Introduction

The main characteristic of acute respiratory distress syndrome (ARDS) is that the capillary endothelial cells of alveolar wall are damaged by various physical, chemical and biological factors, and a large amount of intertissue fluid accumulates in the lung1,2,3. Studies have shown that only 20–30% of the total number of alveoli with normal ventilation function in ARDS4,5, so the maintenance of pulmonary microvascular endothelial cell function is an important intervention for the development of ARDS6,7. The results of the prediction disease potentially effective drugs section of the Traditional Chinese Medicine Systems Pharmacology (https://old.tcmsp-e.com/tcmsp.php, TCMSP) database suggested that Acetyl-11-keto-β-boswellic acid (AKBA) may be an effective drug to restrict the progression of ARDS.

Frankincense (Boswellia Species, the resin extract of the genus Boswellia, has been used as a traditional medicine for hundreds of years to treat various diseases8,9, especially inflammatory diseases, including asthma, arthritis, but also cancer and a number of other conditions10,11,12. Boswellic acid is the active ingredient of frankincense, among which AKBA (3-O-acetyl-11-keto-β-boswellic acid) is the most important and effective acid13,14. Some studies have shown that the use of AKBA can also protect the function of brain microvascular endothelial cells and improve brain edema in animals and humans15,16,17. It seems that AKBA may be used as an alternative natural medicine not only to treat inflammatory diseases, but also to intervene in edema by protecting microvascular endothelial cell function18,19,20. A study aimed at revealing the identification of target bacteria using a high-throughput screening system indicated that AKBA could attenuate sepsis caused by Staphylococcus aureus in mice21. At present, although the inhibition of AKBA on inflammation and other diseases has been partially revealed, the role and mechanism of AKBA in ARDS is unknown.

Bone marrow stromal cell antigen 2 (BST-2, also known as tetherin, CD317 or HM1.24) has recently been identified as a host restriction factor against a variety of pathogens22,23. Studies have shown that Bone marrow stromal cells antigen 2 (BST2) on lipid rafts enrichment, through its N-terminal TM domain and C-terminal glycosyl phosphatidylinositol anchoring protein (Glycosylphosphatidylinositol, GPI) restrict pathogens to the surface of the cell membrane and inhibit their entry into cells24. Rodrigo I Santos et al. also noted an antibody, BDBV223, that induces Bundibugyo virus (BDBV) to accumulate on the plasma membrane, and its inhibitory activity depends on BST225. Some studies have also confirmed that the GPI of BST2 exists in lipid rafts, which can form obvious physical links between viral particles and connect viral particle blocks to the plasma membrane26,27,28. However, the involvement of BST2 in the progression of ARDS alleviated by AKBA-anchored HPMECs remains poorly understood.

Signal transducer and activator of transcription 1 (STAT1) is a typical signal transduction protein, which belongs to one of the members of STAT family and is involved in signal transduction process29,30,31. It has been suggested that activation of STAT1 signaling pathway can regulate Lipopolysaccharide (LPS)-induced M1/M2 polarization and reduce inflammation by maintaining M2 polarization32. Unphosphorylated STAT1 (U-STAT1) and tyrosine phosphorylated STAT1 (YP-STAT1) are two forms of STAT1. Structurally, P-STAT1 is mainly parallel, and U-STAT1 is mainly anti-parallel. Both of them have the function of regulating gene expression or heterochromatin formation in the nucleus33. It has been shown that a large increase in the concentration of U-STAT1 rather than a rapid increase in the concentration of YP-STAT1 drives the expression of regulatory genes (OAS, IFI27 and BST2)34,35. Sensing a delayed sustained increase in expression after a stimulus, U-STAT1 acts as a transcription factor, is present in the nucleus, and maintains or increases the expression of a subset of genes34,36. At present, our transcriptomic and proteomic results suggest that AKBA processes increased STAT1 expression in LPS-induced HPMECs, but the specific role and mechanism remain incomplete.

Our study shows that AKBA can significantly reduce the inflammation and apoptosis of cecum ligation and puncture (CLP) induced lung tissue in mice, inhibit Lipopolysaccharide (LPS) caused the autophagy of HPMECs and protect cell proliferation migration and tube formation. BST2, may as part of the AKBA carrier, has been able to effectively curb the progression of ARDS, while STAT1can effectively promote BST2 expression as a transcription factor. Our research assumed that AKBA may be a valid treatment for the intervention ARDS to have certain potential.

Materials and methods

Ethics approval and consent to participate

After review by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University, the research protocols and informed consent forms submitted in this project conform to the principles of medical ethics and the requirements of the Declaration of Helsinki. All methods were performed in accordance with the Declaration of Helsinki. The research design is scientifically based, does not bring unnecessary danger to the subjects, and protects the safety and privacy of the subjects to the maximum extent. Approve the implementation of the project according to the scheduled plan. The ethics number is (2023 − 348). The animal experiment scheme involved in this project has been reviewed by the Experimental Animal Management and Use Committee of Chongqing Medical University (IACUC-CQMU), which is in line with the principles of animal protection, animal welfare and ethics, and in line with the relevant provisions of the national experimental animal welfare ethics, and it is agreed to conduct experiments according to this scheme. Pentobarbital sodium (150–200 mg/kg) was administered via slow intraperitoneal (IP) bolus injection for humane euthanasia of mice. The ethics number is (IACUC-CQMU-2023-0203).

AKBA

AKBA (CAS No. 67416-61-9, HY-N0892) was sourced from MedChemexpress (MCE, New Jersey, USA) with a purity of 99.93%. Liquid Chromatography Mass Spectrometry (LCMS) and Nuclear Magnetic Resonance Spectroscopy (HNMR) data for AKBA are available in the supplementary materials. HPMECs were plated in 96-well plates at 3 × 10⁴ cells per well in 100 µL of culture medium, reaching approximately 80% confluence. AKBA was dissolved in DMEM containing 2% FBS to obtain final concentrations of 0 µM, 2.5 µM, 5.0 µM, 10.0 µM, 15.0 µM, 20.0 µM, and 25.0 µM. Control wells contained 100 µL of DMEM with 2% FBS but no AKBA. Each concentration was tested in triplicate. After 24 h, the medium was removed, and 100 µL of diluted AKBA was added to each well. The cells were incubated in a CO₂ incubator at 37 °C with 5% CO₂ for 48 h. After incubation, 100 µL of CellTiter-Glo® reagent was added to each well, and the reaction was carried out at room temperature for 5 min. Fluorescence was measured using a multi-mode microplate reader with excitation at 560 nm. The half-maximal effective concentration (EC50) values were determined from the fluorescence measurements.

Cell culture

Passaged human pulmonary microvascular endothelial cells (HPMECs) from ScienCell (San Diego, CA, USA) and incubated at 37 °C in 5% CO2with Dulbecco’s Modified Eagle’s Medium (DMEM; Gibco, Carlsbad, CA, USA) containing with 10% fetal bovine serum (FBS, Gibco, Paisley, UK) and 1% penicillin/streptomycin (Beyotime Shanghai, China). LPS (Escherichia coli O111:B4) (Sigma-Aldrich, Saint Louis, USA) concentration was configured to 2.5 mg/mL with phosphate-buffered saline (PBS, Pricella Biotechnology, Wuhan, China), and a final 10 µg/mL was used to treat HPMECs based on our previous study37,38,39. AKBA (10 mM, MCE, Shanghai, China) was dissolved in DMSO at 10 mM/mL, in which the dose of DMSO was less than 0.05%. Finally, 9.317 µM AKBA (EC50 9.317, 95%CI 8.814–9.829) was used for follow-up study. Cells were transfected with shRNA and plasmids (TsingKe, Beijing, China) using polyethylenimine (Yeasen, Shanghai, China) were screen by 3 µg/mL puromycin (HY-B1743 A; MCE). Sequences applicative were available in Table S1 of supplementary material. In addition, the BST2 recombinant protein HY-P75469 was purchased from MedChemExpress (New Jersey, USA) and used according to the instructions.

Tube formation assay

Pre-cooled overnight Matrigel (Beyotime Shanghai, China), 96-well plates and pipette tips were placed in an ice box in advance, and the plates were then removed and 10 µL Matrigel was added to each well. The well plates were placed in the prepared wet box and placed in the incubator for 30 min while a 2 × 105 cell suspension was prepared. After Matrigel solidified, 50 µL of cell suspension/well was added into. Finally, the images were collected according to the growth rate of cells, and the tube length, coverage area, ring number and node number were measured and recorded, and the statistical analysis was carried out.

Colony formation assay and the 5-ethynyl-2-deoxyuridine (EdU) assay

HPMECs in the logarithmic growth phase were digested and centrifuged, and the sediment was diluted to 1 × 103 cells/mL. Cells were seeded at 200 cells per well in 6-well plates and cultured in an incubator for 2–3 weeks. When macroscopic clones appeared, the culture was stopped, the medium was discarded, and it was carefully washed 2–3 times with PBS. 1 mL of 4% paraformaldehyde was added to each well and fixed for 15 min. 1 mL of 0.1% crystal violet or Giemsa staining solution was added to each well and stained for 10–30 min. Slowly wash off the stain with tap water and let dry. The number of clones containing more than 50 cells in the microscope was counted and photographed, and the average number of clones in the three complex wells was calculated.

Cells in logarithmic growth phase were seeded in 96-well plates, prepared and incubated with EdU medium for 2 h. Cells were washed 2–3 times with PBS, incubated with pre-cooled 4% paraformaldehyde for 30 min at room temperature, and incubated with glycine for 5 min. Add appropriate amount of PBS and wash for 5 min. 0.5%TritonX-100 was added and incubated for 10 min, followed by Apollo staining and DNA staining according to the manufacturer’s instructions. Finally, image acquisition and analysis.

Cell counting kit-8 (CCK-8)

The treated cells were digested with trypsin and counted and seeded into 96-well plates with 100 µL of medium per well at a cell density of 3000 to 7000 cells/well. Cultures were incubated at 37 °C, 5% CO2, and 90% humidity for 24 h ‌. Wells in good growth condition were taken, sample solutions of different concentrations were added, three compound Wells per concentration, and the culture was continued under the same conditions for appropriate times. 10 µL of CCK-8 solution was added to each well and the incubation continued for 0.5–4 h ‌ at 37 °C, 5% CO2, and 90% humidity. The absorbance value of each well was determined using a microplate reader at 450 nm wavelength and the data were processed and analyzed by Excel and Graphpad Prism.

Wound healing assay and transwell migration assay

When the cell density reached 80–90%, a 10 µL pipette tip was used to draw a scratch of appropriate width in a pre-labeled six-well plate at an angle of 65 degrees. HPMECs were rinsed 2–3 times with PBS and images were taken immediately and after 24 h of incubation. Statistical analysis was performed to calculate scratch confluence. 1.5 × 104 HPMECs were incubated with 200 µL of pure medium in transwell inserts (pore size: 8 µM; Corning, NY, USA), the bottom was filled with 500 µL of medium containing 10% FBS. After 24 h, the chambers were removed and fixed with 4% paraformaldehyde for 30 min, stained with crystal violet for 15 min, and then dried before being photographed and analyzed under a microscope, with three replicates per group.

Transmission electron microscopy (TEM)

The medium was discarded, 2.5% glutaraldehyde fixative solution was added, fixed at room temperature for about 5 min, and the cells were gently scraped down in one direction with cell scraping (or flat small squares cut with soft rubber cover). The cell liquid was sucked into the centrifuge tube with a pasteurized straw, put into a centrifuge (no more than 3000 rpm), and centrifuged for about 2 min to obtain the mungbean sized cell mass. After dehydration with gradient ethanol, the cells were immersed in the transmission electron microscope with appropriate acetone. By adjusting the focus and contrast of the electron beam, high-resolution images of the internal structure of cells were obtained, and the number and morphology of autophagosomes in each group were evaluated.

Patient specimen acquisition

This study enrolled 15 patients with septic shock complicated by acute lung injury and 15 healthy controls, all of whom were admitted to the Department of Intensive Care Medicine at the First Affiliated Hospital of Chongqing Medical University between January and December 2023. Informed consent was obtained from all participants. The study protocol was approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University. Based on our previous studies40, peripheral blood samples were collected from the participants, transported under conditions of 4 °C, and then centrifuged at 4 °C at 3000 rpm for 10 min to obtain the supernatant for further analysis.

Animals

C57BL/6 J male mice aged 6–8 week (purchased from the Animal Experiment Center of Chongqing Medical University) were assigned based on six mice per group. 40 µg of plasmid or shRNA (TsingKe, Beijing, China) was diluted into 50 µL of 10% glucose, sterile water was added to 100 µL, gently vortexed and rotated downward. 6.4 µL of in vivo jetPEI® (Polyplus, Strasbourg, France) was diluted into 50 µL of 10% glucose, sterile water was added to 100 µL, gently vortexed as above and rotated downward. Diluted in vivo jetPEI® was immediately added to the diluted plasmid or shRNA, vortexed briefly and rotated downward. The mice were incubated for 15 min at room temperature and injected into the tail vein with the obtained compound 2–3 times at a week. Subsequently, AKBA was intraperitoneally injected at a dose of low dose (5 mg/kg), medium dose (10 mg/kg), high dose (15 mg/kg). CLP was performed 24 h after the last injection and sacrificed 48 h later. All mice were injected with 1% pentobarbital deep anesthesia with high concentration CO2, and then lung tissues were acquired for further tests. All procedures were approved by the Ethics Committee of the First Affiliated Hospital of Chongqing Medical University. We adhere to ARRIVE guidelines describing studies involving live animals and refer to the American Veterinary Medical Association (AVMA) for the use of CO2 to sacrifice all laboratory animals.

Immunohistochemical (IHC) and Immunofluorescence (IF)

For Immunohistochemical (IHC) staining, mouse lung tissue samples were deparaffinized, rehydrated, and antigen recovered. Primary antibody (Bad, 1:100 diluted) was used to stain the sections using a DAB substrate kit (Zhongshan Golden Bridge Biotechnology, Beijing, China) according to the manufacturer’s instructions. The sham group was used for staining intensity normalization.

The slides were washed with PBS, fixed with 4% paraformaldehyde, and permeabilized with 0.5%Triton X-100 (prepared in PBS) for 20 min at room temperature. Blocked by FBS and incubated with primary antibodies (CD31, BST2, STAT1, LC3, 1:250 dilution) at 4 °C overnight. Then, the secondary antibody and DAPI were incubated in the dark, and the plates were sealed with the sealing wave containing anti-fluorescence quencher, and the images were observed and collected under a fluorescence microscope. Lung tissues was fixed with poly-formaldehyde, be transparent after dehydration with alcohol, soaked in paraffin wax in three tanks (60℃) in turn, sectioned and baked according to the instructions. Then sections were dewaxed, followed by antigen repairing and blocking, incubation with primary antibody, staining with secondary antibody (Beyotime Shanghai, China) and DAPI in the dark, and sealing with anti-fluorescence quencher. Finally, the images were observed and collected under a fluorescence microscope.

Hematoxylin-eosin (HE) staining

The lung tissue sections were placed in the prepared fixative solution, dehydrated and transparent, and then embedded in wax for sectioning. Finally, Suzuki scores were used to assess the impairment of lung tissue in line with HE results. The specific operation steps are shown in our preliminary study37.

Wet-dry weight ratio and protein detection in alveolar lavage fluid

Lung tissues were excised, gently blotted to remove surface fluid, and immediately weighed to obtain wet weight (WW). Samples were then dehydrated at 60 °C for 72 h until constant dry weight (DW) was achieved, and the wet-to-dry ratio (WW/DW) was calculated to assess pulmonary edema. For bronchoalveolar lavage fluid (BALF) collection, lungs were lavaged three times with sterile PBS (0.8 mL per wash), and the pooled fluid was centrifuged (1500 ×g, 10 min, 4 °C) to remove cellular debris. Total protein content in the supernatant was quantified using a bicinchoninic acid (BCA) assay, with absorbance measured at 562 nm via spectrophotometry against a bovine serum albumin (BSA) standard curve. All measurements were performed in triplicate, and data were normalized to total BALF volume.

Transcriptomics and proteomics

Total RNA and protein were extracted from LPS + DMSO group (C1, C2, C3) and LPS + AKBA group (K1, K2, K3) accordaing to the protocols of the manufacturer. The concentration and purity of RNA and protein were determined and high-quality RNA and protein samples were screened. The library was constructed on the basis of accurate quantification of the effective concentration (the effective concentration of the library was greater than 2 nM) and the quality of the library was guaranteed. Bioinformatics method was used to analyze the difference and enrichment of the results. In addition, multiple hypothesis corrections have been made to all transcriptomic and proteomic data, especially false discovery rate (FDR) corrections, to minimize the risk of false positives.

RNA extraction and reverse transcription-quantitative (RT-qPCR)

Total RNA from cells and tissues was extracted using TRIzol® reagent (Takara, Kusatsu, Japan) and the RNA extraction kit (TIANGEN, Beijing, China), following retroactive protocols. The extracted RNA was used to synthesize cDNA samples (MCE, Shanghai, China) using reverse transcription kit, according to the manufacturer’s instructions. The expression of target genes was detected by Real-time polymerase chain reaction (RT-qPCR), and β-catenin was used as the reference gene for normalization. The primer sequences were provided in Table S2 of supplementary material.

Western blot

Total proteins were extracted from mouse lung tissues and cells, and western blots were performed according to the manufacturer’s instructions. Primary antibodies BST2 and Bad (1:5000, dilution) from Proteintech (Wuhan, Hubei, China), U-STAT1 and YP-STAT1 (1:3000, dilution) from ABcam (Shanghai, China), β-actin (1:5000, dilution) from ZEN-BIOSCIENCE (Chengdu, Sichuan, China) was operated as described as our previous research41.

Cell thermal shift assay (CETSA) and drug affinity responsive target stability (DARTS)-WB

After cellular AKBA treatment, the cells are lysed, and the total protein extraction process is the same as before. After centrifugation, the supernatant lysate is subjected to gradient heating (40, 44, 48, 52, 56, 60, 64 °C). The protein denatured by heating is removed by centrifuging at 20,000 g at 4 degrees Celsius for 20 min. The target protein in the supernatant is detected and quantified by Western blot, and the CETSA dissolution curve (CETSA meltcurve) is plotted using GraphPad. HPMECs were cultured to 80–90% confluence, treated with AKBA or vehicle control (0.05% DMSO) for 24 h, and lysed in NP-40 buffer. The cell lysate was incubated with 20ug/ml of proteinase from Beyotime (Shanghai, China) at different concentrations of 0, 1:100, 1:300, 1:1000, 1:3000, 1:10000 at room temperature for 30 min to evaluate the stability of the target protein. Proteolysis was halted by adding protease inhibitors, followed by SDS-PAGE and Western blotting to detect protein degradation patterns. Band intensities were quantified using ImageJ, comparing AKBA-treated and DMSO-treated groups to evaluate drug-induced stabilization of target proteins.

Dual-luciferase reporter assay

HPMECs (80% confluence) were co-transfected with BST-wild-type (WT; TTTCTGGGAAA, 50 ng) or BST2-mutant (MUT; CCGACTTAGGC, 50 ng) and OE-STAT1 (1.5 µg, vector pcDNA3.1, NM_007315.4) or OE-NC (1.5 µg, vector pcDNA3.1) using polyethylenimine (Yeasen, Shanghai, China), following the specifical method. After 48 h of incubation, the luciferase activity was assessed using the Dual Luciferase Reporter Gene Assay Kit (Yeasen, Shanghai, China). The results were presented by the Multiskan SkyHigh microplate reader (Thermo, Massachusetts, USA) with Renilla luciferase activity as the internal reference. All plasmids were obtained from Beijing Tsingke Biotech Co., Ltd.

Molecular docking and molecular dynamics simulation

AKBA and presumption of BST2 combination model based on hierarchical FFT HDOC global docking program (http://hdock.phys.hust.edu.cn/). DITSCORERR software with the interaction scoring function was used to evaluate and rank the generated binding patterns. HNADOCK takes the information of base pairing between Rnas into account, which can significantly improve the docking accuracy. The PDB structure and 3D structure of the molecule were provided when we submit the task, the docking process will also include binding site information, that is, restrictions are imposed to ensure that the corresponding nucleotides are located on the same interface or within a certain distance. Molecular dynamics simulations were performed in Amber24 using the ff19SB force field and OPC water model. The AKBA-BST2 complex was solvated in a cuboid water box with electrostatic and van der Waals interaction cutoffs set to 1.0 nm, a 2 fs time step, and long-range electrostatics handled by the PME method. The system was equilibrated at 300 K and 1 bar using the V-rescale thermostat and Parrinello-Rahman barostat. Following energy minimization, sequential equilibration included 200 ps NVE and 100 ps NPT ensembles. A final 100 ns production MD simulation was conducted. Post-simulation analyses (RMSD, RMSF, radius of gyration, and hydrogen bond counts) were performed using CPPTRAJ and Python scripts integrated with Amber Tools.

CHIP-qPCR

Cells were cross-linked with 1% formaldehyde (10 min, RT), quenched with glycine, washed in PBS, and pelleted (3,000×g, 5 min). Pellets were lysed in membrane extraction buffer, nuclei isolated, and chromatin digested with MNase (37 °C, 15 min); fragmented chromatin was sonicated, then immunoprecipitated overnight (4 °C) with anti-STAT1 antibody bound to Protein A/G magnetic beads. After sequential IP buffer washes, complexes were eluted (65 °C, 30 min), treated with NaCl/Proteinase K (65 °C, 1.5 h), and DNA purified via binding buffer and spin columns, eluted in 50 µL buffer for qPCR/sequencing analysis.

Flow cytometry

After the HPMECs density in the six-well plate was 80–90%, the cells were digested by pancreatic enzymes, centrifuged with pre-cooled PBS at 1000 rpm for 5 min, and repeated three times. It was then re-suspended with 500ul PBS and performed according to the standard flow cytometry kit for detecting apoptosis37.

Statistical analysis

SPSS 22.0 (IBM, USA) and GraphPad Prism 8.0 (GraphPad Software, USA) were used for statistical analysis. Data are presented as mean ± standard deviation (SD) to ensure the scientificity of the results. One-way ANOVA analysis was used between multiple groups, while Least Significance Difference was used to compare the difference of two groups. Two-tailed Student’s t-test was used in comparison between different groups. p < 0.05 was considered to be statistically significant.

Results

AKBA alleviates ARDS both in vitro and in vivo

The results of disease and drug prediction in TCMSP database suggest that AKBA can effectively alleviate ARDS, but the specific role and mechanism are still unknown. HPMECs injury is an important factor in the occurrence and progression of ARDS. Therefore, we preliminarily evaluated whether AKBA has a protective effect on LPS-induced HPMECs. As shown in Fig. S1 of supplementary materials, the EC50 value of AKBA is 9.317 μm. Firstly, we evaluated the effect of AKBA on the tubule formation ability of HPMECs using tubule formation assay. The study results in Fig. 1A-B indicated that AKBA effectively inhibits LPS-induced tube formation degradation of HPMECs. The tubule length and the number of branch points in LPS + AKBA group were obviously increased compared with LPS + DMSO and LPS group, and significantly decreased in LPS group compared with control group. The subsequent colony formation assays and EDU assays (Fig. 1C-F) were used to evaluate the proliferation ability of the cells. The results showed that AKBA could effectively alleviate the decreased proliferation ability of HPMECs induced by LPS, and the number of cell clones and EDU fluorescence intensity in LPS + AKBA group were significantly higher than those in LPS and LPS + DMSO groups. In addition, CCK8 results in Fig. 1G also showed that HPMECs activity in LPS + AKBA group was significantly restored compared with LPS group and LPS + DMSO group, and the difference was statistically significant.

Fig. 1
figure 1

Protective effect of AKBA on HPMECs induced by LPS. (A, B) Tubule formation assay was performed to evaluate the tubule-forming ability of HPMECs (100×magnification), n = 3. (C-F) Colony formation assay and EDU assay were used to detect the relief of AKBA on LPS-induced proliferation inhibition of HPMECs (100×magnification), n = 3. (G) Cell viability was evaluated by CCK8 assay, n = 3. (H-K) The migration protection of AKBA on HPMECs was confirmed by scratch assay and transwell assay (40×and 100×magnifications), n = 3. (L, M) The LC3 expression of HPMECs was determined by immunofluorescence assay, using CD31 as an internal reference. (200×magnification), n = 3. (N) Effective inhibition of autophagy by AKBA in HPMECs was elucidated by TEM (15.0k×magnification). Values are presented as mean ± standard deviation (SD), n = 3. Statistical significance levels are denoted as *p < 0.05, **p < 0.01 and ***p < 0.001.

Full size image

The results of wound healing assay and transwell assay in Fig. 1H-K showed that AKBA could significantly alleviate the LPS-induced reduction of HPMECs migration ability. Specifically, the migration ability of HPMECs in LPS + AKBA group was significantly higher than that in LPS + DMSO and LPS groups, and there was no significant difference between the control group and LPS group. The expression of LPS and AKBA on LC3 was illustrated by immunofluorescence experiments. As shown in Fig. 1L-M, LPS-induced HPMECs expressed LC3 at a higher level, and AKBA treatment could effectively alleviate the LPS-induced increase in expression. Transmission electron microscopy was further used to evaluate the effect of LPS or AKBA on the level of autophagy. As shown in Fig. 1N, we can see that LPS-induced HPMECs showed an increase in the number of autophagosomes, and the number of autophagosomes of LPS + AKBA significantly decreased compared with LPS and LPS + DMSO groups. There was no significant difference between the two groups.

The optimal protection of AKBA (10 mg/kg) on the lung tissue of ARDS mice was verified by the mouse lung injury score in Fig. S2 of the supplementary materials. In Fig. 2A-B, we can see that the lung tissue of mice in the CLP induced group showed obvious inflammatory cell infiltration and structural destruction. Compared with CLP and CLP + DMSO groups, the lung tissue injury of mice in CLP + AKBA group was significantly alleviated, which showed that the lung tissue structure was intact and inflammatory cell infiltration was reduced. There was no significant difference between CLP + AKBA group and sham group, CLP group and CLP + DMSO group. We examined the level of LC3 in mice lung tissue use CD31 as an internal reference. The results (Fig. 2C-D) shows that the fluorescence intensity of LC3 of CLP group was significantly higher than that of the sham group, and there was no difference between the CLP + DMSO group, and the LC3 expression of the CLP + AKBA group was significantly suppressed, and the level of the autophagy level was significantly reduced. The expression of Bad in the mice lung tissue was evaluated by IHC. As shown in Fig. 2E-F, the level of Bad in the CLP group was significantly higher than that in the sham group, and there was no difference between the CLP group and the CLP + DMSO group, CLP + AKBA group and sham group. AKBA treatment can significantly reduce Bad level and protect lung tissue from apoptosis. It can be seen that the bad level of CLP + AKBA group is significantly different from that of CLP and CLP + DMSO group.

Fig. 2
figure 2

AKBA significantly inhibited CLP-induced lung injury in mice. (A, B) HE staining was used to evaluate the inflammatory injury of lung tissue in mice. (100×magnification), n = 5. (C, D). The LC3 expression of mouse lung tissue was obtained by immunofluorescence assay, using CD31 as an internal reference. (100×magnification), n = 5. (E, F) The expression of Bad in lung tissue was detected by immunohistochemistry. (100×magnification), n = 5. *p < 0.05, **p < 0.01 and ***p < 0.001.

Full size image

Preliminary study on transcriptomics and proteomics of AKBA protecting ARDS

After partially demonstrating that AKBA protects ARDS both in vitro and in vivo, we utilized the transcriptomics and proteomics of LPS or AKBA-treated HPMECs to reveal its protective mechanism. As shown in Fig. 3A, all samples in LPS + AKBA group and LPS + DMSO group met the requirements of quality inspection. The volcanic map (Fig. 3B) suggested that the LPS + AKBA group had 190 up-regulated genes and 290 down-regulated genes compared to the LPS + DMSO group, and the same description was also shown in the heat map of Fig. 3C. The KEGG analysis of the transcription group is shown in Fig. 3D, suggesting that AKBA may play an important role in protecting ARDS42,43,44.

Fig. 3
figure 3

Transcriptomics of LPS + AKBA compared to LPS + DMSO group. (A) Intra-group and inter-group correlation analysis of samples. (B) A volcano plot of differentially expressed genes between LPS + AKBA group (K1, K2, K3) and LPS + DMSO group (C1, C2, C3. p value < 0.05 and |log2 FC| > 0, p value < 0.05 and |log2 FC| > 1 were established as the thresholds. (C) Heat map of differential genes. (D) The 20 most significantly upregulated items in the KEGG enrichment analysis results.

Full size image

In view of the transcriptomics results, we further explored the AKBA mechanism by proteomics.

As shown in Fig. 4A, it is the result of our sample quality inspection. It can be seen that the intra-group correlation of samples is desirable. According to the volcano diagram of proteomics in Fig. 4B, there were 16 down-regulated proteins and 54 up-regulated proteins in the LPS + AKBA group compared with the LPS + DMSO group. The names and expressions of all differential proteins can be seen in the heat map in Fig. 4C. Transcriptomic and proteomic association analysis in Fig. 4D indicated that BST2 and STAT1 were one of the top 20 differentially expressed genes.

Fig. 4
figure 4

Proteomics of LPS + AKBA compared to LPS + DMSO group. (A) PCA showing sample correlations. (B) Volcano plot of differential proteins. (C) Heat map of differential protein expression and gene ID. (D) Transcriptomic and proteomic correlation analysis of the top 20 KEGG enrichment.

Full size image

AKBA may anchor transmembrane protein BST2 to increase U-STAT1 level and promote BST2 transcription

The results of both transcriptome and proteomics analysis suggested that BST2 may be one of the important targets of AKBA. In view of the above, we first detected the mRNA level of BST2 in the peripheral blood of 15 patients with septic shock and lung injury, and the peripheral blood of 15 normal people as a control, the specific results are presented in Fig. 5A. We can observed that the expression of BST2 in peripheral blood of patients is significantly reduced compared with normal people. Subsequently, the mRNA expression of 6 top genes (ISG15, BST2, OAS2, IFIT1, IFIT3, STAT1) in HPMECs with significantly increased expression was also identified in the LPS + AKBA group, with LPS + DMSO group as the reference. The results are shown in Fig. 5B, BST2 and STAT1 both have significant mRNA expression increase. As shown in Fig. 5C, the molecular docking of AKBA and BST2 shows that AKBA may play a role by entering cells through BST2, and BST2 may be indispensable for AKBA to recognize and anchor HPMECs (Table S3, Docking Score of different prediction models in the supplementary material), and the anchoring sites may be GLN-87, ALA-88 and ASN-92. Based on the above prediction results, we used CETSA technology to verify the interaction between AKBA and BST2. As shown in Fig. 5D-E, the stability of BST2 in LPS + AKBA group was significantly improved compared with LPS + DMSO group. As an important transcription factor, STAT1 is often involved in the regulation of various diseases. In combination with the above verification of AKBA-induced increased STAT1 expression, we plan to reveal the role of STAT1 in AKBA-induced lung protection.

Fig. 5
figure 5

AKBA may activate U-STAT1 expression and promote BST2 transcription by anchoring BST2. (A) Expression of BST2 in peripheral blood of patients with septic shock. (B) Expression of ISG15, BST2, OAS2, IFIT1, IFIT3 and STAT1 in LPS + AKBA group, n = 3. (C) Molecular docking results and site prediction of AKBA and BST2. (D) Top 20 items for differential protein enrichment analysis and a special concentration of STAT1 signaling pathway. (E, F) The interaction between AKBA and BST2 was evaluated by CETSA technique and CETSA dissolution curve, n = 3. (G) mRNA expression levels of STAT1 and BST2 in HPMECs were evaluated using reverse transcription-quantitative PCR after regulation of STAT1 or BST2, n = 3. (H) Relative luciferase activity was assessed using the dual-luciferase reporter assay in HPMECs co-transfected with OE-STAT1 or OE-NC and BST2-WT or BST2-MUT. (I, J) The molecular dynamics simulation of AKBA and BST2 complex system was conducted to investigate the binding stability and kinetic behavior. (K, L) DARTS-WB assesses the stability of AKBA and BST2 complex systems. (M, N) CHlP-gPCR was performed in HPMECs transfected with pcDNA3.1-HA-AP to identify the enrichment of HA-STAT1 onto BST2 promoter region, IgG served as an antibody control. n = 3. *p < 0.05, **p < 0.01 and ***p < 0.001.

Full size image

When using the JASPAR, UniProt, and UCSC databases for target gene transcription factor prediction, we unexpectedly found that STAT1 might be one of the transcription factors for BST2. (https://alggen.lsi.upc.es/cgibin/promo_v3/promo/promo.cgi?dirDB=TF_8.3&idCon=172412375900&getFile=resumSearchRes.html). At the same time, our transcriptomic and proteomic results indicated that STAT1 may play an important role in the protection of ARDS by AKBA (Fig. 5F). Subsequently, we constructed and verified the efficiency of OE-STAT1, OE-BST2, sh-STAT, sh-BST2, as shown in Fig. S3 in supplementary materials, to further explore the mechanism of AKBA protecting ARDS. Compared with the si-NC group, the mRNA expression levels of STAT1 and BST2 were significantly down-regulated in HPMECs transfected with si-STAT1 (Fig. 5G). Nevertheless, whether BST2 overexpression or downregulation, STAT1 mRNA levels were not significantly different compared with the control group. In addition, results of the dual-luciferase reporter assay expounded that relative luciferase activity was significantly elevated in HPMECs co-transfected with BST2-WT promoter reporter and OE-STAT1 compared with the BST2-WT promoter reporter and OE-NC (Fig. 5H). As shown in Fig. 5I-J, molecular dynamics simulations once again confirmed that the complex of AKBA and BST2 has high binding stability and strong binding energy. Finally, we performed DARTS-WB to evaluate AKBA binding to BST2 and promoting its significantly increased stability, and we could see that AKBA could stabilize protease-induced degradation of BST2 (Fig. 5K-L). The results of CHIP-qPCR also confirmed that STAT1 may be a transcription factor of BST2 (Fig. 5M-N). These results suggest that STAT1 transcription factor may regulate BST2 in ARDS.

HE staining was used to evaluate the lung damage in mice induced by CLP induced by AKBA and BST2. Results As shown in Fig. 6A-B, the lung tissue injury of mice in CLP + AKBA + sh-BST2 group was significantly aggravated compared with that in CLP + AKBA + sh-NC group. Compared with CLP + AKBA + sh-BST2 group, the lung tissue injury in CLP + AKBA + sh-BST2 + HY-P75469 group was significantly reduced. The results of wet-dry weight ratio used to evaluate lung tissue injury in mice (Fig. 6C) showed that compared with CLP + AKBA + sh-NC group, the wet-dry weight ratio of lung tissue in CLP + AKBA + sh-BST2 group was significantly increased, suggesting increased permeability as a result of increased alveolar injury. Correspondingly, the wet-dry weight ratio of lung tissue in CLP + AKBA + sh-BST2 + HY-P75469 group was significantly lower than that in CLP + AKBA + sh-BST2 group, suggesting that AKBA may alleviate CLP-induced lung tissue injury and permeability through BST2. Subsequently, we evaluated the protein levels in the alveolar lavage fluid of mice in each group, and the results (Fig. 6D) showed that compared with the CLP + AKBA + sh-NC group, the protein content in the alveolar lavage fluid of mice in the CLP + AKBA + sh-BST2 group was significantly increased, suggesting increased permeability due to aggravated alveolar injury. The protein level in alveolar lavage fluid in CLP + AKBA + sh-BST2 + HY-P75469 group was significantly lower than that in CLP + AKBA + sh-BST2 group, suggesting that AKBA may alleviate CLP-induced lung tissue injury in mice through BST2.

Fig. 6
figure 6

AKBA-mediated BST2 improves in vivo validation of ARDS. (A, B) The lung tissue injury of CLP-induced ARDS model mice was evaluated by HE staining. (100×magnification), n = 5. (C) Wet-dry weight ratio of lung tissue in CLP-induced ARDS model mice, n = 5. (D) Evaluation of proteins in alveolar lavage fluid of CLP-induced ARDS model mice, n = 5. (E, F) The protein levels of BST2 in lung tissue of CLP-induced ARDS model mice were evaluated by immunofluorescence assay, (100×magnification), n = 5. (G) The protein levels of STAT1 in lung tissue of CLP-induced ARDS model mice were evaluated by immunofluorescence assay, (100×magnification), n = 5. (H) The protein levels of ISG15 in lung tissue of CLP-induced ARDS model mice were evaluated by immunofluorescence assay, (100×magnification), n = 5. *p < 0.05, **p < 0.01 and ***p < 0.001.

Full size image

Finally, we detected the levels of CD31, BST2, STAT1 and ISG15 in mouse lung tissue by immunofluorescence experiment to further reveal the mechanism of AKBA’s regulation of BST2. Results As shown in Fig. 6E-H, compared with CLP + AKBA + sh-NC group, the expressions of BST2, STAT1 and ISG15 in CLP + AKBA + sh-BST2 group were significantly decreased, and compared with CLP + AKBA + sh-BST2 group, the expression of BST2, STAT1 and ISG15 was significantly decreased. The expressions of BST2, STAT1 and ISG15 were significantly increased in CLP + AKBA + sh-BST2 + HY-P75469 group.

Based on the above demonstration, we further evaluated the effect of AKBA regulation of BST2 on the function of HPMECs in vitro. The expression and cell localization of U-STAT1 and BST2 were further confirmed by cellular immunofluorescence. Our immunofluorescence results in Fig. 7A-C suggested that the fluorescence intensity of BST2 and STAT1 in LPS group and LPS + DMSO group were lower than that in control group and LPS + AKBA group, and that LPS induction may inhibit nuclear metastasis and cytoplasmic punctate aggregation of STAT1, while AKBA treatment effectively promoted STAT1 nuclear entry and cytosolic deaggregations. The above may suggest that U-STAT1, as a transcription factor of BST2, may promote its transcription. The results of HPMECs cell viability evaluation showed that HPMECs viability in LPS + AKBA + sh-NC group was significantly reduced (Fig. 7D). Subsequently, the permeability of HPMECs was evaluated, and the results were shown in Fig. 7E. Compared with LPS + AKBA + sh-NC group, the permeability of HPMECs in LPS + AKBA + sh-BST2 group was significantly increased. Finally, the apoptosis level of HPMECs was evaluated by flow cytometry (Fig. 7F-G), and it could be seen that compared with LPS + AKBA + sh-NC group, the apoptosis level of HPMECs in LPS + AKBA + sh-BST2 group was significantly increased.

Fig. 7
figure 7

AKBA mediates BST2 to improve HPMECs dysfunction. (A, B) Fluorescence intensity and cellular localization of BST2, (630×magnification). (C) Fluorescence intensity and cellular localization of U-STAT1, (630× magnification). (D) Cell viability was evaluated by CCK8 assay. (E) The permeability of HPMECs was evaluated by the vascular permeability marker Evans Blue. (F, G) The apoptosis level of HPMECs was evaluated by flow cytometry. n = 3. *p < 0.05, **p < 0.01 and ***p < 0.001.

Full size image

AKBA protects LPS-induced HPMECs dysfunction through U-STAT1/BST2

The differential expression of BST2 and STAT1 transcript levels was initially verified, and the protein levels of BST2 and STAT1 in HPMECs were further detected. Combined with the above-mentioned regulatory effects of U-STAT1 and YP-STAT1 in different diseases, the protein levels of BST2, U-STAT1 and YP-STAT1 were detected, and the results were shown in Fig. 8A-D. LPS induction could significantly inhibit the expression of BST2 and U-STAT1 in HPMECs, while the protein level of YP-STAT1 in LPS group was not significantly different from that in control group. In addition, the protein levels of BST2, U-STAT1 and YP-STAT1 were also detected in the lung tissue of mice induced by CLP. The specific results can be seen in Fig. 8E-H. Compared with the sham group, the protein levels of BST2 and U-STAT1 in the CLP group were significantly decreased, and the YP-STAT1 level was not differentially expressed.

Fig. 8
figure 8

AKBA may inhibit ARDS progression through U-STAT1/BST2. (A) BST2, U-STAT1, YP-STAT1 protein levels in HPMECs were quantified by Western blot analysis, n = 3. (B) Statistical analysis of Western blot of BST2 in HPMECs, n = 3. (C) Statistical analysis of Western blot of U-STAT1 in HPMECs, n = 3. (D) Statistical analysis of Western blot of YP-STAT1 in HPMECs, n = 3. (E) Statistical analysis of Western blot of BST2 in mouse lung tissues, n = 5. (F) Statistical analysis of Western blot of U-STAT1 in mouse lung tissues, n = 5. (G) Statistical analysis of Western blot of YP-STAT in mouse lung tissues, n = 5. (H) BST2, U-STAT1, YP-STAT1 protein levels in mice lung tissues were quantified by Western blot, n = 5. (I-J) The LC3 expression of HPMECs was determined by immunofluorescence assay, using CD31 as an internal reference. (200×magnification), n = 3. (K) The autophagy level of HPMECs was evaluated by transmission electron microscopy. (15.0k×magnification), n = 3. *p < 0.05, **p < 0.01 and ***p < 0.001.

Full size image

Subsequently, using the fluorescence intensity of CD31 as an internal reference, we evaluated the immunofluorescence intensity of LC3 in HPMECs after LPS, AKBA, OE-BST2, sh-BST2, OE-STAT1 and sh-STAT1 treatment. The results in Fig. 8I-J showed that AKBA treatment could effectively alleviate LPS-induced increase in LC3 expression, while Transfection of OE-BST2 and OE-STAT1 can significantly reduce the increase of LC3 expression induced by LPS. Transfection of sh-BST2 or sh-STAT1 into LPS-induced HPMECs treated with AKBA significantly weakened the protective effect of AKBA. As shown in Fig. 8K, the autophagy level of HPMECs was observed by transmission electron microscopy. The results showed that AKBA treatment significantly alleviated the increase in the number of autophagosomes induced by LPS, and the number of autophagosomes in LPS + OE-BST2 and LPS + OE-STAT1 groups significantly decreased compared with LPS group, and the difference was statistically significant. The autophagy level of mice in LPS + AKBA + sh-BST2 and LPS + AKBA + sh-STAT1 groups was significantly higher than that in LPS + AKBA group.

As shown in Fig. 9A-D, EDU and colony formation assay were used to evaluate the proliferation of HPMECs. The results shows that compared with LPS group, the cell proliferation of LPS + AKBA group was significantly increased, and the difference was statistically significant. In addition, the proliferation level of LPS + AKBA + sh-BST2 and LPS + AKBA + sh-STAT1 groups was significantly lower than that of LPS + AKBA group. The proliferation level of HPMECs in LPS + OE-BST2 and LPS + OE-STAT1 groups was significantly higher than that in LPS group, and the difference was statistically significant. The tube formation experimental results of HPMECs (Fig. 9E-F) indicated that compared with the LPS group, the tube formation ability of HPMECs in the LPS + AKBA group was significantly increased. In addition, the level of cell tube formation in LPS + AKBA + sh-BST2 and LPS + AKBA + sh-STAT1 groups was significantly lower than that in LPS + AKBA group. The tube formation level of HPMECs in LPS + OE-BST2 and LPS + OE-STAT1 groups was significantly higher than that in LPS group.

Fig. 9
figure 9

AKBA protects LPS-induced HPMECs via U-STAT1/BST2. (A, B) The proliferation of HPMECs was verified by EDU assay, (100×magnification), n = 3. (C, D) The proliferation of HPMECs was verified by colony formation assay, (100×magnification), n = 3. (E, F) The tube forming ability of HPMECs was evaluated by quantifying the number of nodes and the vascular coverage area, (100×magnification), n = 3. (G, H) The migration function of HPMECs was verified by wound healing assay. (100×magnification), n = 3. (I, J) The migration function of HPMECs was verified by transwell assay. (100×magnification), n = 3. (K) Western blot analysis of BST2, U-STAT1, YP-STAT1, Bad, using β-actin expression as the internal reference in HPMECs, n = 3. (L) Statistical analysis of Western blot of BST2 in HPMECs, n = 3. (M) Statistical analysis of Western blot of U-STAT1 in HPMECs, n = 3. (N) Statistical analysis of Western blot of Bad in HPMECs, n = 3. *p < 0.05, **p < 0.01 and ***p < 0.001.

Full size image

The migration of HPMECs was first evaluated by wound healing assay and transwell assay. The results were shown in Fig. 9G-J. The migration level of HPMECs in LPS + AKBA + sh-BST2 and LPS + AKBA + sh-STAT1 groups was significantly lower than that in LPS + AKBA group, were significantly higher in LPS + OE-BST2 and LPS + OE-STAT1 groups than those in LPS group, and the differences were statistically significant. The results of transwell assay were consistent with the results of scratch assay. Specifically, AKBA significantly reduced the migration ability of LPS-induced HPMECs, and the migration level of LPS + OE-BST2 and LPS + STAT1 groups was significantly higher than that of LPS group. The migration ability of HPMECs in LPS + AKBA + sh-BST2 and LPS + AKBA + sh-STAT1 groups was significantly decreased compared with that in LPS + AKBA group. In Fig. 9K-N, BST2, U-STAT1 levels in LPS + AKBA + sh-BST2 and LPS + AKBA + sh-STAT1 groups were significantly lower than those in LPS + AKBA group. However, its levels in LPS + OE-BST2 and LPS + OE-STAT1 groups were significantly higher than those in LPS group. On the contrary, the expression of Bad in LPS + AKBA + sh-BST2 and LPS + AKBA + sh-STAT1 groups was significantly higher than that in LPS + AKBA group, while it was significantly lower in LPS + OE-BST2 and LPS + OE-STAT1 groups than that in LPS group. There was no significant difference in YP-STAT1 expression among those groups.

AKBA alleviates CLP-induced lung tissue damage in mice via U-STAT1/BST2

In vitro, it was revealed that AKBA may resist LPS-induced HPMECs dysfunction through U-STAT1/BST2, and we further explored the specific mechanism of AKBA alleviating CLP-induced lung tissue injury in mice. First of all, as shown in Fig. 10A-B, LC3 fluorescence in lung tissue of CLP-induced mice transfected with OE-BST2 or OE-STAT1 was significantly lower than that of the CLP group, while AKBA-treated CLP-induced mice transfected with sh-BST2 or sh-STAT1 showed increased LC3 expression. The difference was statistically significant. In Fig. 10C-F, BST2, U-STAT1 levels in CLP + AKBA + sh-BST2 and CLP + AKBA + sh-STAT1 groups were significantly lower than those in CLP + AKBA group. However, its levels in CLP + OE-BST2 and CLP + OE-STAT1 groups were significantly higher than those in CLP group. On the contrary, the expression of Bad in CLP + AKBA + sh-BST2 and CLP + AKBA + sh-STAT1 groups was significantly higher than that in CLP + AKBA group, while it was significantly lower in CLP + OE-BST2 and CLP + OE-STAT1 groups than that in CLP group. There was no significant difference in YP-STAT1 expression among the groups. Subsequently, we evaluated the apoptosis level of mouse lung tissue. As shown in Fig. 10G-H, AKBA can effectively alleviate CLP-induced apoptosis of mouse lung tissue, which is represented by a decrease in red fluorescence intensity reflecting apoptosis level. OE-BST2 or OE-STAT1 transfection can produce a similar alleviation on lung tissue damage induced via CLP as AKBA treatment. The re-transfection of sh-BST2 or sh-STAT1 in CLP-induced mice treated with AKBA showed higher apoptosis of lung tissue than CLP + AKBA group.

Fig. 10
figure 10

AKBA alleviates CLP-induced lung tissue damage in mice via U-STAT1/BST2. (A, B) LC3 expression was determined by relative quantification of immunofluorescence, using CD31 as an internal reference, (200×magnification), n = 5. (C) Statistical analysis of Western blot of U-STAT1 in mouse lung tissues, n = 5. (D) Statistical analysis of Western blot of BST2 in mouse lung tissues, n = 5. (E) Statistical analysis of Western blot of Bad in mouse lung tissues, n = 5. (F) Western blot analysis of BST2, U-STAT1, YP-STAT1, Bad, using β-actin expression as the internal reference. n = 5. (G, H) The apoptosis level of mouse lung tissue was evaluated by TUNEL kit, n = 5. *p < 0.05, **p < 0.01 and ***p < 0.001.

Full size image

Discussion

ARDS is a life-threatening form of respiratory failure that accounts for 10% of intensive care unit admissions worldwide, with more than 3 million cases per year45,46. The increased pulmonary vascular permeability and diffuse alveolar injury are the important diagnostic criteria for ARDS, and the dysfunction and intervention of the HPMECs are gradually taken seriously47. Treatment of ARDS is predominantly supportive, even though no pharmacological intervention has yet been shown to be effective, possibly due to the lack of correct identification of the different ARDS phenotypes48,49. As drug therapy for ARDS has been tested in preclinical and clinical studies, pharmacotherapy is confirmed to have positive results50. For example, more and more studies have confirmed that the use of statins to intervene in ARDS may be related to its effect on endothelial function51,52. Plant extracts are known for their high safety and low resistance. At present, there is still a large space for their application in ARDS.

AKBA, as a boswellic acid derivative, is reported to be anti-inflammatory, oxidative stress, involved in metabolic reprogramming and other pathophysiological processes53,54,55. However, there is still a large space to explore its influence and mechanism in ARDS. In our research, the TCMSP database predicts that AKBA may be an effective intervention drug for ARDS. Therefore, we aimed to investigate the effect and mechanism of AKBA on LPS-treated HPMECs and CLP-induced lung tissue in mice. Our results suggest that AKBA can inhibit the apoptosis and autophagy of HPMECs induced by LPS, promote their tube formation and migration, and correspondingly inhibit the inflammatory injury, apoptosis and autophagy of lung tissue induced by CLP in mice. In addition, this study revealed that U-STAT1/BST2 plays an indispensable role in the mechanism of relief action on ARDS in AKBA. Here, we believe that there is still more room for discussion on the mechanism of AKBA alleviating ARDS, including its direct interacting proteins and indirect signaling pathways involved in regulation.

BST2 acts as a direct physical tether to immobilize the virions to the cell membrane and link the virions to each other. Tethered virions can either be internalized by endocytosis and subsequently degraded or remain on the cell surface56,57,58. Studies have shown that BST2 inhibits the replication of pathogenic microorganisms through selective autophagy targeting and degradation of nucleocapsids. In addition, LC3-related pathways are selectively involved in BST2 of viral tethering24,59. At present, two studies demonstrated down-regulated expression of BST2 were verified in the bovine endometrial cell response to LPS with transcriptomic profiling validation60. BST2, as a transmembrane protein, is identified as the surface molecules of the last divided b cells, participating in multiple immune responses61,62. Hyouna yoo and others tested the purified BST2 protein’s external structure domain significantly reduced the adhesion of the single nuclear cells to human umbilical vein endothelial cells (HUVECs)63, indicating that AKBA induced BST expression may be involved in the migration of monocytes to endothelial cells from the blood flow. Our transcriptomic and proteomic analysis showed that AKBA induced an increase in the transcription and protein levels of BST2, and correspondingly, STAT1 signaling pathway was highly enriched. Molecular docking results suggest that BST2 may act as a traction for AKBA-anchored HPMECs. Our research suggests that the relationship between AKBA and BST2 is worth exploring, and STAT1 may be one of the key factors acting as a bridge.

STAT1, a member of the STAT family, is a signal transducer and activator of transcription in response to interferons, cytokines and other growth factors64,65,66. Studies suggest that tyrosine-phosphorylated STAT1 (YP-STAT1) drives the expression of a large number of genes, and its concentration increases rapidly but declines again within a few hours. IFN-stimulated concentrations of unphosphorylated STAT1 (U-STAT1) tended to increase substantially and persisted for several days36,67. Our research suggests that AKBA induces a significant increase in the U-STAT1 expression, which is combined with the BST2 promoter and promotes transcription, and more BST2 as a cross-membrane receptor anchoring AKBA to the HPMECs. Consistent with our study, Hyeon Cheon and George R Stark exogenously administered higher concentrations of U-STAT1 detected increased expression of many immune regulatory genes, such as BST2 and OAS234. In addition, AKBA-induced STAT1 intranuclear transfer was also detected in our results. In addition, the protein expression levels of BST2 and STAT1 were both increased as detected by our immunofluorescence and western blot. According to the analysis of string website, we speculated that this may be related to the participation of ISG15-mediated ISG-acylation. ISG15, also known as ubiquitin-like protein ISG15, may be involved in mediating the ubiquitination degradation of target proteins either by binding to target proteins (ISG-acylation) or by acting as a free protein68,69. At present, we are conducting a series of studies to reveal the role and specific mechanism of ISGs in alleviating ARDS in AKBA, and hope to reveal the relationship between these regulations more clearly in the near future.

Our research confirms the role of AKBA in actively intervening in ARDS and confirms the potential of screening BST2 as a target protein, which can provide strong evidence for drug small molecules interacting with their targets. But our study still has some limitations. Firstly, the CETSA technique was used to validate BST2 as an AKBA interaction target protein, and optimal methods such as surface plasmon resonance may be more convincing. Secondly, the penetration of AKBA into HPMECs by targeting the transmembrane protein BST2 has not been elaborated. Finally, ISG15-mediated ISG-acylation, which maintains the high expression of BST2 and U-STAT1, has only been preliminarily discovered and needs further elucidation.

Conclusion

In summary, this study revealed the positive effect of AKBA intervention on ARDS and partially explained the indispensable role of U-STAT1/BST2 in this, which provides a promising way to improve the treatment strategy of ARDS patients.