The human osteosarcoma cell lines MG-63 and Saos-2 were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, United States) and McCoy’s 5A medium (Procell, China) respectively supplemented with 10% fetal bovine serum (FBS) (Gibco). All the cells were cultured in an incubator at 37°C with 5% CO₂.

Antibodies targeting the following proteins were used in this study: cyclin D1 (WL01435a), CDK2 (WL01543), p27 (WL04174), cleaved caspase-9 (WL01838), cleaved caspase-3 (WL01992), Bcl-2 (WL01556), Bax (WL01637), LC3BII/I (WL01506), beclin-1 (WL02508), p62 (WL02385), p-JNK (WL01813), JNK (WL01295), c-JUN (WL02863), E-cadherin (WL01482), vimentin (WL01960), MMP-2 (WL1579), MMP-9 (WL01580), β-actin (WL01372), goat anti-rabbit IgG-HRP (WLA023) (all from Wanleibio, China), and p-c-JUN (AF3095, Affinity, United States). Purified peiminine (98%) was obtained from Yuanye Bio Co., Ltd (Shanghai, China; B20081). Cell Counting Kit-8 (CCK-8; C0037), the live/dead assay kit, the cell cycle and apoptosis analysis kit, the ROS Kit, NAC, SP600125, and the terminal dUTP nick-end labeling (TUNEL) Kit (C1086) were acquired from Beyotime Institute of Biotechnology (Shanghai, China). Tetramethylrhodamine methyl ester (TMRM; I34361) and Hoechst 3342 (H3570) were sourced from Life Technology (OR, United States). The mRFP-GFP-LC3 adenovirus was acquired from HanBio Technology Co., Ltd (Shanghai, China). The rest of the reagents and experimental materials were purchased from common commercial sources.

MG-63 and Saos-2 cells were treated with different concentrations of peiminine for 24, 48, or 72 h. CCK-8 solution was then added to the cells following the manufacturer’s protocol and incubated for an additional 2 h at 37°C. The absorbance at 450 nm was calculated using a microplate reader (ELX-800, BioTek, China).

Cells were incubated in a medium containing 2.5 μM calcein AM and 4 μM ethidium homodimer 1 (EthD-1) (C2015M, Beyotime) according to the manufacturer’s instructions. Cells were observed and counted under a fluorescence microscope (U-RFL-T, Olympus, Tokyo, Japan).

For cell cycle analysis, cells were collected and fixed in 75% ethanol at 4°C overnight. The next day, the cells were incubated in staining buffer for 15 min and then resuspended. Propidium iodide (PI) and ribonuclease (RNase) (C1052, Beyotime) were subsequently added and the cells were cultured in the dark for 30 min. Finally, the proportion of cells in each phase of the cell cycle was determined by flow cytometry (Becton Dickinson, United States).

For the analysis of apoptosis, Annexin V-FITC/PI double-staining kit (C1062L, Beyotime) was employed. Peiminine-treated cells were harvested, resuspended in 1 × binding buffer, and incubated with Annexin V-FITC and PI in the dark for 15 min according to the manufacturer’s protocol. Finally, the cells were examined by flow cytometry (Becton Dickinson).

Cells were cultured in glass-bottom cell culture dishes and pretreated with different concentrations of peiminine. After discarding the original culture medium, the cells were rinsed with phosphate-buffered saline (PBS) and incubated with TMRM at 37°C with 5% CO₂ for 45 min. After discarding the medium, the cells were rinsed with PBS, incubated with Hoechst 3342 for 15 min, and imaged using a fluorescence confocal microscope (FV10CW3, Olympus). The fluorescence intensity representing the mitochondrial membrane potential was measured using ImageJ software (National Institutes of Health, United States).

Transmission electron microscopy (TEM) was employed to observe the effect of peiminine on cell ultrastructure. Cells were fixed in 2.5% glutaraldehyde and post-fixed in 1% osmium tetroxide. After dehydration with different concentrations of ethanol, the cells were embedded in Epon. Ultrathin sections were observed under a transmission electron microscope (Hitachi TEM system).

OS cells were transfected with mRFP-GFP-LC3 adenovirus. Fluorescent puncta formed by autophagosomes (yellow) and autolysosomes (red) represented cell autophagy. After 24 h of transfection, the cells were treated with peiminine for 48 h, and then viewed and imaged using a fluorescence confocal microscope (FV10CW3, Olympus).

Following the manufacturer’s instructions, cells were first incubated with the DCFH-DA probe (S0033M, Beyotime) for 20 min at 37°C, following which the nuclei were counterstained with Hoechst 3342 for 15 min. Finally, the cells were viewed using a fluorescence confocal microscope (FV10CW3, Olympus). Fluorescence absorbance was measured using an Infinite 200 PRO microplate reader (Tecan, Switzerland).

Cells were cultured as described above and treated. The cells were lysed on ice with cold radioimmunoprecipitation (RIPA) lysis buffer containing 1% protease inhibitor and 10% phosphatase inhibitor for 30 min. After centrifugation at 4°C for 15 min at 13,500 rpm, the supernatants containing total protein contents were collected and the protein levels were quantified using the BCA Protein Assay Kit (P0009, Beyotime). Equal amounts of total protein were separated by SDS–PAGE (8–12%) and then transferred onto nitrocellulose membranes. After blocking with 5% skimmed milk for 2 h, the membranes were first incubated (with shaking) with primary antibodies at 4°C overnight and then with the respective fluorescently labeled secondary antibodies at room temperature in the dark for 1 h. Finally, the target bands were detected using the Odyssey Infrared Imaging System (LI-COR Biosciences, United States).

Confluent cells were treated with peiminine for 48 h. A wound (scratch) was made in the cells using a 100 μL pipette tip, and the detached cells were washed off with serum-free medium and then cultured with 10% FBS (Jin et al., 2020). Wound healing was assessed at different time points (0, 12, and 24 h) under a microscope (CKX41, Olympus).

Matrigel (356234, BD, United States) was precoated on the upper layer of the Transwell chamber (3422, Corning, United States), after which 200 μL of a serum-free cell suspension (2 × 10⁴ cells that had been pretreated with different concentrations of peiminine for 48 h) was added to the upper chamber and 700 μL of medium containing 30% FBS was placed in the lower chamber. After 12 h, cells in the lower chamber (migrated cells) were fixed in 4% paraformaldehyde at room temperature for 20 min and stained with crystal violet for 15 min. Finally, images were acquired under an inverted microscope (TH4-200, Olympus) at × 200 magnification. The numbers of migrated cells were counted using ImageJ software.

Female BALB/c nude mice (4 weeks old) were acquired from the Shanghai Experimental Animal Center of the Chinese Academy of Sciences. MG-63 cells (1 × 10⁶ cells in, 100 μL of PBS) were subcutaneously implanted into each mouse. One week after implantation, all the mice were randomly divided into two groups (n = 6 per group). Mice in the control group were administered 100 μL of vehicle by intraperitoneal injection every other day, while mice in the peiminine treatment group were administered 100 μL of peiminine (2 mg/ kg) (Zhao et al., 2018) by intraperitoneal injection every other day. After seven doses (21 days after tumor implantation), all the mice were euthanized and the tumors were separated, weighed, and fixed for subsequent analysis.

In brief, sections were deparaffinized, hydrated, incubated with proteinase K, permeabilized, incubated with freshly prepared TUNEL reaction mixture, and then with DAPI (C1006, Beyotime) for 5 min in the dark. The sections were imaged under a microscope (U-HGLGPS, Olympus) at × 200 magnification.

Tissue specimens were fixed, embedded, and serially sectioned (4 μm). The primary tumor and major organs (heart, liver, spleen, lungs, and kidneys) were stained with hematoxylin and eosin (H&E) for morphological evaluation. For immunohistochemical staining, sections were dewaxed with xylene and dehydrated using a graded ethanol series. After antigen retrieval in 10 mM sodium citrate, the sections were blocked in 5% BSA, incubated with primary antibodies overnight at 4°C, and then with the appropriate secondary antibodies at room temperature for 1 h. Immunoreactivity was detected using a DAB Kit (P0203, Beyotime).

GraphPad Prism version 8.0 (GraphPad Software Inc., La Jolla, CA, United States) was used for statistical analysis. All data are shown as means ± SD of independent experiments, each repeated at least three times. Tukey’s multiple comparison test or one-way ANOVA was applied to determine whether there were any statistically significant differences among the different groups. A p-value < 0.05 was considered statistically significant.

We first evaluated the effect of peiminine on the proliferative ability of OS cells. MG-63 and Saos-2 cells were treated with various concentrations of peiminine for 24, 48, and 72 h, following which cell viability was measured by CCK-8 assay. The results showed that peiminine significantly inhibited the proliferation of OS cells in vitro in a time- and dose-dependent manner (Figures 1A,B). Treatment with 100 and 200 μM peiminine for 48 h was selected for subsequent experiments based on the IC₅₀ values obtained from the CCK-8 assays (Supplementary Table S1). The live/dead assay also showed that peiminine treatment disrupted the morphology of OS cells and led to a significant increase in the number of dead cells compared with the control group (Figures 1C–F). These results indicated that peiminine exerts a significant inhibitory effect on the proliferative ability of OS cells in a time- and dose-dependent manner.

Considering that cell cycle arrest can suppress cell proliferation, we investigated the effect of peiminine on the cell cycle of OS cells. The results showed that the proportion of OS cells in the G1 phase was significantly increased after peiminine treatment compared with that in the control group (Figures 2A–C). In addition, the expression of key regulatory factors involved in the G0/G1-phase transition was significantly changed. The levels of cyclin D1 and CDK2 were increased, whereas those of p27 were decreased (Figures 2D–F). In summary, these findings indicated that peiminine induces cell cycle arrest at the G0/G1 phase through regulating the levels of cell cycle-related proteins in OS cells.

As many anticancer drugs exert their effects through activating the apoptotic pathway (Baig et al., 2016), we next examined whether the inhibitory effects of peiminine on OS were associated with apoptosis. Annexin V-FITC/PI staining showed that the apoptosis rate of OS cells was significantly higher with than without peiminine treatment (Figures 3A,B). Because the loss of mitochondrial membrane potential is known to be a critical step in apoptosis (Fu et al., 2020), we evaluated the impact of peiminine treatment on the mitochondrial membrane potential in OS cells as well as the expression of essential proteins involved in the mitochondrial apoptosis pathway. The results showed that the mitochondrial membranes of OS cells underwent significant depolarization after peiminine treatment (Figures 3C–F). Furthermore, the levels of the mitochondrial apoptosis pathway-related proteins caspase-9, caspase-3, and Bax were significantly upregulated, whereas those of Bcl-2 were decreased (Figures 3G–I). These observations suggested that peiminine induces OS cell apoptosis via the mitochondrial pathway.

Autophagy has been implicated in tumor cell death (Li et al., 2020). Here, we investigated whether the inhibitory effect of peiminine on OS cells was partly mediated via autophagic processes. Compared with the control group, the numbers of autophagosomes were significantly increased after peiminine treatment as determined by TEM (Figure 4A). Consistent with these results, we observed that the numbers of autophagosomes and autolysosomes were both significantly increased in OS cells transfected with mRFP-GFP-LC3 adenovirus (Figures 4B–E). Additionally, western blotting analysis showed that peiminine treatment led to an increase in the expression of LC3B II and beclin-1 and a decrease in that of p62 in OS cells (Figures 4F–H). Taken together, these results indicated that peiminine induced the formation of autophagosomes and promoted autophagic flux in OS cells. In the context of cancer treatment, it is known that autophagy can promote either tumor cell survival or tumor cell death. To determine which effect applied to peiminine treatment, we exposed OS cells to chloroquine (CQ), an autophagy inhibitor (Duffy et al., 2015). We found that CQ (20 μM) could attenuate the inhibitory effect of peiminine on OS proliferation (Figures 4J,K). Meanwhile, western blotting results showed that CQ suppressed the peiminine treatment-mediated upregulation of the levels of apoptosis-related proteins (Figure 4I). These findings suggested that peiminine-induced autophagy promotes apoptosis and cell death in OS.

We have previously shown that ROS have a major impact on tumor cell fate (Zou et al., 2017). In this study, we assessed ROS intracellular levels using a DCFH-DA probe (Figures 5A,B; Supplementary Figure S1). The results showed that peiminine (200 μM) promoted a significant increase in intracellular ROS levels in OS cells, whereas NAC application could block this effect (Figure 5C). The JNK signaling pathway is a known ROS target (Wang et al., 2016a). Here, we found that peiminine could activate the JNK pathway in OS cells (Figure 5D). Moreover, both SP600125 (30 μM) and NAC (5 mM) treatment could prevent the peiminine-induced activation of the JNK pathway (Figures 5E,F). Taken together, these results demonstrated that peiminine activates the ROS/JNK axis in OS cells.

Next, we investigated whether peiminine-induced apoptosis and autophagy in OS cells were dependent on the ROS/JNK pathway. We found that exposure to either SP600125 or NAC could attenuate the inhibition of OS cell proliferation mediated by peiminine as determined by the CCK-8 assay (Figure 6A). Flow cytometric analysis further indicated that SP600125 (30 μM) or NAC (5 mM) treatment prevented the peiminine-induced (200 μM) increase in the apoptotic rate in OS cells (Figure 6B), while western blot results demonstrated that both treatments also suppressed the expression of proapoptotic-related proteins (Figures 6C,D). Additionally, in peiminine-treated OS cells, the number of mRFP-GFP-LC3 puncta and the levels of autophagy-associated proteins were also reduced following SP600125 and NAC treatment (Figures 6E–G). Taken together, these results suggested that peiminine induces OS cell apoptosis and autophagy through the ROS/JNK signaling pathway.

We also examined the migratory and invasive abilities of OS cells after peiminine treatment. The results of the wound healing and Transwell assays showed that, compared with the control condition, the invasive and migratory capacity of OS cells was significantly reduced with peiminine treatment (Figures 7A–F). Because epithelial–mesenchymal transition (EMT) and matrix metalloproteases (MMPs) are reported to be associated with tumor invasion and metastasis, we subsequently measured the levels of EMT-related proteins and MMPs in control and peiminine-treated cells by western blotting.) The results showed that the expression levels of vimentin, MMP-2, and MMP-9 were decreased, and that of E-cadherin was increased with increasing peiminine concentrations (Figures 7G–K). In summary, these data showed that peiminine treatment could suppress the migration and invasion of OS cells.

To investigate the antitumor function of peiminine in vivo, we generated OS nude mouse xenograft model using MG-63 cells. Tumor volume and tumor weight analysis showed that peiminine significantly inhibited OS growth in the mice (Figures 8A–C). Moreover, no significant body weight loss was observed compared with that in mice in the control group (Supplementary Figure S2). Consistent with that seen in vitro, TUNEL staining results suggested that peiminine could induce the apoptosis of OS in vivo (Figure 8D). Immunohistochemical results further showed that cleaved caspase-3, LC3B-II, and p-JNK levels were significantly increased whereas that of proliferating cell nuclear antigen (PCNA) was decreased in the tumor tissues after peiminine treatment (Figure 8E). Moreover, histological examination of hematoxylin and eosin (H&E)-stained sections showed that peiminine did not cause major organ-related toxicity (Figure 8F). Overall, our findings clearly indicated that peiminine has strong antitumor activity with low toxicity in vivo.