Abstract

Background

   Acute lung injury (ALI), a critical respiratory condition, often
   escalates into acute respiratory distress syndrome, which is associated
   with significant morbidity and mortality. Bauhinia championii, a
   botanical drug used in traditional Chinese medicine, is reputed for its
   antioxidative and anti-hypoxia effects. However, the active metabolites
   within B. championii and their mechanisms of action in alleviating ALI
   remain to be elucidated.

Methods

   A comprehensive literature review and database search within Chemistry
   Database were conducted to compile a complete profile of the
   metabolites identified in B. championii. Utilizing network analysis, we
   predicted potential targets of metabolites in B. championii (MBC) for
   ALI treatment. A protein-protein interaction (PPI) network was
   constructed using Cytoscape 3. 9. 1, complemented by GO annotations and
   KEGG pathway enrichment analyses via the DAVID online platform. The
   isolation and characterization of polymethoxyflavones (PMFs) from B.
   championii were performed using HPLC and confirmed by LC-MS. In vivo
   pharmacological assessments were executed to substantiate the network
   analysis predictions. Moreover, the Autodock software facilitated
   molecular docking studies to elucidate the role of endoplasmic
   reticulum (ER) stress modulation in ALI treatment by PMFs.

Results

   17 known MBC were identified in which 7 active metabolites of
   flavonoids were used as predictive targets. 122 target genes associated
   with both MBC and ALI were tested for KEGG and GO enrichment analyses,
   which indicated these target genes involvement in antioxidant,
   anti-inflammatory, and anti-apoptotic pathways. The PMFs were extracted
   from B. championii and identified as 5, 6, 7, 3′,
   4′-pentamethoxyflavone, 5, 6, 7, 3′, 4′, 5′-hexamethoxyflavone, 5, 7,
   3′, 4′, 5′-pentamethoxyflavone, 5, 6, 7, 5′-tetramethoxy-3′,
   4′-methylenedioxyflavone and 5, 7, 5′-trimethoxy-3′,
   4′-methylenedioxyflavone. PMFs were effective in alleviating
   LPS-induced pulmonary inflammatory responses for releasing ALI. In
   addition, PMFs inhibited the secretion of GSH-Px and CAT, reduced the
   accumulation of HYP and MDA as well as the infiltration of inflammatory
   cells, not to mention alleviated LPS-induced apoptosis by inhibiting
   the Caspase 3-mediated apoptosis pathway. Furthermore, the PMFs can
   spontaneously bind to multiple ER stress targets to exert the effect of
   calming ER stress to alleviate ALI.

Conclusion

   PMFs inhibited the expression of inflammatory cytokines and reduced
   oxidative stress injury to resist apoptosis in lung. Moreover, PMFs
   attenuated LPS-induced ER stress activation by regulating ER stress
   related targets, which in turn alleviated ALI.

   Keywords: polymethoxyflavones, Bauhinia championii, acute lung injury,
   endoplasmic reticulum stress, network analysis

1 Introduction

   Acute lung injury (ALI) manifests as an acute hypoxic respiratory
   insufficiency, precipitated by damage to alveolar epithelial cells and
   capillary endothelial cells from a spectrum of direct and indirect
   insults. This damage leads to widespread interstitial and alveolar
   edema. In its severe form, ALI can progress to acute respiratory
   distress syndrome (ARDS), with an incidence rate approaching 40% and
   associated with a mortality rate of 40%–50% ([42]Wang et al., 2020). In
   addition to increased mortality during and after hospital discharge,
   patients who survive ARDS are likely to have serious long-term sequelae
   ([43]Shaw et al., 2019). Currently, available therapies for ALI/ARDS
   are categorized into supportive therapies and pharmacologic
   interventions. The lung protection strategy of mechanical ventilation
   is presently considered the only supportive therapy that effectively
   improves survival ([44]Luh and Chiang, 2007). Physiologically-based
   pharmacotherapy is the use of medicine that affects ventilation, and
   diffusion or perfusion is used to alleviate the symptoms of ALI. In
   fact, there is no effective pharmacologic therapy for ALI/ARDS that
   substantially reduces mortality and improves patient quality of life
   ([45]Fan et al., 2018; [46]Ortiz-Diaz et al., 2013).

   It is well known that ALI/ARDS is an inflammatory disease of the lungs.
   The inflammatory response is widely considered an essential feature
   contributing to a variety of lung diseases ([47]Lee et al., 2021). The
   release of inflammatory signals causes lung injury characterized by
   inflammatory cell infiltration, which induces apoptosis of APCs
   involved in the early pathological process of ALI ([48]Martin et al.,
   2005). Severe pneumonia increases the production of pro-inflammatory
   cytokines such as IL-6, Il-1β, TNF-α and IFN-γ, which attract
   neutrophils and increase the production of reactive oxygen species
   (ROS) to exacerbate respiratory airway disease ([49]Takeda and Akira,
   2005). Lipopolysaccharide (LPS), a major component of the cell wall of
   Gram-negative bacteria, has been shown to induce ALI ([50]Yu et al.,
   2014). Upon entry into the biological organism, LPS contributes to lung
   injury primarily through the perception of innate immune cells and
   initiates the secretion of inflammatory mediators. This process
   destroys the alveolar epithelial and endothelial barriers to impair the
   normal respiratory function of the host ([51]Kolomaznik et al., 2017).
   Therefore, LPS-induced animal models of ALI are commonly used to
   explore the mechanisms of the lung inflammatory response.

   Endoplasmic reticulum (ER) stress is defined as the accumulation of
   unfolded or misfolded proteins in the ER and the subsequent triggering
   of the unfolded protein response (UPR). It is facilitated by three
   transmembrane ER signaling proteins: pancreatic endoplasmic reticulum
   kinase (PERK), inositol-requiring enzyme 1 (IRE1), and activating
   transcription factor 6 (ATF6) ([52]Hosoi and Ozawa, 2009). Accumulation
   of unfolded peptides diverts GRP78 away from these three “stress
   sensors” thus inducing changes in downstream signals such as XBP1,
   CHOP, and eIF2α, which leads to upregulation of various UPR-targeted
   genes to restore ER homeostasis ([53]Wang et al., 2012). Recent studies
   have indicated that ER stress is involved in LPS-stimulated airway
   epithelial cells in vitro and activation of the ATF4/eIF2a/CHOP
   signaling axis to promote cell death ([54]Kim et al., 2013). It is not
   coincidence that inhibition of NF-κB inflammatory signaling is
   effective in suppressing oxidative stress and ER stress to alleviate
   ALI ([55]Yang et al., 2020). However, since the role of ER stress in
   LPS-induced lung inflammation has not been fully elucidated, there is
   still no suitable agent targeting ER stress for the treatment of ALI.

   Traditional Chinese medicines are often remarkably effective in
   treating ALI/ARDS. Based on in vivo and in vitro results, a variety of
   natural metabolites with multiple anti-inflammatory activities and lung
   protective effects, such as flavonoids, alkaloids, and terpenoids, have
   been proposed for treating ALI ([56]He et al., 2021). Studies have
   shown that Bauhinia championii (Benth.) Benth. (Leguminosea), extracts
   have antitumor, antioxidant, and ER stress-alleviating effects
   ([57]Chen et al., 2020; [58]Liao et al., 2016). Our previous study also
   demonstrated that extracts of B. championii alleviates neuronal
   apoptosis by modulating ER stress ([59]Huang et al., 2022). A network
   analysis showed that flavonoids in B. championii have a potential
   anti-inflammatory function. In the present study, we extracted,
   isolated and characterized a class of polymethoxyflavones (PMFs) from
   B. championii. Our objectives were to explore the mitigating effects of
   PMFs on LPS-induced ALI and to explain the mechanism by which PMFs
   reduce apoptosis in lung cells by modulating endoplasmic reticulum (ER)
   stress in lung cells to reduce inflammatory responses and reactive
   oxygen species (ROS) accumulation.

2 Materials and methods

2.1 Drugs and reagents

   The stem of B. championii was collected from Fujian Province (on 6 May
   2021). It was identified and gifted by Professor Chen Jianzhong (Fujian
   University of Traditional Chinese Medicine). A voucher specimen was
   deposited in the Herbarium of Hangzhou Medical College.

   Silica gel (100–200 mesh) was purchased from Qingdao Ocean Chemical
   Co., Ltd. (China). ODS-A-HG (50 μm) and the YMC-Pack ODS-A-HG column
   (250 mm × 10 mm i. d., 10 μm) were purchased from YMC. Co., Ltd.
   (Japan). Agilent ZORBAX Extend-C18 column (250 mm × 4.6 mm i. d., 5 μm)
   was purchased from Agilent Technologies, Inc. (Agilent, United States).
   Methanol, and formic acid (all LC-MS grade) were purchased from Thermo
   Fisher (United States). CD[3]Cl (D, 99.8% + 0.03 v/v TMS) was purchased
   from Tenglong Weibo Technology (China). The reference standard of
   sinensetin (No. FY1653-B) was purchased from Nantong Jingwei Biological
   Technology Co., Ltd. (China).

   Malondialdehyde (MDA), glutathione peroxidase (GSH-Px), catalase (CAT)
   and hydroxyproline (HYP) detection kits were purchased from Nanjing
   Jiancheng Bioengineering Institute (China). The dihydroethidium (DHE)
   kit, TdT-mediated dUTP Nick-End Labeling (TUNEL) kit, BCA kit, ECL kit
   and Wright-Giemsa staining kit were obtained from Shanghai Biyuntian
   Biological Co., Ltd. (China). Primary antibodies targeting the
   following proteins were used: IKKα, IκB, p-IκB, NF-κB, p-NF-κB, PERK,
   p-PERK, eIF2α, p-eIF2α, JNK, and p-JNK (CST, United States), IRE1α,
   p-IRE1α, and GRP78 (Abcam, United States), ATF4, CHOP, TRAF2, and ATF6
   (Proteintech Group, United States), and GAPDH (Hangzhou Xianzhi,
   China). HRP-conjugated goat anti-rabbit IgG (H + L) was purchased from
   Proteintech. Other reagents were purchased from Thermo Scientific
   (United States).

2.2 Purification of PMFs

   Dried slices of B. championii were ground, and 1.5 kg of powder was
   extracted with 15 L MeOH under ultrasonication for 1 h at room
   temperature (repeated three times). The MeOH extract was concentrated
   under reduced pressure to yield a crude residue (248.8 g), which was
   suspended in water and successively partitioned with ethyl acetate and
   n-butyl alcohol. After concentration, the ethyl acetate fraction
   (93.6 g) was chromatographed over silica gel and eluted with gradient
   mixtures of petroleum ether/ethyl acetate (4:6–1:9) to yield 5
   fractions. Fr. Four was rechromatographed by an ODS C18 column and
   eluted with gradient mixtures of MeOH/H[2]O (1:9–6:4) to afford 4.5 g
   PMFs. Appropriate quantities of PMFs was separated by prep-HPLC to
   afford compounds PMF1-5 (4, 12, 15, 7 and 5 mg, respectively). The
   purity of PMF1-5 was detected by the peak area normalization method
   using high-performance liquid chromatography (HPLC), all of which were
   greater than 98%.

2.3 Identification of the main metabolites in PMFs

   Metabolites PMF 1–5 (2–5 mg) were dissolved in 0.6 mL CDCl[3] and
   transferred to NMR tubes. 1H-NMR spectra (600 MHz) were recorded on a
   Bruker AV-600 spectrometer. Chemical shifts were recorded in δ units
   (ppm) and coupling constants (J) in Hz. HR-ESI-MS experiments were
   carried out using an Agilent 1260 HPLC/6,530 Q-TOF mass spectrometer.
   An Agilent ZORBAX Extend-C18 column (250 mm × 4.6 mm, i. d. 5 μm) was
   used for chromatographic separation. The mobile phase consisted of 0.1%
   formic acid in H[2]O (A) and methanol (B). The elution program was as
   follows: 0–5 min, 10% B; 5–30 min, 10%–20% B; 30–55 min, 20%–35% B;
   55–70 min, 35%–55% B; 70–95 min, and 55%–65% B. The flow rate was
   1 mL/min and the injection volume was 5 μL. The column compartment was
   maintained at 30°C. The mass spectrometry (MS) spectra were recorded
   under electrospray ionization (ESI) positive mode. The MS parameters
   were as follows: scan range, m/z 100–1,400; nitrogen flow rate,
   8 L/min; drying gas temperature, 300°C; nebulizer pressure, 35 psi;
   capillary voltage, 3500 V. A collision energy of 30 eV was used in
   MS/MS.

2.4 HPLC analysis of PMFs

   The HPLC analysis was performed on a Waters Acquity Arc
   ultra-high-performance liquid chromatography (UHPLC) system equipped
   with a Waters 2,998 photodiode array (PDA) detector. An Agilent ZORBAX
   Extend-C18 column (250 mm × 4.6 mm i. d. 5 μm) was used for the
   chromatographic separation. The mobile phase consisted of 0.1% formic
   acid in H[2]O (A) and methanol (B). The elution program was as follows:
   0–2 min, 55% B; 2–17 min, 55%–70% B; 17–20 min, 70%–90% B; and
   20–22 min, 90% B. The flow rate was 1 mL/min and the injection volume
   was 5 μL. The ultraviolet (UV) detection wavelengths were set at 332 nm
   for PMF 1, 320 nm for PMF 2, 327 nm for PMF 3, 334 nm for PMF 4 and
   339 nm for PMF 5. The column compartment was maintained at 30°C.

   The PMFs (2.04 mg) were dissolved in methanol in a 1 mL volumetric
   flask. Precisely 0.5 mL of the solution was transferred into a 25 mL
   volumetric flask and filled to the mark with methanol. The solution was
   filtered through a 0.22 μm polytetrafluoroethylene (PTFE) filter for
   the HPLC analysis.

   The PMF 1–5 compounds were accurately weighed to be 1.62, 1.36, 1.57,
   1.28, and 1.92 mg and dissolved in methanol in a 1 mL volumetric flask
   to make the standard stock solutions. Appropriate volumes of the stock
   solutions of PMF 1–5 were carefully transferred into a 10 mL volumetric
   flask to obtain a mixture of standard stock solutions of PMF 1–5 of
   1.62, 27.2, 31.4, 12.8, and 1.92 μg/mL. Five calibration curves of PMF
   1–5 were prepared withed six appropriate dilutions of the mix stock
   solution. An aliquot of 5 μL of the sample and standard solution was
   used for the HPLC analyses.

2.5 Predicting potential targets for MBC

   The known metabolites in B. championii (MBC) were obtained by reading
   the literature and searching on the Chemistry Database
   ([60]https://organchem.csdb.cn), and the known metabolites were entered
   into the Pubchem ([61]https://pubchem.ncbi.nlm.nih.gov/) database to
   obtain their canonical simplified molecular-input line-entry system
   (SMILES). These were then sequentially imported into SwissADEM
   ([62]http://www.swissadme.ch/) and screened for active metabolites with
   scores for gastrointestinal (GI) absorption of “high” and drug likeness
   with at least two “Yes”. Subsequently, the canonical SMILES of the
   screened metabolites were entered into the SwissTargetPrediction
   ([63]http://www.swisstargetprediction.ch/) to predict the targets, and
   a “probability>0” was used as the filtering condition to obtain the
   potential targets of the Nine Dragons Vine.

2.6 Predicting potential targets for ALI

   The OMIM ([64]https://www.omim.org/), GeneCards
   ([65]https://www.genecards.org/) and DisGeNET
   ([66]https://www.disgenet.org/) databases were searched using the
   keyword “acute lung injury”. To ensure the reliability of the targets,
   the targets obtained from the three databases were exported after
   removing the duplicate genes to create a database of targets related to
   acute lung injury.

2.7 Identifying core targets and the protein-protein interaction (PPI)
network construction

   The targets of acute lung injury and the targets of MBC were compared
   using an online tool ([67]http://www.liuxiaoyuyuan.cn/) to obtain the
   intersection targets of MBC and acute lung injury. We set the species
   as “Homo sapiens” and uploaded the intersecting targets into the STRING
   ([68]https://string-db.org/) database for the PPI network construction.
   Cytoscape 3.9.1 was utilized for mapping.

2.8 GO and KEGG enrichment analysis

   “H. sapiens” was set as the species, a Geno Ontology (GO) and Kyoto
   Encyclopedia of Genes and Genomes (KEGG) enrichment analysis were
   performed on the intersecting targets using the Database for
   Annotation, Visualization, and Integrated Discovery (DAVID)
   ([69]https://david.ncifcrf.gov/), and the top 20 enrichment results
   were plotted as visualized bubble plots.

2.9 Molecular docking

   The relevant targets of the endoplasmic reticulum stress pathway were
   obtained using the GeneCards ([70]https://www.genecards.org/) database.
   The intersection was found using the targets of five PMFs, the targets
   of the endoplasmic reticulum stress pathway, and the targets of acute
   lung injury to obtain the key targets of PMFs for the treatment of
   acute lung injury through endoplasmic reticulum stress. The key targets
   were entered into the Protein Data Bank (PDB)
   ([71]https://www.rcsb.org/) database to obtain their structures and
   molecularly docked with PMFs using Autodock to observe the binding
   energies of each group.

2.10 Animals and treatments

   Seventy-two specific pathogen-free (SPF) male C57BL/6 mice, weighing
   16–20 g, were purchased from the Zhejiang Provincial Laboratory Animal
   Center. The mice were housed in a temperature-controlled environment
   (22°C ± 2°C) under humidity-controlled conditions (55% ± 5%) with a
   12 h light/dark cycle and free access to water and food. Mice were
   housed in cages (six mice per cage) and all mice were acclimatized to
   the living environment for 3°days before the experiment. All animal
   experiments in this study were approved by the Institutional Animal
   Care and Use Committee. Institutional Animal Care guidelines issued by
   the Chinese Ministry of Science and Technology were followed. All
   animal experiments were performed according to the protocol approved by
   the Experimental Animal Ethics Committee (2021-162) of Hangzhou Medical
   College.

   Mice were randomly divided into six groups (12 mice per group): (i)
   mice in the control group received 50 μL of 0.9% saline intranasally to
   mimic nasal administration, and gavaged with 0.5% CMC-Na to simulate
   oral administration; (ii) mice in the LPS group were given 20 μg of LPS
   dissolved in 50 μL saline intranasally on the fifth day, and simulated
   gavaged with 0.5% CMC-Na on day 1, day 3, and 4 h after LPS treatment;
   (iii) mice in the LPS + Dex group received LPS and 5 mg/kg of
   dexamethasone was given intraperitoneally 4 h after LPS administration
   as well as 1 day before LPS treatment; (iv) mice in the low-dose PMF
   group [LPS + PMFs(L)] were given PMFs at a dose of 100 mg/kg by gavage
   on day 1, day 3, and 4 h after LPS treatment; (v) mice in the
   medium-dose PMF group [LPS + PMFs(M)] were given PMFs at a dose of
   300 mg/kg by gavage on day 1, day 3, and 4 h after LPS treatment; and
   (vi) mice in the high-dose PMF group [LPS + PMFs(H)] were given PMFs at
   a dose of 900 mg/kg by gavage on day 1, day 3, and 4 h after LPS
   treatment. The gavage volume for all animals was 0.2 mL per 10 g body
   weight. At 24 h after LPS administration, all mice were weighed,
   anesthetized, and executed. Lung wet and dry weight were recorded from
   six animals in each group; the lungs from three animals in each group
   were used for histological examination with alveolar lavage; the lung
   tissues of the remaining three mice from each group were used for
   Western blot and PCR assays. The detailed experimental schedule is
   shown in [72]Figure 3A.

FIGURE 3.

   [73]FIGURE 3
   [74]Open in a new tab

   Effects of PMFs in LPS-induced acute lung injury. (A) Experimental
   schedule for study design. (B) Body weight (n = 12). (C) Lung
   weight/body weight ratio (n = 6). (D) Lung wet-dry weight ratio (W/D)
   (n = 6). Scale bar = 200 μm. (E) Alveolitis score for evaluation of
   lung lesion. (F) H&E staining to detect histological changes in the
   lung tissues of mice. Data are presented as mean ± SD, *p < 0.05, **p <
   0.01, ***p < 0.001, ****p < 0.0001.

2.11 Collection of BALF and production of lung tissue slices

   At 24 h after LPS treatment, mice in each group were executed by
   cervical dislocation. Then the thoracic cavity of the sacrificed mice
   was clipped, and when the trachea was exposed, a T-shaped incision was
   cut in the trachea, and the line trachea was inserted. In addition, the
   left lung was ligated at the left main bronchus. Finally, 1.5 mL of
   physiological saline was used to irrigate the right lung three times.
   The recovered fluid (first >80%) was used as bronchoalveolar lavage
   fluid (BALF).

   The left lung was cut and collected after lavage and fixed in 10%
   formalin overnight. The next day, the fixed lungs were rinsed with
   water and dehydrated with serially diluted alcohol. Paraffin-embedded
   samples were sectioned at a thickness of 4–5 μm using sledge microtome.

2.12 BALF analysis

   BALF (50 μL) was observed under a microscope and 500 μL BALF was used
   for antioxidant index detection. The remaining lavage fluid was
   centrifuged at 1,200 rpm for 5 min, and the supernatant was discarded.
   The cell precipitates were stained with Wright-Giemsa (Item No. C0131)
   staining solution and applied to the cell counting plates. Total cells,
   macrophages and neutrophils were counted under an optical microscope.

2.13 Antioxidant indices

   The collected BALF was centrifuged at 1,200 rpm. The supernatants were
   used for the GSH-Px, CAT, MDA, and HYP assays after quantification of
   the total protein by a BCA kit (Item No. P0009). Samples treated with a
   GSH-Px kit (Item No. A005-1), CAT kit (Item No. A007-1), MDA kit (Item
   No. A003-1), and HYP kit (Item No. A030-2) were analyzed using a
   microplate reader at 412, 240, 532, and 550 nm, respectively. The
   levels of GSH-Px, CAT, MDA, and HYP in each sample were also calculated
   according to the colorimetric method following the manufacturer’s
   instructions.

2.14 H&E staining and immunohistochemistry

   Lung slices were subjected to xylene deparaffinization and hydration
   with graded concentrations of ethanol. Slices were stained with
   hematoxylin, differentiated with 1% ethanol hydrochloride, and stained
   with 1% weak aqueous ammonia and eosin. Histopathology was observed
   with a microscope (Leica DM4000, Germany) by a pathologist. The degree
   of lung injury was demonstrated by the alveolar score as previous
   described ([75]Wei et al., 2022). For immunohistochemistry, slices were
   also subjected to immunohistochemistry analysis. Endogenous peroxidase
   activity was quenched with 3% hydrogen peroxide. After washing with
   PBS, sections were blocked with 1% bovine serum albumin for 1 h at room
   temperature and then incubated with anti-GRP78 overnight at 4°C. The
   next day, sections were washed with PBS, followed by incubation with
   HRP-conjugated secondary antibody for 1 h. Next, tissue sections were
   incubated with diaminobenzidine, counterstained with hematoxylin,
   dehydrated, and mounted at room temperature. Sections were observed and
   imaged using an optical microscope. Integrated optical density (IOD)
   values were calculated using ImageJ (version 1.51).

2.15 TUNEL staining

   A TUNEL kit (Item No. C1089) was used to detect apoptosis in lung
   sections following the manufacturer’s instructions. Briefly, paraffin
   sections were defatted and gradually rehydrated in xylene and a series
   of concentrations of ethanol. Then slices were incubated with
   proteinase K at 37°C for 30 min, followed by incubation with TUNEL
   reaction solution at 37°C for 1 h in the dark. Finally, samples were
   incubated with DAPI (Invitrogen, United States) for 5 min at room
   temperature to visualize the nuclei. Images were acquired using a
   fluorescence microscope (Olympus X51, Japan). TUNEL-positive cells were
   counted using ImageJ.

2.16 DHE staining

   DHE staining was performed to measure ROS generation in lung tissues.
   Paraffin sections with a thickness of 5 μm were incubated with DHE
   solution (Item No. S0063) for 30 min at 37°C in the dark. Images were
   acquired using a fluorescence microscope and the fluorescence intensity
   of positive cells (red staining) was quantified using ImageJ.

2.17 Quantitative real-time PCR (RT-qPCR)

   Lung tissues were cut into pieces, and total RNA was extracted using
   TRIzol^® reagent. Prime Script RT Master Mix (Takara, Japan) was used
   to reverse transcribe RNA and TB Green^® Premix Ex Taq™ (Takara, Japan)
   was used for the quantitative real-time PCR. Fusion curve analysis
   confirmed the specificity of PCR products. Relative quantification of
   gene expression was performed using the 2^−ΔΔCt method. Gene expression
   was normalized to β-actin. All primers (IL-6: forward,
   5′-TCC​TAC​CCC​AAT​TTC​CAA​TGC​T-3′, reverse,
   3′-TGG​TCT​TGG​TCC​TTA​GCC​AC-5′; IL-1β: forward,
   5′-TGC​CAC​CTT​TTG​ACA​GTG​ATG-3′, reverse,
   3′-ATG​TGC​TGC​TGC​GAG​ATT​TG-5′; TNF-α: forward,
   5′-CCA​CGT​CGT​AGC​AAA​CCA​CC-3′, reverse,
   3′-CCC​TTG​AAG​AGA​ACC​TGG​GAG-5′; IFN-γ: forward,
   5′-GAG​GTC​AAC​AAC​CCA​CAG​GT-3′, reverse,
   3′-GGG​ACA​ATC​TCT​TCC​CCA​CC-5′; β-actin: forward,
   5′-AGG​GAA​ATC​GTG​CGT​GAC​AT-3′, reverse,
   3′-GGA​AAA​GAG​CCT​CAG​GGC​AT-5′) were designed using GenBank
   ([76]https://www.ncbi.nlm.nih.gov/genbank/).

2.18 Western blot analysis

   Lung tissue was homogenized in RIPA buffer and then centrifuged at
   12,000 × g for 15 min at 4°C. The pellet was discarded and the total
   protein concentration in the supernatant was determined using a BCA
   kit. Proteins were separated by SDS-PAGE and transferred to
   polyvinylidene difluoride membranes, which were blocked in 5% skimmed
   milk for 30 min and incubated with primary antibodies at 4°C overnight.
   The immunoreactive protein bands were detected using a Bio-Rad ChemiDoc
   MP system (Bio-Rad, United States) after dropwise addition of ECL
   developer.

2.19 Statistical analysis

   Statistical analysis was performed using GraphPad Prism 8. Values are
   expressed as mean ± standard deviation (SD), and P values less than
   0.05 were considered significant. Two-tailed Student’s t-test and
   one-way analysis of variance (ANOVA) followed by Dunnett’s post hoc
   test were used to compare between two groups and among three or more
   groups, respectively.

3 Results

3.1 Network analysis of MBC for treating ALI

   To predict the active ingredients in B. championii and their molecular
   targets in the treatment of ALI, we conducted a network analysis using
   in silico simulations. Based on literature and Chemistry Database
   search, a total of 17 metabolites were identified in B. championii,
   among which 7 flavonoids and 1 organic acid were screened by SwissADEM
   as potential active metabolites ([77]Table 1). The 7 flavonoids,
   including 5, 6, 7, 3′,4′-pentamethoxyflavone, 5, 6, 7, 3′, 4′,
   5′-hexamethoxyflavone, 5, 7, 3′, 4′, 5′-pentamethoxyflavone, 5, 6, 7,
   5′-tetramethoxy-3′, 4′-methylenedioxyflavone, 5, 7, 5′-trimethoxy-3′,
   4′-methylenedioxyflavone, 5, 7, 4′-trimethoxyflavone and 5, 7, 3′,
   4′-telramethoxyflavone, considered as active metabolites predicted a
   total of 154 targets ([78]Figure 1A). A total of 4,776 targets related
   to ALI were screened, of which 122 targets were also present in the
   MBC’s targets ([79]Figure 1B), and these 122 targets were considered as
   potential targets of MBC for the treatment of ALI. The STRING platform
   was utilized to construct the PPI network of potential targets, and
   based on the screening of network topological properties, the top 20
   nodes in the network in terms of degree were found to include AKT1,
   BCL2L1, and PIK3R1 ([80]Figure 1C). In addition, the results of the GO
   and KEGG enrichments showed that MBC could treat ALI through
   anti-inflammatory, antioxidant and anti-apoptotic approaches
   ([81]Figures 1D,E).

TABLE 1.

   The chemical names, Molecular formulas and SMILES of 8 potential active
   metabolites found in Bauhinia championii.
   Chemical name Molecular formula SMlLES References