The crude total flavonoids of Oldenlandia diffusa (FOD) was prepared as follows: the raw materials were extracted with 80% ethanol, and then the extract was treated with lime cream and H₂SO₄ to exclude sediments and other compositions. After filtering, adding water, sealing, and sterilization, FOD was obtained. And each 1 ml of liquid contains 0.25 mg total flavonoids of Oldenlandia diffusa (Anhui Fengyang Keyuan Pharmaceutical Co. LTD., China, lot number 210514). 4-Phenylbutyric acid (4PBA) is purchased from MedChemExpress (Shanghai, China). Chloroquine (CQ) is purchased from Abmole (America). Rapamycin is purchased from MedChemExpress (America). Dulbecco’s Modified Eagle Medium (DMEM), fetal bovine serum (FBS), penicillin-streptomycin, and phosphate buffered saline (PBS) were purchased from Gibco. FITC Annexin V/PI Apoptosis Detection kit were purchased from Yeasen Biotechnology Co. Ltd. (Shanghai, China). Monodansylcadaverine (MDC), Lyso-Tracker Red (cat. no. L8010), Hoechst 33258 (cat. no. C0021) and DAPI (cat. no. C0065) were purchased from Beijing Solarbio (Beijing, China). Rabbit Anti-Phospho-PERK (Thr980) antibody, Goat Anti-rabbit IgG H&L/FITC antibody, Rabbit Anti-Mouse IgG-Fc/PE antibody, Mouse Cleaved caspase-3 antibody, Rabbit Anti-ATF4 antibody and Rabbit Anti-Phospho-EIF2S1 (Ser51) antibody were purchased from Beijing Boosen Biotechnology Co., LTD. LC3B, GAPDH, β-actin were purchased from Cell Signaling Technology (Boston, MA, United States). SQSTM1/p62 Mouse Monoclonal Antibody, CHOP antibody (Mouse mab), DDIT3/CHOP Rabbit Polyclonal Antibody and Ad-mCherry-GFP-LC3B were purchased from Beyotime Biotechnology (Shanghai, China).

Human HCC cells HepG2, Hep3B, HCCLM3 and mouse hepatocellular carcinoma cells H22 were chosen for the following experiments, purchased from National Biomedical experimental cell resource bank (Beijing, China). The cells were incubated in DMEM medium supplemented with 10% FBS, 100U/ml penicillin-streptomycin and maintained at constant temperature 37°C in a sterile incubator with 5% CO₂ as the normoxic condition. The digestion with 0.25% tryspin-EDTA was selected for use when the cells at about 70%–80% confluency. The cells whose growth cycle is in the logarithmic growth phase were selected for further experiments.

Hep3B, HepG2 and HCCLM3 cells were digested by 0.25% trypsin when the cells were in the logarithmic growth phase. The cells were collected for cell counting after centrifugation and were inoculated into each well of 96-well plate with 6 × 10³/well, 100 ul per well. This was cultivated for 24 h, and the cells were treated with FOD at different concentrations for 24 h and 48h, respectively. 10 μL CCK-8 detection reagent was added to each well and incubated for 2 h. The absorbance (A) of each well at 450 nm was measured using a microplate reader. Experiments were performed parallelly in triplicate.

Apoptosis was analyzed using an Annexin V-FITC/PI Apoptosis detection kit (YEASEN, Shanghai, China). 1.5 × 10⁵ HepG2 cells were inoculated in each well of the 6-well plates, were treated with FOD (12.5, 20, 25 μg/ml) for 24 h. Then the experiments were carried out according to the instructions. FACSVerse flow cytometer (BD Biosciences, San Jose, CA, United States) was used to detect the apoptotic cells. Data acquisition and analysis were performed using the Flowjo software (BD Biosciences, San Jose, CA, United States).

HepG2 cells were treated with FOD (12.5, 20, 25 μg/ml) for 24 h and then the original medium was removed and washed twice with PBS. 1ml of Hoechst 33258 (Solarbio) with a concentration of 5 μg/ml was added to each well of the six-well plate for 10 min at 37°C, and then PI dye solution was added to the final concentration of 15 μg/ml. The dye was stained for 10 min at 4°C, and then observed and photographed under a fluorescence microscope.

HepG2 cells treated with FOD (0, 12.5, 20, 25 μg/ml) for 24 h were collected and washed twice with ice-cold PBS. Cells were then fixed in 70% ethanol overnight at 4°C. With twice washing of PBS, cells were stained in solution with PI and RNAse according to the manufacturer’s operating instructions. A total of 30,000 events per sample were acquired by using flow cytometry (FACSCalibur, Becton Dickinson, San Jose, CA, United States), and cell cycle distribution were analyzed accordingly.

HepG2 were exposed to FOD (0, 12.5, 20, 25 μg/ml) for 24 h before harvest. Cleaved caspase-3, p-EIF2α and CHOP were stained with fluorescent labeled antibodies according to Feng, et al. reported method (Qian and Montgomery, 2015). Cells were then detected using a ImageStreamX MkII instrument (Amni, Luminex), and analyzed with IDEAS Software.

We separated equal amounts of proteins from cells using sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred them to polyvinylidene fluoride membranes. A 5% nonfat dry milk buffer was used to block membranes for 2 h at room temperature. Following blocking, membranes were incubated with appropriate primary antibodies at 4°C overnight, followed by 1 h at room temperature incubation with secondary antibodies. The proteins were visualised using SuperSignal West Pico PLUS reagents (ThermoFisher Scientific, United States), under the LAS-4000 mini luminescent image analyser (GE, PA, United States). Normalization was performed using the reference proteins. The results are displayed as the ratio of the target protein to internal reference normalized to the model group (the y-axis is the fold average of model values).

HepG2 cells (1 × 10⁶ cells/well) were seeded in 6-well plates. After treatment with FOD for 24 h, cells were incubated with 1 ml of DCFH-DA (Reactive oxygen species assay kit, Solarbio, China) reagent at a concentration of 10 μM in each well of a six-well plate and incubated at 37°C for 20 min. The DCF fluorescence intensity was then immediately detected using a fluorescence microscopy.

All animal experiments were approved by the Experimental Animal Ethics Committee of Capital Medical University. Six-week-old female BALB/c athymic (Nu/Nu) mice were purchased from the Beijing Weitong Lihua Laboratory Animal Technology Co., LTD. and were acclimated to the institutional animal care facility for 1 week. Mice were injected subcutaneously with H22 cells (1 × 10⁶ per mouse) and were randomly divided into three groups: 6 mice in the control, 4 mice in the model, and 5 mice in the FOD (0.4 mg/kg/d). The FOD group was treated daily by intraperitoneal injection starting the day after tumor cell inoculation. Tumors take rate was 100%. Tumor growth was then measured for 2 days. The length and width of the tumors (mm) were measured three times a week using calipers. The tumor volume was calculated using formula (L ×  W²)  × 0.5, where L and W represent the length and the width, respectively.

All experiments were repeated three times independently. Data were presented as means ± standard deviation (SD). Difference between groups was analyzed by one-way univariate analysis of variance (ANOVA) by Prism 8.0 software (San Diego, CA, United States), and the difference was considered significant when p value < 0.05 (marked as *).

To investigate the effectiveness of FOD to inhibit HCC growth in vivo, 0.4 mg/kg/d FOD was administered to murine models injected with 1 × 10⁶ H22 cells. Studies have shown that FOD treatment can inhibit tumor proliferation. The tumor sizes in the groups of FOD treatment were significantly smaller model group, and the expression level of Ki67 was significantly decreased in the FOD group (Figures 1A,B,K). FOD has anti-inflammatory effect in mice, and the expressions of IL-6 and TNFα in serum of mice in FOD group were lower than those in model group (Figures 1C,D). The body weight, Alanine Aminotransferase (ALT), Aspartate transaminase (AST), UREA, CREA (Creatinine) and HE-stained tissue of liver in the FOD treatment groups were not significantly different from the model group (Figures 1E–J).

To investigate in more detail the mechanism by which FOD inhibits HCC proliferation, we focused on endoplasmic reticulum stress. To further examine the FOD-induced ER stress, Western blotting was used to assess the expression of ER stress-related proteins. The results showed that FOD can significantly increase the expression levels of p-PERK, p-EIF2α, ATF4 and CHOP in tumor tissues of FOD group compared with model group (Figures 2A,B). CHOP is a pro-apoptotic transcription factor that stimulates cell apoptosis when ER stress is unresolved (Han et al., 2013; Hecht et al., 2021). Therefore, cleaved caspase-3 protein expression levels was detected, the results showed that the expression level of cleaved caspase-3 in FOD group was significantly higher than that in model group. Studies have highlighted that ER stress and autophagy are strictly interconnected (Kouroku et al., 2007; Jia et al., 2019; Fang et al., 2021). Next, we examined whether FOD triggers autophagy. As shown in Figures 2C,D, FOD significantly increased the protein expression level of autophagy-ralated molecules, including LC3-II and P62 in tumor tissues of FOD group.

To determine the role of FOD in HCC, HepG2 cell was treated with FOD at different concentrations and for different lengths of time. The results showed that the cell morphology and growth of cells changed gradually with the increase of drug concentration and the extension of treatment time (Figure 3A). In order to further explore the effect of FOD on the proliferation of HCC cells,CCK-8 experiment were performed on HepG2, Hep3B and HCCLM3 cell lines. As shown in Figure 3A and Supplementary Figure S1A, FOD significantly inhibited the growth of HepG2, Hep3B, HCCLM3, and H22 in dose-and time-dependent manners. The expression of Ki67, a proliferation-related marker, was significantly decreased in HepG2 cells treated with FOD (Figure 3B). Furthermore, FOD induced cell cycle arrest in HepG2 cells (Figure 3C). These results suggest that FOD can inhibit HCC cell proliferation.

To explore the effect of FOD on the apoptosis of HCC cells, in the present investigation, HepG2 cells were treated with different concentrations of FOD (12.5, 20, 25 μg/ml) for 24 h. In addition, HepG2 cells were treated with FOD at a concentration of 20 μg/ml for 6 h, 12 h, 24 h, and 48 h. We utilized flow cytometry to analyse apoptosis of HCC cells, the results revealed that FOD can induced apoptosome occurrence in the HepG2 cells in dose-and time-dependent manners (Figures 4A,C). Supplementary Figure S1B shows that FOD could also induce apoptosis in mouse HCC cells H22. Hoechst33258/PI staining was used to detect apoptotic HepG2 cells. There were almost no apoptotic cells in the control group, but atrophic, hyperchromatic and pyknotic nuclei were observed in the FOD group (Figure 4B). We next studied cleaved caspase-3 expressions in HepG2 cells treated with different concentrations of FOD using Flow cytometry. As shown in Figure 4D, FOD treatment elevated the expression of cleaved caspase-3 protein.

In a subsequent study, we examined the effect of FOD on autophagy. By marking intracellular autophagosomes with MDC and Lyso-Tracker, we found that FOD could increase the generation of autophagosomes and the number of acid lysosomes in HCC cells (Figures 5A,B). We further confirmed the expression of autophagy signature protein LC3B-II by FOD through Western blotting experiment. As shown in Figures 5C,D, with the increase of FOD administration time and concentration, the expression of LC3B-II and P62 increased. In order to confirm the role of FOD in autophagy flux, HepG2 cells were transfected with the designed fusion protein mCherry-GFP-LC3B by adenoviral vector. Due to the superposition of GFP and mCherry signals, autophagosomes were marked as yellow. The autophagosome was marked red as a GFP signal quenched by low lysosomal pH (Song et al., 2019b). As shown in Figure 5E, most of the cells treated with FOD lost the GFP signal and retained the mCherry signal. However, in the control group, the signal expression of GFP and mCherry signal was very weak, indicating that the expression of autophagosome was very little and autophagy was not activated. Additionally, we introduced the autophagy promoter Rapamycin (RAPA) and autophagy inhibitor Chloroquine (CQ) to verify the effect of FOD on autophagy. As we know, CQ plays a role in inhibiting autophagy by decreasing autophagosome-lysosome fusion and blocking the autophagic flux (Mauthe et al., 2018). HepG2 cells were pretreated with RAPA and CQ for 2 h, and then incubated for 24 h with cells in the presence or absence of FOD. According to the results (Figure 5F), LC3-II levels were higher when FOD and CQ were combined than when CQ was used alone. Therefore, FOD treatment increases autophagy-related membrane synthesis and activates the process, which is similar to the results of classical experiments that detect autophagy flux (Klionsky et al., 2021). Overall, our findings suggest that FOD can activate autophagy.

Numerous studies have found a link between autophagy and ER stress (Yu et al., 2019; Zhang et al., 2021a; Chipurupalli et al., 2021; Gámez-García et al., 2021; Jahangiri et al., 2022), so we focused on ER stress in this study. p-PERK and p-EIF2α are ER stress-related proteins, and the results show that FOD increases their expression (Figures 6A,C; Supplementary Figure S1C). ATF4 is a key link between ER stress response pathway and autophagy gene expression, because it directly binds to the promoters of several autophagy genes (MAP1LC3B, ATG12 and BECN1) and upregulates their expression (Luhr et al., 2019; Muñoz-Guardiola et al., 2021). Western blot analysis showed that the expression of ATF4 increased dose-dependently and time-dependently after FOD treatment (Figures 6A,B). Study have reported that ATF4 can promote cell apoptosis by regulating the expression of CHOP, which encodes pro-apoptotic protein (Park et al., 2022). By imaging flow cytometry, CHOP expression in HepG2 cells treated with different concentrations of FOD was detected. The results showed that a higher level of CHOP expression was observed after exposure to FOD (Figures 6D,E). Previous studies have shown that excessive activation of ERO1α by CHOP during ER stress increases ROS production (Hetz, 2012). ROS were detected in HepG2 cells treated with FOD at different concentrations and for different periods of time. As shown in Figures 6F,G, FOD could increase the ROS level in HCC cells in a concentration and time-dependent manner. Therefore, we hypothesized that ER stress is important for FOD-induced apoptosis of HCC cells.

To further validate the effect of ER stress on FOD-induced apoptosis and autophagy activation, 4-Phenylbutyric acid (4PBA), a putative ER stress inhibitor, was used to demonstrate. We pretreated HepG2 cells with 4PBA(1 mM) for 2 h and then treated cells with FOD. The results showed that compared with the group treated with FOD alone, the FOD group treated with 4PBA increased the survival rate of HepG2 cells and decreased the apoptosis rate (Figures 7A,B). These results indicated that FOD-induced apoptosis was mediated by ER stress. WB and flow cytometry results showed that pretreatment of HCC cells with 4PBA not only reduced the expression of ER stress-related proteins induced by FOD, but also decreased the expression of autophagy marker LC3B-II (Figures 7C,D). The expression of cleave-caspase3, an apoptosis-related protein, was further detected in HCC cells after inhibition of PERK/EIF2α/ATF4 pathway. As shown in Figure 7E, inhibition of ER stress pathway could reduce the expression of Cleave-Caspase3 after FOD treatment. Overall, these results suggest that FOD induces apoptosis and autophagy in HCC cells by inducing ER stress.