Baicalein was obtained from Sigma-Aldrich Chemical Co. (St. Louis, MO, USA). Dulbecco's modified Eagle's medium (DMEM), penicillin/streptomycin, 0.25% trypsin-EDTA, and dimethyl sulphoxide (DMSO) were purchased from Gibco (Grand Island, NY, USA). Anti-Ezrin (1:500) and anti-phosphor-Ezrin (1:500) antibodies were acquired from Abcam (Cambridge, MA, USA). In addition, anti-GAPDH (1:1,000) and secondary antibodies (1:3,000) were obtained from the Cell Signaling Technology (CST; Beverly, MA, USA).

Human OS cell lines MG-63 and Saos-2 were purchased from China Center for Type Culture Collection (Wuhan, China), and human osteoblast cell line hFOB1.19 was obtained from the Chinese Cell Bank of the Chinese Academy of Sciences (Shanghai, China). MG-63 and Saos-2 cells were maintained in complete medium containing DMEM, 10% fetal bovine serum (FBS; Biolnd, Israel), 50 U/ml penicillin/streptomycin in a humidified incubator with 5% CO₂ at 37°C. hFOB1.19 cells were cultured in complete medium consisting of DMEM/F12 (1:1) medium, 10% FBS (Biolnd, Israel), and G418 (0.3 mg/ml) with 5% CO₂ at 34°C. Sebsequently, these cells were digested by 0.25% trypsin-EDTA solution for all in vitro experiments and for cell pass and 27age unless stated otherwise after cells reached at least 80% confluence.

Baicalein was dissolved in DMSO solution to a final concentration of 4 mM. MG-63, Saos-2 and hFOB cells were seeded at density 5,000 cells/well in 96-well plates and cultured for 24 h. Cells were treated with baicalein at different concentrations (0, 25, 50, 75 and 100 µM) at 0 h, and then were cultured for 24, 48 or 72 h. Subsequently, the proliferation assay was performed using a Cell Counting kit-8 (CCK-8; Dojindo Molecular Technologies, Inc., Kumamoto, Japan) according to the manufacturer's instructions. A microplate reader (Thermolex Molecular Device Co.) was utilized to detect the absorbance at 450 nm.

Following maintenance in culture by different treatments, cell apoptosis was assessed using an Annexin V/fluorescein isothiocyanate (FITC) Apoptosis Detection kit I (BD Biosciences, San Jose, CA, USA). The untreated or treated MG-63 and Saos-2 cells were harvested after indicated treatments with 0.25% trypsin-EDTA, and a single cell suspension was prepared. Cells were washed twice with cold PBS, centrifugated at 1,500 rpm for 5 min and resuspended in 1X Binding Buffer so that cell density reached 1×10⁶ cells/ml. 5 µl Annexin V/FITC was added to 300 µl the above resuspended solution and incubated for 10 min, followed by adding 5 µl propidium iodide (PI) for a 10 min incubation at room temperature in the dark. Cell apoptosis were then detected using flow cytometry (BD Biosciences). The results were analyzed by cell quest software (BD Biosciences) to determine the rate of apoptosis in the lower right quadrant.

Cell invasion assay was carried out using Millicell Hanging Cell Culture Insert (24-well, pore size 5 µm; Merck Millipore, Billerica, MA, USA). After treatment, MG-63 and Saos-2 cells resuspended to a density of 5×10⁴ cells/ml in serum-free DMEM medium were transferred into upper chamber of the inserts coated with 30 mg/cm² of Matrigel (BD Biosciences), DMEM medium with 20% FBS were added to the well out of the insert and then incubation for different concentration or time based on experimental program. Ultimately, cells on the upper membrane surface were removed through careful wiped off a cotton swab, and the cells on lower surface were fixed with 95% ethanol for 30 min and stained with 0.2% Crystal Violet solution for 30 min. Five vision fields per chamber were randomly selected and then counted the number of invasion cells under an inverted microscope. The general procedure of the migration assay was performed the same as that of the invasion assay described above, except that Matrigel was not applied.

Total RNAs were extracted from either MG-63 or Saos-2 cells using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc., Waltham, MA, USA) according to the manufacturers' protocol. The purity and concentration of RNAs were detected by a NanoDrop ND-1,000 spectrophotometer (Thermo Fisher Scientific, Inc., Wilmington, DE, USA). The reverse transcription reactions were conducted by a miRcute miRNA Reverse Transcription kit (TianGen Biotech Co., Ltd., Beijing, China) and PrimeScript RT Master Mix (Perfect Real Time; Takara Biotechnology Co., Ltd., Dalian, China). All reactions were performed using a 7500 Real-Time PCR System (Thermo Fisher Scientific, Inc.) in a reaction volume of 20 µl. The levels of miR-183 and Ezrin were normalized to U6 small nuclear RNA and GAPDH, respectively. The relative quantification was calculated by applying the 2^−∆∆Cq method (17). The PCR primers were designed and synthesized by Shanghai Sangon Biological Engineering Technology Co., Ltd., (Shanghai, China) (Table I).

Total proteins were extracted from MG-63 or Saos-2 cells using RIPA Lysis Buffer (Beyotime, Shanghai, China). The protein concentrations were measured via Enhanced BCA Protein Assay kit (Bio-Rad, Richmond, CA, USA). Lysates of Total protein were separated by a 10% sodium dodecyl sulfate-polyacrylamide (SDS-PAGE) gel and transferred to a polyvinylidene fluoride (PVDF) membrane (Merck Millipore). The membrane was blocked by 5% (v/v) skim milk in Tris-buffered saline Tween-20 (TBST) at room temperature for 2 h, followed by incubation with the primary antibodies at 4°C overnight. After washing, the membrane was incubated with HRP-conjugated secondary antibody at room temperature for 1 h. The proteins were visualized using a chemiluminescence method (electrochemiluminescence and western blot detection system; Amersham Biosciences, Foster City, CA, USA). The results were normalized using GAPDH.

MiR-183 mimics/negative controls and inhibitors/negative controls were supplied by RiboBio (Guangzhou, China). The whole DNA fragment of Ezrin was amplified from the OS cells. The obtained products were cloned into recombinant plasmid pcDNA3.1(+)-EGFP to overexpress Ezrin (pcDNA3.1-Ezrin; Invitrogen; Thermo Fisher Scientific, Inc.), with the empty vector as a negative control (pcDNA3.1-Mock). The constructs were identified by DNA sequencing. Both MG-63 and Saos-2 cells were seeded into a 12-well plate at a 1×10⁵/ml density, and then transfected with 20 nM miR-183 mimic/negative controls or 40 nM miR-183 inhibitors/negative controls or pcDNA3.1-Ezrin using Lipofectamine 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions.

A fragment of 3′UTR of Ezrin containing the putative miR-183 binding site was amplified by RT-PCR. The amplified product was then subcloned into a pMIR-REPORT luciferase reporter vector applying restriction sites XhoI (Promega, Madison, WI, USA). The mutated sequences of predicred miR-183 binding located in the 3′UTR of Ezrin mRNA were introduced using the QuikChange kit (Stratagene, La Jolla, CA, USA). All plasmid constructs were verified by DNA sequencing (Biology Engineering Corp., Shanghai, China). Successively, the 20 nM miR-183 mimics/negative controls or 40 nM miR-183 inhibitor/negative controls and 0.1 µg pMIR-REPORT-Ezrin (WT) or pMIR-REPORT-Ezrin (MUT) constructs were co-transfected into the cells using Lipofectamine 2000. After transfection for 48 h, the Dual-Luciferase Reporter Assay System (Promega) was used to detect the luciferase activity. The Renilla luciferase values were regarded as a correction factor, and the firefly/Renilla ratio was obtained. Luciferase activity was averaged from three replicates for each transfection.

All data were presented as the means ± standard deviation (SD), and all of graphs were produced by GraphPad Prism 5.0 Software (Graph Pad Software, Inc., La Jolla, CA, USA). SPSS 22.0 software (IBM Corp., Armonk, NY, USA) was used for all statistical analyses. Results were analyzed using Student's t-test for comparison between two groups or one-way ANOVA followed by the Tukey's test for comparison of multiple groups. P<0.05 was considered to indicate a statistically significant difference.

To investigate the effect of baicalein on OS cell proliferation, cells were treated with various concentrations of baicalein for the indicated time, followed by the CCK-8 assay. Our results showed that baicalein inhibited the proliferation of MG-63 and Saos-2 cells in a concentration- and time-dependent manner (Fig. 1A and B), but did not affect hFOB cell growth (Fig. 1C). Given that 100 µM baicalein had the optimal efficacy, we selected this dose for subsequent experiments.

Given the fact that the inhibition of apoptosis contributed to cell proliferation in various type of tumor cells, next we employed fluorescence activated cell sorter (FACS) analysis via Annexin V-FITC and PI double-staining to evaluate the apoptosis rate of MG-63 and Saos-2 cells exposed to 100 µM baicalein for 48 h. As shown in Fig. 1D, the apoptosis rate of MG-63 and Saos-2 cells was low in control group but significantly elevated in baicalein-treated group, suggesting an inductive effect of baicalein on MG-63 and Saos-2 cell apoptosis.

It is well known that migration and invasion are the two pivotal stages in malignant progression and metastasis for tumors. Thus, the present study further evaluated the effects of baicalein on the aggressive potential of MG-63 and Saos-2 cells using cell migration and invasion assays. After stimulation of MG-63 and Saos-2 cells with 100 µM baicalein for 48 h, the migration ability of cells was measured. In Fig. 2A, we found that the migration ability of MG-63 and Saos-2 cells was obviously inhibited. Aside from migration test, we also employed the cell invasion assay to measure the invasion ability of MG-63 and Saos-2 cells. As shown in Fig. 2B, the similar effect on invasive ability was also observed in parallel invasive test, compared with the control group. Altogether, the above results reveal that baicalein can suppress the migration and invasion ability of MG-63 and Saos-2 cells.

It has been reported that miRNAs are important regulators in the aggressive behavior of tumor, including OS (18). Our previous study has been proposed that Ezrin as a direct target of miR-183 promotes the migration and invasion of OS MG-63 cells through increased N-cadherin and activating ERK signaling (16). However, whether baicalein regulates the activation of miR-183/Ezrin pathway in MG-63 and Saos-2 cells remains unclear. In order to explore this hypothesis, we validated again the relationship between Ezrin and miR-183 in the current research. Treatment of MG-63 and Saos-2 cells with miR-183 mimic/negative control or miR-183 inhibitors/negative control both exerted a higher transfection efficiency (Fig. 3A). The pairing site of Ezrin 3′-UTR with miR-183 were predicted by TargetScan (an open sourced software) (Fig. 3B). A luciferase reporter gene with 3′-UTR of Ezrin downstream (WT-Ezrin), its mutant version (MUT-Ezrin) via the binding site mutagenesis were also constructed (Fig. 3C). As shown in Fig. 3D, the luciferase activity of cells transfected with miR-183 mimic was dramatically decreased in wild type, but not mutant. Furthermore, the RT-qPCR and western blot analyses showed that overexpression of miR-183 noteworthily reduced the levels of Ezrin mRNA and protein compared with control group, whereas an opposite effect appeared in response to miR-183 inhibitor (Fig. 3E and F). Overall, these results demonstrate that miR-183 directly targets the 3′-UTR of Ezrin and negatively regulates its expression.

We firstly detected the contents of miR-183 and Ezrin in MG-63, Saos-2 and hFOB cells, respectively. RT-qPCR showed that miR-183 levels in MG-63 and Saos-2 cells were lower than those in hFOB cells (Fig. 4A). However, the expression of Ezrin mRNA and protein displayed an opposed tendency in above cells (Fig. 4A and B). These data indicate that abnormal expressions of miR-183 and Ezrin might be correlated highly with the pathogenesis and metastasis of OS.

Next, the underlying effects of baicalein on miR-183/Ezrin pathway were investigated using RT-qPCR and western blot in MG-63 and Saos-2 cells treated by 100 µM baicalein for 48 h. As demonstrated in Fig. 4C, baicalein-stimulated MG-63 and Saos-2 cells displayed a significant increase in miR-183 level in comparison with control group. However, the alterations in the mRNA and protein levels of Ezrin presented an opposite tendency in above parallel groups (Fig. 4D and E). In conclusion, these data corroborate that stimulation of baicalein is able to promote the generation of miR-183 and then retard the expression of Ezrin in MG-63 and Saos-2 cells.

As evidenced above experiments, baicalein represses the proliferation and migration and invision abilitiess, and participates in the activation of apoptosis in MG-63 and Saos-2 cells. Meanwhile, it as well negatively regulates the Ezrin expression via upregulating miR-183 level in MG-63 and Saos-2 cells. Therefore, we hypothesized that the miR-183/Ezrin pathway might be responsible for the antitumor effects of baicalein on OS cells. To verify the assumption, MG-63 and Saos-2 cells were co-cultured with baicalein and miR-183 inhibitors or pcDNA3.1-Ezrin, which showed a higher transfection efficiency (Fig. 5A). As expected, miR-183 inhibitor significantly reversed baicalein-induced high apoptosis rate and low proliferation, migration and invasive ability in MG-63 and Saos-2 cells, compared with baicalein group (Fig. 5B-E). Besides, we also found that overexpression of Ezrin by pcDNA3.1-Ezrin exerts similary effect in stumilation of baicalein with MG-63 and Saos-2 cells (Fig. 5B-E). Taken together, these observations indicate that baicalein induces apoptosis and attenuates proliferation and migration and invasion abilities of MG-63 and Saos-2 cells via activating miR-183/Ezrin pathway.