Twenty four apolipoprotein E-deficient (ApoE^−/−) male mice (approximately 8 weeks old) were purchased from GemPharmatech Co., Ltd. (Jiangsu, China). All mice were kept in the animal house of China State Institute of Pharmaceutical Industry, where they were maintained in a 12 h day/night cycle at a temperature of 20°C–22°C and relative humidity of 40%–70%, with free access to food and water ad libitum. All the experiment procedures were conducted in accordance to the guidelines of the Animal Ethics Committee of China State Institute of Pharmaceutical Industry to minimize the suffering of the animals.

All the mice were randomly divided into control group, model group, alisol A group (A) and Atorvastatin (AT) group (n = 6). The control group was fed with standardized chow (BEIJING HFK BIOSCIENCE CO., LTD., H10010, China), the model group was given high-fat chow (BEIJING HFK BIOSCIENCE CO., LTD., H10141, China), and A group was given high-fat chow mixed with alisol A (Self-made, dosage 150 mg/kg/d) in high-fat chow, and the AT group was given high-fat chow mixed with atorvastatin (dose 15 mg/kg/d). All groups of mice were fed for 16 weeks.

Serum was collected from each group of mice and the concentrations of total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), and high density lipoprotein cholesterol (HDL-C) were measured according to the protocols of the kit (RECOM BIO, XZ017801, XZ027801, XZ037801, Nanjing Jiancheng Bioengineering Institute, A112-1-1,China), which was used to evaluate the levels of blood lipids.

Experimental animals were killed by cervical dislocation. The thoracic cavity of mice was rapidly opened, and the aorta of mice was obtained under stereomicroscope. The aortic arch and heart were taken and stored in 4% paraformaldehyde, stained with Sudan IV staining solution for 6 min, decolorized in 80% ethanol for 3 min, finally washed in PBS for 2 min. The aortic vessels were fixed with a small needle to spread out. The atheromatous plaques in the aortic arch and bifurcation were photographed under a stereomicroscope. The images were stitched together to form a complete picture by Photoshop software, and the ratio of the plaque area to total vessel area was calculated by Image Pro software.

Human aortic endothelial cells (HAECs) were purchased from YUCHI BIOLOGY (Shanghai, China). The cells were cultured in endothelial cell medium (ECM, ScienCell, 1001, United States),supplemented with 5% fetal bovine serum (FBS, ScienCell, 0025, United States), 1% dual anti-penicillin/streptomycin (P/S, ScienCell, 0503, United States) and 1% growth supplementation factor (ECGS, ScienCell, 1052, United States), at 37°C and 5% CO₂ atmosphere.

The cell viability was detected by CCK8 method. HAEC cells in logarithmic growth phase (no more than 10 passages) were inoculated into 96-well cell culture plates, given medium containing different concentrations of alisol A for incubating 24 h, 48 h and 72 h, and then operated according to the protocols of the kit (Cell Counting Kit-8, Beyotime, C0038, China). Light absorbance was measured at 450 nm wavelength and the cell viability was assessed.

The migration characteristics were evaluated using the scratch assay. HAEC cells in logarithmic growth phase (no more than 10 passages) were inoculated into 24-well cell culture plates, and scratched vertically with a 200 μL pipette tip after 80% cell fusion. Then, cells were incubated in medium (serum-free treatment) containing different concentrations of alisol A for 8 h. The cells were observed and photographed under a microscope, and the ratio of migratory areas and distances were calculated by ImageJ software.

HAEC cells in logarithmic growth phase (no more than 10 passages) were inoculated into 96-well cell culture plates covered with Matrix-Gel™ (Beyotime, C0372, China), and given medium (serum-free treatment) containing different concentrations of alisol A for incubating 6 h. The cells were observed and photographed microscopically, and the number of junctions and the total length of the tubes were analyzed and compared with each other by using the Angio Tool software, to evaluate the angiogenesis.

The Griess method was used to determine the secretion of NO. HAEC cells in logarithmic growth phase (no more than 10 passages) were inoculated into 96-well cell culture plates, and then divided into control group, high-fat stimulation group (FFA, 200 μM), and different concentration of alisol A treated group for incubation for 48 h. The cell culture medium was measured according to the protocols of the kit (Total Nitric Oxide Assay Kit, Beyotime, S0021S, China). The absorbance was measured at 545 nm, and the NO concentration was calculated according to the standard curve.

Cellular release of LDH was measured using a colorimetric assay to evaluate the integrity of the cell membrane. HAEC cells in logarithmic growth phase (no more than 10 passages) were inoculated into 96-well cell culture plates, divided into control group, oxidative damage group (hydrogen peroxide, 100 μM), and different concentration of alisol A treated groups for incubation for 48 h. The cell culture medium was measured according to the protocols of the kit (LDH Cytotoxicity Assay Kit, Beyotime, C0017, China), and the absorbance were measured at 492 nm to calculate the LDH release rate.

Human microvascular endothelial cells (HMEC-1) were inoculated in 60 mm culture dishes, and divided into control group, inflammation injury group (LPS, 1 μg/ml) and alisol A treatment group. In the A group, the cells were treated with alisol A (10 μM) for 72 h, then LPS was added, and after 6 h of stimulation, the cells were scraped off with a spatula and collected. Then the total RNA was extracted with TRIzol reagent (Beyotime, R0016, China). Among them, the control group was labeled KB, the A group was labeled A, and the model group was labeled Model.

The concentration and purity of the extracted RNA were examined by Nanodrop (2000). RNA completeness was detected by agarose gel electrophoresis, and RNA quality number (RQN) was determined by Agilent 5,300. The following requirements should be met: the amount of RNA should be greater than 1 μg; the concentration of RNA should be higher than 30 ng/μL; RQN should greater than 6.5; the ratio of OD260/280 should be between 1.8 and 2.2.

Eukaryotic mRNA sequencing was performed on a NovaSeq X Plus platform. Fastp quality control and RSEM software were used for expression analysis, and DESeq2 software for differential expression analysis. Enrichment analysis was performed by Goatools and Python scipy software, using the Fisher’s exact test.

To evaluate the binding energy and interaction mode between alisol A and its targets, the 3D structures of key targets were downloaded from the PDB database (https://www.rcsb.org/) and the molecular structure of alisol A was obtained from the PubChem compound database (https://pubchem.ncbi.nlm.nih.gov/). The target protein and alisol A were dehydrogenated and hydrogenated, and the docking pocket was set as a square of 30 A × 30 A × 30 A with a lattice distance of 0.05 nm. Molecular docking was fitted by using AutodockVina 1.2.2 software, and the binding energies were calculated, finally PyMOL software was used for visualization.

Each experiment was repeated at least three times. Data were expressed as mean ± standard deviation. One-way analysis of variance (ANOVA) was used for multiple group comparisons, and t-test was used for two-group comparisons. Bar graphs were plotted using GraphPad Prism 8.0 software, while p < 0.05 was considered statistically significant.

Sudan IV staining demonstrated obvious lipid deposition in arterial lumen and severe atherosclerotic plaques in the model group. Plaque staining area was significantly reduced after alisol A treatment. Quantitative analysis showed that plaque area significantly increased from (1.68 ± 0.49)% to (13.62 ± 6.18)% in the model group after high-fat feed feeding (****P < 0.0001), and the plaque area was significantly reduced to (5.66 ± 1.50)% compared with the model group after alisol A treatment (**P < 0.01) (Figure 2B).

There were no significant differences in serum concentrations of total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), and highdensity lipoprotein cholesterol (HDL-C) between groups of experimental animals (data not shown).

Cell proliferation is one of the important physiological functions of living cells, and cell viability can indirectly reflect cell proliferation (Khalef et al., 2024). Taking 24 h as an example, the cell viability of control group was (99.80 ± 13.87)%, and there was no significant change in cell viability after incubating with low concentration of alisol A. When the concentration was increased up to 40 μM, the cell viability decreased to (72.24 ± 15.68)% (control group vs. 40 μM group, **P < 0.01), indicating significant inhibition of cell viability (Figure 3C). The change trends of cell viability at 48 h and 72 h were consistent results. It can be inferred that alisol A has an inhibitory effect on the proliferation of vascular endothelial cells, and its inhibitory effect is related to the concentration of alisol A (when the concentration reaches 40 μM, it shows a significant decrease), and is not related to the incubation time.

Mobility is a property of vascular endothelial cells, and wound healing experiments are often used to evaluate the migratory ability of cells (Montoro-García et al., 2023). As shown in the figure, the percentage of migratory distance in control group was (71.97 ± 4.24)% after 8 h. However, the presence of alisol A decreased the percentage of migratory distance of the cells, and the percentage of migratory distance was further decreased with the increase of concentration to (53.97 ± 4.37)%, (44.92 ± 3.53)%, and (33.08 ± 4.11)% (control group vs. 5 μM group **P < 0.01, control group vs. 10 μM group ***P < 0.001, control group vs. 20 μM group ****P < 0.0001). The trend of migration area is consistent with the distance (Figure 3A).

In the physiological state, vascular endothelial cells exhibit angiogenesis properties (Kelley et al., 2022). After inoculation of low density of cells, the morphology of the cells was observed, and the number of bifurcation points and the length of the total tube formation were measured to evaluate the status of angiogenesis. As shown in Figure 3B, the formation of reticular vascular morphology in the field of view was inhibited after 6 h of alisol A treatment. The statistics showed that with the increase of the concentration of alisol A, the length of the total tube was further reduced from (62405.26 ± 5214.90), to (54307.70 ± 3138.88), (46699.08 ± 1953.67), and (37266.45 ± 2663.71) (control group vs. 10 μM group, **P < 0.01, control group vs. 20 μM group ****P < 0.0001). The number of junctions decreased from (520.33 ± 7.09) to (483.00 ± 9.54), (377.67 ± 15.04) and (275.33 ± 40.50) (control group vs. 10 μM group, ***P < 0.001, control group vs. 20 μM group, ****P < 0.0001). It can be inferred that slisol A has the effect of inhibiting angiogenesis.

Endothelial cells have a secretory function, among which NO is an endothelial factor that is secreted outside to exert vasoprotective effects (Jiang et al., 2021). The concentration of NO in the culture medium can be used as an indicator to evaluate the secretory function of cells (Jiang et al., 2021). As can be seen from Figure 3E, FFA treatment caused the NO concentration in the cell culture medium to decrease from (3.67 ± 0.61) μM to (0.83 ± 0.73) μM. (control group vs. FFA group, **P < 0.01). However, there was a tendency for NO concentration to increase in the presence of alisol A. The concentration of NO in the presence of alisol A tended to increase. When the concentration of alisol A reached 20 μM, the NO concentration increased to (3.63 ± 0.95) μM, which was significantly different compared with the FFA group (FFA group vs. 20 μM group, **P < 0.01). These data suggest that alisol A has a restorative effect on decreasing the NO secretion caused by FFA.

When the cell membrane structure is damaged, intracellular lactate dehydrogenase is released into the extracellular space, so LDH release is viewed as an important indicator of the integrity of the cell membrane and is widely used in cytotoxicity assays (Kuma et al., 2018). Hydrogen peroxide treatment caused cellular LDH release to increase from (18.06 ± 2.33)% to (53.79 ± 6.74)% (control group vs. H₂O₂ group, ***P < 0.001). However, the LDH release rate tended to decrease in the presence of alisol A. The LDH release rate further decreased to (44.75 ± 4.71)%, (36.81 ± 3.93), even (29.78 ± 2.43)% with increasing concentration (H₂O₂ group vs. 5 μM group, *P < 0.05, H₂O₂ group vs. 10 μM group, ***P < 0.001, H₂O₂ group vs. 20 μM group ****P< 0.0001) (Figure 3D). These data suggest that alisol A has a protective effect against cell membrane breakdown brought about by hydrogen peroxide.

To explore the effects of alisol A on cell proliferation, migration, and tube-forming properties, we performed transcriptome analysis of vascular endothelial cells in the medium-dose group. Based on the expression quantification results, differentially expressed genes (DEGs) analysis was performed using DESeq2, screening for up and downregulated multiples greater than 2, p-value less than 0.05, with a total of 4,086 DEGs, of which 3,378 were upregulated (red) and 708 were downregulated (gray) (Figure 4B). The DEGs distribution can be represented by the volcano diagram (Figure 4C). The closer the points to the two sides and the upper side, the more significant the drug modulation effect. In the clustering heatmap, we can observe the expression value of each gene in different samples (red represents higher expression, while gray represents lower expression), and the genes clustered significantly under the effect of alisol A (Figure 4D). The PCA analysis showed good parallelism of biological replicates between samples within a group, and significant differences in data between groups (Figure 4D).

Functional enrichment analysis was performed on the genes to obtain the main functions of the genes in the set. KEGG enrichment analysis categorized and enriched the genes from the perspective of signaling pathways (Kanehisa et al., 2017). Figure 5A lists the top 20 pathways, of which the highest number of enriched genes was TNF pathway, suggesting that alisol A may intervene in plaque progression by modulating the effects of the TNF pathway. GO enrichment analysis encompasses three ontologies, which describe the molecular function, cellular components, and biological processes involved in genes (Mo et al., 2023). The result was shown in Figure 5B.

The enriched genes were further annotated and analyzed, i.e., genomic information and functional information were integrated, and the genes were classified according to signaling pathways or biological functions. The results showed that alisol A plays an important role in cell growth and apoptosis, immune system diseases, cardiovascular diseases, infectious diseases, and other processes. Further indications suggest that alisol A may intervene in the development of plaques through anti-inflammatory and other effects (Figures 5C, D).

At the micro level, we focused on five genes, TNF, CSF2, IL1B, SULT2B1 and MALAT1. TNF, CSF, and IL-1β are cytokines with pro-inflammatory effects, which act on vascular endothelial cells to promote blood vessel formation and cell proliferation (Zhou et al., 2020; Zhou et al., 2021). Alisol A showed a significant inhibitory effect on these genes. The results suggested that the alisol A exerted its inhibitory effect on vascular endothelial cell proliferation and tube formation by inhibiting the expression of the genes for TNF, CSF, and IL-1β. (Figures 5E–G). In addition, SULT2B1 and MALAT1 have been extensively studied in the field of cancer in recent years. Hangyu Pan found that inhibition of SULT2B1, reduced macrophage inflammatory response, stabilized plaque, and delayed the progression of atherosclerosis (Yin and Chen, 2020). Katharina M found that silencing of MALAT1 promotes endothelial cell migration, which indicated that MALAT1 inhibited endothelial cell migration (Michalik et al., 2014). The expression of MALAT1 was upregulated and that of SULT2B1 was downregulated in endothelial cells following incubation with alisol A (Figures 5H, I). The synergistic effect of them exerted the effect of inhibiting endothelial cell migration.

To validate the transcriptomics findings, we performed molecular docking analysis on the targets TNF, CSF, IL-1β and SULT2B1 (MALAT1 is a noncoding gene and does not encode proteins) and assessed the effects of the alisol A on the above targets (Figure 6B). When the ligand binds to the target to form a conformationally stable structure, the lower the energy, the more stable the structure. When the binding energy is less than 0 kcal/mol, the molecular ligand can spontaneously bind to the protein receptor. If the binding energy is less than −5 kcal/mol, it indicates that their binding ability is stronger and the possibility of interaction is higher (Hsin et al., 2013). According to Figure 6B, alisol A has a low binding energy with all four targets, and has the lowest binding energy of −10.117 kcal/mol with TNF, indicating the best stability. TNF stimulates the cellular production of IL-1β, and it also enhances the CSF production, according to the docking results, alisol A can directly bind to and inhibit TNF, and then inhibit its downstream factors CSF and IL-1β production; it can also directly act on CSF and IL-1β to play an inhibitory role.