DLT was manufactured by Jilin Connell Pharmaceutical Co. Ltd., China, following the Chinese Pharmacopoeia 2015 (Commission of Chinese Pharmacopoeia, 2015). Conioselinum anthriscoides “Chuanxiong” (52 g), Curcuma aromatica Salisb (52 g) and Alisma plantago-aquatica subsp. orientale (Sam.) Sam. (138 g) are ground into fine powders, then sieved and mixed. Paeonia lactiflora Pall (52 g), Trichosanthes kirilowii Maxim (86 g), and Allium macrostemon Bunge (40 g) are extracted twice by heat reflux with 70% ethanol for 1.5 h. The ethanol extract is collected and filtered, then condensed in vacuo to a final relative density of 1.25 and 1.30 (65°C). Pueraria montana var. lobata (Willd.) Maesen and S.M. Almeida ex Sanjappa and Predeep (138 g) and Salvia miltiorrhiza Bunge (138 g) (individually packed) are extracted three times with ethanol by heat reflux for 1 h. Extracts are collected and filtered, then condensed in vacuo to final relative densities between 1.25 and 1.30 (65°C). Astragalus mongholicus Bunge (114 g), Davallia trichomanoides Blume (26 g), and the extracted residue from Salvia miltiorrhiza Bunge are extracted twice with water for 1.5 h. Extracts are collected and filtered, then condensed in vacuo to final relative densities between 1.25 and 1.30 (65°C). The condensed extracts are mixed with the fine powder, desiccated in vacuo, ground, and pelletised. The mixture is made into 1,000 tablets (0.3 g per tablet) and film-coated. DLT-containing serum was added to the medium for the treatment of cells in culture. DLT-containing serum was prepared as follows: Twenty healthy male SD rats (Guangdong Medical Laboratory Animal Center) were randomly divided into two groups: Rats in the DLT group were administered the drug, 1,400 mg/kg/d. Normal saline was administered to rats in the blank-controlled group. Volumes administered were the same for both groups of animals. Drug or saline was given by gavage for 7 days. Rats were sacrificed 2 h after the last oral administration. Blood samples were collected from the abdominal aorta, centrifuged at 4°C, 3,000 rpm for 20 min. Supernatants were filtered, sterilized, and stored at −80°C for later use.

Six-week-old male C57BL/6 mice and ApoE^−/− mice were purchased from Guangdong Medical Laboratory Animal Center. DLT was purchased from Jilin Cornell Pharmaceutical Co., LTD (Jilin, China). The mice were housed in plastic cages and fed with food and water at an ambient temperature of 23 ± 2°C. All of the mice were divided into six groups (n = 8) randomly after a week: (1) normal control group (NC), (2) atherosclerosis model group (AS model), (3) atherosclerosis model group with low-Danlou tablet (AS model + Low-DLT), (4) atherosclerosis model group with medium-Danlou tablet (AS model + Med-DLT), (5) atherosclerosis model group with high-Danlou tablet (AS model + High-DLT), and (6) atherosclerosis model group with rosuvastatin calcium tablet (AS model + RST). C57BL/6 mice were used as a normal group and fed with a normal diet. ApoE^−/− mice were used to establish an atherosclerosis model, fed with a high-fat diet and different doses of drug: Low-DLT (700 mg/kg/d); Med-DLT (1,400 mg/kg/d); High-DLT (2,800 mg/kg/d); rosuvastatin calcium tablet (10 mg/kg/d), via intragastric gavage, for 10 weeks. DLT powder was suspended and diluted with physiological saline and then administered to mice. After the last drug administration, the mice were euthanized by inhaling an overdose of ether; then, blood and aortic sinus samples were collected for testing. All the animal procedures were approved by the Ethics Committee of The First Affiliated Hospital of Henan University of Chinese Medicine (YFYDW2019-041).

The mice were euthanized, and the sections of aortic sinuses were harvested. The tissues were immediately fixed in 4% paraformaldehyde. Paraffin-embedded tissues were sectioned (4 µm in thickness), placed on poly-l-lysine-coated slides, and incubated for 1.5 h at 60°C. Conventional hematoxylin and eosin staining was performed (H&E), and the degree of inflammation was graded. The fibrosis level was evaluated by Masson staining of collagen accumulation according to the manufacturer’s protocol (Solarbio, China). Oil red O staining was performed using the lipid staining kit (Oil Red O) according to the manufacturer’s instructions. The stained sections were observed under a light microscope (Nikon, Ci-E).

Serum was harvested from the peripheral blood of the mice. The HDL level was assessed using ELISA kit (Elabscience, E-EL-M1402c) following the manufacturer’s instructions. The TC and LDL levels were assessed using the total cholesterol colorimetric test kit (Elabscience, E-BC-K109-M; NJJCBIO A113-1-1). The TG level was assessed by the triglyceride test kit (NJJCBIO, A110-1-1), according to the manufacturer’s instructions. A commercially available ELISA kit was used to measure the serum levels of inflammatory markers: IL-6 (MEIMIAN, MM-0163M1) and TNF-α (MEIMIAN, MM-0132M2).

Human vascular adventitial fibroblasts (HVAFs) were obtained from WUHAN PROCELL LIFE SCI&TECH CO., LTD. HVAFs were inoculated into a Petri dish pre-coated with polylysine and incubated at 37°C under 5% CO₂. The culture medium was changed 48 h later for the first time and then every 3 days. The cells were then overgrown for use. OX-LDL was used to induce an atherosclerotic cell model.

Cell proliferation was determined by the cell counting kit 8 assay (CCK8, Dojindo) and EdU staining assay (RIBOBIO, Cat. No. C10327) following the manufacturer’s protocol. Different groups of cells were treated as indicated in the figure legend for 3 days. At different time intervals (0, 1, 2, and 3 days) after plating, CCK-8 and EdU solutions were added to each well. CCK-8 assay was measured at 450 nm using a microplate reader (SpectraMax M3, Molecular Devices). The cells in the EdU assay were fixed in 4% paraformaldehyde overnight at room temperature and photographed under an inverted fluorescence microscope (Leica, Germany).

After being seeded in 6-well plates at a confluence of about 80–90% after 24 h of incubation, the cells were wounded using a 200-µL pipette tip to scratch the monolayer of the subconfluent cell. Then, the cells underwent different drug treatments for 24 h. For the next 48 h, the cells were cultured in a serum-free medium and allowed to migrate. The images of cell migration were captured using an inverted microscope. The migration speed was calculated by dividing the length of the gap by the wound areas in the captured images.

Millicell chambers with polycarbonate microporous membranes and artificial matrigel were used. The cells were digested, collected, and washed with phosphate buffered saline (PBS) once and suspended with a serum-free medium to adjust the concentration to 2×10⁵/ml; 800 μL of 10% serum medium was added to the bottom chamber, and then 100–150 μL of cell suspension was added to the upper chamber. The culture continued for 12 h in the incubator. The chamber was retrieved and fixed with methanol for 30 min at room temperature. Next, the chambers were moved to a well with 800 μL of crystal violet solution to be stained at room temperature for 15 min. The cells on the membrane surface at the bottom of the upper chamber were carefully wiped off using a wet cotton swab. Then the membrane was carefully removed with forceps and dried upward. The membrane was transferred to a slide and sealed with neutral resin, and photographs were taken under an inverted fluorescence microscope. Nine random fields were selected and counted.

The cell slides were fixed with 4% paraformaldehyde solution at room temperature. After 15 min, the cells were washed with PBS three times and permeabilized by immersion in 0.3% Triton X-100 in PBS for 10 min. Then the slides were blocked with 2% bovine serum albumin (BSA) in PBS for 1 h at room temperature and incubated with the primary antibodies Vimentin (ab8978, 1:500, Abcam) and LC3B (ab229327, 1:200, Abcam) at 4°C overnight. Afterward, the slides were stained with fluorescently-labeled secondary antibody for 1 h at room temperature and mounted by VECTASHIED solution (vector) with DAPI. The images were taken under a confocal fluorescence microscope (Olympus, Japan) and analyzed by NIS elements imaging software.

The total protein extraction was performed with RIPA lysis buffer (mixed with phosphatase and protease inhibitor), and protein quantification was carried out with a BCA kit. The samples were separated on 4–20% SDS-PAGE gel and then transferred to a nitrocellulose membrane. After blocking by 5% skimmed milk for 1 h, the membrane was incubated overnight at 4°C with specific antibodies: p-mTOR (#5536, 1:1,000, CST), mTOR (#2983, 1:1,000, CST), AKT (#4691, 1:1,000, CST), p-AKT (#4060, 1:1,000, CST) p-S6 (#2215, 1:1,000, CST), LC3A/B (#12741, 1:1,000, CST), P62 (#5114, 1:1,000, CST), Beclin-1 (ab210498, 1:1,000, Abcam), ICAM-1 (ab53013, 1:1,000, Abcam), VCAM-1 (ab115135, 1:1,000, Abcam), and GAPDH (ab8245, 1:5,000, Abcam). After being washed with TBST buffer three times, the membrane was further incubated with the secondary antibody for 1 h, washed three times with TBST buffer, and then exposed in ECL developer in Image lab.

The samples were dropped on a carbon support membrane copper mesh for 3–5 min, and then a filter paper was used to absorb excess liquid. Then, 2% phosphotungstate was dropped on the carbon support membrane copper mesh for 2–3 min and air-dried at room temperature. The images were observed under a transmission electron microscope (HT7700 HITACHI) and captured for analysis.

GraphPad Prism software was used to perform statistical analyses. The experimental data were expressed as the mean ± SD, analyzed by one-way ANOVA, followed by Student’s t-tests. Differences with p < 0.05 were considered statistically significant (*p < 0.05, **p < 0.01, and ***p < 0.001).

To detect the influence of the medical tablet on AS in vivo, we first established an AS mouse model using the ApoE^−/− mice by a high-fat diet, and significant AS plaque was observed as previously described (Emini Veseli et al., 2017). To further elucidate the therapeutic effect of Danlou tablet, different doses of Danlou tablet and rosuvastatin calcium tablet were administrated, and AS plaque tissues were harvested from the mice, followed by the assessment for the degree of the lesion. Histological examinations demonstrated thicker plaque in ApoE^−/− mice than the control mice. While Danlou tablets significantly reduced the degree of pathological lesions in a dose-dependent manner based on H&E staining (Figure 1A), they exhibited efficacy comparable to traditional rosuvastatin calcium tablets. In addition, lipid accumulation and fibrosis formation were alleviated in the Danlou tablet-treated ApoE^−/− mice according to the Oil Red O and Masson trichrome staining, red lipid droplets accumulated mostly at the edge of AS lesions (Figures 1B,C).

In order to investigate the biochemical basis that leads to the pathological results, we collected serum samples from the mice to determine the blood total cholesterol (TC), triglyceride (TG), high-density lipoprotein (HDL), and low-density lipoprotein (LDL) levels and inflammatory factors, like IL-6 and TNF-α. Interestingly, Danlou tablets significantly reduced TC (Figure 2A), TG (Figure 2B), and LDL (Figure 2C) levels but did not alter the HDL (Figure 2D) levels in ApoE^−/− mice. Of note, two inflammatory factors, IL-6 (Figure 2E) and TNF-α (Figure 2F), were strongly decreased by high-dose treatment of Danlou tablets. Collectively, these data underline the capacity of Danlou tablets to alleviate AS in vivo.

LDL could accumulate in fibroblasts at the edge of some AS lesions, which might be an early indication of atherosclerosis (Lees et al., 2001). We used a human vascular adventitial fibroblasts (HVAFs) cell line to identify the mechanism of Danlou tablets (DLT) in alleviating AS. We confirmed the high expression of vimentin, a well-known marker of HVAFs, by immunofluorescence cytometry (Figure 3A). Next, an AS cell model was established by adding ox-LDL to the HVAFs culture medium. Therefore, the effect of DLT was evaluated in vitro by this AS cell model. As indicated in Figure 3B, DLT-containing serum notably reduced the growth of HVAFs, which were significantly stimulated by ox-LDL through the CCK8 experiment. The anti-proliferation ability of DLT was also similar to rosuvastatin (RST), consistent with the EDU staining assay (Figure 3D). The cell wound healing assay and transwell assay illustrated that DLT significantly down-regulated cell migration and invasion ability compared to RST treatment (Figures 3C,E). After activation, HVAFs can secrete a large amount of intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecules (VCAM-1) (Liu et al., 2010), chemokines, and inflammatory factors to accelerate the chemotaxis of monocytes and lymphocytes and mediate the inflammatory response. Therefore, we used western blotting to detect the ICAM-1 and VCAM-1 expression after ox-LDL treatment, and the results were consistent with the phenotype of HVAFs (Figure 4E). In short, these data suggest that DLT might slow down the formation of AS by suppressing the activity of HVAFs.

It is well established that vascular endothelial cells undergo autophagy to protect their structures from inflammation and oxidative stress when stimulated by ox-LDL in AS. Here, we speculated that the reduced activity of HVAFs might result from the alteration of autophagy. To directly determine the degree of autophagy in HVAFs, we detected the LC3B level by immunofluorescence cytometry. As shown in Figure 4A, DLT increased the expression of LC3B, indicating a high rate of autophagy. Moreover, we also observed the extensive formation of phagosomes under the transmission electron microscope (Figure 4B). Consistent with this finding, increased expression of Beclin1 and LC3 I-to-LC3 II ratio further supported our hypothesis (Figure 4D). In addition, the increased expression of p62 in the autophagosome degradation stage is usually regarded as a sign of inhibited autophagy activity. We also showed via Western blot that DLT reduced the expression of p62 (Figure 4D). Altogether, we concluded that DLT activates autophagy to protect HVAFs from damage caused by AS.

To explain our findings, we analyzed the signal molecular mechanisms involved in autophagy as previous studies have demonstrated that the mTOR pathway is a key link of the autophagy process. PI3K/Akt and MAPK signaling pathways induce the activation of mTOR to inhibit autophagy. For instance, the activated PI3K-I generates PIP3 in cells, activating Akt with the assistance of PDK1, inhibiting the TSC1/2 complex and activating mTORC1 (Dieterle et al., 2014). Ribosomal protein S6 is a downstream target of mTOR. Levels of phospho-S6 are a frequently used marker of activation of the mTOR pathway. Phosphorylation of S6 exerts an inhibitory effect on autophagic proteolysis (Hay and Sonenberg, 2004; Lu et al., 2019). To exploit the contribution of mTOR pathways to the alteration of autophagy in HVAFs, we analyzed the indicated signaling pathways by western blotting. As the level of phosphorylated mTOR, Akt, and S6 were notably decreased by the administration of DLT (Figure 4C), these findings indicate that DLT activated autophagy partly through inhibiting the mTOR via the PI3K/Akt pathway.

To verify the influence of DLT on the mTOR signaling pathway, we further performed 3-MA, an mTOR pathway agonist (Wu et al., 2013), to study whether activated autophagy could be reversed. As expected, the decreased proliferation of HVAFs was distinctly restored by 3-MA (Figures 5A,B). Of note, the efficiency of DLT in the cells was comparable to Rapa, a conventional antagonist of the mTOR signaling pathway (Cai et al., 2018). The results presented in Figures 5C,D also indicated that 3-MA increased the motility of HVAFs. The raised expression of ICAM-1 and VCAM-1 via western blotting (Figure 6E) further manifested the recuperative activity of HVAFs.

3-methyladenine (3-MA) is a widely used inhibitor of autophagy due to its inhibitory effect on PI3K and suppression of the conversion of LC3-I to LC3-II (Zhang et al., 2020). Here we observed reduced autophagy by the lower expression of LC3B by immunofluorescence cytometry (Figure 6A) and decreased formation of phagosomes under the transmission electron microscope in the 3-MA group (Figure 6B). The lower expression of Beclin1 and LC3 I -to-LC3 II ratio by western blotting were consistent with the phenotype. In addition, we also demonstrated increased expression of P62 via western blotting (Figure 6D). Data from western blotting also confirmed that 3-MA reversed the activated autophagy by DLT by stimulating the mTOR via the PI3K/Akt pathway (Figure 6C). Treatment with DLT decreased p-mTOR protein levels, increased conversion of LC3-I to LC3-II, and activated autophagy compared with the ox-LDL group, but the effect could be reversed by 3-MA (Figures 6 B–D). In conclusion, the therapeutic effect of DLT could be reversed by inhibiting autophagy through the mTOR pathway agonist, indicating the pivotal role of the PI3K/Akt/mTOR pathway in this process.