control group. Either polydatin or dexamethasone treatments significantly attenuated this signal, with high-dose polydatin demonstrating inhibitory efficacy comparable to dexamethasone (Fig. 5A). Transcriptomic profiling revealed a significant upregulation of Bim and Bax mRNA levels (as indicated by average FPKM values) in the CSE group. Bcl-2 expression was characterized by an opposing trend (Fig. 5B). WB results confirmed that CSE treatment augmented Bim protein abundance in lung, whereas polydatin and dexamethasone effectively counteracted this effect equally (Figs. 5C-D). This was in alignment with the immunofluorescence observations. Bim serves as a pivotal initiator of the apoptotic process. Consequently, we further characterized the essential molecules within the apoptotic cascade. TUNEL staining of lung sections demonstrated that CSE markedly enhanced the intensity of green fluorescence signals co-localized with cell nuclei, indicating increased DNA fragmentation. Conversely, administration of polydatin or dexamethasone significantly attenuated this trend, suggesting a potential anti-apoptotic effect (Fig. 6A). CSE challenge induced an apoptosis-promoting polypeptide signature encompassing Bax plus Cleaved-Caspase 3 as quantified via immunoblotting. Inversely, it diminishing anti-apoptotic Bcl-2 expression. Remarkably, the aforementioned observations became conspicuously counteracted following application of dexamethasone or polydatin (Figs. 6B-C).

Fig. 5.

Polydatin inhibits CSE-induced Bim expression. (A) Immunofluorescence staining of Bim in alveolar septa. Scale bar, 50 μm. (B) Heatmap illustrating mRNA expression levels (FPKM) of Bim, Bax, and Bcl-2 from transcriptomic profiling (n = 3). (C–D) Western blot (C) and quantification (D) of Bim protein levels in lung tissues (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 vs. CSE group. Data are expressed as mean ± standard deviation (SD)

Fig. 6.

Polydatin attenuates CSE-induced lung apoptosis. (A) TUNEL staining of lung sections. Scale bar, 50 μm. (B–C) Western blot (B) and quantification (C) of pro-apoptotic (Bax, Cleaved-Caspase 3) and anti-apoptotic (Bcl-2) proteins (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 vs. CSE group. Data are expressed as mean ± standard deviation (SD)

Polydatin promotes the phosphorylation of AKT1 and FoxO3 and diminishes Bim in vitro

CCK-8 assays conducted on MLE12 cells following a 6-hour exposure to gradient concentrations of CSE exhibited a concentration-responsive decline in cell viability. Notably, 40% CSE induced exhaustive cell death, whereas 1% CSE yielded a viability of roughly 70% compared with the untreated controls (Figure S1A). Accordingly, a concentration of 1% CSE was adopted for subsequent modeling. Additionally, MLE12 cells exposed to polydatin at concentrations lower than 10 µM for a 24-hour period exhibited a viability exceeding 80% (Figure S1B). Consequently, concentrations of 0.1 µM and 0.5 µM were designated as the optimal dosages for subsequent cellular investigations. Consistent with the in vivo findings, CSE exposure significantly diminished P-AKT1 (Thr308) and P-FoxO3a (Ser253) levels without compromising the total expression of AKT and FoxO3. This inhibitory effect was subsequently mitigated by polydatin supplementation in a concentration-dependent manner. Additionally, CSE-induced elevation of Bim protein levels in MLE12 cells was effectively countered by polydatin concomitant intervention (Figs. 7A-C).

Fig. 7.

Polydatin regulates Bim expression via the AKT-FoxO3 axis in MLE12 cells. (A-C) WB and quantification of AKT, P-AKT (Thr308), FoxO3a, P-FoxO3a (Ser253) and Bim in MLE12 cells treated with CSE and varying concentrations of polydatin (n = 3). (D–E) WB and quantification showing the effects of SC79 (AKT activator) and GDC-0068 (AKT inhibitor) (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 vs. CSE group. Data are expressed as mean ± standard deviation (SD)

Polydatin-mediated suppression of Bim is strictly mediated through the AKT-FoxO3 cascade

To investigate the involvement of the P-AKT1 (Thr308)-FoxO3 pathway in polydatin-induced Bim modulation, we utilized the specific Akt agonist SC79 as a positive control. Furthermore, the specific Akt inhibitor GDC-0068 was co-administered with polydatin to determine if Akt inhibition could counteract the pharmacological effects. To ensure cytocompatibility, the safe dosage ranges of SC-79 and GDC-0068 were preliminarily screened and confirmed via CCK-8 analysis (Figures S1C-D). SC-79 at concentrations below 5 µM and GDC-0068 below 0.15 µM exhibited no discernible cytotoxicity in MLE12 cells. Consequently, 2.5 µM SC-79 and 0.03 µM GDC-0068 were selected for all subsequent experimental procedures. WB data indicated that SC79 and polydatin effectively mitigated CSE-induced alterations in P-AKT, P-FoxO3a, and Bim expression (Figs. 7D-E). The observation that the polydatin-induced decrease in Bim was fully reversed by GDC-0068 suggested that the regulatory role of polydatin in Bim expression was strictly dependent on the AKT-FoxO3 axis.

Polydatin modulates the subcellular distribution of FoxO3 in vitro

The nuclear import of FoxO3 is contingent upon its phosphorylation state [30]. Our initial IF results demonstrated that FoxO3 exhibited a balanced nucleocytoplasmic distribution in baseline MLE12 cells. While CSE stimulation significantly enhanced the nuclear accumulation of FoxO3, this translocation was effectively suppressed by polydatin or SC79. Notably, the inhibitory effect of polydatin on FoxO3 nuclear localization was completely abrogated by the AKT inhibitor GDC-0068 (Fig. 8A). Consistent with the immunofluorescence findings, WB analysis of nucleocytoplasmic fractionation revealed that SC79 and polydatin facilitated the nuclear export of FoxO3. Notably, the polydatin-induced shift in FoxO3 localization was entirely abrogated by co-treatment with GDC-0068(Figs. 8B-C). Above data suggested that polydatin-mediated regulation of FoxO3 nucleocytoplasmic partitioning is AKT-dependent.

Fig. 8.

Polydatin regulates FoxO3 nucleocytoplasmic translocation in an AKT-dependent manner. (A) Representative immunofluorescence images of FoxO3 localization in MLE12 cells. Scale bar, 50 μm. (B–C) WB and quantification of FoxO3 levels in nuclear and cytoplasmic fractions (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001 vs. CSE group. Data are expressed as mean ± standard deviation (SD)

Polydatin modulates the biochemical composition of the intestinal milieu

Heatmap and cluster analyses initially indicated that biological replicates within each group exhibited a high degree of compositional uniformity via UM analysis of intestinal contents. Simultaneously, substantial inter-group variations were clearly delineated (Fig. 9A). We further characterized the specific chemical categories of the differential compounds to evaluate the impact of polydatin administration following CSE exposure. As depicted in the pie chart, the differential metabolic profile induced by polydatin was characterized by an increased representation of prenol lipids, carboxylic acids and derivatives, organooxygen compounds, and steroids and steroid derivatives. These categories comprised a cumulative 50% of all significantly altered compounds in the intestinal contents (Fig. 9B). Polydatin intervention following CSE treatment yielded 13 significantly upregulated and 5 significantly downregulated compounds (Fig. 9C). The top three most significantly upregulated compounds were 4’-Deoleandrosyl-6,8a-seco-6,8a-deoxy-5- (PubChem CID: 139585883), 3,4,5 − Trihydroxy − 6−{[2,3,4 − Trihydroxy − 5−(Hydroxymethyl)Oxolan − 2 − Yl]Methoxy}Oxane − 2 − Carboxylic Acid (PubChem CID: 131832081) and glabric acid (PubChem CID: 46173993) (Fig. 9D). Functional annotation via KEGG enrichment analysis demonstrated that the intestinal metabolic variations across the three cohorts were predominantly linked to oxidative phosphorylation, the citrate cycle (TCA cycle), and lipoic acid metabolism (Fig. 9E).

Fig. 9.

Polydatin modulates the intestinal metabolic profile in CSE-exposed mice. (A) Heatmap and cluster analysis of intestinal metabolites based on untargeted metabolomics (n = 3). (B) Proportional distribution of differential metabolites. (C) Volcano visualization of significant metabolite alterations in the intergroup comparison. (D) Bar chart of significantly altered intestinal metabolites. (E) Bubble diagram of KEGG pathway enrichment analysis based on intergroup differential metabolites

Polydatin treatment reshapes the intestinal microbial profile and rehabilitates the intestinal epithelial barrier function

16S rRNA gene sequencing was employed to analyze the colonic mucosal microbiota across the three cohorts, aiming to delineate their respective microbial profiles. Assessment of alpha diversity via the Chao1 index indicated that CSE exposure led to a decline in microbial richness. This effect was mitigated by PD intervention, demonstrating the capacity of polydatin to restore intestinal microbial diversity (Fig. 10A). Concomitantly, partial least squares discriminant algorithm (PLS-DA) served to demarcate beta diversity, unveiling divergence amidst tripartite microbiota’s organizational architectures (Fig. 10B). To ascertain the differential abundance of microbial taxa across the study groups, we employed Wilcoxon rank-sum tests at the genus level. CSE exposure resulted in a significant elevation of g__Pantoea and g__Enterococcus enrichment in the colonic mucosa. Inversely, these taxonomic shifts were markedly reversed following polydatin supplementation (Figs. 10C-D). Notably, the abundance of g__Saccharopolyspora was slightly increased following CSE treatment and was further substantially augmented by polydatin intervention in the colonic mucosa (Fig. 10E). Based on the integrated findings from above analyses, we subsequently performed a histopathological assessment of colon. H&E staining identified structural impairments in the CSE group, characterized by mucosal irregularities and crypt disarray. Both of which were attenuated by PD treatment. Given that ZO-1 and Occludin are pivotal to maintaining the intestinal epithelial barrier, their expression levels were assessed via IHC. The CSE-induced depletion of these junctional proteins was significantly countered by oral PD administration, suggesting a protective effect of polydatin on barrier integrity (Fig. 10F).

Fig. 10.

Polydatin restores intestinal microbiota diversity and maintains colonic barrier integrity in CSE-exposed mice. (A) Alpha diversity assessed by the Chao1 index (n = 3). (B) Beta diversity visualized via Partial Least Squares Discriminant Analysis (PLS-DA) (n = 3). (C) Comparative analysis of intestinal bactirium-spectrum across groups at the family and genus levels. (D-E) Relative abundance of differential microbial genera in colonic mucosa based on Wilcoxon rank-sum tests (n = 3). (F) Representative H&E-stained images and immunohistochemical staining of ZO-1/Occludin in intestinal tissues (n = 3). Scale bar, 100 μm

Discussion

Accumulating data from numerous studies highlight polydatin as a promising agent with significant potential for treating respiratory-related ailments. Polydatin has demonstrated potent anti-fibrotic activity in bleomycin-challenged mice. Mechanistically, it mitigates bleomycin-induced pulmonary fibrosis through the downregulation of fibrotic markers (Collagen I and α-SMA) and the alleviation of systemic and local inflammatory responses. The therapeutic contribution of the microbiota was validated by fecal grafting, demonstrating that polydatin-modified microbiota is sufficient to alleviate fibrogenesis in antibiotic-treated mice [31]. As a common etiology of pulmonary fibrogenesis, chemotherapy-induced above damage can be mitigated by polydatin. Evidence suggests that polydatin dose-dependently ameliorates methotrexate-induced pulmonary fibrosis by augmenting intracellular glutathione homeostasis [32]. In alignment with previous reports, our results revealed that polydatin treatment at non-toxic concentrations suppressed CSE-induced pro-inflammatory signaling while enhancing the antioxidant capacity of lung tissues. Epidemiological data suggest a negative correlation between smoking and BMI, with smokers exhibiting a 3.31 kg/m² lower average data compared to non-smokers [33]. Correspondingly, our study showed that CSE-induced weight loss was significantly ameliorated following oral supplementation with polydatin. These observations are potentially inextricably linked to alterations in both the respiratory and intestinal microenvironments. Consequently, the modulatory role of polydatin on gut homeostasis warrants significant consideration and should not be overlooked.

The role of polydatin in governing disease progression via the regulation of apoptotic cascades has long been a well-established and prominent research theme. Investigations into polydatin-mediated modulation of apoptotic cascades are predominantly focused on oncological contexts [34]. The polydatin-mediated pro-apoptotic effect in human osteosarcoma cells is driven by the STAT3, concurrently involving the modulation of autophagy indicators including ATG14 and LC3 [35]. Polydatin precipitates intensified endoplasmic reticulum proteotoxicity mediated via the ATF4/CHOP axis and profound genomic instability. Above process ultimately enhanced the chemotherapeutic potency of oxaliplatin [36]. Collectively, the positive modulation of apoptotic effect by polydatin serves as a pivotal mechanism for impeding multifarious tumors’ progression. Conversely, in specific contexts where the preservation of physiological homeostasis mandates the attenuation of apoptosis, polydatin demonstrates a discernible cytoprotective efficacy. Polydatin abrogates hepatocyte apoptosis-associated indicators release through the downregulation of the TLR4 cascade, concurrently restraining macrophage polarization and the resultant pro-inflammatory cytokine profile [37]. The mitigation of cardiomyocyte programmed cell death by polydatin following myocardial I/R injury precipitates a significant diminution in necrotic area and fibrotic remodeling, ultimately restoring left ventricular functional integrity [38]. Therefore, pleiotropic apoptotic effect of polydatin performs a disease‑dependent or condition‑dependent effect. The correlation between polydatin and the pro-apoptotic mediator Bim remains unidentified in current literatures, particularly regarding the underlying regulatory networks. Herein, we provide initial evidence of their putative regulatory relationship, substantiated by multi-level data derived from transcriptional and translational perspectives.

The AKT/PKB family is characterized by three evolutionarily conserved and structurally homologous isoforms (AKT1, AKT2, and AKT3). Each of them shares a common architecture defined by three highly similar and conserved domains [28]. Ubiquitously expressed across various tissues, Akt1 serves as a key mediator of cell growth and survival. As a central signaling hub, Akt governs multiple biological functions, including but not limited to metabolic homeostasis, cell proliferation, and angiogenesis [39]. Above successful readjustments are facilitated through the serine/threonine phosphorylation of AKT downstream-effectors [40]. Prior investigations have preliminarily demonstrated that polydatin functions as a modulator of the AKT to achieve its pharmacological efficacy. Polydatin has been reported to exert protective effects against diabetic glucolipotoxicity by augmenting p-Akt expression, thereby restoring metabolic homeostasis [41]. Furthermore, evidence has confirmed that polydatin attenuates transient renal reperfusion injury driven by AKT suppression. This process may involve the recruitment of PI3K [42]. However, it is noteworthy that the majority of these preliminary studies were confined to identifying initial evidence of AKT modulation, without further elucidating the specific downstream substrates targeted by AKT. To overcome the aforementioned constraints, in-depth studies have further explored the downstream effectors regulated by polydatin via the AKT. AKT-mediated phosphorylation via polydatin of GSK-3β at Ser9 relieves the suppression of Nrf2, leading to increased Nrf2 levels and subsequent neuroprotective potential in Parkinson’s disease models [43]. Evidence indicates that Polydatin promotes apoptosis in hepatocellular cancer by inhibiting AKT activity, which subsequently suppresses the phosphorylation of FoxO1 [44]. Despite limited supporting evidence, the specific downstream cascades of the polydatin-AKT axis remain largely unexplored. In this study, by integrating transcriptomic profiling with molecular biological rescue assays, we identified FoxO3 as the pivotal downstream effector regulated by polydatin, providing novel insights for future investigations.

The AKT cascade functions as a pivotal inhibitory upstream regulator of FoxO3. Specifically, AKT-mediated phosphorylation dictates the subcellular sequestration and transcriptional potency of it. Upon its nuclear localization, FoxO3 orchestrates a comprehensive pro-apoptotic transcriptional program by upregulating critical intrinsic apoptotic effectors such as Bim and Bcl-6 [45]. The canonical correlation between FoxO3 nuclear sequestration and Bim expression has been extensively substantiated in prior literature. Piperlongumine suppresses AKT phosphorylation, thereby facilitating the nuclear translocation of FoxO3 and its subsequent recruitment to the Bim gene promoter. This regulatory cascade culminates in the transcriptional activation of Bim, which serves as a critical driver for the induction of intrinsic apoptosis across diverse neoplastic cell lines [30]. Previous studies have established Regorafenib as a potent tumor suppressor with significant pro-apoptotic activity. It has been observed that Regorafenib-mediated nuclear translocation of FoxO3 is markedly attenuated by SC79, which subsequently abrogates the upregulation of Bim [29]. A study have found that glucocorticoids can regulate apoptosis in thymoma cells solely through modulating the transcription and translation of Bim, independent of Bax and Bcl-2 [46]. These findings are in high concordance with our current observations, thereby reinforcing the empirical validity and robustness of our research conclusions.

The gut microbiota plays a multifaceted and pivotal role in the ontogeny, modulation, and preservation of homeostatic immune responses. Consequently, precision medicine frameworks informed by metagenomic profiling of the intestinal microbiome have emerged as a sophisticated and efficacious strategy for the clinical management of respiratory pathologies [47]. Metabolic byproducts derived from gut commensals play dual roles in immunomodulation, exerting either stimulatory or suppressive inflammatory effects [48]. Clinical data substantiate this by showing a marked enrichment of histamine-producing bacterial taxa in the fecal secretome of asthmatic patients relative to non-asthmatic volunteers [18]. Thus, integrating the analysis of intestinal content composition with microbial profiling is essential to fully capture the homeostatic mechanisms governing the enteric microenvironment. Our research first attests polydatin effectively counteracts cigarette smoke-induced enrichment of intestinal Pantoea. Existing clinical literature indicates that Pantoea is a pathogen capable of triggering severe systemic infections in neonates, with primary manifestations including pulmonary hemorrhage and respiratory distress syndrome. These findings highlight a putative link between the dysbiosis of intestinal Pantoea and the pathogenesis of respiratory disorders [49]. A similar inhibitory profile was observed for polydatin regarding the expansion of Enterococcus populations in this study. Dense intestinal colonization by Enterococcus often occurs at the expense of beneficial indigenous flora. Disruption of the gastrointestinal mucosal barrier serves as a critical conduit for Enterococcus bacteremia, enabling the bacteria to migrate from the enteric environment into the bloodstream, thereby predisposing the host to multi-organ infections at distal sites [50]. This study represents the first comprehensive investigation into the regulatory effects of polydatin on intestinal homeostasis. By elucidating a suite of beneficial modulatory mechanisms, our findings provide novel insights for future explorations into its therapeutic potential in gut-related pulmonary disorders.

There are also some limitations to this study. In this study, the mouse model was not established using conventional cigarette smoke exposure. Although this approach allows precise control over the dosage and concentration of smoke, thereby minimizing inter-animal variability in inhalation and promoting a more uniform distribution of pulmonary lesions, the use of cigarette smoke extract (CSE) may induce substantial systemic effects that cannot fully replicate the comprehensive pathophysiology of inhalational injury. Thus, both modeling strategies present distinct advantages and inherent limitations. While the underlying mechanisms were verified by AKT inhibitor treatment in vitro, no in vivo rescue experiments were performed for further confirmation. Simultaneously, due to the inevitable off-target effects associated with natural plant products, substantial inter-individual variability may exist. Thus, structural modification of the natural product by introducing additional chemical groups could represent an effective approach to overcome this issue.

Conclusion

In conclusion, polydatin confers significant protection against CSE-induced pulmonary injury via the targeted activation of the AKT1-FoxO3 axis. Mechanistically, polydatin facilitates the site-specific phosphorylation of AKT1 (Thr308) and FoxO3 (Ser253). This post-translational modification promotes the nuclear-to-cytoplasmic translocation of FoxO3, thereby downregulating the expression of the pro-apoptotic mediator Bim. This regulatory cascade subsequently suppresses the activation of Bax and cleaved caspase-3 while restoring the levels of the anti-apoptotic protein Bcl-2, ultimately mitigating pulmonary cell apoptosis. Furthermore, polydatin modulates the intestinal microbiota landscape and fortifies the gut epithelial barrier integrity through the upregulation of tight junction proteins ZO-1 and Occludin, highlighting the potential involvement of the gut-lung axis (Fig. 11). Taken together, these findings identify polydatin as a promising candidate for CSE-induced lung injury operating through a distinctive AKT/FoxO3-dependent anti-apoptotic mechanism.

Fig. 11.

Graphical abstract illustrating the mechanisms by which polydatin mitigates cigarette smoke-induced pulmonary inflammation and apoptosis by modulating the gut microenvironment and regulating Bim via FoxO3 post-translational modifications

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Longfei Lv wrote the manuscript. Longfei Lv, Tianyu Lu, Rui Wang, and Xiang Li, performed the experiments. Youbin Cui, Li Ren, and Jie Zhang contributed to the conception and design of the study.

Funding

This work was supported by the Science and Technology Development Project Foundation of Jilin Province (20260206014ZP).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The animal study was approved by Animal Ethics Committee of College of Basic Medical Sciences, Jilin University (approval number: 2025 − 322). All experimental protocols were performed in accordance with institutional animal welfare guidelines.

Competing interests

The authors declare no competing interests.