UHPLC -Q-Orbitrap HRMS Integrated with Network Pharmacology Reveals the Anti-COPD Mechanism of Jianzhong Peiyuan Decoction and Functional Validation of Core Component Luteolin

Abstract

Objective

Jianzhong Peiyuan Decoction, based on Shenling Baizhu Powder, is commonly used for stable COPD in Guizhou. However, its chemical composition and pharmacodynamics are not well understood. This study aims to explore its pharmacodynamic basis and multi-target pathways in COPD treatment and to confirm the intervention effect of its key active ingredient, luteolin, through in vitro cell experiments.

Methods

UHPLC-Q-Orbitrap HRMS identified the chemical constituents of Jianzhong Peiyuan Decoction. Network pharmacology screened core components, identified drug-disease targets, and conducted PPI network, GO function, and KEGG pathway analyses. Molecular docking and dynamics simulations assessed core components’ binding stability to key targets. 16HBE cell injury model induced by cigarette smoke extract was used to examine the expression of key targets in the SRC/MAPK1/Akt/STAT3 and PPARγ/CD36/ABCA1 pathways via RT-qPCR.

Results

Study identified 497 chemical components in Jianzhong Peiyuan Decoction, including 9 key active ones such as Longikaurin A, kaempferol, and luteolin. It found 675 potential targets for COPD treatment, with six core targets—SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR—highlighted through PPI network analysis. KEGG pathway enrichment pointed to the lipid and atherosclerosis pathway as crucial. Molecular docking and simulations showed strong binding of the active ingredients, especially luteolin, to the core targets. In vitro experiments on 16 HBE cells in a CSE-induced COPD model showed overexpression of SRC, AKT1, MAPK1, STAT3, EGFR, and CD36 mRNA, while ESR1, PPARγ, and ABCA1 mRNA were reduced. Luteolin was found to normalize these gene expressions in a dose-dependent manner.

Conclusion

Jianzhong Peiyuan Decoction aids in COPD treatment through multiple components, targets, and pathways. Luteolin, a key ingredient, mitigates cigarette smoke-induced bronchial injury by modulating inflammatory and lipid metabolism pathways. This study clarifies the Decoction’s material basis and molecular mechanism in treating stable COPD, offering experimental and theoretical support for its clinical use and quality standards.

Introduction

Chronic obstructive pulmonary disease (COPD) is a condition characterized by chronic bronchitis and emphysema, leading to airway remodeling, oxidative stress, and persistent inflammation. This disease is associated with prolonged respiratory symptoms, including cough, excessive sputum production, and dyspnea.¹ According to the most recent data published by the World Health Organization (WHO) in 2024, COPD ranks as the fourth leading cause of mortality globally, responsible for approximately 5% of all deaths worldwide, with 3.5 million fatalities recorded in 2021. Given the global trend of an aging population, the prevalence of COPD is anticipated to increase in the future.² Smoking and air pollution are key risk factors for COPD, causing oxidative and carbonyl stress that damage airway cells. While only 15%-20% of smokers develop COPD, and quitting smoking does not fully halt its progression,³ this indicates a persistent chronic inflammatory response in susceptible individuals. Harmful stimuli trigger ongoing secretion of pro-inflammatory factors like LTB4 and IL-8, repeatedly activating lung neutrophils and creating a cycle of oxidative stress and chronic inflammation, worsening lung damage.⁴ Currently, the clinical management of COPD is primarily categorized into three treatment modalities: long-acting anticholinergic agents, long-acting β2-adrenergic agonists, and inhaled corticosteroids. These interventions predominantly aim to alleviate symptoms rather than address the underlying pathophysiological mechanisms, such as chronic inflammation, oxidative stress, airway remodeling, and the progressive decline in pulmonary function.⁵ Furthermore, prolonged use of these medications is associated with cumulative adverse effects and organ toxicity. Additionally, there are significant limitations in prescribing these treatments to patients with comorbid conditions. Consequently, there is a pressing need to develop safe and efficacious therapeutic strategies that are readily acceptable to patients.^6,7

Traditional Chinese medicine (TCM) has a longstanding history in the management of COPD.⁸ Within TCM theory, COPD is often referred to as “Lung Distention”, “Wheezing”, and “Cough” based on the specific characteristics of its symptoms.⁹ Treatment strategies are tailored according to the clinical manifestations and progression of COPD.¹⁰ Jianzhong Peiyuan Decoction, derived from the classical Shenling Baizhu Powder, is extensively utilized in the treatment of stable COPD. This Decoction comprises Codonopsis Radix (Codonopsis pilosula) 20g, Astragali Radix (Astragalus membranaceus) 15g, Atractylodis Macrocephalae Rhizoma (Atractylodes macrocephala) 15g, Poria (Poria cocos) 12g, Coicis Semen (Coix lacryma-jobi var. ma-yuen) 12g, Dioscoreae Rhizoma (Dioscorea polystachya) 12g, Lablab Semen (Lablab purpureus) 12g, Schisandrae Chinensis Fructus (Schisandra chinensis) 15g, Psoraleae Fructus (Psoralea corylifolia) 15g, Angelicae Sinensis Radix (Angelica sinensis) 12g, Salviae Miltiorrhizae Radix et Rhizoma (Salvia miltiorrhiza) 15g, and Glycyrrhizae Radix et Rhizoma (Glycyrrhiza uralensis) 3g. According to traditional Chinese medicine theory, while COPD primarily manifests externally through symptoms such as cough, asthma, and other lung-related syndromes, its pathogenesis is not confined to the lungs alone. The underlying cause is attributed to the dysfunction of three organs: the lung, spleen, and kidney, as well as the disharmony in the movement of qi throughout the body. Zongqi plays a crucial role in respiration and the circulation of blood within the vessels. When the qi of these three zang-organs is deficient, Zongqi becomes biochemically inactive, leading to impaired function and abnormal water metabolism, which results in the production of phlegm. This phlegm obstructs the lung collaterals and airways, ultimately culminating in the characteristic clinical manifestations of COPD, such as cough, expectoration, wheezing, and shortness of breath.¹¹ Jianzhong Peiyuan Decoction focuses on strengthening the qi of the lungs, spleen, and kidneys, and regulating overall qi movement to address the root cause of phlegm in COPD. It builds on Shenling Baizhu Powder by adding ingredients to enhance kidney function, aligning with the chronic lung disease pathogenesis, and effectively benefits stable COPD patients. It has been shown to enhance symptoms, improve lung function, and elevate the quality of life in patients with COPD.¹² Extensive pharmacological research has provided robust evidence supporting the therapeutic efficacy of qi-tonifying and primordial-consolidating formulas similar to our decoction. Shen Qi Wan has been demonstrated to inhibit the OPN/CD44/PI3K positive feedback loop, thereby reducing airway inflammation and oxidative stress in murine models of COPD.¹³ The Shenqi Wenfei Formula has been shown to alleviate pulmonary inflammation in COPD rat models by modulating the NLRP3/GSDMD pathway and partially restoring dysregulated gut microbiota.¹⁴ Furthermore, the Shenqi Tiaoshen Formula mitigates airway inflammation and ameliorates COPD symptoms in rat models by influencing the METTL16-m6A-MALT1-NF-κB signaling axis.¹⁵ Collectively, these studies underscore that TCM formulas employing a similar tonifying strategy exert significant effects in suppressing airway inflammation, a central pathological feature of COPD.

In this study, UHPLC-Q-Orbitrap HRMS technology, along with network pharmacology, molecular docking, and molecular dynamics simulations, was employed to analyze the chemical constituents of Jianzhong Peiyuan Decoction. Subsequently, in vitro cell experiments were used for verification. The objective is to elucidate the mechanism and pharmacodynamic material basis of Jianzhong Peiyuan Decoction at a systemic level, to comprehend its protective mechanism in COPD, and to offer a reference for the quality standards and clinical application of Jianzhong Peiyuan Decoction. This study provides a significant scientific foundation for further uncovering the pharmacodynamic material basis of Jianzhong Peiyuan Decoction and for the development of related standardized quality control methodologies.

Materials and Methods

Extract of Jianzhong Peiyuan Decoction

The herbal medicine of Jianzhong Peiyuan Decoction was purchased from the First Affiliated Hospital of Guizhou University of Traditional Chinese Medicine. The dried drug was ground into powder using a grinder and then immersed in 70% ethanol (v/v) at room temperature for 1 hour. The mixture was ultrasonically treated for 1 hour and repeated twice. The combined filtrate was collected, concentrated under reduced pressure using a rotary evaporator, and then freeze-dried to obtain a reddish-brown powder for further analysis.

UHPLC-Q-Orbitrap HRMS Technology Compounds Identification in Jianzhong Peiyuan Decoction

Accurately weigh 52mg Jianzhong Peiyuan Decoction granules to 2mL centrifuge tube, add a grinding bead with a diameter of 6mm; add 400μL extract [methanol:water= 4:1(v:v)], containing four internal standards [L-2-chlorophenylalanine (0.02mg/mL),etc.]; frozen tissue grinder grinding 6min (−10°C,50Hz); low temperature ultrasonic extraction 30min (5°C,40KHz); the samples were placed at −20°C for 30min. Centrifugation for 15min (13000g, 4°C), the supernatant was transferred to the injection vial with intubatton for analysis.¹⁶

The instrument platform for LC-MS analysis was the UHPLC-Q Exactive system of ultra-high performance liquid chromatography tandem Fourier transform mass spectrometry.¹⁷ C18 column chromatographic conditions: chromatographic column was ACQUITY UPLC BEH C18 (100mm×2.1mm i.d, 1.7μm; waters, Milford, USA); the mobile phase A was 2% acetonitrile water (containing 0.1% formic acid), the mobile phase B was acetonitrile (containing 0.1% formic acid), the injection volume was 3μL, and the column temperature was 40°C. Mass spectrometry conditions: The samples were ionized by electrospray ionization, and the mass spectrometry signals were collected by positive and negative ion scanning modes. The specific parameters are shown in Table 1.

Table 1.

Mass Spectrometric Parameters

Description	Parameter
Scan type (m/z)	70-1050
Sheath gas flow rate (arb)	50
Aux gas flow rate (arb)	13
Heater temp (°C)	450
Capillary temp (°C)	320
Spray voltage (+) (V)	3500
Spray voltage (-) (V)	−3000
S-Lens RF Level	40
Normalized collision energy (%)	20,40,60
Resolution (Full MS)	70000
Resolution (MS2)	17,500

Network Pharmacology Analysis of Jianzhong Peiyuan Decoction

Screening of Main Active Ingredients

The chemical constituents of Jianzhong Peiyuan Decoction were preliminarily characterized via UHPLC-Q-Orbitrap HRMS detection and further validated against the Traditional Chinese Medicine Systems Pharmacology Database (TCMSP, https://tcmsp-e.com/). Two critical pharmacokinetic indicators, oral bioavailability (OB) and drug-likeness (DL), were adopted to screen potential bioactive compounds. OB quantifies the fraction of an orally administered compound that enters systemic circulation, while DL evaluates whether a chemical entity possesses drug-like structural properties to eliminate unqualified candidates. Compounds satisfying the thresholds of OB ≥ 30% and DL ≥ 0.18 were retained as candidate active ingredients. Subsequently, all predicted protein targets of qualified compounds were collected, and duplicate entries were removed to generate a non-redundant target list for Jianzhong Peiyuan Decoction. To unify gene nomenclature for subsequent bioinformatic analysis, all target protein names were standardized to official human gene symbols by cross-referencing the UniProt database (https://www.uniprot.org/).

Prediction of Drug Target

The SMILES file was imported into the Swiss Target Prediction database (http://www.swisstargetprediction.ch/) using “Human” as the species. Compounds with high Gastrointestinal absorption (GI) absorption and at least two “Yes” for drug-likeness were selected to identify potential targets of Jianzhong Peiyuan Decoction. Targets with Probability>0 were identified, and their human gene data were standardized using the Uniprot database (https://www.uniprot.org/) and stored in an Excel file.

Potential Target Gene Screening and PPI Network Construction

Using “chronic obstructive pulmonary disease” as the keyword, databases like GeneCards (https://www.genecards.org/), OMIM (https://omim.org/), Disgenet (https://disgenet.com/), and Drugbank (https://go.drugbank.com/) were searched for disease targets, and results were mapped using Venny (https://bioinfogp.cnb.csic.es/tools/venny/) to identify COPD targets. Jianzhong Peiyuan Decoction-COPD potential target genes were then analyzed in the STRING database (https://string-db.org/) to create a PPI network, illustrating protein interactions. This network was visualized and analyzed in Cytoscape 3.8.0.

GO Enrichment Analysis and KEGG Pathway Analysis

The primary targets were analyzed in the Metascape database (https://metascape.org/) for GO function and KEGG pathway enrichment. Statistically significant pathways and biological processes (p<0.05) were identified and visualized using the WeiShengXin online tool (https://www.bioinformatics.com.cn/).

Molecular Docking and Molecular Dynamics Simulation Verification

To assess the binding affinity between core ligands and target receptors, molecular docking was performed with systematic negative controls. Core compounds’ 2D structures were obtained from PubChem and converted to mol2 files using Chem3D for optimization. Human target crystal structures were sourced from the RCS PDB database, with water, redundant ligands, and heteroatoms removed using AutoDockTools. Polar hydrogens were added, and Gasteiger charges calculated to prepare PDBQT files. Grid boxes were centered on the protein’s active site, covering the binding cavity. Two negative controls were used:¹ blank control with an empty grid, and.² negative ligand control with molecules known not to bind the target proteins. AutoDock Vina was used to calculate binding free energy (ΔG), with binding energy ≤ −5.0 kcal/mol indicating moderate affinity and ΔG ≤ −7.0 kcal/mol indicating strong binding. Docking conformations with the lowest binding energy were visualized via PyMOL 2.5 and BIOVIA Discovery Studio 2021 to analyze hydrogen bonds, hydrophobic interactions and other intermolecular forces. The heatmap of minimum binding energy for all ligand-hub target pairs was plotted using the ggplot2 package in R 4.2.3.

From the pool of candidate core compounds exhibiting pharmacokinetic properties, those compounds demonstrating the highest rankings according to the compound-target network topology index were selected for further analysis. Specifically, compounds with a robust binding affinity (ΔG < −8.0 kcal/mol) to six COPD-related therapeutic targets (SRC, ESR1, AKT1, MAPK1, STAT3, EGFR) were chosen for molecular dynamics simulations. The optimal ligand-protein complex underwent molecular dynamics (MD) simulation using the CHARMM36 force field. The complex was placed in a TIP3P water box with Na⁺/Cl⁻ ions to achieve a 0.15 M physiological ion concentration. The simulation involved four stages: energy minimization, 2 ns NVT heating from 0 K to 300 K, 2 ns NPT at 1 bar, and a 100 ns production run. A 2 fs timestep was used, with the SHAKE algorithm applied to hydrogen bonds and PME for long-range electrostatics. Trajectories were saved every 100 ps. Analyses included RMSD, RMSF, radius of gyration, SASA, and hydrogen bonds. Binding free energy was calculated using MM-PBSA on stable trajectories from 70–100 ns.

Cell Experiment

Cell Culture and Treatment

Human bronchial epithelial cells (16HBE, Mingzhou Bio, MZ-1420) were cultured at 37 °C in a DMEM high glucose medium (Thermo Fisher, 11320082) supplemented with 10% fetal bovine serum (FBS) (Thermo Fisher, A5256701) and 1% penicillin-streptomycin. The culture environment was maintained at 5% CO₂ with saturated humidity. Once the cells reached 80%-90% confluency, they were subjected to digestion and passaging using 0.25% trypsin (Thermo Fisher, 25200056). Cells in the logarithmic growth phase were then selected for subsequent experimental procedures. This study complies with the ethical standards set forth in the Declaration of Helsinki. The research exclusively utilized the commercial 16HBE cell line for cellular experiments, with no involvement of human subjects, identifiable clinical data, or human biological materials. In accordance with Article 32, Items 1 and 2, of the *Measures for Ethical Review of Life Science and Medical Research Involving Human Subjects* (2023), this study is exempt from institutional ethical review, thereby negating the necessity for Institutional Review Board (IRB) approval and informed consent.

Cell Viability Assay

The cell viability was assessed using the CCK-8 assay. For this purpose, 16HBE cells were seeded into 96-well plates at a density of 1×10^4 cells per well and incubated overnight, with three replicates per group. Each experimental group was treated with Luteolin (Sigma, 440025–5MG) at final concentrations of 0, 5, 10, 20, 40, and 80 μg/mL. Following a 24-hour incubation period, 10 μL of CCK-8 solution (Biosharp, BS350E) was added to each well, and the plates were incubated in the dark for an additional 2 hours. Absorbance was measured at 450 nm using a microplate reader, and the cell survival rate was subsequently calculated.

Cell Model Establishment and Luteolin Intervention

The logarithmic phase 16 HBE cells were seeded into 6-well plates and divided into the following experimental groups: blank control group (Control group), COPD model group (CSE group), low-dose luteolin group (L-L group, 10 μmol/L), medium-dose luteolin group (L-M group, 20 μmol/L), and high-dose luteolin group (L-H group, 40 μmol/L). The cells were cultured for 24 hours to allow for adherence. Subsequently, luteolin was administered at the respective concentrations for 1 hour, followed by the addition of 20% CSE for 24 hours. The blank control group was maintained under standard culture conditions. Upon completion of the incubation period, both the cells and the cell supernatant were collected for further analysis of specific indices.

RT-qPCR

Total RNA was extracted from each group of cells cultured in 6-well plates using Trizol reagent. The purity and integrity of the RNA were subsequently assessed. Following extraction, RNA was reverse transcribed into complementary DNA (cDNA) in accordance with the instructions provided in the reverse transcription kit. Specific primers were designed for each target gene, including SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR, as well as for molecules involved in lipid metabolism and the atherosclerosis pathway, such as CD36, PPARγ, and ABCA1. GAPDH served as the internal reference gene (Table 2). Real-time fluorescence quantitative PCR was employed to evaluate the relative mRNA expression levels in each group. Furthermore, the study analyzed the regulatory effects of Luteolin on the expression of target genes and pathway-related mRNAs.

Table 2.

List of Primers for Real-Time PCR

Gene Name	Primer Sequence
SRC	Forward	5’-TGAAGATGGTGAAGGTGCTG-3’
	Reverse	5’-GGTGATGGTGTTCAGGTTGT-3’
ESR1	Forward	5’-TGGCTACAAGGTCAACAGCA-3’
	Reverse	5’-CCAGCAGCAGGTAGTGGAGT-3’
AKT1	Forward	5’-TGGCACCTTCATTGGCTACA-3’
	Reverse	5’-GATGATGAAGGTGTTGGGTG-3’
MAPK1	Forward	5’-CCTGCTGCTGGACATGGAGA-3’
	Reverse	5’-GGTGGTGTTGATGAAGGTCT-3’
STAT3	Forward	5’-CAGATGGCCCAATGGAACAG-3’
	Reverse	5’-GGGTCTTGATGAAGGTGAAG-3’
EGFR	Forward	5’-GTGAGCAAGATGGCTCTTGA-3’
	Reverse	5’-AGGGTCTTGATGGTGAAGAG-3’
CD36	Forward	5’-GGAAGTGGTGATGTGGTGAA-3’
	Reverse	5’-CAGTGTATGTTGCTGAGGGT-3’
PPARγ	Forward	5’-GAAGACCACTCGCATTCCTT-3’
	Reverse	5’-TCAGCGGGTGAAGACTCAT-3’
ABCA1	Forward	5’-TTCCAGGATGGAGATGTTGA-3’
	Reverse	5’-AGATGAGGTTGAGGATGGTG-3’
GAPDH	Forward	5’-GGAGCGAGATCCCTCCAAAAT-3’
	Reverse	5’-GGCTGTTGTCATACTTCTCATGG-3’

Statistical Analysis

All experimental data were expressed as mean ± standard deviation (SD) of at least three independent replicates. One-way ANOVA was used to analyze the experimental data, and then Tukey’s post hoc test was used to evaluate the differences between multiple groups. Statistical analysis was performed using GraphPad Prism 8 software. The difference was statistically significant at p< 0.05.

Results

Analysis of Chemical Constituents of Jianzhong Peiyuan Decoction

The UHPLC-Q-Orbitrap HRMS technology was employed to generate the total ion chromatogram (Figure 1A and B), leading to the identification of 497 compounds within the Jianzhong Peiyuan Decoction granules. This analysis revealed the presence of 76 flavonoids, 56 terpenoids, 35 lipids, 32 phenolic acids and their derivatives, 24 coumarins and their derivatives, 22 amino acids and their derivatives, 15 carbohydrates and their derivatives, 10 organic acids and their derivatives, 9 tannins, 8 steroids and their derivatives, 7 lignans and their derivatives, 6 alkaloids and their derivatives, 4 indoles and their derivatives, 3 vitamins, 2 stilbenes, 3 quinones, 1 nucleotide and its derivatives, and 184 other types of compounds (Figure 1C and D). Detailed information regarding the mass-to-charge ratio, retention time, fragmentation score, and identification results can be found in Table S1.

Figure 1.

Identification of active ingredients in Jianzhong Peiyuan Decoction. (A) Base peak diagram of positive ion mode detection; (B) Base peak diagram of negative ion mode detection; (C) The relative abundance of different types of compounds; (D) Number of different types of compounds.

The Main Active Ingredients and Targets in Jianzhong Peiyuan Decoction

The active components of Jianzhong Peiyuan Decoction were checked in the TCMSP database. The scores of OB≥30%, DL≥0.18, intestinal absorption (GI) were “high”, and drug-likeness was screened by at least 2 “Yes”. A total of 53 main active components of Jianzhong Peiyuan Decoction were screened (Table 3). Subsequently, according to the oral bioavailability (OB%) and drug similarity drug-likeness (DL) values, nine key compounds-Longikaurin A, Kaempferide, Chrysoeriol, Vitrofolal A, Diosmetin, Luteolin, Coumesterol, Kaempferol and Hydroxygenkwanin were screened from high to low.

Table 3.

Main Active Ingredients of Jianzhong Peiyuan Decoction

Metabolite	Number	Class	m/z	Formula	OB%	DL
Fisetin	MOL013179	Unclassified	287.0549	C15H10O6	52.6	0.24
Kaempferide	MOL004564	Flavonoids	301.0706	C16H12O6	73.41	0.27
Salvianolic acid G	MOL007141	Unclassified	341.0654	C18H12O7	45.56	0.61
Marmesin	MOL001944	Coumarins and derivatives	247.0944	C14H14O4	50.28	0.18
Licochalcone B	MOL004841	Flavonoids	287.0912	C16H14O5	76.76	0.19
Berberine	MOL001454	Unclassified	336.1226	C20H18NO4+	36.86	0.78
Calycosin	MOL000417	Flavonoids	285.0755	C16H12O5	47.75	0.24
7-Acetoxy-2-methylisoflavone	MOL004991	Flavonoids	295.0964	C18H14O4	38.92	0.26
Chrysoeriol	MOL003044	Flavonoids	301.0705	C16H12O6	35.85	0.27
Alisol C	MOL000854	Terpenoids	487.3415	C30H46O5	32.7	0.82
Wighteone	MOL003673	Unclassified	339.1226	C20H18O5	42.8	0.36
16alpha-Hydroxydehydrotrametenolic acid	MOL000273	Terpenoids	471.3469	C30H46O4	30.93	0.81
Nobiletin	MOL005828	Flavonoids	403.1386	C21H22O8	61.67	0.52
Magnolignan A	MOL008539	Unclassified	301.1432	C18H20O4	32.21	0.2
(+)-Ganodermanondiol	MOL011241	Terpenoids	439.3569	C30H48O3	37.64	0.8
Corydine	MOL004197	Unclassified	342.1698	C20H23NO4	37.16	0.55
Triptonide	MOL003244	Unclassified	359.1485	C20H22O6	68.45	0.68
Obacunone	MOL013352	Terpenoids	496.2304	C26H30O7	43.29	0.77
Icaritin	MOL004373	Flavonoids	369.133	C21H20O6	45.41	0.44
Piperine	MOL001592	Alkaloids and derivatives	286.1437	C17H19NO3	42.52	0.23
Isoxanthohumol	MOL003217	Flavonoids	355.1538	C21H22O5	56.81	0.39
Moracin E	MOL003859	Flavonoids	341.1383	C19H16O4	56.08	0.38
Bavachin	MOL000448	Flavonoids	325.1432	C20H20O4	54.44	0.32
Isosinensetin	MOL013277	Flavonoids	355.1174	C20H20O7	51.15	0.44
Crebanine	MOL006971	Alkaloids and derivatives	372.1804	C20H21NO4	34.64	0.75
Moracin D	MOL003858	Flavonoids	309.112	C19H16O4	60.93	0.38
Vitrofolal A	MOL011938	Unclassified	339.1228	C20H18O5	79.17	0.36
Glabrone	MOL004912	Flavonoids	337.1069	C20H16O5	52.51	0.5
Medioresinol	MOL002058	Lignans and derivatives	353.1383	C21H24O7	57.2	0.62
Cryptotanshinone	MOL007088	Terpenoids	297.1483	C19H20O3	52.34	0.4
Tigloylgomisin P	MOL008957	Tannins	515.2275	C28H34O9	30.71	0.83
Tanshinone	MOL007154	Terpenoids	295.1328	C19H18O3	49.89	0.4
Schisandrin C	MOL008992	Unclassified	385.1648	C22H24O6	46.27	0.84
Estrone	MOL010921	Unclassified	271.1692	C18H22O2	53.56	0.32
Andrographolide	MOL008232	Unclassified	349.2029	C20H30O5	46.96	0.36
Isoimperatorin	MOL001942	Unclassified	269.0825	C16H14O4	45.46	0.23
5,7,3’-Trihydroxy-4’-methoxyflavanone	MOL002341	Flavonoids	301.0724	C16H14O6	70.31	0.27
Norizalpinin	MOL002563	Flavonoids	269.0461	C15H10O5	45.55	0.21
Diosmetin	MOL002881	Flavonoids	299.0567	C16H12O6	31.14	0.27
Luteolin	MOL000006	Unclassified	285.0411	C15H10O6	36.16	0.25
Genkwanin	MOL005573	Flavonoids	283.0618	C16H12O5	37.13	0.24
Coumesterol	MOL012976	Unclassified	267.0305	C15H8O5	32.49	0.34
Kaempferol	MOL000422	Flavonoids	285.0411	C15H10O6	41.88	0.24
Hydroxygenkwanin	MOL005530	Unclassified	299.0567	C16H12O6	36.47	0.27
(-)-Farrerol	MOL012432	Flavonoids	299.0931	C17H16O5	42.65	0.26
Longikaurin A	MOL004624	Terpenoids	347.1871	C20H28O5	47.72	0.53
Liquiritigenin	MOL001792	Flavonoids	255.0667	C15H12O4	32.76	0.18
Deoxyelephantopin	MOL008210	Terpenoids	343.1193	C19H20O6	105.32	0.4
Formononetin	MOL000392	Flavonoids	267.0668	C16H12O4	69.67	0.21
Medicarpin	MOL002565	Flavonoids	269.0824	C16H14O4	49.22	0.34
Pectolinarigenin	MOL005842	Flavonoids	313.0724	C17H14O6	41.17	0.3
Emodin	MOL000471	Quinones	269.0461	C15H10O5	83.38	0.24
Isobavachin	MOL000448	Flavonoids	323.1296	C20H20O4	54.44	0.32

Jianzhong Peiyuan Decoction in the Treatment of COPD Targets and PPI Network Analysis

The associated targets for COPD were sourced from the GeneCards, OMIM, DisGeNET, and DrugBank databases, resulting in the identification of 9,300 disease target genes after the removal of duplicates.^18,19 Jianzhong Peiyuan Decoction’s active components were found to correspond with 816 disease target genes. By determining the intersection between the targets of Jianzhong Peiyuan Decoction’s active components and the COPD targets, 675 potential target genes were identified (Figure 2A). These 675 common targets were subsequently input into the STRING database to construct a “drug-disease” visual protein-protein interaction (PPI) network. This PPI network comprises 673 nodes, 2,204 edges, and an average node degree of 6.5 (Figure 2B). To further elucidate the mechanism by which Jianzhong Peiyuan Decoction exerts its therapeutic effects on COPD, the PPI network was analyzed using the CytoNCA plug-in. Based on the median values of Betweenness, Closeness, Degree, Eigenvector, Local Average Connectivity (LAC), and Network metrics, a refined core PPI network was generated, consisting of 6 nodes and 13 edges (Figure 2B–E). The key targets identified in this network include SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR, which may play crucial roles in the therapeutic action of Jianzhong Peiyuan Decoction.

Figure 2.

Network pharmacological analysis of Jianzhong Peiyuan Decoction in the treatment of COPD. (A) Identification of overlapping genes between Jianzhong Peiyuan Decoction and COPD; (B–E) The nodes represent proteins, and the color gradient from light to dark indicates the binding degree of target proteins; (F–H) Gene Ontology enrichment analysis of overlapping genes; (I) Kyoto Encyclopedia of Genes and Genomes pathway enrichment analysis of overlapping genes.

GO and KEGG Enrichment Analysis

To comprehensively elucidate the mechanism of Jianzhong Peiyuan Decoction in the treatment of COPD, GO and KEGG enrichment analyses were conducted on 675 overlapping targets. The GO enrichment analysis identified enrichment in 3,375 biological processes, 257 cellular components, and 540 molecular functions. The top ten items were selected based on their Enrichment Score (Figure 2F–H). The most significantly enriched biological process was peptidyl-serine phosphorylation (GO:0018105), encompassing 88 overlapping targets. The cellular component-related projects predominantly involved the membrane raft (GO:0045121), membrane microdomain (GO:0098857), and integral component of the presynaptic membrane (GO:0099056). In terms of molecular function, protein serine/threonine kinase activity (GO:0004674), transmembrane receptor protein tyrosine kinase activity (GO:0004714), and protein tyrosine kinase activity (GO:0004713) were highlighted. KEGG identified the top 20 pathways based on their p. adjust values. The analysis revealed that the therapeutic mechanism of Jianzhong Peiyuan Decoction in the treatment of COPD primarily involves the Lipid and Atherosclerosis pathway (hsa05417), the Calcium Signaling Pathway (hsa04020), and the AGE-RAGE Signaling Pathway in Diabetic Complications (hsa04933) (Figure 2I).

Molecular Docking

Utilizing network pharmacology analysis, molecular docking studies were conducted to assess the interactions between the six primary targets of Jianzhong Peiyuan Decoction for the treatment of COPD and its principal active constituents: Longikaurin A, Kaempferide, Chrysoeriol, Vitrofolal A, Diosmetin, Luteolin, Coumesterol, Kaempferol, and Hydroxygenkwanin (Figure 3A–I). The heat map analysis of docking scores revealed that these compounds exhibit strong binding affinities with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR, with scores less than −5.0 kcal/mol, suggesting that these active components are potential therapeutic targets for Jianzhong Peiyuan Decoction in COPD management (Figure 3J). Notably, the binding energies for Luteolin-SRC (−9.07 kcal/mol), Coumesterol-SRC (−9.06 kcal/mol), Chrysoeriol-ESR1 (−8.19 kcal/mol), Luteolin-AKT1 (−8.61 kcal/mol), Kaempferide-MAPK1 (−8.44 kcal/mol), Chrysoeriol-STAT3 (−8.46 kcal/mol), and Kaempferide-EGFR (−8.16 kcal/mol) are all below −8.0 kcal/mol, indicating a particularly strong binding affinity.

Figure 3.

Binding conformations between key COPD targets and core active ingredients of Jianzhong Peiyuan Decoction. (A) Binding conformations of Longikaurin A with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (B) Binding conformations of Kaempferide with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (C) Binding conformations of Chrysoeriol with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (D) Binding conformations of Vitrofolal A with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (E) Binding conformations of Diosmetin with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (F) Binding conformations of Luteolin with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (G) Binding conformations of Coumesterol with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (H) Binding conformations of Kaempferol with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (I) Binding conformations of Hydroxygenkwanin with SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR; (J) Heat map of molecular docking scores.

MD Simulation

To advance the investigation of protein-ligand interaction stability, MD simulations were conducted on six protein-luteolin complexes: Luteolin-SRC, Luteolin-ESR1, Luteolin-AKT1, Luteolin-MAPK1, Luteolin-STAT3, and Luteolin-EGFR. The RMSD metric was employed to assess the stability of the simulation system. Equilibrium is considered to be achieved when the RMSD values oscillate around a stable mean without exhibiting significant deviations. Therefore, the RMSD values of all complexes tended to be stable and maintained in a small fluctuation range during the simulation process (Figure 4A), indicating that all systems reached equilibrium and the trajectory was stable and reliable. The RMSF results showed that the overall fluctuation of each protein residue was at a low level (Figure 4B), indicating that the overall structure of the protein skeleton maintained good conformational stability after ligand binding, and no large area of disordered or unfolded regions appeared. The radius of gyration (Rg) used to analyze the receptor-ligand binding tightness remained stable during the simulation of all complexes, with a small fluctuation (Figure 4C), indicating that the overall structure of the protein after ligand binding remained compact and did not undergo significant expansion or looseness. The SASA, which reflects the degree of solvent exposure of protein, also showed a stable fluctuation trend in each system (Figure 4D), indicating that the interaction mode between protein and solvent did not change dramatically, and the overall structure remained stable The results of hydrogen bond counting analysis showed that each complex formed a continuous hydrogen bond interaction during the simulation process (Figure 4E), indicating that there was a stable polar interaction between the ligand and the protein, which helped to maintain the binding state of the complex. The two-dimensional and three-dimensional mapping results of the FEL show that each system forms an obvious low-energy stable conformational region (Figure 4F and G), indicating that the ligand-protein complex can stably occupy the low-energy conformation during the simulation process, forming an energy-favorable binding state. The results of residue energy decomposition further showed that multiple key amino acid residues contributed significantly to the binding free energy (Figure 4H), which verified that the interaction between ligand and protein had clear site specificity. The binding free energy (ΔGbind) was calculated using the MM/PBSA method, and the lower ΔGbind value corresponded to a stronger binding affinity. The snapshot is extracted from the equilibrium trajectory for MM/PBSA energy calculation. The results showed that the ΔGbind values of the six complexes were as follows: Luteolin-SRC (−5.9 kcal/mol), Luteolin-ESR1 (−8.5 kcal/mol), Luteolin-AKT1 (−3.4 kcal/mol), Luteolin-MAPK1 (−8.1 kcal/mol), Luteolin-STAT3 (−13.7 kcal/mol) and Luteolin-EGFR (−9.9 kcal/mol) (Figure 4I). These results showed that there was a stable binding between Luteolin and six target proteins, among which the Luteolin-STAT3 complex showed the strongest binding affinity, while the Luteolin-AKT1 binding affinity was relatively weak, and the other complexes showed moderate to strong binding ability, suggesting that Luteolin can form a stable interaction with multiple target proteins.

Figure 4.

Molecular dynamics simulation (MDS) results of Luteolin binding with six target proteins. (A) RMSD values of each target protein-Luteolin complex. (B) RMSF values during Luteolin simulation. (C) Rg curves of Luteolin-bound complexes. (D) SASA of Luteolin-protein complexes during the 100ns simulation. (E) Hydrogen bond dynamics observed during molecular dynamics simulation. (F–G) Two-dimensional and three-dimensional mapping of the free energy landscape. (H) Residue energy decomposition of Luteolin-protein binding. (I) Binding free energy (ΔGbind) between Luteolin and target proteins calculated by MDS.

Abbreviations: RMSD, root mean square deviation; RMSF, root mean square fluctuation; Rg, radius of gyration; SASA, solvent accessible surface area.

Cell Experiment

Luteolin concentrations ranging from 5 to 40 μg/mL exhibited no discernible cytotoxic effects on 16HBE cells, maintaining a cell viability rate exceeding 85%. However, at a concentration of 80 μg/mL, a significant reduction in cell viability was observed, indicating that higher concentrations of Luteolin possess cytotoxic properties (Figure 5A). Results from RT-qPCR analysis demonstrated that, relative to the Control group, the mRNA expression levels of SRC, AKT1, MAPK1, STAT3, EGFR, and CD36 were significantly up-regulated in the CSE model group (p<0.01). Conversely, the mRNA expression levels of ESR1, PPARγ, and ABCA1 were significantly down-regulated (p<0.05, p<0.01), indicating that CSE stimulation effectively induced the activation of inflammation-related pathways and disruptions in lipid metabolism. In comparison to the CSE group, varying concentrations of Luteolin were able to modulate the aberrant expression of the aforementioned genes in a dose-dependent manner. Specifically, the mRNA expression levels of SRC, AKT1, MAPK1, STAT3, EGFR, and CD36 were significantly reduced, whereas the mRNA expression levels of ESR1, PPARγ, and ABCA1 were significantly elevated (Figure 5B–J). These findings suggest that Luteolin ameliorates CSE-induced injury in 16HBE cells by modulating the expression of targets associated with inflammation and molecules involved in lipid metabolism pathways.

Figure 5.

Effects of Luteolin on cell viability and mRNA expression of related genes in 16HBE cells. (A) Effects of different concentrations of Luteolin (0–80 μg/mL) on the viability of 16HBE cells. (B–J) The mRNA expression levels of SRC, ESR1, AKT1, MAPK1, STAT3, EGFR, CD36, PPARγ, and ABCA1 in 16HBE cells of different treatment groups (Control group, CSE model group, low-dose Luteolin group, medium-dose Luteolin group, and high-dose Luteolin group). *p<0.05, **p<0.01.

Discussion

COPD represents a significant global public health challenge, with primary risk factors including prolonged exposure to cigarette smoke, ambient air pollution, industrial particulate matter, and secondary factors such as recurrent pulmonary infections, developmental lung anomalies, and genetic predispositions.²⁰ Jianzhong Peiyuan Decoction, a modification of the traditional Shenling Baizhu Powder, is extensively utilized in Guizhou clinical practice for treating stable COPD patients exhibiting lung-spleen qi deficiency syndrome.¹² Although this herbal formula has demonstrated efficacy in alleviating clinical symptoms and enhancing quality of life during routine treatment, the comprehensive profile of its bioactive components and the systematic molecular regulatory mechanisms remain insufficiently understood. This gap in knowledge drives the current integrated research, which combines chemical profiling, computational pharmacology, and in vitro cell validation.

In this study, a total of 497 chemical constituents were qualitatively characterized using UHPLC-Q-Orbitrap HRMS. The identified components predominantly include flavonoids (such as kaempferol, luteolin, and liquiritin), terpenoids (including tanshinone, oleanolic acid, and glycyrrhizin), unsaturated lipids (such as stearidonic acid and 9(S)-HOTrE), as well as a substantial presence of phenolic acid derivatives (such as chlorogenic acid and ferulic acid). Given that the formulation comprises twelve distinct medicinal herbs, the complex multicomponent composition aligns with the intrinsic characteristics of multi-target herbal medicines.²¹ In agreement with previous pharmacological studies, several key compounds identified in this analysis demonstrate lung-protective effects against COPD-related injury. Specifically, kaempferol inhibits pulmonary ferroptosis by preventing NCOA4-mediated ferritin degradation and restoring GPx4 antioxidant activity, thereby disrupting lipid peroxidation cascades.^22,23 Additionally, the combination of liquiritin and licochalcone B reduces pulmonary inflammation and fibrosis by inhibiting HCK signaling, thus alleviating oxidative stress.²⁴ Luteolin, identified as the principal monomer in our cell assays, mitigates airway oxidative damage and inflammatory responses by modulating the TRPV1/SIRT6, CYP2A13/NRF2, and NOX4/NF-κB pathways.^25,26 Concurrently, tanshinone increases intracellular heme levels, thereby suppressing pro-oxidant and pro-inflammatory cascades in pulmonary parenchymal cells and macrophages.²⁷ It is important to acknowledge that our ingredient screening utilized only in OB and DL thresholds, without validation through serum pharmacochemistry. Consequently, we cannot confirm which components are capable of entering systemic circulation following oral administration; this limitation will be addressed in subsequent studies.

Through the integration of network pharmacology prediction, protein-protein interaction (PPI) topological screening, and KEGG pathway enrichment analysis, it has been collectively demonstrated that SRC, ESR1, AKT1, MAPK1, STAT3, and EGFR function as central hub genes. Additionally, three signaling pathways—namely, the lipid and atherosclerosis pathway, the calcium signaling pathway, and the AGE-RAGE pathway—serve as the principal regulatory modules underpinning the anti-COPD effects of Jianzhong Peiyuan Decoction. These predicted targets and pathways exhibit biological relevance to established pathological cascades associated with COPD. Specifically, exposure to cigarette smoke extract activates SRC kinase in small airway epithelial and alveolar macrophage cells, leading to the upregulation of MMP-9/MMP-12, cathepsin K, and pro-inflammatory mediators such as IL-17, TNF-α, MCP-1, and KC, thereby exacerbating airway inflammatory infiltration and lung tissue destruction.²⁸ Furthermore, hyperactivation of the EGFR/MAPK signaling pathway accelerates the abnormal proliferation of airway smooth muscle cells and perpetuates inflammatory stimuli. Persistent activation of AKT1 transactivates downstream NF-κB, resulting in the excessive secretion of TNF-α, IL-6, and IL-17, as well as the continuous recruitment of neutrophils and mononuclear macrophages.²⁹ Which collectively contribute to the formation of a self-amplifying inflammatory microenvironment in the airways of COPD patients.^30,31 Furthermore, dysregulated lipid metabolism synergistically exacerbates pathological remodeling processes. Specifically, aberrant LOX-1/NF-κB/AKT signaling pathways induce epithelial apoptosis, excessive mucus production, and increased smooth muscle contractility, collectively resulting in the narrowing of the bronchial lumen.^32,33 CD36, PPARγ, and ABCA1, which are three critical mediators of lipid homeostasis enriched within the lipid and atherosclerosis pathway, also contribute to the progression of chronic obstructive pulmonary disease (COPD). Exposure to cigarette smoke upregulates CD36 expression in alveolar macrophages, facilitating the uptake of oxidized low-density lipoprotein and free fatty acids, thereby promoting foam cell formation and intracellular lipid accumulation.³⁴ PPARγ exhibits protective effects by inhibiting the JAK-STAT, MAPK, and NF-κB signaling cascades, thereby suppressing inflammation and airway remodeling.³⁵ In contrast, smoke-induced downregulation of ABCA1 impairs cellular cholesterol efflux, leading to lipid accumulation and hyperactivation of the NLRP3 inflammasome.³⁶

Subsequent molecular docking and 100 ns molecular dynamics simulations validated stable binding interactions between luteolin and all six hub proteins (SRC, ESR1, AKT1, MAPK1, STAT3, EGFR).³⁷ Our in vitro CSE-induced 16HBE cell model further provided preliminary transcriptional evidence to support these computational predictions. CCK-8 cytotoxicity testing confirmed that luteolin exerted no obvious cellular toxicity within the concentration range of 5–40 μg/mL.³⁸ RT-qPCR results demonstrated that luteolin dose-dependently reversed CSE-triggered transcriptional disorders: it downregulated the elevated mRNA levels of pro-inflammatory hub genes (SRC, AKT1, MAPK1, STAT3, EGFR) and lipid uptake receptor CD36, while restoring the suppressed expression of ESR1, PPARγ and ABCA1. Such transcriptional changes are highly consistent with the target-pathway network predicted by network pharmacology, preliminarily verifying that luteolin mediates dual regulation of inflammatory cascades and lipid metabolic homeostasis.

Integrating all above results, we tentatively propose a potential mechanistic axis to interpret the therapeutic effect of Jianzhong Peiyuan Decoction and its core component luteolin against stable COPD. Cigarette smoke stimulus triggers sustained overactivation of the SRC/MAPK1/AKT/STAT3 pro-inflammatory axis in bronchial epithelial cells, which drives persistent inflammatory cell infiltration, protease over-secretion and irreversible airway remodeling.^39–41 Meanwhile, hyperactive inflammatory signaling represses PPARγ transcriptional activity, further disturbing the balance of its downstream lipid transporters CD36 and ABCA1; this vicious cycle causes intracellular lipid accumulation, foam cell generation and aggravated oxidative injury.^42–45 Luteolin, as a representative bioactive constituent of Jianzhong Peiyuan Decoction, can simultaneously intervene in both cascades to normalize abnormal gene expression, thereby alleviating CSE-mediated bronchial epithelial damage in vitro. It is important to clarify that this regulatory mechanism is only supported by mRNA-level in vitro data, and more multi-layered experimental evidence is required to fully validate the conclusion.

While this study offers valuable insights into the potential of Jianzhong Peiyuan Decoction for improving COPD, several limitations warrant further consideration. Firstly, the chemical analysis conducted was limited to the qualitative identification of herbal components and did not include the quantitative determination of luteolin content within Jianzhong Peiyuan Decoction. The concentration of luteolin used for cell intervention was based on existing literature rather than quantitative data derived from the entire formula. Secondly, the study employed network pharmacology and molecular docking, which are computational predictive analyses that lack validation through in vivo animal models. Additionally, all cellular functional assays were conducted using the 16HBE cell line, which does not fully replicate the complex pathological microenvironment of airway tissue in COPD patients. Thirdly, the in vitro verification was limited to detecting changes at the mRNA transcription level of key genes, without assessing protein expression, inflammatory factor secretion, or intracellular oxidative stress indicators. This resulted in a relatively narrow molecular evidence chain for the proposed multi-target regulatory axis. Fourthly, this study focused solely on the intervention effects of the single active component luteolin, necessitating further investigation into the synergistic regulatory mechanisms of the multiple components present in Jianzhong Peiyuan Decoction. In forthcoming research, we intend to develop a murine model of cigarette smoke-induced COPD to assess the efficacy and safety of luteolin in vivo. Furthermore, we aim to conduct an in-depth investigation into the molecular mechanisms underlying the action of Jianzhong Peiyuan Decoction and to elucidate the dose-response relationship of the complete formulation.

Conclusions

This study used UHPLC-Q-Orbitrap HRMS to identify 497 chemical constituents in Jianzhong Peiyuan Decoction, including flavonoids and terpenoids. From these, 53 bioactive ingredients meeting OB ≥ 30% and DL ≥ 0.18 were screened using the TCMSP database, with nine core compounds, such as luteolin, selected for further analysis. By intersecting drug targets with COPD disease genes from four databases, 675 therapeutic targets were identified. PPI network analysis highlighted six hub proteins (SRC, ESR1, AKT1, MAPK1, STAT3, EGFR), with GO/KEGG enrichment revealing key pathways. Molecular docking showed all nine compounds had binding energy below −5.0 kcal/mol with hub targets, and 100 ns MD simulation confirmed stable binding between luteolin and the six proteins. In vitro experiments on 16HBE cells used one-way ANOVA and Tukey’s post hoc test for comparing mRNA data across multiple groups. RT-qPCR data indicated that luteolin at 5–40 μg/mL could dose-dependently reverse CSE-induced abnormal gene expression related to inflammation and lipid metabolism without significant cytotoxicity, suggesting its protective effect on bronchial epithelial cells. Although current research offers only computational predictions and mRNA-level in vitro evidence, it establishes a strong basis for further exploration of Jianzhong Peiyuan Decoction. Future studies will develop COPD mouse models induced by cigarette smoke for in vivo efficacy assessments, perform quantitative component analysis to determine dose-effect relationships, and investigate the synergistic interactions among the active ingredients for effective COPD treatment.

Funding Statement

This research was supported by the Guizhou Science and Technology Cooperation Support Program [2021] General Project No. 014.

Data Sharing Statement

The data generated in this study are available from the corresponding author upon request.

Author Contributions

Amei Tang: Conceptualization, Methodology, Writing – original draft.

Ke Lu: Investigation, Writing – original draft.

Yang Liu: Data curation, Writing – review and editing.

Guoxiang Tang: Data curation, Writing – review and editing.

Feng Cao: Methodology, Writing – review and editing.

Sufang Zhou: Supervision, Methodology, Writing – review and editing.

All authors gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work.

Disclosure

No conflicts of interest in this work.