BALB/c nude mice (male, four-week-old) were purchased from Shanghai Laboratory Animal Center (Shanghai, China). All mice were kept in the facilities of Ninth People’s Hospital, Shanghai Jiao Tong University School of Medicine under pathogen-free conditions. The experiments were carried out according to the ethical regulations of the hospital and the Guide for the Care and Use of Laboratory Animals (The Ministry of Science and Technology of China, 2006).

Erianin was obtained from ChemFaces Biochemical (Wuhan, China). 3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) and chloroquine were bought from Sigma-Aldrich (St. Louis, MO, USA). Ad-mCherry-p62 and Ad-mCherry-GFP-LC3B were purchased from Beyotime (Nantong, China). Apoptosis detection kit and propidium iodide (PI)/RNase staining buffer were purchased from BD Pharmingen (San Diego, CA, United States). Antibodies for Cyto C, caspase 9, cleaved caspase 3, GSDMD, CDC25C, CDC2, phospho-CDC2 (p-CDC2, Thr161), cyclin B1 and phospho-4EBP1 (p-4EBP1) were obtained from Cell Signaling Technology (Danvers, MA, United States); antibodies for PARP, NLRP3, 4EBP1, mTOR, phospho-mTOR (p-mTOR), RHEB, LC3B, beclin-1 and p62 were obtained from Proteintech Group (Wuhan, China); antibodies for cleaved-caspase 1, GSDME and PPT1 were from Abcam (Cambridge, MA, United States); and the antibody for β-actin was from Sigma-Aldrich. The secondary antibodies were purchased from Zhong Shan Golden Bridge Biotech Co. (Beijing, China).

WSU-HN4 cells were provided by the Key Laboratory of Stomatology, Shanghai Ninth People’s Hospital. SCC-9 (ATCC: CRL-1629) and CAL-27 (ATCC: CRL-2095) cells were purchased from Zhongyuan Biotech Co. (Beijing, China). WSU-HN4 and CAL-27 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco-BRL, Rockville, MD, United States) supplemented with 10% fetal bovine serum (FBS; Gibco-BRL), penicillin (100 μ/mL; Beyotime, Nantong, China), and streptomycin (100 μg/ml; Beyotime). Comple culture medium for SCC-9 cells was DMEM and Ham’s nutrient mixture F12 (DMEM/F12; Gibco-BRL) medium supplemented with 400 ng/ml hydrocortisone (Sigma-Aldrich), 10% FBS, 100 μ/mL penicillin and 100 μg/ml streptomycin. All cells were kept at 37°C in a humidified 5% CO2 atmosphere.

3×10³ cells/well were plated into 96-well plates. After the overnight, the cells were exposed to erianin at various concentrations. MTT solution (0.5 mg/ml in PBS) (Sigma) was added into each well at different time points. After an additional 4 h of incubation, the culture medium was discarded, the formazan crystals was dissolved by DMSO and the optical density was measured with a microplate reader. For the clonogenic assay, 800 cells/well were plated in 6-well plates. After the overnight, cells were treated with or without erianin. The culture medium was changed every 3 days for 2 weeks. Colonies were stained by crystal violet and were counted under a microscope.

2×10⁵ OSCC cells/well were seeded into 6-well plates. The cells were treated with or without erianin after overnight incubation. Tumor cells were collected 24 h after erianin treatment. For cell cycle analysis, the cells were fixed with cold 70% ethanol and stained for total DNA content with PI/RNase staining buffer. For apoptosis analysis, cells were double stained with FITC-conjugated annexin V and PI. Cell cycle distribution and apoptosis were analyzed by flow cytometry (Becton Dickinson, San Jose, CA, United States).

After erianin treatment, the cell culture medium was collected. LDH activity was then determined by using an LDH detection kit. The optical density was detected at 450 nm using a microplate reader. For the determination of IL-1β, we used an Immulite 1,000 chemiluminescence instrument and matching reagents (Siemens, Germany). Six multiple wells were set for each concentration, and the experiment was repeated three times.

Ad-mCherry-GFP-LC3B and Ad-mCherry-p62 were employed to monitor autophagic flux. In brief, OSCC cells (2×10⁴ cells/well) were added into a 12-well plate. At the second day, the cells were infected with Ad-mCherry-GFP-LC3B (multiplicity of infection = 50) or Ad-mCherry-p62 (multiplicity of infection = 50) for 24 h. Then the culture medium was changed with fresh medium and the cells were cultured for another 24 h. The cells were treated with erianin afterwards. Finally, the cells were washed with PBS for five times and observed with a fluorescence microscope (Olympus, Tokyo, Japan). For Ad-mCherry-GFP-LC3B transfection, both fluorophores fluoresced at neutral pH, indicating the presence of autophagosomes (yellow). When an autophagosome fuses to a lysosome during autolysosome maturation, the lower pH in the lysosome quenches the GFP signal, which shows red fluorescence (autolysosome). For cells transfected with Ad-mCherry-p62, mCherry-p62 usually diffuses within the cell under normal conditions. The formation of autophagosomes typically increases the puncta or speckles of mCherry-p62, while autolysosome maturation leads to p62 degradation and reduces the number of red speckles. When autolysosome maturation is inhibited, the red spots or patches increase in size and number.

OSCC cells were seeded on glass cover slips and treated with erianin as described previously. After treatment for 24 h, the cells were fixed by 4% formaldehyde and permeabilized by 0.2% Triton X-100. Then the cells were blocked with 1% BSA. Later the cells were incubated with the mixture of rabbit anti-RHEB polyclonal antibody and mouse anti-mTOR monoclonal antibody at 4°C overnight. After washing for three times, the cells were incubated with the secondary antibodies conjugated with the oligonucleotides of anti-rabbit PLUS (Sigma-Aldrich, DUO92003) and anti-rabbit MINUS (Sigma-Aldrich, DUO92005), followed by the addition of ligase-ligase solution. The cells were finally subjected to microscopic analysis after the addition of amplification-polymerase solution.

OSCC cells were lysed and total protein were collected and denatured as described previously. Then, the samples were electrophoretically separated on SDS-PAGE gels (EpiZyme, Shanghai, China) and transferred to a PVDF membrane (Bio-Rad, Hercules, CA, United States). After blocking with 5% nonfat milk in TBS/Tween 20 (TBS/T), the membrane was incubated with primary antibodies overnight. Immunoblots were analyzed by the Odyssey Infrared Imaging system (LI-COR Biosciences, Lincoln, NE, United States).

Commercially available PPT1 overexpression lentivirus particles were purchased from GeneChem Biotech Co. (Shanghai, China). The lentivirus-based expression plasmid Ubi-MCS-3FLAG-CBh-gcGFP-IRES-puromycin constructs was used to carry to PPT1 sequences. The primers for PPT1 were 5′-CTC​TAG​AGG​ATC​CCG​CCA​CCA​TGG​CGT​CGC​CCG​GCT​GCC​TG-3′ (forward) and 5′-AGT​CCA​TAC​CTC​CAA​GGA​ATG​GTA​TGA​TGT​GGG​CAT​A-3′ (reverse). Cells were transfected with PPT1 and vector (multiplicity of infection = 10) lentivirus-based plasmid constructs diluted with enhanced infection solution (GeneChem). Cells were collected for further processing 72 hours after transfection.

5×10⁶ WSU-HN4 cells were subcutaneously injected into the backs of nude mice next to the forelimb and permitted to grow until the tumors were palpable. Then, the mice were randomly grouped. For the treatment group, mice were injected with 25 mg/kg or 50 mg/kg erianin intraperitoneally (i.p.) every other day for a total of 15 days. Mice in the control group received vehicle injection accordingly. With a vernier caliper, tumors were measured every 3 days. And tumor volumes were calculated as followed: tumor volume (mm³) = a×b²×0.52. In the formula, a is the longest diameter and b is the shortest diameter. Body weight of all the mice was also recorded each time. When the experiment came to the end, blood samples of the mice were obtained. Tumors were weighed after being separated, fixed with 4% paraformaldehyde and finally embedded in paraffin. Serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) activities AST activities were measured on the Dimension RXL MAX Chemistry Analyzer (Dade Behring Inc., Newark, United States).

To determine the expression of PPT1 and p-mTOR in tumor tissues of nude mice, IHC staining was performed. Briefly, five-μm-thick paraffin tumor sections were deparaffinized and rehydrated. After heat-induced antigen retrieval and blocking with normal goat serum, endogenous peroxidase activity was removed. Primary antibodies were added onto the sections and were allowed to incubate overnight at 4°C. HRP-conjugated secondary antibody was incubated at room temperature the next day. Sections were finally developed in DAB solution and counterstained with hematoxylin.

All experiments were repeated at least three times, and representative data are shown. Student’s t-test was used to calculate the significance of the differences, and data was reported as the means ± SD. The value of p < 0.05 was considered statistically significant.

To confirm the antitumor potential of erianin in OSCC cells, WSU-HN4, SCC-9 and CAL-27 were employed for erianin treatment, and their viability was evaluated by the MTT assay. Generally, the viability of all three OSCC cells was reduced by erianin treatment, which exhibited negative correlation with the dose of erianin (Figure 1B). The IC50 was approximately 50 nM. The colony formation assay was adopted to determine whether erianin inhibited the growth of OSCC cells in vitro (Figure 1C). As expected, erianin significantly reduced the colony numbers of WSU-HN4, SCC-9 and CAL-27 cells in a dose-dependent manner (Figure 1D). Therefore, erianin surppressed the proliferation of OSCC cells in vitro.

To determine the manner in which erianin inhibited OSCC growth, the cell cycle distribution of OSCC cells with or without erianin treatment was analyzed by flow cytometry. The results indicated that erianin increased cell numbers at G2/M phase, accompanied by reduced cell numbers at G0/G1 and/or S phases in the OSCC cell lines (Figures 2A,B). These data suggested that erianin cuased G2/M-phase arrest in OSCC cells.

In regulating the progression of the cell cycle, cyclins and CDKs play important roles (Grana and Reddy, 1995). The progression of G2/M transition and mitotic phase is under the regulation of cyclin B1. Cyclin B1 increases in the G2 phase, peaks in metaphase, and is degraded during anaphase. By coupling with cyclin B1, Cdc2 promoted maturation promoting factor (MPF) production, thus regulate the G2 phase transition to M phase. The cyclin B1/Cdc2 complex is regulated by Cdc25c via removing the inhibitory phosphorylation of Thr14/Tyr15 residues on Cdc2 and also by MO1518 via activating the phosphorylation of Cdc2 at the Thr161 residue. Our results indicated that cyclin B1 expression was upregulated, while Cdc25c and p-Cdc2 (Thr161) levels were downregulated (Figures 2C,D). Therefore, erianin modulated the expression of G2/M transition-related proteins, thus blocking cell cycle progression.

Next, erianin-induced apoptosis in OSCC cells was investigated. OSCC cells were treated with erianin for 24 h, and apoptosis was examined through analysis of annexin V-FITC- and PI-stained cells. With erainin treatment, annexin V-FITC-positive cells increased significantly in all three OSCC cell lines (Figures 3A,B).The expression levels of apoptosis-related proteins, including Cyto C, caspase 9, cleaved caspase 3 and PARP, in erianin-treated OSCC cells were examined by western blot assay. We found that erianin increased Cyto C expression, induced the activation of caspase 9 and caspase 3, and increased the cleavage of PARP, which is the mediator of apoptosis (Figures 3C,D). These data indicated that apoptosis was involved in the decrease of OSCC cell viability induced by erianin.

Recently, pyroptosis has recently attracted great attention owing to its contributory role in cancer progression or anti-cancer therapy. Pyroptosis is a kind of lytic programmed cell death that is mainly mediated by the gasdermin family, which usually share a pore-forming domain (Liu et al., 2021). To figure out whether pyroptosis participated in erianin-induced decrease of cell viability, we first checked the expression of GSDMD, which is the most well characterized executioner of pyroptosis (Liu et al., 2021). We also examined the expression of cleaved-caspase 1, the key regulator of GSDMD cleavage. As NLRP3 inflammasome activation usually leads to the autocleavage of caspase-1 (Xue et al., 2019) and has pro-tumor growth effect in OSCC cells (Huang et al., 2017; Wang et al., 2018), we further detected the expression of NLRP3. Results showed that erianin didn’t induce the cleavage of GSDMD, neither affect the expression of NLRP3 nor cleaved-caspase 1. However, erianin treatment significantly increased the GSDME-N fragments in OSCC cells (Figures 4A,B). GSDME has been identified to be cleaved by activated caspase-3 to generate GSDME-N fragment, which executes pyroptosis by forming pores in the plasma membrane (Chen et al., 2021; Wang et al., 2021). Besides, there was markedly increased LDH and IL-1β in the culture medium of erianin-treated OSCC cells (Figures 4C,D). Therefore, we concluded that erianin induced GSDME but not GSDMD dependent pyroptosis.

Autophagy has been reported to modulate cancer cell survival or death during cancer progression and therapy (Niklaus et al., 2021; White et al., 2021). To make it clear whether erianin triggered autophagy in OSCC cells, we included the autophagy inhibitor chloroquine (CQ) (Bonam et al., 2020; Klionsky et al., 2021) and performed a series of experiments. By western blot assay, we found that CQ treatment induced the accumulation of LC3B II, Beclin 1 and p62 in OSCC cells. CQ was identified to impede autophagy by inhibiting the degradation of autophagosomes in autolysosomes (Bonam et al., 2020; Klionsky et al., 2021). Similar expression patterns of these proteins were found in OSCC cells treated with erianin (Figures 5A,B). Thus, we speculated that erianin also impaired autophagic flux in OSCC cells at late phase of autophagy.

To make this result more intuitive, adenoviruses with mCherry-GFP-LC3B and mCherry-p62 were transfected into OSCC cells, and the cells were observed by fluorescence microscopy. Under normal condition, mCherry-GFP-LC3B was mainly dispersed in the cytoplasm and due to the combined effects of mCherry and GFP, the control cells usually exhibited dispersed yellow fluorescence. Yellow puncta in the cytoplasm generally indicate the formation of autophagosomes. When the autophagosomes and lysosomes fuse (autolysosome formation), GFP fluorescence is gradually quenched and LC3B is displayed in the form of red spots. Interestingly, we found increased red spots in OSCC cells treated with 50 nM erianin and more yellow puncta in 100 nM erianin-treated cells, indicating the autolysosome formation impairment (Figure 5C). P62 is usually degraded in autolysosomes during autophagy. With mCherry-p62 transfection, we found an increase in the size and number of red spots in OSCC cells treated with erianin for 24 h, which suggested inhibition of autolysosomes (Figure 5D). Therefore, these data demonstrated that erianin impeded autophagic flux.

We next went further to determine how erianin impaired autolysosome function in OSCC cells. As PPT1 is reported to be a crucial regulator of lysosomal function that modulates autophagy and mTOR signaling (Rebecca et al., 2017), we first checked PPT1 expression in OSCC cells. Significantly decreased PPT1 expression was found in erianin-treated OSCC cells (Figures 6A,B). Moreover, the phosphorylation of mTOR and downstream 4EBP1 (eukaryotic initiating factor 4E binding protein 1) was notably inhibited by erianin, indicating the impairment of mTOR signaling (Figures 6A,B). To make this conclusion more intuitive, a proximity ligation assay was performed. As shown in Figure 6C, many red dots were observed in control OSCC cells, suggesting the vibrant interaction between mTOR and RHEB and activation of the mTOR pathway. However, erianin reduced this interaction, resulting in a marked decrease in the number of red dots in OSCC cells. The less mTOR interacts with RHEB, the less mTOR is phosphorylated. Therefore, these data implied that PPT1 reduction and subsequent mTOR signaling impairment may be involved in erianin-induced growth inhibition in OSCC cells.

PPT1 has been reported to be overexpressed in cancer and is considered a valid target to control tumor growth (Rebecca et al., 2019; Sharma et al., 2020). To identify whether erianin-induced PPT1 reduction contributed to reduced OSCC cell viability, we first analyzed the expression and potential function of PPT1 in OSCC using online databases. Higher expression levels of PPT1 were found in OSCC epithelia than in normal squamous epithelia, with a fold change of 2.577 and a p-value of 2.14E-7 (Oncomine database, Toruner Head-Neck, Figure 7A). Moreover, Kaplan-Meier survival analysis demonstrated that female OSCC patients with higher PPT1 expression had a worse prognosis (hazard ratio = 2.56, p = 0.0024, Figure 7B), suggesting that PPT1 promoted OSCC progression and may be a good target in OSCC therapy. We next overexpressed PPT1 in OSCC cells with lentivirus particle transfection. The transfection efficiency was confirmed by western blot, as shown in Figure 7C. OSCC cells transfected with vector or PPT1 were later treated with erianin and cell viabilities were determined by MTT assay. The results showed that PPT1 overexpression significantly reversed erianin-induced inhibition of cell viability (Figure 7D). Hence, these data indicated that PPT1 reduction contributed to erianin-induced growth inhibition in OSCC cells.

To investigate the therapeutic potential of erianin in OSCC in vivo, the xenograft model was established. The animals were inoculated, grouped and treated as described in Materials and Methods Tumor sizes were measured every 3 days and volumes were calculated. Body weight of the mice was recorded. After sacrificed, the tumors were separated and weighed. As shown in Figures 8A–C, erianin treatment significantly reduced the tumor volume and weight. No obvious differences were found in the body weight or serum aminotransferases levels of mice from the three groups (Figures 8D,E). Also, there was no visible differences in the appearance or movement of the mice. To identify the in vivo effects of erianin on the expression of PPT1 and p-mTOR, IHC staining of tumor sections from different groups were performed. Markedly, the expression of PPT1 and p-mTOR decreased in tumor cells derived from erianin-treated mice, as shown by the reduced brown staining (Figures 8F,G). So, erianin showed potentially therapeutic effect of OSCC in vivo.