Network pharmacology analysis

An overview of the network pharmacology workflow is shown in Fig. 1. Key screening criteria and parameter settings for each step are specified in Sect. Obtaining targets associated with Ligustilide and ALI– Molecular Docking Validation.

Fig. 1The alternative text for this image may have been generated using AI.

Schematic workflow of the network pharmacology analysis of ligustilide in ALI. The pipeline includes target collection, target intersection, PPI analysis, GO/KEGG enrichment, network construction, and molecular docking

Obtaining targets associated with ligustilide and ALI

The PubChem database (https://pubchem.ncbi.nlm.nih.gov/) was utilized to obtain the SMILES name of “Ligustilide” [15], which was then imported into the SwissTargetPrediction database (http://swisstargetprediction.ch/) to obtain Ligustilide’s corresponding targets (probability > 0) with species parameter set to “Homo sapiens” [16]. The chemical structure file and potential targets of ligustilide were retrieved from the TCMSP database (http://tcmspw.com/tcmsp.php) using “Ligustilide” as the search term [17]. Concurrently, the structure file was uploaded to PharmMapper (https://lilab-ecust.cn/pharmmapper/submitfile.html) for complementary target prediction. The obtained Uniprot IDs were standardized using the Uniprot database (https://www.uniprot.org/). All predicted targets were aggregated and deduplicated to establish the comprehensive ligustilide-related target dataset. To minimize potential false-negative results, a relatively inclusive probability threshold was applied at the initial target prediction stage, allowing comprehensive coverage of potential ligustilide-associated targets for subsequent network analysis.

Obtaining targets associated with ALI

Disease-associated targets were systematically collected from three repositories: OMIM (https://www.omim.org/), GenCards (https://www.genecards.org/) (Relevance score exceeding the median value) [18], and TTD (https://db.idrblab.net/ttd/) using “Acute Lung Injury” as the query term. ALI-related targets from all databases were consolidated and deduplicated to generate the dataset. This strategy ensured the inclusion of disease-relevant targets while reducing potential bias from individual databases.

Protein-Protein interaction (PPI) network construction

The ligustilide-related targets and ALI-related targets were imported into the bioinformatics visualization tools of Sangerbox 3.0 (http://sangerbox.com/home.html) to identify overlapping targets. These targets were submitted to the STRING database (https://cn.string-db.org/) with parameters: “Homo sapiens” species, minimum interaction score ≥ 0.7, and hidden disconnected nodes [19]. A high-confidence interaction threshold was selected to enhance the reliability of the constructed PPI network. The resultant PPI network was imported into Cytoscape 3.8.0 for topological analysis. Core targets were identified using CytoNCA plugin through betweenness Degree Centrality (DC) and Betweenness Centrality (BC) (exceeding twice the median values) [20]. Nodes meeting these criteria were considered to play central roles in the network and were therefore defined as core targets for subsequent functional analysis.

Gene ontology (GO) and Kyoto encyclopedia of genes and genomes (KEGG) pathway enrichment analysis

GO functional annotation (biological process (BP), molecular function (MF), cellular component (CC)) and KEGG pathway enrichment were performed using R packages “clusterProfiler” and “enrichplot”. Significant terms were filtered (adjusted P < 0.05, q < 0.05) with the top 30 KEGG pathways visualized as bubble plots. A “Compound-Disease-Pathway-Target” network was reconstructed in Cytoscape 3.8.0 for integrative visualization.

Molecular docking validation

The 3D structure of ligustilide (SDF format) was downloaded from PubChem and energy-minimized in Chem3D. The crystal structure of PIK3R1 (PDB ID: 4L23) was obtained from RCSB PDB (https://www.rcsb.org/). AutoDock Vina 1.5.7 performed molecular docking after protein preparation (water removal, hydrogen addition, charge calculation). Binding conformations were visualized using PyMOL with ligand-receptor interactions analyzed.

Validation by in vitro cell experimentsCulture and passaging of HPAEpiCs

HPAEpiCs (batch No. HUM-iCell-a002) were purchased from iCell Bioscience Inc. (Shanghai, China). They were maintained at 37 °C in a humidified incubator with 5% CO₂. For passaging, the medium was removed and cells were gently rinsed twice with PBS, followed by brief trypsinization for 1 min. Detachment was monitored under a microscope, and trypsinization was terminated when 90% of cells became rounded and started to detach by adding complete medium. Cells were gently resuspended, collected by centrifugation (1000 rpm, 5 min), resuspended in fresh complete medium, and subcultured into new T25 flasks as required. Passage numbers were recorded throughout the study.

Establishment of ALI cell model

HPAEpiCs were randomly divided into four groups and 10 µg/ml LPS (batch No.039M4004V, sigma, USA) was applied for 0 h, 4 h, 24 h and 48 h respectively to determine the optimal duration of LPS action based on cellular activity and inflammatory factor release.

Detection of cell activity by MTT method

MTT kit was purchased from BBI Life Sciences Corporation (Shanghai, China). HPAEpiCs in the logarithmic growth phase were seeded into 96-well plates. After the indicated treatments, 10 µL of MTT solution was added to each well and incubated at 37 °C. Cell viability was assessed by measuring the absorbance at 562 nm using a microplate reader (CMax Plus, Molecular Devices, USA). Five replicate wells were included for each group.

ELISA was used to detect changes in the levels of IL-6 and TNF-α in the supernatant of each group of HPAEpiC cultures

The cell culture supernatant was extracted, centrifuged for 20 min (2000–3000 rpm) using Micro17R cryogenic high-speed centrifuge (Thermo Scientific, USA), and then placed in a container and kept at -80 °C. Following the manufacturer’s instructions, ELISA kits (Jiangsu Enzyme Free Industry Co. Ltd., China) were used to measure the levels of IL-6 and TNF-α. Six times were added to the test.

Ligustilide concentration selection

HPAEpiCs were split into 11 groups: a blank control (NC) and 10 experimental groups that underwent ligustilide (HPLC ≥ 98%, Shanghai Yuanye Bio-Technology Co., Ltd., China, batch No. B2042) interventions at various doses (0.78, 1.56, 3.125, 6.25, 12.5, 25, 50, 100, 200, and 500 µg/ml). A ligustilide stock solution (50 mg/ml; 1 mg ligustilide dissolved in 20 µL DMSO) was prepared and subsequently diluted with complete culture medium to the desired working concentrations. Changes in cell activity were monitored using the MTT method previously described. Cells were incubated with ligustilide for 24 h, after which cell viability was assessed using the MTT assay as previously described. Data were collected from five technical replicate wells per group.

Cell viability assay under LPS and ligustilide co-treatment

To further assess the cytoprotective effect of ligustilide under LPS-induced injury, we performed an additional MTT assay under co-treatment conditions. HPAEpiCs were seeded in 96-well plates and allowed to adhere overnight. Cells were assigned to the following groups: (i) Control; (ii) LPS only (10 µg/ml); (iii) LPS + ligustilide co-treatment, where ligustilide (25 µg/ml) and LPS (10 µg/ml) were added simultaneously; and (iv) ligustilide pretreatment, where cells were pre-incubated with ligustilide (25 µg/ml) for 4 h prior to LPS (10 µg/ml) exposure, with ligustilide maintained during LPS stimulation. Cell viability was assessed using the MTT protocol described above. Six replicate wells were included for each group.

Cell grouping and treatment

Well-grown HPAEpiC were injected in 12-well plates at a concentration of 1 × 10^5 cells/ml after being digested with 0.25% trypsin (batch No.SH30042.01, Cyclone, USA), and then they were randomly separated into four groups. The marks are as follows for the purpose of records: (i) Control group; (ii) model group (LPS); (iii) ligustilide group (LPS + ligustilide); and (iv) Anti-IL-17 A neutralizing antibody group (LPS + ligustilide + Anti-IL-17 A neutralizing antibody).

Ligustilide (25 µg/ml) and anti-IL-17 A neutralizing antibodies (1:500, batch No.DF6127, Affinity Antibodies, USA) were added at the ideal concentration and incubated for 4 h after HPAEpiC expanded to 60–70% density. Pretreatment therefore allows sufficient time to activate biological pathways before the injury is initiated. 10 µg/ml of LPS was added after incubation, and cells were harvested after 24 h. 1 ml/well of Trizol reagent (batch No. B511311, Bioengineering Shanghai Co., Ltd., China) was added to each well to extract total RNA and protein. After the best action time for usage, the cell supernatant was collected.

Determination of IL-17 A, IL-6 and TNF-α in the supernatant of each group of cells by ELISA.

Using the previously stated ELISA technique, the experiment was carried out 10 times.

Real-Time PCR method to determine the mRNA expression levels of Beclin-1 and MAP1LC3B (LC3B) in each group of cells

RNA was isolated from the cells by Trizol one-step extraction, agarose gel electrophoresis was performed to identify its purity by EPS300 electrophoresis instrument (Shanghai, China Tanon Technology Co., Ltd.), nucleic acid concentration was measured by mixing with RNA-free deionized water, the amount of nucleic acid was calculated, and cDNA was synthesized by reverse transcription (batch No. CW2569, Kangwei Century Biotechnology Co.). Reaction conditions: 42 °C, 15 min; 85 °C, 5 min.

According to the primer design principle, the DNA and mRNA of the test genes, Beclin-1 and LC3B, as well as the sequence of the internal reference gene, GAPDH, were located in the PUBMED database. Primer Premier 5.0 was used to construct the upstream and downstream primers (Table 1). Then perform real-time fluorescence quantitative PCR reactions (CFX Connect, BIO-RAD, USA). Conditions: 95 °C, 10 min denaturation; 95 °C, 15 s; 60 °C, 60 s; 40 cycles. All experiments were performed in triplicate.

Table 1 Primer sequences for PCRImmunofluorescence detection of Beclin-1 and LC3 expression in various groups of cells

The plates were seeded at 60–70% cell density, and the drugs were individually added after walling, fixed with 4% paraformaldehyde (PFA) for 10 min, and rinsed 3 times with pH 7.2–7.4 phosphate buffer (PBS) (Cyclone, USA) for 5 min each time; 1–2 ml of 0.5% Triton X-100 was added and treated for 2 min to permeabilize the cell membrane, rinsed in the same way, and closed with an appropriate amount of 3% BSA blocking solution for 0.5 h and incubated the primary antibody. Repeated the rinsing step; incubated the secondary antibody and kept all subsequent steps away from light. Rinse twice, stained with DAPI, rinse once, and added 1 ml of PBS to keep wet. Added an appropriate amount of blocking agent Mounting media, and capillary aspirated nail polish to seal the film; observed and photographed under the inverted fluorescent microscope (Ts2-FC, Nikon Corporation, Japan). Fetal Bovine Serum (batch No.11011 − 8615) was purchased from Zhejiang Tianhang Biotechnology Co. (China) and Goat anti-rabbit IgG H&L (Alexa Fluor® 488) (batch No. ab150077) from Abcam Plc. (UK). All experiments were independently repeated three times.

Transmission electron microscope observation of autophagy in each group of HPAEpiC

Collected cells were fixed in 2.5% glutaraldehyde solution for 4 h, rinsed 4 times in PBS for 15 min each, fixed in 1% osmium acid solution for 1–2 h, rinsed 3 times; dehydration in a gradient of ethanol solution (50%, 70%, 80%, 90%, 95% and 100%) for 15 min in each gradient except 100% treatment for 20 min, and finally treated with pure acetone for 20 min. Samples were treated with an embedding agent acetone mixture, (V/V = 1/1) ratio for 1 h, (V/V = 3/1) for 3 h, overnight at room temperature; placed in 0.5 ml dry centrifuge tubes, heated and polymerized overnight at 70 °C; trimmed and sectioned (50–70 nm) on a Leica EM UC 7 ultrathin sectioning instrument; plastic embedding moulds, 50% Uranium acetate 50% ethanol saturated solution 100 µL staining (15 min–1 h), double-distilled water rinse with 100 µL of lead citrate for 15 min; electron microscopy (H-7650, Hitachi Limited, Japan) photographed. Data represent three independent experiments.

Western blot assay for the protein expression levels of Beclin-1, LC3, Akt, p-Akt, mTOR and p-mTOR in cells

Discarded supernatant. Cells were washed twice with PBS before centrifuging at 1000 rpm for 5 min, the supernatant removed, lysed on ice with RIPA lysis solution for 30 min, centrifuged at 12,000 g, 4 °C for 5 min, supernatant taken as protein sample; BCA method (batch No.pc0020, Solarbio, Beijing) was used to determine the total protein concentration of the samples; 10% separation gel and 5% concentration gel were configured according to the instructions, injected 1× Tris-Gly’s electrophoresis buffer, add the sample and pre-stained protein marker (batch No.PR1910, Solarbio, Beijing), and keep the gel surface in equilibrium with 1× SDS loading buffer. The initial voltage was 80 V, and the protein sample was lifted to 120 V after entering the separation gel for electrophoresis. After transferring the membrane well (VE186, Tanon, China) marked, the membrane was washed 10 min x 3 times with TBST. Conditions: Regulated current 200 mA, 120 min.

The PVDF membrane (GE Healthcare Life, USA) was closed with 5% skimmed milk powder in a blocking solution and shaken for 1.5–2 h. The membrane is washed as above; the membrane was incubated overnight at 4 °C in an incubator containing primary antibody dilution (according to the instructions) and shaken at room temperature; the next day the membrane was removed and shaken at room temperature for 30 min, the primary antibody (Anti-rabbit IgG, CST, USA) was aspirated and discarded and the membrane was washed; the secondary antibody (HRP-linked Antibody, CST, USA) was diluted in blocking solution and shaken at room temperature for 1–2 h. After the reaction, the secondary antibody was recovered and the membrane was washed. The PVDF membranes were exposed to the ECL reagent for 3 min; images were taken using chemi capture software, chemical light, and automatic exposure. All assays were independently repeated three times. Beclin-1, LC3, Akt, p-Akt, mTOR, p-mTOR and GAPDH Antibodies were purchased from Affinity Biosciences Pty Ltd. (USA). The LC3 antibody recognizes both LC3-I and LC3-II forms, as confirmed by the presence of two bands in WB.

Statistical analysis

Statistical analyses were performed using SPSS 16.0. Data are expressed as mean ± SD. Comparisons among multiple groups were conducted using one-way ANOVA followed by Dunnett’s post hoc test. A value of P < 0.05 was considered statistically significant.