Cepharanthine inhibits the proliferation and epithelial-mesenchymal transition of oral squamous cell carcinoma via HMGA2/FOXL2 axis

Abstract

Background

Cepharanthine (CEP), a kind of isoquinoline alkaloid extracted from stephania, is applied in the treatment of cancer. This study aimed to explore the antitumor effects and specific mechanism of CEP on oral squamous cell carcinoma (OSCC) in vitro and in vivo.

Methods

The anticancer effects of CEP were studied by evaluating cell apoptosis, viability, migration, and invasion of OSCC cells. The epithelial-mesenchymal transition (EMT) related proteins levels were detected using qRT-PCR and western blot methods. Moreover, the N6-methyladenosine (m6A) modification of high mobility histone A2 (HMGA2) was determined by methylated RNA immune-precipitation (MeRIP) assay.

Results

We found that CEP suppressed the proliferation and EMT of OSCC cells. Moreover, CEP treatment increased the expression of METTL14 and suppressed m6A modification of FOXL2. Additionally, Overexpression of METTL14 reversed the effects of CEP and induced the aggressiveness of OSCC cells.

Conclusion

CEP impeded the proliferation and EMT of OSCC cells via m6A-induced inactivation of HMGA2/FOXL2 axis, relieving the carcinogenic behaviors of OSCC.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40360-025-01028-5.

Introduction

Oral squamous cell carcinoma (OSCC) is characterized by morbidity. Approximately, one in 100,000 adults worldwide are diagnosed with oral cavity cancer every year [1, 2]. The recurrence rate of OSCC within 5 years reached 42.3% [3]. The invasion, metastasis, clinicopathological features and inadequate treatment of OSCC are considered to be the reasons for poor prognosis of OSCC patients [4]. The pathogenesis of OSCC is associated with environmental and genetic factors [5, 6]. In epigenetic research, N6-methyladenosine (m6A) methylation modification is the most common form of mRNA modification [7]. At present, it has been confirmed that m6A methylation can lead to the up-regulation of multiple oncogenes [8]. Epigenetic changes are one of the important characteristics of cell carcinogenesis [9]. Therefore, m6A methylation in OSCC cells may be used as an important basis for early diagnosis of OSCC.

Epithelial-mesenchymal transition (EMT) is the primary step of embryonic development, wound healing, and tumor metastasis [10–13]. EMT is mainly manifested by the loss of polarity of epithelial cells, and the decline of tight junctions and adhesion junctions between cells. As a result, epithelial cells have acquired infiltration and migration capabilities, and evolved into cells with mesenchymal cell morphology and characteristics [14, 15]. The forkhead box (FOX) family of proteins are a large group of master transcriptional factors with important roles in chemo-resistance and EMT. FOXC2, FOXC1, FOXQ1, and FOXM1 regulate EMT-associated molecules in different cancers [16–19]. FOXL2, another member of the FOX protein family, directly regulates the metastasis and EMT of chemoresistant gastric cancer [20]. However, the effects of FOXL2 on EMT and malignant phenotypes of OSCC remain unclear. Therefore, to investigate the molecular mechanism of FOXL2 meidated EMT in OSCC can not only provide new research direction for the biological behavior of OSCC metastasis, but also provide potential strategies for the treatment of OSCC.

Cepharanthine (CEP) is a kind of isoquinoline alkaloid extracted from stephania, which has various pharmacological effects [21]. CEP promotes the increase of white blood cells by stimulating the reticuloendothelial system, activating hematopoietic tissue, and promoting the proliferation of bone marrow tissue [22]. Clinically, it is mainly used in the treatment of silicosis and leukopenia [21]. In recent years, the anti-tumor effects of CEP have been extensively explored. A growing number of studies have confirmed that CEP prevents tumor growth in vivo and in vitro [23, 24]. However, the effects of CEP on OSCC remains limited.

The signaling pathway between high mobility histone A2 (HMGA2) and FOXL2 has been shown to play a pivotal role in regulating EMT and metastasis in chemoresistant gastric cancer [20]. Given that HMGA2 is a well-established oncogene in OSCC [25, 26] and its expression is known to be regulated by m6A modification [27], we sought to investigate whether this axis is also critical in OSCC. Therefore, we hypothesized that CEP might exert its antitumor effects by influencing the m6A modification of HMGA2, thereby inactivating the HMGA2/FOXL2 axis and relieving the carcinogenic behaviors of OSCC cells.

Materials and methods

Cell culture and treatment

The human OSCC cell lines CAL27 and UM1 were purchased from Cell Bank of Chinese Academy of Sciences (Shanghai, China). Cells were incubated with Dulbecco’s Modified Eagle Medium: Nutrient Mixture F-12 (DMEM/F12) with 10% (v/v) fetal bovine serum at 37℃ in an incubator with 5% CO₂. OSCC Cells were plated in a 96-well plate, with typically 5,000 cells per well. CEP (S4238, Selleck, Houston, TX, USA) at concentration of 10 µmol/L (LD), 20 µmol/L (MD) and 30 µmol/L (HD) were added into the plate [28]. After 24 h, the culture medium containing CEP was replaced, and the incubation continues for another 24 h.

Cell transfection

A total of 2 × 10⁵ OSCC cells were seeded into six-well cell culture plates in DMEM/F12 free of serum and antibiotics. Eighty pmol of the small interfering RNA (siRNA) targeting METTL14 (5’-AGGAAAGCTATGAGGATACTCTGTT-3’) were transfected into the cells by using jetPRIME reagent (Polyplus-transfection, Shanghai, China) according to the manufacturer’s instructions. Briefly, the siRNAs and jetPRIME reagent were diluted in jetPRIME dilution buffer (Polyplus Co., Illkirch, France) and incubated for 30 min at room temperature. Subsequently, the mixtures, containing the siRNA, were added to each well containing cells in opti‐MEM medium. The cell culture plates were then incubated for 5–7 h at 37 °C in a CO₂ incubator. After that, DMEM/F12 with 20% fetal bovine serum was added into each well with transfected cells. To generate OSCC cells that overexpressed HMGA2, lentiviral vectors (Ubi-CMV-SOCS2-SV40-puro) were purchased from GenePharma (Shanghai, China). The empty vector was used as the negative control. OSCC cells were cultured with lentivirus for HMGA2 transfection and the antibiotic-resistant transfected cells were selected and enriched by treatment of puromycin (Hanyin Biotechnology) at a final concentration of 2 µg/mL in culture medium.

MTT assay

Cells were seeded in a 96-well (5 × 10³ cells/well). After treated with 10 µmol/L, 20 µmol/L and 30 µmol/L (HD Cepharanthine for 24, 48, and 72 h. Afterwards, cells were cultured with 20 µl of 5% MTT solution (Beyotime, China). Subsequently, results were determined using a microplate reader (Invitrogen) at 490 nm. Cell viability inhibition rate (%) = ((A) control group - (A) experimental group) / (A) control group.

Flow cytometry

Cells were plated in 6-well plates (2 × 10⁵ cells/well). Next, cells were digested with trypsin and resuspended by 400 µL binding buffer. Afterwards, cells were stained with 4 µL annexin V-FITC and 4 µL PI using Annexin V-FITC apoptosis detection kits (C1062S, Beyotime, China). Cells were cultured in dark for 15 min. Cell apoptosis was detected by flow cytometry, FL1 (Annexin V-FITC) and FL2 (PI) channels were detected at 488 nm. By gating forward scattering (FSC) and lateral scattering (SSC), cell debris and non-target cell populations can be excluded. FlowJo software was used for data analysis and graph rendering.

Transwell assays

Cells were plated into 96-well plates (5 × 10³ cells/well) in upper chamber precoated with or without Matrigel. The cells in lower chamber were incubated with 500 µL medium. After 24 h, cells in upper chamber were wiped off. The migrated or invaded cells were fixed with 4% polyformaldehyde and stained with 0.5% crystal violet and captured using a microscope.

qRT-PCR

The mRNA expression levels in OSCC cells were determined. Trizol (reagent purchased from Invitrogen company) was used to extract the RNA from cells according to the instructions. Nanodrop1000 was used to determine the purity and concentration of the extracted RNA, and the integrity of the RNA was confirmed by 10 g/L agarose gel electrophoresis. After reverse transcription according to the instructions of PrimeScript™, the cDNA was stored at − 20 ℃. SYBR^® Premix Ex Taq™ Kit (Takara biological company, Dalian, China) was used to performed qRT-PCR. The PCR procedure was: 95 ℃ 30 s, 95 ℃ 5 s, 60 ℃ 31 s, 40 cycles. GAPDH was used as loading control. The relative expression level of the target gene was calculated by the 2^−ΔΔCt method. The primer sequences were used as follows: E-cadherin: Forward 5′-ATTTTTCCCTCGACACCCGAT-3′,  Reverse 5′-TCCCAGGCGTAGACCAAGA-3′;

Vimentin: Forward 5′-GACGCCATCAACACCGAGTT-3′,  Reverse 5′-CTTTGTCGTTGGTTAGCTGGT-3′;

ZEB1: Forward 5′-TTACACCTTTGCATACAGAACCC-3′,  Reverse 5′-TTTACGATTACACCCAGACTGC-3′;

N-cadherin: Forward 5′-AGCCAACCTTAACTGAGGAGT-3′,  Reverse 5′-GGCAAGTTGATTGGAGGGATG-3′;

HMGA2: Forward 5′-ACCCAGGGGAAGACCCAAA-3′,  Reverse 5′-CCTCTTGGCCGTTTTTCTCCA-3′;

FOXL2: Forward 5′-GGTCGCACAGTCAAGGAGC-3′,  Reverse 5′-CGCGATGATGTACTGGTAGATG-3′;

ITGA2: Forward 5′-GGGAATCAGTATTACACAACGGG-3′,  Reverse 5′-CCACAACATCTATGAGGGAAGGG-3′;

FTO: Forward 5′-GCTGCTTATTTCGGGACCTG-3′,  Reverse 5′-AGCCTGGATTACCAATGAGGA-3′;

METTL3: Forward 5′-TTGTCTCCAACCTTCCGTAGT-3′,  Reverse 5′-CCAGATCAGAGAGGTGGTGTAG-3′;

METTL14: Forward 5′-GAACACAGAGCTTAAATCCCCA-3′,  Reverse 5′-TGTCAGCTAAACCTACATCCCTG-3′;

YTHDF1: Forward 5′-ATACCTCACCACCTACGGACA-3′,  Reverse 5′-GTGCTGATAGATGTTGTTCCCC-3′;

YTHDF2: Forward 5′-CCTTAGGTGGAGCCATGATTG-3′,  Reverse 5′-TCTGTGCTACCCAACTTCAGT-3′;

YTHDF3: Forward 5′-GGTGTATTTAGTCAACCTGGGG-3′,  Reverse 5′-AAGAGAACTAGGTGGATAGCCAT-3′;

YTHDC1: Forward 5′-GAGGGCCAAATCTCCTACGC-3′,  Reverse 5′-GTCTCATGGTCAGAGCCATATTC-3′;

YTHDC2: Forward 5′-CAAAACATGCTGTTAGGAGCCT-3′,  Reverse 5′-CCACTTGTCTTGCTCATTTCCC-3′;

GAPDH: Forward 5′-ACAACTTTGGTATCGTGGAAGG-3′,  Reverse 5′-GCCATCACGCCACAGTTTC-3′;

Western blot

Total protein was extracted using RIPA dissolution buffer (Beyotime, China). Protein concentration was detected using BCA kits (Beyotime, China). 60 µg protein was separated using 12% SDS-PAGE. Electrophoresis. The separated protein was moved onto PVDF membranes, which were sealed with 5% skimmed milk. Then the membranes incubated overnight with primary antibody (E-cadherin, 1:500, ab314063; Vimentin, 1:1000, ab92547; ZEB1, 1:1000, ab203829; ZEB2, 1:1000,ab138222; HMGA2, 1:500, ab207301; FOXL2, 1:500, ab246511; ITGA2, 1: 1000, ab181548; GAPDH, 1:2000, ab9485; Abcam). Next, the membranes were incubated with secondary goat-anti-rabbit antibody (1: 5000). The photos were taken by las-4000 system (Tokyo, Japan) and analyzed quantitatively by gel Pro Analyzer 4.0 software.

Methylated RNA immune-precipitation (MeRIP) assay

The RNA was isolated from OSCC cells. The MeRIP assay was conducted with a RiboMeRIP m6A Transcriptome Profiling Kit (Ruibo Biotechnology Co., Ltd, Guangzhou, China). Briefly, the total RNA was mixed with immunoprecipitation (IP) buffer (4 °C, 12 h) and incubated with 30 µL protein A/G agarose. Then the RNA was centrifuged and washed with IP buffer, m6A modified RNA was enriched and analyzed by qRT-PCR. The IgG was selected as the negative control.

Bioinformatic analysis

SRAMP, A sequence-based m6A modification site predictor, was used to predict the potential m6A modification site of HMGA2.

Dual luciferase gene reporter assay

Dual-reporter promoter clones or controls were transfected into UM17 and CAL27 cells. The indicated scrambled and stable METTL14 inhibition cells were transfected with the pEZX-PG04-HMGA2 promoter Gaussia luciferase/secreted alkaline phosphatase. After 24 h, these cells were treated with rapamycin (0.1 µM) for another 24 h. The Secrete-Pair Dual Luminescence Assay kit (GeneCopoeia, SPDA-D010) was used to detect HMGA2 promoter luciferase activity.

Orthotopic mouse model establishment

To establish constitutive expression, a lentivirus carrying the HMGA2 protein coding region was acquired from GeneChem Biotechnology. To achieve constitutive effects, puromycin selection was conducted to obtain stable cell lines. The BALB/c nude mice (6 to 8 weeks old, about 22 g) were obtained from Vital River Laboratories (Beijing, China), and were divided into negative control and stable HMG2 inhibition groups (n = 6 per group). 2 × 10⁶ cells (scrambled CAL27 cells or stable HMG2 inhibition CAL27 cells) were injected into the caudal vein of nude mice. The injected mice were euthanized after 4 weeks. Isolated subcutaneous tumours were cut into pieces (2 mm³) and kept on ice prior to orthotopic implantation surgery. Finally, tumour tissues were fixed in formalin, embedded in paraffin, sectioned and stained with haematoxylin and eosin (H&E). The protein expression of Ki-67 were evaluated using immunohistochemistry (IHC). The Ki-67 primary antibody was obtained from Abcam (1:500, ab15580).

Statistical analysis

The data in current study was analyzed by SPSS 20.0. The results are expressed as mean ± standard error. The difference was evaluated with student t-test and ANOVA. P < 0.05 was interpreted as significant difference.

Results

CEP promoted the apoptosis and decreased the viability Inhibition rate of the OSCC cells

Figure 1A showed molecular structure of CEP. The IC50 of CEP on OSCC cells was evaluated using MTT method, and the results suggeted that the concentration of CEP at about 20 µmol/L had half inhibitory effect on cell viability (Figure S1). Then we explored the effects of CEP on the apoptosis and viability inhibition rate of OSCC cells with flow cytometry and MTT assay. The results showed that CEP significantly promoted the apoptosis (Fig. 1B) and suppressed the proliferation of OSCC cells in a dose-dependent manner (Fig. 1C).

Fig. 1.

Cepharanthine (CEP) promoted the apoptosis of oral squamous cell carcinoma (OSCC) cells. Molecular structure formula of CEP. (B) The apoptosis rate of OSCC cells. (C) MTT assay was performed to measure the inhibition rate of OSCC. *P < 0.05, **P < 0.01, vs. NC group. n = 3

CEP inhibited the viability, migration and invasion of OSCC cells

We further detected the the effects of CEP on cellular functions of OSCC cells. As shown in Fig. 2A, CEP decreased cell viability of OSCC cells in a dose-dependent manner (Fig. 2A-B). This was paralleled with the results from transwell assay. CEP suppressed the migration and invasion ability of OSCC cells (Fig. 2C-D).

Fig. 2.

CEP suppressed the viability, invasion and migration of OSCC cells. B) The viability of OSCC cells. (C-D) The migrated and invaded ability of OSCC cells. *P < 0.05, **P < 0.01, vs. NC group. n = 3

CEP regulated EMT markers in OSCC cells

Next, we determined the expressions of EMT related genes. CEP upregulated E-cadherin, and downregulated Vimentin, ZEB1 and N-cadherin both at mRNA and protien expression (Fig. 3A and B).

Fig. 3.

Effects of CEP on the mRNA and protein expressions of epithelial-mesenchymal transition (EMT) related genes of OSCC cells. qRT-PCR (A) and Western blot (B) were utilized to examine the expressions of E-cadherin, Vimentin, N-cadherin and zinc finger E-box-binding protein 1 (ZEB1) of OSCC cells. *P < 0.05, **P < 0.01, vs. NC group. n = 3

CEP inhibited HMGA2/FOXL2 signaling pathway

The activation of HMGA2/FOXL2 axis is deeply associated with pathogenesis of cancer. Bioinformatics analysis dictated that high expression of HMGA2 or FOXL2 was related to poor survival rates of OSCC patients (Fig. 4A). We therefore, hypothesized that CEP may suppress the development of of OSCC via regulating HMGA2/FOXL2 signaling pathway. As shown in Fig. 4, the mRNA (Fig. 4B) and protein (Fig. 4C) expressions of HMGA2, FOXL2 and ITGA2 were down-regulated by CEP. Thererfore, we selected high-dose CEP (30 µmol/L) for rescue experiments.

Fig. 4.

Effects of CEP on the mRNA and protein expressions of high mobility group AT-hook 2 (HMGA2), L2 forkhead box L2 (FOXL2) and integrin alpha 2 (ITGA2) of OSCC cells. The survival rates of OSCC patients. qRT-PCR (B) and Western blot (C) were utilized to examine the expressions of HMGA2, FOXL2 and ITGA2 of OSCC cells. *P < 0.05, **P < 0.01, vs. NC group. n = 3

Overexpressed HMGA2 alleviated the effects of CEP and induced aggresiveness of OSCC cells

To further verify HMGA2/FOXL2 signaling pathway in the progression of OSCC, cells were transfected with HMGA2 overexpression plasmids. As showed in Fig. 5A, HMGA2 was significantly up-regulated by HMGA2 OE plasmids, suggesting cells were successfully transfected. Overexpression of HMGA2 significantly alleviated the effects of CEP and promoted the cell viability of OSCC cells (Fig. 5B). Moreover, up-regulated HMGA2 significantly promoted the migration and invasion ability of OSCC cells (Fig. 5C and D). In addition, upregulated HMGA2 significantly upregulated the protein expressions of FOXL2, Vimentin, ZEB1 and N-cadherin, and downregulated the E-cadherin protein expression of OSCC cells (Fig. 5E and F).

Fig. 5.

Overexpression of HMGA2 reversed the effects of CEP treatment on the expressions of EMT related genes and metastasis of OSCC cells. (A) The mRNA expression of HMGA2. (B) The viability of OSCC cells. (C-D) The migrated and invaded ability of OSCC cells. (E-F) The mRNA and protein expressions of FOXL2, E-cadherin, Vimentin, N-cadherin and ZEB1 in OSCC cells. **P < 0.01, vs. NC group; #P < 0.05, vs. si-HMGA2 + OE-NC group. n = 3

METTL14 m6A modified HMGA2

To systematically investigate the mechanism by which CEP influences m6A modification, we first screened the expression of key m6A regulatory genes, including writers (METTL3, METTL14), erasers (FTO), and readers (YTHDF1-3, YTHDC1-2). Notably, among all the regulators tested, METTL14 exhibited the most significant upregulation in OSCC cells upon CEP treatment (Fig. 6A). This finding directed our focus towards METTL14 as a primary mediator of CEP’s effects. Since HMGA2 is a critical oncogene in OSCC and a reported target of METTL14-mediated m6A modification, we next explored whether CEP affects the m6A modification of HMGA2. CEP significantly increased m6A level of HMGA2 in OSCC cells (Fig. 6B). The HMGA2 m6A methylated levels were increased by OE-METTL14, and decreased by si-METTL14 (Fig. 6C). Furthermore, knockdown of METTL14 significantly suppressed the effects of CEP and decreased HMGA2 m6A methylated levels (Fig. 6D). Moreover, knockdown of METTL14 significantly antagonized the effects of CEP and up-regulated HMGA2 (Fig. 6E). Knockdown of METTL14 antagonized the decrease in mRNA stability of HMGA2 induced by CEP and suppressed the mRNA stabiltiy of HMGA2 (Fig. 6F). The potential m6A modification site of HGMA2 was then verified using dual luciferase gene reporter assay, and the results suggested that inhibition of METTL14 significantly downregulated the relative luciferase activity of wildtype HMGA2 gene while there was no significant difference between mutant groups (Fig. 6G and H). Afterwards, the effects of METTL14 on regulatory effects of HMGA2 on OSCC cells were investigated. As showed in Fig. 7, inhibition of METTL14 further enhanced the effects of HMGA2 overexpression on cell viability, invasion, and migration (Fig. 7A-G).

Fig. 6.

Methyltransferase like 14 (METTL14) m6A modified HMGA2. (A) The mRNA expression of fat mass and obesity-associated (FTO), methyltransferase like 3 (METTL3), METTL14, YTH N6-methyladenosine RNA binding protein (YTHDF)1–3 and YTH domain containing (YTHDC)1–3 in the SCC25 cells. (B) HMGA2 m6A methylated levels of the SCC25 and CAL27 cells were determined after CEP treatment. (C) HMGA2 m6A methylated levels of the SCC25 and CAL27 cells were determined after si-METTL14 and OE-METTL14 transfection. (D) HMGA2 m6A methylated levels of the SCC25 and CAL27 cells were measured after CEP treatment and si-METTL14 transfection. (E) HMGA2 expression of the SCC25 and CAL27 cells were detected after CEP treatment and si-METTL14 transfection. (F) Changes of HMGA2 expression of the SCC25 cells after CEP treatment and si-METTL14 transfection. (G) The potential m6A modification site of HMGA2 predicted by SRAMP datatbase. (H) Dual luciferase gene reporter assay was performed to verifiy the predcited m6A modification site of HMGA2 in OSCC cells with downregulation of METTL14. *P < 0.05, **P < 0.01, vs. NC group; #P < 0.05, ##<0.01, vs. si-NC group or Cepharanthine group. n = 3

Fig. 7.

Inhibition of METTL14 enhanced the effects of overexpression of HMGA2 on the expressions of EMT related genes and metastasis of OSCC cells. The viability of OSCC cells. (B-C) The migrated and invaded ability of OSCC cells. (D-G) The mRNA and protein expressions of FOXL2, E-cadherin, Vimentin, N-cadherin and ZEB1 in OSCC cells. *P < 0.05, vs. OE-NC + si-NC group; #P < 0.05, vs. OE-HMGA2 group. n = 3

Inhibition of HMGA2 in vivo suppressed tumorigenicity

In addition, a mouse model was established to evaluate the effects of HMGA2 inhibition. We first downregulated the HMGA2 gene in CAL27 cells, and subcutaneously transplanted CAL27 cells with stable downregulation of HMGA2 into nude mice. It was found that inhibition of HMGA2 in vivo significantly reduced the size, volume, and weight of the tumors (Fig. 8A-C). Furthermore, H&E and IHC staining showed that inhibition of HMGA2 relieved the inflammatory infiltration of the tumor, and decreased the expression of Ki-67 (Fig. 8D and E).

Fig. 8.

Inhibition of HMGA2 in vivo suppressed tumorigenicity. Representative images, (B) tumor volume, and (C) tumor weight of tumor tissues. (D) Haematoxylin and eosin (H&E) and (E) immunohistochemistry (IHC) images of tumor tissues. **P < 0.01, vs. sh-NC group. n = 6

Discussion

In this study, we demonstrated that CEP suppresses the proliferation and EMT of OSCC cells by targeting the METTL14/HMGA2/FOXL2 axis. Our rationale for investigating this specific axis was twofold. First, the HMGA2/FOXL2 pathway is a well-documented driver of EMT in multiple cancers, including OSCC, as supported by its prognostic value and prior literature [20, 25, 26]. Second, and more directly, our initial screening of m6A regulators revealed that METTL14 was the most prominently upregulated factor in response to CEP treatment. This observation, coupled with the established role of METTL14 in depositing m6A modifications on oncogenes like HMGA2, provided a compelling mechanistic link. Our subsequent data confirmed that CEP up-regulates METTL14, enhances m6A modification of HMGA2 transcript, reduces its stability, and consequently inactivates the HMGA2/FOXL2 signaling pathway, ultimately suppressing the malignant phenotypes of OSCC.

CEP exhibits a variety of biological activities such as anti-inflammatory, immune regulation, ect. CEP has been widely used in chemoprevention and treatment of a variety of diseases, such as leukopenia [21]. In recent years, the antitumor effect of CEP has been continuously revealed. For example, Anna Kiyomi et al. (2020) found CEP has a significant inhibitory effect on breast cancer cells [23]. In ovarian cancer, CEP promotes the apoptosis of ovarian cancer [24]. In addition, CEP inhibits the proliferation and promoted cell death of human choroidal melanoma cells via inducing the production of ROS and activating c-JNK [29]. In this study, CEP significantly inhibited the agressiveness of OSCC cells, implying that CEP is an promising strategy for OSCC. The potential mechanism was then investigated.

EMT is a process of the degradation of epithelial characteristics and the acquisition of mesenchymal features [10, 30]. In addition, EMT is often accompanied by the change of cytoskeletal protein types, such as the increase of vimentin expression [31]. Furthermore, ZEB proteins, including ZEB1 and ZEB2, are a vital regulatory factor in EMT [32]. Many studies have confirmed that EMT leads to the metastasis and deterioration of tumor [15]. In present study, CEP enhanced epithelial characteristics and suppressed the acquisition of mesenchymal features, manifested by the overexpression of E-cadherin, and decrease in Vimentin, ZEB1 and N-cadherin. These results indicated that CEP may relieve the metastasis of OSCC cells through regulating the EMT development. However, the specific subtype of EMT affected by CEP remains unclear, and should be investigated in the future.

The HMGA2/FOXL2 signaling pathway plays an important role in cancer research, especially in the metastasis and EMT of drug-resistant gastric cancer [20]. In addition, HMGA2, as a non-histone chromosomal protein, is activated in a variety of tumors [33]. High expression of HMGA2 promote EMT, invasion and metastasis of various tumors [20]. FOXL2, a member of Fox transcription factor family, is a key regulator of cell growth, proliferation and differentiation. Dysregulated FOXL2 promotes tumorigenesis [34]. In recent years, Fox families regulate the EMT processes of tumor cells. In our study, HMGA2 and FOXL2 predicted poor prognosis of OSCC. Moreover, CEP suppressed the proliferation and EMT of OSCC cells by downregulating HMGA2 expression, and inactivating HMGA2/FOXL2 pathway. Moreover, inhibition of HMGA2 also suppressed the aggressiveness of OSCC cells in vivo. Thence, activation of HMGA2/FOXL2 axis may play a tumor promoter role in OSCC, which is consistent with previous studies [20, 33, 34]. Moreover, CEP may be a promising drug for blocking HMGA2/FOXL2 axis in OSCC.

m6A motificaiton is closely related to the progression of cancer [8]. The related modifications of m6A are co-regulated by methyltransferase complexes (METTL3 and METTL14), demethylase (FTO) and corresponding readers (YTHDF1/2/3, YTHDC1) [35]. In present study, we found the expression of METTL14 was significantly up-regulated after CEP treatment in OSCC cells. However, the roles of METTL14 varies with cancer and cell types. For instance, knockdown of METTL14 promotes malignant phenotype of glioblastoma by up regulating the expression of oncogenes (ADAM19) and down regulating the expression of tumor suppressor genes (CDKN2A) [36]. However, up-regulation of METTL14 promoted the development of pancreatic cancer through increasing the PERP m6A methylated levels [37]. In this study, CEP up-regulated the HMGA2 m6A methylated levels. In addition, METTL14 mediated m6A modification of HMGA2 suppressed its mRNA stability. CEP downregulated HMGA2 via recruiting METTL14. Therefore, CEP may suppressed the proliferation and EMT of OSCC via regulating METTL14/HMGA2/FOXL2 axis.

Conclusion

To sum up, CEP suppressed the EMT of OSCC cells via METTL14/HMGA2/FOXL2 axis. This findings may provide an alternative for OSCC treatment.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

CEP

Cepharanthine

EMT

Epithelial-mesenchymal transition

FOXL2

L2 forkhead box L2

FTO

Fat mass and obesity-associated

HMGA2

High mobility group AT-hook 2

IP

Immunoprecipitation

ITGA2

Integrin alpha 2

MeRIP

Methylated RNA immune-precipitation

METTL14

Methyltransferase like 14

METTL3

Methyltransferase like 3

MTT

3-(4,5)-dimethylthiahiazo (-z-y1)-3,5-di- phenytetrazoliumromide

NC

Negative control

OE

Overexpression

OSCC

Oral squamous cell carcinoma

qRT-PCR

Quantitative real time polymerase chain reaction

RIPA

Radio-immunoprecipitation assay

siRNA

Small interfering RNA

YTHDC

YTH domain containing

YTHDF

YTH N6-methyladenosine(m6A)RNA binding protein

ZEB1

Zinc finger E-box-binding protein 1

ZEB2

Zinc finger E-box-binding protein 2

Author contributions

YH and JH conceived the study; YH and XZ conducted the experiments; XZ analyzed the data; YH and JH ware a major contributor in writing the manuscript. All authors read and approved the final manuscript.

Funding

This study was supported by Research Project of TCM Bureau of Guangdong Province (Project Name: Study on the mechanism of inhibiting invasion and metastasis of oral squamous cell carcinoma by reducing the m6A modification level of miR-99a-5p by downregulating METTL3, Project Number: 20222084).

Data availability

No datasets were generated or analysed during the current study.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of MDKN Biotechnology Co., Lt. All methods were carried out in accordance with relevant guidelines and regulations.

Consent for publication

All authors approved the final manuscript and the submission to this journal.

Competing interests

The authors declare no competing interests.