Roswell Park Memorial Institute medium-1640 (RPMI-1640), 10% fetal bovine serum (FBS) and trypsin-EDTA were from Gibco-BRL (Grand Island, NY, USA). Jak2, p-Jak2 (Y1007/1008), p-STAT3 (Y705) and p-STAT3 (S727) antibodies were from Cell Signaling (Cell Signaling Technologies, MA, USA). STAT3, Bax, Bcl-2, Bcl-X_L, TATA binding protein (TBP), Caspase-3, cytochrome c, β-actin antibodies and secondary antibody (rabbit, goat anti-mouse IgG-horseradish peroxidase) were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The enhanced chemiluminescence plus (ECL Plus) detection kit, RT-PCR Premix kits, oligo(dT), Bcl-2, Bcl-X_L, Bax and 18S primer for RT-PCR were from Bioneer (Daejeon, Korea). DiOC₆ was from Sigma (St. Louis, MO, USA). Mitochondria isolation kit and Coomassie (Bradford) protein assay kit were from Thermo Scientific (Thermo Scientific, MA, USA). Restore™ Western Blot Stripping Buffer and NE-PER kits were from Pierce (Rockford, IL, USA). RNeasy mini kit, and the Qiaprep spin miniprep kits were from Qiagen (Hilden, Germany). The electrophoretic mobility shift assay (EMSA) kit and oligonucleotide probes (STAT3) were from Panomics (Redwood City, CA, USA). Vybrant FAM poly-caspases assay kit was from Molecular probes (Eugene, OR, USA) and CaspGLOW™ fluorescein active caspase-3 staining kit was from eBioscience (San Diego, CA, USA).

YD-38 cell lines were cultured and maintained in RPMI-1640 medium containing 10% serum and 1% penicillin/streptavidine, respectively. Unless otherwise specified, cells were grown in 10-cm dishes to ~80% confluence before being placed in serum-free media for 18–24 h. Serum-deprived cells were treated as specified in the figure legends.

Cell proliferation was analyzed using the crystal violet assay. The YD-38 cells were seeded on to 6-well plates and incubated overnight at ambient condition. After 24-h incubation, the cells were treated with increasing concentration of TA (20–100 μM) for 24 or 48 h. The cells were washed with PBS and incubated with crystal violet. Excess amount of crystal violet was washed off with water and the dye captured by the cells were dissolved using 1% SDS. The final colour formed was measured colorimetrically at 570 nm.

The DNA content of TA or other chemical combinations treated and non-treated YD-38 cells were determined by BD Cycletest Plus DNA reagent kit (BD Biosciences, CA, USA) following the manufacturer's protocol. Briefly, ~5×10⁵ cells were induced, or not induced with TA or other chemical combinations for indicated time. The cells were separated, washed twice with PBS and permeabilized using trypsin buffer. The RNA interaction with PI was neutralized by treating the cells with trypsin inhibitor and RNase buffer. These samples were then stained with propidium iodide for 30 min in the dark and analyzed using FACSCalibur (BD FACSCalibur, CA, USA).

Fluorescein-conjugated Annexin V (Annexin V-FITC) was used to quantify the percentage of cells undergoing apoptosis. The necrotic cells were counter stained with propidium iodide (PI). The cells treated or not were washed twice with cold PBS and resuspended in the binding buffer at a concentration of 1×10⁶ cells/ml. Five microliters each of Annexin V-FITC and PI were added to the cell suspension. After incubation at room temperature in the dark for 15 min, the percentages of apoptotic cells were analyzed by flow cytometry (BD FACSCalibur). Cells treated with 10 μM camptothecin served as positive control.

The YD-38 cells were treated with TA for determined times and lysed on ice with radioimmunoprecipitation assay (RIPA) lysis buffer, containing 1X BD baculogold protease inhibitor cocktail (BD Bioscience) and 1X PhosSTOP phosphatase inhibitors (Roche, NJ, USA). Protein concentrations were detected using the Bradford method. Equal amounts of protein obtained by total lysis were separated using SDS-PAGE and blotted onto a nitrocellulose membrane. Blocking was done with either 5% skim milk or BSA in TBS-T buffer. The membranes were probed with primary antibodies followed by specific HRP conjugated secondary antibodies. Antibody detection was done by using enhanced chemiluminescence (ECL) plus detection kit.

Total RNA from YD-38 cells were prepared using RNeasy Mini kit (Quiagen, CA, USA) according to the manufacturer's instructions. Equal amount of RNA from each sample reverse transcribed using AccuPower RT-premix kit (Bioneer, Korea) and oligo(dT) primers. PCR was performed using 2 μl of the reverse transcription product. The PCR reactions were performed in 25–30 cycles of denaturation 94–95°C, annealing 56–60°C and an extension of 72°C. The primers used for the amplification are listed in the Table I. After amplification, the products were visualized in 1.2% agarose containing ethidium bromide.

STAT3 DNA binding activity was detected using EMSA (19). Gingival cancer cells were grown to ~80% confluence and nuclear protein extracts were prepared using the Nuclear Extraction kit (Affymetrix, CA, USA). EMSA was performed with EMSA gel shift kit (Panomics) according to the manufacturer's protocol. Briefly, the nuclear proteins prepared were subjected for hybridization with a double-stranded, biotin-labeled oligonucleotide probe containing the consensus-binding site for STAT3 (sense strand, 5′-CATGTTATGCATATTCCTGTAAGTG-3′). The protein-DNA complexes were resolved in a 6% non-denaturing PAGE gel and transferred to Pall Biodyne B nylon membrane (Pall Life Sciences, NY, USA) and detected using streptavidin-HRP and a chemiluminescent substrate.

Changes of mitochondrial membrane potential (ΔΨm) were measured by DiOC₆ staining method. Briefly, gingival cancer cell lines treated or not treated with TA were washed and suspended in 0.1 μM DiOC₆ solution. Cells were then incubated at 37°C for 20 min and washed with pre-warmed DPBS and analysed using FACSCalibur.

Mitochrondria from TA-treated and non-treated cells were isolated using mitochondria isolation kit (Thermo scientific, USA) following the manufacturer's protocol. Briefly, 2×10⁶ cells were treated with mitochondria isolation reagent and incubated on ice. Following incubation, Reagent B and C were added with mixing and incubation on ice between each addition. The mixture was centrifuged at 700 × g for 10 min and the supernatant was collected and re-centrifuged. The supernatant was collected as the cytosol and the mitochondrial pellet obtained was washed with Reagent C and used for downstream applications.

Activation of caspases were studied using Vybrant FAM poly-caspase assay kit, following the manufacturer's protocol. Briefly, gingival cancer cells treated or non-treated with TA or other chemical combinations were suspended at a concentration of 1×10⁶ cells/ml culture media. From this 300 μl was mixed with 10 μl 30X FLICA and incubated 1 h at 37°C and 5% CO₂. Following this, the cells were washed multiple times using 1X wash buffer. The cells were then analyzed using FACSCalibur.

Caspase-3 activation studies using CaspGLOW™ fluorescein active caspase-3 staining kit, following the manufacturer's protocol. Apoptosis is induced using 60 μM TA. In order to confirm the role of caspase-3 in TA mediated apoptosis, the cells were pre-treated with Z-VAD-FMK. Following this, 300 μl of cell suspension containing 1×10⁶ cells/ml was made and active caspases-3 stained using FITC-DEVD-FMK. The mixture was incubated at 37°C with 5% CO₂ for 1 h. Then the cells were washed with 1X wash buffer and subjected for FACS analysis.

All experiments were repeated at least three times and the results expressed as mean ± SEM. Statistical analysis was performed with ANOVA and Student's t-test of SAS 9.3 program. One-way analysis of variance (ANOVA) was performed with Duncan's multiple range test. P<0.05 was considered statistically significant.

The ability of tannic acid to inhibit the proliferation of YD-38 cells were studied at concentrations ranging from 20 to 100 μM. TA inhibited the cell viability of YD-38 cells with IC₅₀ values ranging from 50 to 70 μM/l for 48-h treatment (Fig. 1). The effects of TA on cell viability occurred very slowly. Following 48-h treatment, 60 μM TA decreased cell viability by 50%. Analysis of cell viability showed, TA did not induced cell death up to 24 h, rather, it inhibited the proliferation by inducing cell cycle arrest.

Studies conducted to investigate the role of TA on gingival cancer cell proliferation showed a prominent growth inhibition (Fig. 1). In order to uncover the mechanism of growth arrest, gingival cancer cells were treated with TA at different concentrations (40 and 60 μM). Following TA treatment, the cells were stained using PI and the distribution of nucleous analyzed using FACSCalibur. The study revealed that, in cells treated with 40 μM TA, there is an accumulation of cells in the G1 phase with a decrease in percentage of cell population on the G2 phase (Fig. 1B). In case of 60 μM TA treated cells (61%; ^***P<0.001), the percentage accumulation of cells in the G1 phase was comparatively higher than that of the 40 μM treated cells (51%; ^**P<0.01), showing a concentration-dependent G1 phase arrest on the gingival cancer cells (Fig. 1C). Based on the ability of this concentration to induce cell cycle arrest, the concentration was used for the subsequent experiments to elucidate the signaling events involved in TA mediated G1 arrest.

We next focused on whether TA has the capacity to induce apoptosis in gingival cancer cells as it does in breast and AML cells (21). The study revealed that, in cells treated with TA, there was an accumulation of cells in the apoptotic phase (Fig. 1D). In case of 60 μM TA-treated cells (27.30%, Fig. 1E; ^***P<0.001), the percentage of apoptosis was similar to that of the positive control, camptothecin 10 μM treated cells (27.91%, Fig. 1E; ^***P<0.001), confirming induction of apoptosis in the gingival cancer cells.

The role of STAT3 in the induction of apoptosis was reported previously (15). In our study also, TA inhibited the expression as well as phosphorylation of STAT3 (Fig. 2A). In the cytoplasm, STAT3 is phosphorylated by upstream kinases Jak2. Following phosphorylation it forms homo- or hetero-dimers and translocate to nucleus there it controls transcriptional functions of multiple genes. Western blot analysis of whole cell extracts showed a reduction in tyr705 phosphorylated STAT3 together with total STAT3. Phosphorylation of STAT3 is primarily relying on Jak2 phosphorylation. Hence the Jak2 (Y1007/Y1008) phosphorylation levels were checked in these cell lines. As expected, TA regulated Jak2 phosphorylation (Fig. 2A). The relative expression analysis showed a concentration-dependent and significant inhibition on the Jak2, STAT3 expression as well as their phosphorylation (Fig. 2B; ^**P<0.01 and ^***P<0.001).

STAT nuclear translocation and DNA binding activities are influenced by STAT tyrosine phosphorylation rather than serine phosphorylation (12,13). Translocation of initiated STATs to the nucleus follows its binding to a specific response elements in the target gene promoters, and transcriptionally activates the genes. As shown in Fig. 2C, there was a decrease in the nuclear level of pSTAT3 in TA-treated cells, when compared with the control cells. The transcriptional functions of STAT3 is dependent on its ability to bind with specific response element in the target genes. EMSA analysis specific to the STAT3-TF showed a decline in DNA binding activity upon TA treatment (Fig. 2D).

The treatment of gingival cancer cells with TA induced G1 phase arrest. The inhibition of G1/S phase transition is primarily dependent on the p21^Waf1/Cip1 and p27^Kip levels (22) and its loss leads to uncontrolled cell proliferation. Our studies revealed that, treatment with TA intensified the expression of p21^Waf1/Cip1 and p27^Kip transcriptionally (Fig. 3A). The elevation in the expression followed a dose-depended pattern and after a period of 24 h the increase was significant (Fig. 3B; ^***P<0.001). Moreover, the activation of p21^Waf1/Cip1 and p27^Kip showed a statistically significant pattern.

Members of cyclins and CDKs are important mediators of the cell cycle. The translational level expression of cyclin D1, cyclin E and CDK-4 were concentration-dependently inhibited by TA (Fig. 3C). The level of CDK-4 was inhibited at protein level (Fig. 3D; ^***P<0.001) but non-significantly inhibited at transcriptional level (Fig. 3B).

RT-PCR studies were carried out in YD-38 cells treated with increasing concentrations of TA by random priming of total RNA. Bcl-2 and Bcl-X_L was amplified using gene specific primers. As TA declined the STAT3 and pSTAT3 levels and their DNA binding activity, it also downregulated the STAT3 target gene products such as Bcl-2, and Bcl-X_L (Fig. 4A). By modulating these molecules, TA can target anti-apoptotic mediators and induce apoptosis.

Western blotting studies were carried out in YD-38 cells treated with increasing concentrations of TA. The whole cell lysates were prepared and subjected for the detection of Bcl-2 and Bcl-X_L. TA exposure led to downregulation of STAT3 target gene products, Bcl-2, and Bcl-X_L (Fig. 4B). The p53 expression was activated by the treatment with TA (Fig. 4B). The inhibition of anti-apoptotic genes and activation of p53 gene were concentration-dependent and statistically significant (Fig. 4C; ^***P<0.001).

Mitochondria were isolated from the TA-treated and non-treated gingival cancer cells and the cytosol fraction was collected. Western blot analysis of the isolated mitochondria showed inhibition in the level of both Bcl-2 and Bcl-X_L (Fig. 4D). Increase in mitochondrial localization of Bax was observed with exposure to TA (Fig. 4D) with increased Bax in the cytosol. Expression levels of Bax also found increased at transcriptional level (Fig. 4A). Similarly, the mitochondrial localization Bcl-2 and Bcl-X_L decreased. Whole cell lysates showed decreased mitochondrial pore factors confirming our previous findings of TA on these factors.

Whole cell lysates were prepared from TA-treated and non-treated YD-38 cells and subjected for western blot analysis. We found a concentration-dependent increase on the cytochrome c level (Fig. 5A). In order to find the localization of cytochrome c and release of cytochrome c to the cytosol, mitochondrial and cytosolic fractions of YD-38 cells were prepared from TA-treated and non-treated cells and subjected for western blot analysis. In which, we detected a decrease of cytochrome c in mitochondrial fractions and increased levels on the cytosolic fractions (Fig. 5B) indicating the release of cytochrome c to the cytosol and a loss of mitochondrial membrane potential.

The release of cytochrome c from mitochondria is also usually preceded or accompanied by a reduction in the ΔΨm. To address whether the TA-induced alteration in pore factors were associated with the change of ΔΨm, gingival cancer cells were treated with TA for pre-determined time and stained with DiOC₆ to access ΔΨm (Fig. 5C). Treatment of TA reduced the ΔΨm significantly (Fig. 5D; ^***P<0.001), showing that the mechanism of apoptosis induction by TA was through the mitochondria-dependent pathway in YD-38 cells.

Mitochondrial apoptotic pathways require active caspases to ensure apoptosis. Whole caspase activation on TA challenged cells were analyzed using poly-caspase assay kit, and showed a prominent increase in caspase activation (Fig. 6A). After observing a significant increase in whole caspase (Fig. 6B), we analyzed the activation of caspase-3. Western blot analysis of TA-treated YD-38 cells showed cleaved form of caspase-3 confirming the role of caspases in TA-induced apoptosis (Fig. 6C). In-order to confirm the role of caspase-3 in TA-induced apoptosis, cells were treated with the caspase specific inhibitor Z-VAD-FMK prior to TA treatment. The results showed an increase in active caspase-3 in TA-treated cells comparing to the non-treated control cells (Fig. 6D). Moreover, inhibition of caspase-3 prior to TA treatment inhibited apoptosis and caspase-3 activation significantly (Fig. 6E). These data confirmed that TA induces caspase-dependent apoptosis and activation of caspase-3 is an essential step for apoptosis induced by TA.