SKOV-3 (p53 negative), OVCAR-3 (mutant type p53) and PA-1 (wild type p53) cells were obtained from the Korean Cell Line Bank, Seoul National University, Seoul, Korea. All cells were cultured in DMEM (Gibco, Rockville, MD), supplemented with 5% heat-inactivated FBS (Fetal Bovine Serum) (Gibco), 0.37% sodium bicarbonate (Sigma, St. Louis, MO), and streptomycin/penicillin (Gibco), at 37℃ in a 5% CO₂ incubator.

For the cell proliferation inhibition assay, EGCG (a kind gift from Dr. Yukihiko Hara of Mutsui Norin Co., Fujied, Japan) was diluted 50% in DMSO and stored at -20℃ before use. For viable cell counting, 5×10³ cells per well were treated with EGCG ranging from 6.25 to 100 µM for various times. Cells were counted using a hemacytometer under a microscope after trypsinization. The cell viability was determined by the trypan blue exclusion assay. From the analysis of three replicates, the typical measurement deviations were observed to be less than±3.0 percent for each assay.

Each cell line was divided into 5×10⁵ cells/100 mm dish plate. After 24 hr incubation, the cells were added with different amounts of EGCG. After 24 or 48 hr, the cells were centrifuged lysis buffer [0.5% SDS, 0.1 M NaCl, 0.01 M EDTA, 10 mM Tris-HCl (pH8.0)], was added to the subsequent cell pellets followed by the addition of 20 µg/ml proteinase K and 10mg/ml RNase A (Sigma). This mixture was incubated for 4 hrs at 56℃. DNA was extracted by phenol/chloroform treatment. Five µg of the extracted DNA was analyzed on a 2% agarose gel, containing EtBr (ethidium bromide) (0.1 µg/ml), and the DNA ladder formation visualized under UV light.

Cell suspensions were treated with several concentrations of EGCG for 1 and 2 days. The cells were then trypsinized, washed twice with phosphate-buffered saline (PBS) and stained with propidium iodide (100 µg/ml in PBS). Samples were then analyzed using FACS (Becton Dickinson, San Jose, CA). For DNA contents, the cell debris and fixation artifacts were gated out and the G₀/G₁, S and G2/M populations quantified using the CellQuest program.

Cell suspensions were centrifuged and resuspended in phosphate-buffered saline (PBS) to a concentration of 5×10⁵ cells/100 mm dish. After treatments with several concentrations of ECGC, the cells were incubated with 5 µl Annexin V-FITC and 10 µl propidium iodide (PI) in the dark at room temperature for 10min, followed by fixation with 2% formaldehyde. The stained cells were analyzed for DNA content by fluorescence-activated cell sorting (FACS) using a FACScan (Becton Dickinson, San Jose, CA). The forward and side scatter gates were set to exclude any dead cells from the analysis; at least 10000 events were collected for each sample.

The total cellular RNA of SKOV-3 was isolated by the acid-guanidinium-thiocyanate-phenol chloroform method, using the Tri-Reagent (Molecular Research, Inc., Cincinnati, OH). The apoptosis pathway GEArray Q Series Human Cell Cycle Gene Array (http://www.superarray.com) was obtained from SuperArray Inc. (Bethesda, MD). This kit determines the differential expression levels of multiple genes involved in a biological pathway. Briefly, the total RNA from the respective samples was used as a template to generate cDNA probes, using the GEAprimer mix as a reverse transcriptase primer. The cDNA probes, which represent an abundance of the mRNA population, were then denatured, and hybridization conducted in GEHybridization solution to two nylon membranes spotted with gene-specific cDNA fragments. The membranes were then washed in 2×SSC and 1% SDS twice for 15 min each, followed by 0.1×SSC and 0.5% SDS twice for 15 min each. The membranes were then exposed to ChemiDoc (BioRad Laboratories, Hercules, CA). The relative expression level of each gene was then determined through the GEArray Analyzer program (Superarray) from duplicate measurements on two separated arrays.

Total RNA was isolated from cells using Tri-reagent (Gibco) and used as templates. Reverse transcription was carried out using RNase (Invitrogen, Carlsbad, California), and performed at 22℃ for 10 min and then at 42℃ for 20 min, using 1.0 µg of RNA per reaction. The primers used in the PCR reactions for the genes analyzed are summarized in Table 1. PCR was performed on the EGCG-treated cell lines under 35 cycles at 94℃ for 1 min, 57℃ for 1 min and 72℃ for 1 min. The amplified products were subjected to electrophoresis on a 2% agarose gel. The gel was then stained with EtBr and photographed. The specific primers (forward: 5'-TGACGGGGTCAC CCACACTGTGCCCATCTA-3' and reverse: 5'-CTAGA AGCATTT-GCGGTGGACGGATGGAGGG-3') were used as control gene β-actin. The amplification reaction involved denaturation at 94℃ for 30 sec and annealing at 72℃ for 30 sec and 72℃ for 45 sec for 20 cycles.

The cells were grown in 6-well culture dishes (5×10⁵) and treated with several concentrations of EGCG. After 24 or 48 hr, the cells were lysed. The protein concentration of the clear supernatant collected by means of centrifugation was evaluated using the BioRad Protein- Assay kit (Bio-Rad, Hercules, CA), which was adjusted to give a final concentration of 2 mg/mL. After the addition of 2-mercaptoethanol (2%), the samples were boiled for 5 minutes and used in the experiments. 40 µg of each protein sample was subjected to electrophoresis for 2h with SDS-PAGE at 10-mA, and the blotting was performed with a Hybond-ECL membrane (Amersham, Uppsala, Sweden) at 100 volts. The blotted membrane was blocked with 5% skimmed milk and reacted with a primary polyclonal antibody (rabbit CDK2, E2F-1, E2F-4, CDK4; Santa Cruz Biotechnology Inc., Santa Cruz, Ca) and a primary monoclonal antibody (mouse P21, Bax, Rb, Cyclin D1, BclXl, PCNA, Actin; Pharmingen, Franklin Lakes, NJ) at 1mg/ml for 4h at room temperature. After washing with Tris-buffered saline, containing 0.1% Tween 20, the membrane was then incubated with horseradish peroxidase-conjugated secondary antibody (Jackson Immunoresearch, Bar Harbor, MA) at a dilution of 1:1000. The protein bands were visualized using an ECL Kit (Amersham, Arlington Heights, IL).

Statistical analysis was performed using ANOVA. The values between the different groups were compared and P values of less than 0.05 considered significant. All the data shown in the study for cell cycle analysis and cell growth inhibition are representative of three independent studies.

To determine the anti-tumor effects of EGCG in the three different ovarian cancer cell lines, SKOV-3, OVCAR-3 cells and PA-1, the cells were treated with EGCG at concentrations ranging from 10 to 100 µM. As shown in Fig. 1, the EGCG showed significant growth inhibitory effects in the ovarian cancer cell lines over the specified treatment times, which were also dose dependent.

The cell cycle arrests in the ovarian cancer cells were also tested. To determine if the EGCG treatment had any effects on cell cycle perturbations, a cell cycle analysis was performed by propidium iodide staining and flow cytometric analysis. As shown in Table 2, the percentage distribution of cells in the different cell cycle phases after treatment with EGCG, showed significant alterations in the cell cycle progression in each cell line, which suggests EGCG induces cell growth arrest in ovarian cancer cells. In the control cells, the percentage distributions of the PA-1 in the different cell cycle phases were different compared to the other cell lines after 24 to 48 hrs of culturing. At each EGCG concentration, the percentage of SKOV-3 cells in the G₁ phase was significantly increased for up to 48 hrs, which suggests the cell cycle was arrested during the G₁ phase. The G₁ phase arrest profiles of EGCG were consistent over time in the OVCAR-3 cells, where the G₁ phase are significantly increased for up to 48 hrs, with the exception of a marked increase in the cell population in the G₂/M phase that was observed for up to 48 hrs. The EGCG treatment resulted in a dramatic increase in the G₁ cell population, even for up to 24 hrs. In contrast, there was a significant increase in the cell population of the G₂/M phase for up to 24 hrs, and the S phase for up to 48 hrs in the PA-1 cells, with a compensatory decrease in the population in phases S and G₂/M. After the EGCG treatment, the cell population in the S phase was especially significantly increased for up to 48 hrs compared to the other cell lines. These data suggest that the EGCG treatment induced cell cycle arrest in the G₁ phase in the SKOV-3 and OVCAR-3 cells, and the time-dependent shift of cell population form the G₂/M to the S phase in PA-1 cells.

The EGCG treatment has been shown to mediate cell death in ovarian cell lines. The uptake of EGCG prevents cells from undergoing proliferation, and increases the efficacy of the antitumor effect in vitro. To check whether the cell death was accompanied by the development of an apoptotic or necrotic process, the phenotypic changes characteristic of apoptotic cells were analyzed and quantified by double staining the ovarian cancer cells with annexin V and propidium iodide (PI). Exponentially growing control cells were also exposed to EGCG treatment, and then harvested at the predetermined times. Figure 2 shows that the cell death increased significantly after EGCG treatment in the SKOV-3 cells. That is, a greater induction of cell death (late apoptosis or necrosis) resulted from EGCG uptake in the SKOV-3 cells for up to 24 hrs at all EGCG concentrations, suggesting later stages of apoptosis and cells undergoing rapid secondary necrosis under culturing. The cell death rates increased sharply during the first day, but showed a slower decrease thereafter, changing the cell death patterns into early apoptosis for up to 48 hr after EGCG treatment. Conversely, a lower sensitivity to EGCG was shown in the OVCAR-3 compared to the SKOV-3 cells, but a distinct population of early apoptotic cells was not shown. Early apoptotic populations of cells were observed in the SKOV-3 and PA-1 cell lines. In contrast, the PA-1 cells underwent early apoptosis for up to 48 hrs. The cells exhibited a much higher susceptibility to early apoptosis in response to EGCG treatment in comparison to later apoptosis or necrosis. As expected, there was no detectable cell death effect with the control cells. These results showed that EGCG induced apoptotic cell death in each ovarian cancer cell line.

To determine the roles of the molecules responsible for EGCG function, the changes in the levels of transcripts regulated by EGCG were studied by macroarray analyses following EGCG treatment of the ovarian cancer cell lines. In the SKOV-3 cells, several transcripts were upregulated or down-regulated after EGCG uptake. The expression profiles were especially more prominent for up to 48hr than those for up to 24hr. The transcripts commonly up- or down-regulated at least two fold by EGCG treatment over time (24, 48 hr) are listed in Table 3. As shown in Tables 3 and 4, the expressions of several transcripts were down-regulated more than 2 fold after 48 h of treatment with EGCG. The p55cdc, cyclin B1 and cyclin A transcripts levels were found to significantly decrease at 48 hr post-EGCG uptake. In particular, during this period, a 43.4-fold decrease in p55cdc was detected, suggesting this was the most sensitivity to EGCG treatment. However, only BCL2-associated X protein was found to be highly upregulated at 48hr, showing a 2.7-fold decrease. In contrast, the expressions of 12 transcripts were down-regulated more than 2 fold by 24 hr of treatment with EGCG. Of these, the transcripts encoding for E2F transcription factor 1 and DP-1 were found to be significantly downregulated by EGCG for 24hr. The cullin 3, E2F transcription factor 2, cyclin B1 and CDC28 protein kinase 2 were also shown to be downregulated after 24h of treatment. In contrast, the cell division cycle 2 and G₁/S-specific cyclin C were only shown to be upregulated for 24 hr. These data suggest that EGCG can influence the gene expression pattern in the SKOV-3 cells. Furthermore, it is likely that these regulated genes might possibly be associated with resistance/sensitivity to EGCG in this cell line.

Northern and Western blot analyses of several proteins associated with growth arrest and/or apoptosis have been performed with SKOV-3 cells uptaken by EGCG at 25 µM for 24 and 48hr. As shown in Figure 3, several up- and down-regulated transcripts confirmed the patterns obtained from the macroarray, showing the consistency of our experiment. Several differentially expressed transcripts, such as p21, Bax and E2F-4, were reported to be similarly expressed with protein expressions, respectively. As shown in Figure 4, the human β-actin analysis showed that the amount of protein applied to each lane was essentially the same. Moreover, only one protein band was detected in each lane, suggesting there was no degradation of the β-actin. The P21 protein was highly expressed after treatment the Bax was upregulated And the BCL-X_L and Proliferating cell nuclear antigen (PCNA) reached the same levels at 24h, and started to significantly decrease at 48 h. A CDK2 band was observed, but this was slightly weaker compared to the other proteins bands. Although both CDK4 and cyclin D1 were detected, the levels of these proteins were found to remain unchanged upon EGCG treatment.