A biflavonoid‐rich extract from Selaginella moellendorffii Hieron. induces apoptosis via STAT3 and Akt/NF‐κB signalling pathways in laryngeal carcinoma

Abstract Selaginella moellendorffii Hieron. (SM), a perennial evergreen plant, has been used in the treatment of acute infectious hepatitis, thoracic and hypochondriac lumbar contusions, systemic oedema and thrombocytopaenia. However, the role of a biflavonoid‐rich extract from SM (SM‐BFRE) in anti‐larynx cancer has rarely been reported. In this study, the in vitro and in vivo anti‐laryngeal cancer activity and potential mechanisms of SM‐BFRE were investigated. An off‐line semipreparative liquid chromatography‐nuclear magnetic resonance protocol was carried out to determine six biflavonoids from SM‐BFRE. In vitro, MTT assay revealed that SM‐BFRE inhibited the proliferation of laryngeal carcinoma cells. A wound healing assay indicated that SM‐BFRE suppressed the migration of laryngeal cancer cells. Hoechst 33 258 and Annexin V‐FITC/PI double staining assays were performed and verified that SM‐BFRE induced apoptosis in laryngeal carcinoma cells. The Hep‐2 bearing nude mouse model confirmed that SM‐BFRE also exhibited anticancer effect in vivo. In addition, Western blot analysis demonstrated that SM‐BFRE exerted its anti‐laryngeal cancer effect by activating the mitochondrial apoptotic pathway and inhibiting STAT3 and Akt/NF‐κB signalling pathways. All results suggested that SM‐BFRE could be considered as a potential chemotherapeutic drug for laryngeal cancer.

30%-40% of them still die from tumour recurrence or metastasis. 6 Therefore, to improve the survival ratio of patients and alleviate the toxic side effects, finding excellent anticancer drugs and designing a more comprehensive treatment plan are of utmost priority.
Numerous studies have shown that natural products are useful in the treatment of cancer and constitute an important area of cancer drug research. 7 For example, paclitaxel is one of the most famous cancer drugs. 8 Approved by the FDA, it is believed to be useful in the treatment of Kaposi's sarcoma, as well as lung, breast and ovarian carcinomas. 9 It has also been reported that camptothecin has a significant inhibitory effect on leukaemia. 10 Several natural products play a vital role in the prevention of cancer and hardly exhibit obvious side effects. 11 Therefore, it might be feasible to combat the human laryngeal cancer using natural products.

| Plant material
The whole parts of SM were purchased from the herb market of Nanning City, Guangxi Zhuang Autonomous Region, China, in August 2018. A voucher specimen (SC0064) is deposited in School of Pharmaceutical Sciences, South-Central University for Nationalities (SCUN), Wuhan, China. The specimen of SM could be found in Figure S1.

| Preparation and Chemical elucidation of SM-BFRE
Dried whole plant of SM (500 g) was frozen with liquid nitrogen and ground with a YB-2000A high speed pulverizer (Yongkang Sufeng Industry and Trade Co., Ltd, Yongkang City, China). The crushed residue was filtered through 200 mesh sieve, and the uniform powder was then extracted sequentially by maceration with 95% EtOH four times (2.5 L each) at room temperature. The solvent was evaporated under reduced pressure to obtain a crude extract (61.5 g). The extract was suspended in warm water and then partitioned successively with petroleum ether (PE), ethyl acetate (EtOAc) and n-butyl alcohol (n-BuOH) to afford PE fraction (4.3 g), EtOAc fraction (9.4 g) and n-BuOH fraction (23.7 g), respectively. The ethyl acetate extract (8 g)  The chemical profiling of SM-BFRE was performed by HPLC-PDA method as previously described. 14 The gradient elution was optimized for analysis with a Phenomenex Gemini-NX C18 HPLC column (5 μm, 4.6 × 250 mm) (acetonitrile in water with 0.1% formic acid). The gradient program was as follows: 0-13 minutes, 40%-70% acetonitrile; 13-13.05 minutes, 70%-90% acetonitrile; 13.05-20 minutes, 90%-95% acetonitrile; and 20-25 minutes, 95% acetonitrile.
The flow rate of the analysis was 1.0 mL/min. The separation of SM-BFRE was performed by the procedure as previously described. 15 100 mg of SM-BFRE was dissolved in 2.0 mL of methanol, and the solution was filtered. 100 μL of SM-BFRE was subjected to a semipreparative HPLC with a Phenomenex Gemini C18 HPLC column (5 μm, 10 × 250 mm) (acetonitrile in water with 0.1% formic acid from 40% to 60% for 16 minutes, from 60% to 90% for next 0.05 minutes, from 90% to 95% for next 8.95 minutes, 95% acetonitrile holding for 5 minutes). The flow rate of the analysis was 4.0 mL/min. A total of 6 peak-based fractions were collected manually, and 20 injections were repeated to yield compounds 1 (52.7 mg), 2 (7.8 mg), 3 (4.7 mg), 4 (2.3 mg), 5 (8.2 mg) and 6 (2.5 mg). Compounds 1-6 were dissolved in DMSO-d6 for NMR with the amount range of 8.5-2.3 mg for NMR tests.

| Cell culture
Human laryngeal carcinoma cell line Hep-2 was purchased from the American Type Culture Collection (Manassas, VA, USA). Human laryngeal cancer cell TU212 and HBE cell (normal laryngeal epithelial cells originated from human) were obtained from the Shanghai Institute of Biochemistry and Cell Biology, Chinese Academy of Sciences (Shanghai, China). The cells were cultured in DMEM supplemented with 10% (v/v) FBS and 1% penicillin/streptomycin. All cells were maintained in 25 cm 2 or 75 cm 2 cell culture flasks and cultured in an incubator containing a humidified atmosphere with 5% CO 2 at 37°C.

| MTT assay
Hep-2, TU212 and HBE cells in logarithmic growth phase were in-

| Wound healing assay
Hep-2 and TU212 cells in logarithmic phase were inoculated at 2 × 10 5 cells/well in 6-well plates and cultured until about 80% fusion was achieved. After treatment with 0, 5, 10 and 20 μg/mL SM-BFRE, wounds were created by scratching the plate surface with a sterile 200 μL pipette tip. Next, under an inverted phase contrast microscope (Soptop ICX41, Ningbo, China), the scratch wound images were photographed by the camera at 20 × magnification following 0, 12 and 24 h of SM-BFRE incubation. The measurements of the wound surface area were calculated by ImageJ software.

| Annexin V-FITC/PI double staining assay
Laryngeal cancer cells in logarithmic phase were inoculated at 2 × 10 5 cells/well in 6-well plates and cultured until about 80% fusion was achieved. After incubation with 0, 10, 30 and 50 μg/mL SM-BFRE for 24 hours, the cells were digested with trypsin without ethylenediaminetetraacetic acid (EDTA) and centrifuged to collect the suspended cells. Next, Annexin V-FITC/PI staining was performed following the protocol described in the kit (BestBio, Shanghai, China). Immediately after staining, the proportion of apoptotic cells was measured by flow cytometry (Guava easyCyte, United States) and qualitatively observed with a fluorescence microscope. The percentage of apoptotic cells was analysed by FlowJo V10 software.

| Protein preparation and Western blot analysis
Hep-2 and TU212 cells in logarithmic phase were inoculated at 6 × 10 5 cells/dish in 100 mm × 20 mm dishes and allowed to grow for 48 hours. After incubation with 0, 10, 30 and 50 μg/mL SM-BFRE for 24 hours, the cells were collected by centrifugation and lysed in radio immunoprecipitation assay (RIPA) lysis buffer (Beyotime) to extract the proteins. The supernatant containing the protein was collected, and the concentration of the protein was measured by the bicinchoninic acid (BCA) kit (Beyotime). After denaturation, the protein was stored in a −80°C freezer until required. Protein extraction from the tumour tissues was performed following the same protocol. Equivalent amounts of the protein (40 µg) were separated by SDS-PAGE gel electrophoresis and transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad). The membranes were blocked by 5% skim milk formulated with tris-buffered saline with 0.5% Tween 20 (TBST) for 2 hours at room temperature. The membranes were then incubated with the primary antibody at 4°C overnight and incubated with the secondary antibody at room temperature for 1-2 hours. The HRP ECL system (Bio-Rad) was used to detect the protein band, and the grey value was calculated by Image Lab software.

| Terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL)
Tumours were embedded in paraffin and cut into 5 μm sections after pre-treatment with cold saline. They were then fixed in a 10% formalin buffered neutral solution. After the sections were stained with the TUNEL assay kits (Roche Biotechnology, Basel, Switzerland), their photographs were taken at 200 × magnification. The brown-yellow staining of the nucleus represented cell apoptosis.

| Immunohistochemical examinations of transplanted tumour tissues (IHC)
Tumours were embedded in paraffin and cut into 5 μm sections.
Their slices were then incubated with the desired primary antibody and visualized by corresponding secondary antibodies conjugated with horseradish peroxidase. After staining and sealing, photographs were taken at 200 × magnification. Positive staining was quantified using ImageJ software.

| Statistical analysis
All data are shown as mean ± SD from at least three independent experiments. GraphPad Prism 5.0 software was used for all statistical analyses. Statistical differences were analysed by one-way analysis of variance (ANOVA), and P values < .05 were considered significant (*P < .05, **P < .01 or ***P < .001).

| Chemical characterization of SM-BFRE
A HPLC-PDA method was carried out for chemical analysis of the main chemical constituents of SM-BFRE ( Figure 1A). A semipreparative HPLC chromatogram was used for large-scale purification of the main chemical principles with the optimized gradient conditions ( Figure 1B). Six peaks were collected, and 20 injections were repeated to give 6 pure compounds. The six compounds were elucidated as amentoflavone (1), 16 ginkgetin (2), 16 hinokiflavone (3), 17 podocarpusflavone A (4), 18 bilobetin (5) Figure 2D). The cell viability and IC 50 of two laryngeal cancer cells after treatment with 6 pure compounds for 12, 24 and 48 hours could be found in Figure S2. In order to verify that the activity of SM-BFRE was mainly contributed by 6 major compounds and the minor compounds had almost no effect on the anti-laryngeal cancer activity of the extract, we used a semipreparative HPLC to prepare the mixture of 6 major compounds and the mixture of minor compounds in SM-BFRE. SM-BFRE, the mixture of 6 major compounds, and the mixture of minor compounds were all diluted with DMEM to 100 μg/mL and their inhibitory rate on the two laryngeal cancer cells within 24 hours was tested by MTT assay. The results are shown in Table S1 in the Supplement Material. It showed that the inhibitory rate of the mixture of 6 major compounds on Hep-2 and TU212 cells within 24 hours accounted for 95.17% and 93.92%, respectively, of SM-BFRE under the same condition. The inhibitory rate of the mixture of minor compounds on Hep-2 and TU212 cells was 3.24% and 5.26% of SM-BFRE, respectively. It suggested that the activity of the extract was mainly contributed by 6 major compounds.

| Effect of SM-BFRE on the migration of laryngeal carcinoma cells
The effect of SM-BFRE on laryngeal cancer cells was investigated through wound healing assay. When the SM-BFRE concentration was more than 30 μg/mL, the proliferation of two laryngeal carcinoma cells was significantly inhibited. Therefore, SM-BFRE concen-

| Regulation of STAT3 and Akt/NF-κB signalling pathway by SM-BFRE in vitro and in vivo
Signal transduction and activator of transcription 3 (STAT3) is highly phosphorylated in almost all head and neck tumours. 21 Studies have shown that STAT3 can regulate the genes that inhibit the apoptotic pathway and ultimately contribute to the development of tumours. 22 In vitro and in vivo Western blot analysis showed that SM-BFRE dose-dependently inhibited the expression of p-STAT3 (Tyr705), but did not inhibit the protein level of total STAT3 (Figure 6A, B).
This suggested that SM-BFRE might induce the apoptosis of laryngeal cancer cells in vitro and in vivo through the inhibition of STAT3 signalling.
In cancer cells, nuclear factor κB (NF-κB) can be regulated by the PI3K/Akt signalling pathway and eventually enters the nucleus and inhibits tumour cells apoptosis. 23 Western blot assay in vitro and in vivo results showed that SM-BFRE dose-dependently inhibited the expression of NF-κB and p-Akt (Ser473), but did not inhibit the protein level of total Akt ( Figure 6C, D). This suggested that SM-BFRE induced the apoptosis of laryngeal cancer cells through the inhibition of Akt/ NF-κB signalling in vitro and in vivo.

| D ISCUSS I ON
Laryngeal carcinoma is a common type of tumour in the respiratory system. In the past 40 years, the five-year survival ratio of sufferers has decreased from 66% to 63%. 24 However, the current treatment of laryngeal cancer makes its prognosis unsatisfactory and there is an urgent need for finding more effective and feasible methods to treat laryngeal cancer.
Increasingly, studies have shown that natural products derived from plants play a crucial role in the treatment of malignant tumours. 25 Such natural products have the advantage of killing cancer cells directly, in addition to less toxicity and side effects, and prolonging the life span of patients. 26 Biflavonoids have been found to possess extensive pharmacological activity and are the main material basis of the anti-tumour effect of Selaginella plants. 27,28 The present research indicated that SM-BFRE suppressed the multiplication of laryngeal carcinoma cells in a dose-and time-dependent manner. However, it had no obvious toxic effects on normal laryngeal epithelial cells (Figure 2).

Metastasis of cancer cells is a critical step in the development
of most malignant tumours, which represents a poor prognosis for patients and leads to a more advanced stage of cancer. 29 Tumour cell migration and invasion are the initial stages of carcinoma metastasis. 30 Through the scratch assay, treatment with SM-BFRE significantly reduced the wound healing ability of laryngeal carcinoma cell lines in a dose-and time-dependent manner (Figure 3), suggesting that the inhibitory ability of SM-BFRE for the metastasis of laryngeal cancer was positive and obvious.
Programmed cell death (PCD), commonly known as apoptosis, is of vital importance in clearing damaged cells and maintaining cellular homeostasis, and its disorders may lead to cancer. 31 Apoptosis is accompanied by some typical features, such as cell shrinkage and F I G U R E 4 Effects of SM-BFRE on apoptosis of laryngeal carcinoma cells. (A) Changes in two laryngeal carcinoma cells incubated with SM-BFRE (0, 10, 30 and 50 μg/mL) for 24 h and stained with Hoechst 3328 under a fluorescence microscope. (B) Hep-2 and TU212 cells were incubated with SM-BFRE (0, 10, 30 and 50 μg/mL) for 24 h, cells were stained with Annexin V-FITC and PI, and the proportion of apoptosis was detected by the flow cytometry. (C) Apoptosis of laryngeal carcinoma cells incubated with SM-BFRE (0 and 50 μg/mL) was observed under a fluorescence microscope after detection by the flow cytometry. (D) The relative expression of Bcl-2/Bax, cleaved caspase-9 and caspase-3 and PARP was detected by Western blot assay of two laryngeal carcinoma cells incubated with SM-BFRE (0, 10, 30 and 50 μg/mL) for 24 h. β-actin was used as a control. *P < .05, **P < .01 or ***P < .001 compare to control (0 μg/mL served as control) floating, nuclear fragmentation, chromatin condensation and production of apoptotic bodies. 32,33 In this study, Hoechst 33 258 and Annexin V-FITC/PI double staining assay indicated that SM-BFRE treatment dose-dependently resulted in the apoptosis of laryngeal carcinoma cells (Figure 4A-C).
The two basic apoptotic signalling pathways are the extrinsic and intrinsic pathways. 29 Activation of caspase by cell surface death receptors initiates the extrinsic pathway in response to homologous death ligands. 34 The intrinsic apoptotic pathway (the mitochondrial pathway) is regulated by Bcl-2 protein family (including anti-apoptotic protein Bcl-2 and pro-apoptotic protein Bax). 32,35 A ratio of Bcl-2/Bax was usually use as the basis for determining apoptosis, and an increased ratio represents anti-cell apoptosis. 36 The Bcl-2 family controls the release of cytochrome c (Cyt-c) by modulating the mitochondrial membrane permeability. 29 Caspases are a group of proteases that have a similar structure in the cyto- in an inactive state. When stimulated, the IκB protein is degraded by phosphorylation and ubiquitination, and the NF-κB dimer is released and finally transferred to the nucleus, which binds to the target gene to promote its transcription. Studies have indicated that the activation of the PI3K/Akt signalling pathway is also involved in the activation of NF-κB. 42 Akt has two phosphorylation sites (Thr308 and Ser473) that are activated by dual phosphorylation. Phosphorylation of the Thr308 activates the Akt moiety, but the entirety of the Akt functional activity requires phosphorylation of Ser473. 43 The decrease in phosphorylated Akt (p-Akt Ser473) expression is a key indicator of the activation of PI3K/Akt pathway. As mentioned earlier, NF-κB has a regulatory effect on apoptosis, and NF-κB is generally thought to inhibit cell apoptosis. Both in vitro and in vivo results were consistent with our conjecture that SM-BFRE-induced apoptosis in laryngeal carcinoma cells might be attributed to the inhibition of the Akt/NF-κB signalling pathway ( Figure 6C, D).
In summary, HPLC analysis revealed that SM-BFRE contains a large amount of biflavonoids that might play a major role in the anti-laryngeal cancer effects of SM-BFRE. Pharmacological studies showed that SM-BFRE significantly inhibited the proliferation and migration of laryngeal cancer cell lines with no obvious toxicity towards normal laryngeal epithelial cells. Experiments conducted in vitro and in vivo demonstrated that SM-BFRE could induce apoptosis in laryngeal carcinoma cells, and further verified that its anti-laryngeal cancer effect that involved activation of the mitochondrial apoptotic pathway and inhibition of the STAT3 and Akt/NF-κB signalling pathways (Figure 7). The results of our current study suggest that SM-BFRE can be considered as a potential chemotherapeutic drug for laryngeal carcinoma.

ACK N OWLED G EM ENTS
The work was financially supported by National Natural Science Foundation of China grant (81774000), the Fundamental Research F I G U R E 5 Inhibitory effect of SM-BFRE on the growth of transplanted tumours and the effect on apoptosis in vivo. (A and B) After Hep-2-bearing mice were treated with SM-BFRE (0, 45 and 90 mg/kg/day) for 4 weeks, the size of tumours in vivo was measured, and then tumours were removed. (C) The volume of the transplanted tumours was measured with a vernier calliper every 4 days during the experiment, and when the experiment was over, the tumours were removed and weighted. (D and E) Tumour sections were prepared and subjected to TUNEL and IHC staining, and the proportion of apoptotic cells was calculated. (F) Proteins were extracted from transplanted tumour tissues, and the relative expressions of Bcl-2/Bax, cleaved caspase-9 and caspase-3 and PARP were detected by Western blot. β-actin was used as a control. *P < .05, **P < .01 or ***P < .001 compare to control (0 μg/mL or 0 mg/kg served as control)

CO N FLI C T O F I NTE R E S T
The authors confirm that there are no conflicts of interest.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

R E FE R E N C E S
F I G U R E 6 Regulation of STAT3 and Akt/NF-κB signalling pathway by SM-BFRE in vitro and in vivo. (A) The relative expression of p-STAT3 (Tyr705) was detected by Western blot in two laryngeal carcinoma cells incubated with SM-BFRE (0, 10, 30 and 50 μg/mL) for 24 h. (B) Proteins were extracted from transplanted tumour tissues, and the relative expression of p-STAT3 (Tyr705) was detected by Western blot analysis. (C) The relative expression of p-Akt (Ser473) and NF-κB was detected by Western blot in two laryngeal carcinoma cells incubated with SM-BFRE (0, 10, 30 and 50 μg/mL) for 24 h. (D) Proteins were extracted from transplanted tumour tissues, and the relative expression of p-Akt (Ser473) and NF-κB was detected by Western blot analysis. β-actin was used as a control. *P < .05, **P < .01 or ***P < .001 compare to control (0 μg/mL or 0 mg/kg served as control) F I G U R E 7 A schematic summary for the mechanisms of SM-BFRE-induced laryngeal carcinoma cells apoptosis in the present study