Materials

Lipopolysaccharide (LPS, No. L2880, 055: B5) was purchased from Sigma (St. Louis, MO, USA). Dexamethasone (DXMS, No. H37021898) was purchased from Chenxin Pharmaceutical Co, Ltd (Jining, Shandong Province, China). Mouse PIP3 ELISA (enzyme-linked immunosorbent assay) Kit (No. ZC-38629) was purchased from ZCIBIO Technology Co, Ltd. Platycodon grandiflorum (No.101206) was purchased from the Anhui Fengyuan Tongling TCM Decoction Pieces Co, Ltd (Tongling, Anhui Province, China). Sichuan Weikeqi Biological Technology Co, Ltd provided the reference standard deapio dealio platycodin D, platycodin D2 and platycoside D. (Chengdu, Sichuan Province, China). Avanti Polar Lipids Company provided SM (17:0) (batch number: 170SM-13) and PE (17:0/17:0) (batch number: LM170PE-19) (Alabaster, AL, USA). ROE Co. provided isopropanol, MTBE, ammonium formate and ammonium acetate with a mass spectrometry purity of 99.8%. (Oradell, NJ, USA). Merck provided methanol, acetonitrile and formic acid with a mass spectrometry purity of 99.8%. (Darmstadt, Germany). Ultrapure water filtered via using a Milli-Q water purification system was used to make all aqueous solutions (Millipore, Milford, MA, USA).

Apparatus

Qualitative analysis of PAE: ACQUITY UPLC® H-Class System (Waters Co, Milford, MA, USA) and LTQ Orbitrap XL™ hybrid FT mass spectrometer (Thermo Fisher Scientific Inc, Bremen, Germany). Lipidomics: UltiMate® 3000 ultra-high-performance liquid chromatography system (DIONEX, Sunnyvale, CA, USA) and Q Exactive™ Hybrid Quadrupole-Orbitrap™ Mass Spectrometer (Thermo Fisher Scientific Inc, Bremen, Germany).

Preparation of PAE

Seventy-five grams of Platycodon grandiflorus were boiled for 1.5 h after being soaked in 0.9 L water for 30 min. After pouring out the first decoction, 0.75 L of water was added and boiled for 1 h. Finally, two decoctions were combined, filtered through gauze and concentrated to achieve PAE concentrations of 0.151, 0.3775, or 0.755 g/mL.

Qualitative analysis of PAEPAE sample pretreatment

The liquid from the preparation of PAE was diluted with water to 150 mg/mL and centrifuged at 18,000 rpm for 10 min. Then, the supernatant was filtered by a 0.22 μm membrane for LC-LTQ-Orbitrap tandem mass spectrometry qualitative detection. The reference standards used were deapio platycodin D, platycodin D2, and platycoside D.

Chromatographic conditions

The chromatographic column and column temperatures were Hypersil GOLD C18 (100 mm × 2.1 mm, 3 μm) and 40 °C respectively. The mobile phase was composed of A (acetonitrile) and B (0.1% formic acid–water) using a gradient elution of 23% A at 0–18 min, 29% A at 18–22 min, 90% A at 22–23.5 min and 23% A at 23.5–28 min. The flow rate was set at 0.3 mL/min.

MS conditions

All MS experiments were performed in the negative ion modes. The source and ion transfer parameters applied were as follows: spray voltage 3.5 kV and the capillary voltage 35 V. The sheath gas, aux gas, atomization temperature, and the capillary temperature were maintained at 35 arb, 15 arb, 350 °C, and 300 °C, respectively. The tube lens level was set at 110 V and the resolution of FT was 6000.

Network pharmacology

Based on the qualitative analysis of PAE, the Canonical SMILES format of chemical components was downloaded from PubChem Database (https://pubchem.ncbi.nlm.nih.gov) and the targets of chemical components were selected from the SwissTargetPrediction database [19,20,21] (http://www.swisstargetprediction.ch/), with the species limited to “Homo sapiens” [22,23,24]. Furthermore, ALI-associated targets were acquired from Gene Cards Database (https://www.genecards.org/), OMIM Database (https://omim.org/,) and DisGeNET (https://www.disgenet.org/) which were searched using the keywords “Acute Lung Injury” [25, 26]. Venny 2.1 (https://bioinfogp.cnb.csic.es/tools/venny/) was used to take the intersection of PAE and ALI targets [27, 28], which were defined as potential preventive targets.

A “drug-ingredient-target-disease” regulatory network was established based on potential preventive targets by Cytoscape3.7.2 software. The protein–protein interaction (PPI) was further created based on potential preventive targets using String Database (https://string-db.org/). R version 3.6.0 and Bioconductor software packages were applied to analyze Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways. Finally, the top 20 related KEGG pathways were selected for further network analysis. An integrated “target-composition-pathway” network was constructed by Cytoscape 3.7.2.

Animals and LPS-induced ALI

C57BL/6 J male mice [license: SYXK (Su) 2018–0049] were obtained from Shandong Provincial Laboratory Animal Center (Jinan, Shandong Province, China). Before the experiment, all mice were acclimatised to the laboratory settings for three days. The mice were kept in separate cages and subjected to conventional circumstances such as 12 h light–dark cycles, 45% relative humidity and a constant temperature of 23 ± 2 °C. With a weight range of 18–22 g (8–10 weeks old), those mice were randomly divided into the following six groups (n = 6): Control, LPS, LPS + DXMS (10 mg/kg) and LPS + PAE (1.51, 3.775, 7.55 g/kg/d, respectively) groups. DXMS was used as a positive control. According to the Chinese Pharmacopoeia, the maximum clinical dose of Platycodon grandiflorum was 10 g, and the equivalent clinical dose for mice was 1.51 g/kg. The dose group set in this study was equivalent, 2.5 and 5 times the clinical dose, respectively. Studies have shown the non-toxicity of PAE in above dose [29, 30]. After 7 days, LPS, LPS + DXMS and LPS + PAE groups were induced by intratracheal administration of 1 mg/mL LPS (3 mg/kg). Firstly, mice were anesthetized with sodium pentobarbital (60 mg/kg) by intraperitoneal injection and placed in a supine position, and then LPS was injected at 3 mg/kg into the LPS, LPS + DXMS and LPS + PAE groups mice’s trachea [21]. Six hours later, all the mice were sacrificed. Biological samples and lung tissues were collected.

Histopathologic evaluation of the lung tissue

To assess histological alterations in the lung tissues, the tissues were fixed in 4% paraformaldehyde, then paraffin-embedded and stained with hematoxylin and eosin (H&E). A light microscope was used to examine pathological alterations in the lung tissues.

Reverse-transcription quantitative polymerase chain reaction (RT-qPCR)

To begin, total RNA was isolated from mouse lung tissues using Trizol reagent (Invitrogen, Carlsbad, CA, United States). Second, a one-step RT kit (Takara Biotechnology, Dalian, Liaoning Province, China) was used to reversely transcribe total RNA to first-strand cDNA for a 1 g sample. Finally, the reactions were carried out in 10-L quantities with GAPDH (Sangon Biotech, Shanghai, China) serving as an internal control. To achieve the mean value, experiments were done in triplicate for each sample. The relative expression of mRNA was determined using the 2−ΔΔCT method. Apoptotic cytokines including Bcl-2 and Bax as well as inflammatory cytokines such as IL-1β, IL-6 and TNF-α were measured. The primer sequences used in this study are listed in Table 1.

Table 1 Primer sequences RT-qPCRLipidomics analysis of lungSample preparation

Lung tissues (20 mg) were weighed into a 2 mL centrifuge tube with two ball mills and 200 μL ultrapure water for homogenization. Then lung tissues homogenate (20 µL) was added to a 1.5 mL centrifuge tube followed by the addition of 225 µL ice methanol solution with the internal standard [SM (17:0) and PE (17:0/17:0)]. The concentration of the internal standard is around 5 μg/mL. For 10 s, the mixture was vortexed. Seven hundred and fifty microliters of methyl tert-butyl ether (MTBE) were added, and the mixture was shaken for 10 min at 4 °C. The samples were vortexed for 10 s before being centrifuged for 2 min at 14000 rpm at 4 °C after being added 188 L of ultrapure water. In the organic phase, mainly lipids were transferred to fresh tubes and dried in a vacuum centrifuge. Finally, the residue was reconstituted with 110 µL methanol: toluene (9:1) for analysis.

Lipidomic analysis

A reversed-phase Waters Acquity UPLC CSH C18 (100 mm × 2.1 mm, 1.7 µm) column was used for chromatographic separation and maintained at 65 °C. The positive ion mode mobile phase was composed of acetonitrile/water (6:4, v/v) (eluent A) and isopropanol/acetonitrile (9:1, v/v) containing 0.1% formic acid and 10 mM ammonium formate (eluent B). The mobile phase of the negative ion mode was the same as that of the positive ion mode except that no formic acid was added. Both the positive and negative ion modes used a gradient elution of 15% B at 0 min, 30% B at 0–2 min, 48% B at 2–2.5 min, 82% B at 2.5–11 min, 99% B at 11–12 min and 15% B at 12–15 min. The flow rate was set at 0.6 mL/min.

Q Exactive Hybrid Quadrupole-Orbitrap Mass Spectrometer was used for both positive and negative ion modes. Parameters of mass spectrometry: spray voltage was 3.5 kV (positive) and 3.0 kV (negative); for both ionization modes, sheath gas, aux gas, capillary temperature, and heater temperature were maintained at 35 arb, 15 arb, 325 °C and 300 °C, respectively; scan range was m/z 215–1800.

Data processing

Raw spectra files of all samples (including QC) from both ion modes were converted to ABF format by using Reifycs ABF Converter (http://www.reifycs.com/AbfConverter/). For each ion mode, ABF files were imported into MS-DIAL software to match the primary and secondary fragmentation ions with the built-in Lipid Blast database. The results exported from MS-DIAL were manually checked. The unknown annotations, annotations without MS2 information and duplicated annotations were deleted. To reduce instrumental errors and retain biological errors, the data matrix was normalized by R language. Based on the smallest RSD of QCs, the data matrix was imported into SIMCA-P 13.0 for Partial Least Squares Discriminant Analysis (PLS-DA). At the same time, R language was used to conduct a non-parametric test (Kruskal–Wallis test), and the fold change value was calculated according to the median peak height of each annotation, with criteria set at P < 0.05 and fold change (FC) ≥ 1.2 or ≤ 0.8333 to screen differential metabolites [32, 34]. Further, the bubble diagram was used to cluster different lipids in total ion mode by ChemRICH (http://chemrich.fiehnlab.ucdavis.edu/) and cluster analysis was performed on the different lipids in positive and negative ion modes via using a heatmap. Lastly, GraphPad Prism 8.0 was used to observe the expression of different PIs in the three groups.

Western blotting

Proteins were isolated from lung tissues in each group, and the total protein content was determined by using a BCA protein assay kit (Thermo Scientific, Rockford, IL, USA). Then, using SDS-PAGE, equal amounts of proteins were separated, transferred to PVDF membranes (0.45 m, Millipore, Milford, MA, USA), and blotted overnight at 4 °C with primary antibodies against PI3K (1:1000), p-PI3K (1:1000), Akt (1:1000), p-Akt (1:2000), Bcl-2 (1:1000), Bax (1:5000), and β-actin (1:10,000), followed by secondary antibodies incubation. The protein bands were quantified using the Image Lab software (Bio-Rad Co, Hercules, CA, United States) and the ChemiDoc™ MP Imaging System (Bio-Rad Co, Hercules, CA, United States). At least three times, each experiment was carried out.

Immunohistochemical analysis

Cross-immunohistochemical staining in paraffin sections was employed to reveal the presence of phosphorylated phosphatidylinositol 3-kinase (p-PI3K), phosphorylated protein kinase B (p-Akt), B-cell lymphoma-2 (Bcl-2), and Bcl-2 associated X protein (Bax).

TdT-mediated dUTP nick-end labelling (TUNEL) staining

TUNEL labelling was performed to analyze the apoptosis rate of lung cells after embedding in paraffin and cutting into 40 µm sections, as is the standard approach [35]. Fluorescence staining was carried out according to the manufacturer's instructions using Roche Diagnostics GmbH's in-situ Cell Death Detection Kit (Mannheim, Germany). A light microscope was used to view the apoptotic cells, with the results represented as the average number of TUNEL-positive stained cells per 500 magnification field.

Enzyme-linked immunoassay of phosphatidylinositol triphosphate (PIP3)

Solid-phase antibody was prepared by coating microporous plates with purified mouse PIP3 antibody, and then PIP3 was added in rapid succession into the microporous plates coated with monoclonal antibody. The above one was then bound with a PIP3 antibody labelled by human haptoglobin-related proteins (HRP) to form an antibody-antigen-enzyme labelled antibody complex. After thoroughly washing the complex, the substrate tetramethylbenzidine (TMB) was added to it to develop its colour. TMB was first converted into blue under the catalysis of the HRP enzyme, and then into the final yellow under the action of acid. The intensity of the colour is positively correlated with the phosphatidylinositol triphosphate (PIP3) in the sample. So, the absorbance (OD value) was measured with a microplate reader at 450 nm. The concentration of mouse phosphatidylinositol triphosphate (PIP3) in the sample was calculated from the standard curve.