Paired isogenic cisplatin-sensitive human ovarian cancer cell line A2780 and its cisplatin-resistant clone C200 were used in this study (23). As the control cells an immortalized normal human ovarian epithelial cell line (IOSE) was used. All cells were generously provided by Dr S. Ramakrishnan (Minnesota University). The cancer cells were maintained in RPMI-1640 supplemented with 10% fetal bovine serum (FBS, Gibco), insulin and 1% penicillin/streptomycin at 37°C in a humidified 5% CO₂ atmosphere. IOSE was maintained in 50% M199, 50% MCDB105 medium supplemented with 5% FBS and 0.1% NaHCO₂. Primary antibodies against eIF4E (C46H6, rabbit mAb, #9742), eIF4G (C65H5, rabbit mAb, #9742), 4E-BP₁ (53H11, Rabbit mAb, #9644), PARP (46D11, rabbit mAb, #9532) and goat peroxidase (HRP)-conjugated secondary antibodies (#7074) and antibody against β-actin (D6A8, rabbit mAb, #8457) were all purchased from Cell Signaling Technology.

eIF4E/eIF4G interaction inhibitor 4EGI-1 was purchased from Selleck Chemicals (Selleck China, S7369). 4egi-1 was dissolved in DMSO at concentration of 10 mM, and stored at −80°C. Cisplatin was purchased from Sigma-Aldrich (P4394), and was dissolved in 0.9% NaCl solution to make a 1 mM stock solution. Chemiluminescence substrate was obtained from Pierce. The DC Bio-Rad protein quantization reagents were from Bio-Rad.

A total of 5×10⁶ cells were lysed in 300 μl lysis buffer. The supernatants were incubated with 20 μl Sepharose beads (Amersham Pharmacia Biotech) at 4°C for 1 h, and washed twice in 1 ml PBS buffer. Also, 10 μl of m⁷-GTP-Sepharose beads were washed with 500 μl of PBS buffer three times. Then m⁷-GTP-Sepharose beads were added into the cell extracts pre-cleared with Sepharose beads, and rotated overnight at 4°C. The beads were washed three times with PBS buffer. Then beads were denatured, and the supernatants were resuspended in Laemmli sample buffer with 2% β-ME and were resolved on SDS-PAGE for western blot analysis.

The pGIPZ shRNA plasmid and recombinant pGIPZ shRNA plasmid targeting eIF4E and non-silencing negative control shRNA plasmid were kindly provided by Dr S. Ramakrishnan (Minnesota University). DNA template oligonucleotides corresponding to eIF4E gene (GenBank NM_001968.3) was designed as follows: eIF4E-shRNA1 (sense, 5′-AAGCAAACCUGCGGCUGAUCU-3′), eIF4E-shRNA2 (sense, 5′-ACAGCAGAGACGAAGUGAC-3′), and a non-specific shRNA (sense, 5′-GGACGUGGUCGCCA CCCUGCCC-3′). These recombinant plasmids were named pGIP-4e1, pGIP-4e2 and pGIP-NS. For temporal transfection, 2×10⁵ skov-3 cells were seeded in 60-mm plates. The cells (60–70% confluent) were transfected with 2 μg of plasmid using Lipofectamine 2000 (Invitrogen) according to the manufacturer's instructions.

Cells (2×10⁵) were plated into 6-well plates. Total 2 μg of pGIP-4e and vector plasmid of eIF4E was, respectively, transfected into C200 and A2780 cells using Lipofectamine 2000 (Invitrogen) and cultured for 24 h without antibiotic selection. Then the cells were cultured in medium containing 5 μg/ml puromycin (Sigma) until all the non-transfected cells were killed. The antibiotic-resistant cells were pooled and passaged in medium containing 5 μg/ml puromycin. Stable transfectants were termed as C200-4e1 or A2780-4e1 (transfected with pGIP-4e1 plasmid), C200-NS or A2780-NS (transfected with pGIP-NS plasmid) respectively.

Approximately 5×10⁵ cells were harvested and boiled in 200 μl lysis buffer containing protease inhibitors. Samples were subjected to SDS-PAGE and electrophoretically transferred to a PVDF (polyvinylidene difluoride) membrane. Membranes were incubated with the primary antibodies (1:1,000) at 4°C overnight, washed with TBST buffer, and incubated again with an appropriate HRP-conjugated secondary antibody (1:5,000) at room temperature for 1 h. The membranes were washed and examined by chemiluminescence detection.

The cell proliferation was analyzed with 5-bromo-2′-deoxy-uridine (BrdU) incorporated into the newly synthesized DNA of replicating cells. In brief, cells were cultured in a 96-well plate for 24 h and pulse-labeled with BrdU for 4 h according to the manufacturer's instructions. The BrdU label in the DNA was detected using a peroxidase and FITC-conjugated anti-BrdU second antibody subjected to immunodetection and immunofluorescence assays (24). For peroxidase-conjugated anti-BrdU immunodetection, the values of relative proliferation was quantified with a peroxidase substrate and measured at 370 nm with a wavelength of 540 nm. For BrdU-FITC immunofluorescence assay, cells were cultured on cover slips until reaching 60% confluence, and then BrdU incorporated into cellular DNA was detected after shRNA plasmid transfection for 24 h with immunofluorescence assay kit (Roche Applied Science, USA). The cells were analyzed using a Nikon confocal microscope, at the wavelength of excitation and emission of 488 and 525 nm, respectively. The positive BrdU cells were counted from five randomly selected fields by direct counting cells in each sample in a blinded manner. The percentage of positive cells was calculated as the number of positive cells divided by the number of total cells. Each assay was repeated at least 3 times.

For cell colony formation assay on plastic surface, cells were trypsinized and plated in triplicate in 6-well culture plates at 500 cells per dish. Cells were allowed to grow at 37°C, 5% CO₂ with media changes every 3–4 days until colonies were visible by eye. After 14-day incubation, cells were fixed with 4% paraformaldehyde at 4°C, then stained with 0.1% crystal violet for 20 min. Cells were washed and plates were scanned, and colonies consisting of ≥30 cells were counted using ImageJ software.

Cells were collected and fixed with 70% ice-cold ethanol overnight at −20°C. Cells were centrifuged, resuspended in 1 ml of PBS mixed with propidium iodide (PI) and RNaseA (10 μg of propidium iodide, 10 μg of RNase A, and 0.5% Tween-20), and incubated at 37°C for 40 min. Cell cycle distribution was determined by analyzing 10,000–20,000 cells using a FACScan flow cytometer and CellQuest software (Becton-Dickinson, San Jose, CA, USA). The percentage of apoptotic cells was determined by the subG₁ peak in the DNA histogram.

For the cisplatin-siRNA combination experiments, a 3-(4,5-dimethythazolz-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay was used to analyzed the effects of eIF4E-shRNA on cell viability in vitro. All cells were pre-treated with eIF4E-shRNA plasmid for 24 h before cisplatin treatment. Drug-treated cells were incubated for 48 h and then MTT was added to the cultures for an additional 2 h. The results were assessed in a 96-well microreader (Bio-Rad Co.) by measuring the absobance at a wavelength of 570 nm. The inhibitory rate of cell growth was calculated as: inhibitory rate = (1−A_treated group/A_{untreated group}) × 100%. The IC₅₀ was determined based on the dose response curves by Graphpad Prism software. Combination index (CI) for drug interaction was calculated using CompuSyn software (CompuSyn, Inc.). The experiments were performed three times independently.

In vitro cell migration and invasion assays were performed using Transwell chambers. For the migration assays, 2×10⁴ cells were added to the upper chamber of 8-μm pore size Transwells (BD Biosciences, Franklin Lakes, NJ, USA). For the invasion assays, 1×10⁵ cells were added to the upper chamber of 8-μm pore size Tranwells precoated with Matrigel (BD Biosciences). In these assays, cells were suspended in RPMI-1640 medium containing 10%, FBS was added to the upper chambers, and RPMI-1640 medium containing 10% FBS was placed in the lower well, and then incubated at 37°C in 5% CO₂. After 24 h of incubation, non-migrated or non-invaded cells were removed carefully. The filters were then fixed in 4% paraformaldehyde and stained with crystal violet. Five random fields were counted per chamber using an inverted microscope (Olympus Corp., Tokyo, Japan) at ×200 magnification for each membrane. All of the analysis was performed in triplicate.

All statistical analyses were performed using SPSS (version 16, SPSS Inc., Chicago, IL, USA). The values were expressed as the means ± SE and statistical significance was analyzed using two-tailed Student's t-test. A p-value of <0.05 was considered as significant and indicated by asterisks in the figures.

We first detected eIF4E and 4E-BP_1/2 protein expression in A2780 and C200 cells in comparison with an immortalized normal human ovarian epithelial cell line (IOSE) with western blot analysis. As presented in Fig. 1A, we found there were significant differences between IOSE cells and carcinoma cell lines, the levels of eIF4E were substantially increased in A2780 and C200 cell lines. Specially, we detected much lower amounts of 4E-BP_1/2 protein in C200 cells. It indicates that C200 cells possessed much higher levels of eIF4E than A2780 cells, and an increasing activity of eIF4F in C200 cells. Interestingly, we found that all cells expressed similar levels of eIF4G in addition to eIF4E (Fig. 1A).

To further compare the pattern and integrity of eIF4F in IOSE and the other two carcinoma cell lines, we assessed formation of the m⁷GTP-eIF4F complex in an m⁷-GTP cap binding assay. The levels of cap-bound eIF4E, eIF4G and 4E-BP_1/2 were analyzed by immunoblotting. The relative amounts of cap captured eIF4E and eIF4G in cells could be regarded as indicator of the integrity and functional potency of eIF4F. The amounts of cap-associated 4E-BP₁ showed negative impact on eIF4E assembly and function. The cap-bound fraction from IOSE and A2780 cells contained significant amounts of 4E-BP₁ (Fig. 1B), indicating that in these cells, cap-dependent translation was under strong 4E-BP₁-mediated negative control. In contrast, the cap-bound complexes from C200 cells were at a significantly lower level of 4E-BP_1/2 indicating that cisplatin-resistant cells existed in a translationally activated state. These results clearly indicated that eIF4E expression was upregulated in platinum-resistant ovarian cancer cells. Moreover, eIF4E and 4E-BPs displayed a deregulated activity in platinum-resistant ovarian cancer cells.

If elevated eIF4E was critical for the protein synthesis of ovarian cancer cells, we hypothesized that downregulation of eIF4E would result in inhibition of the growth of ovarian cancer cells. To verify this, we employed two different sequences of eIF4E shRNA plasmid pGIP-4e1 and pGIP-4e2 to knock down eIF4E expression and determined its impact on the growth in our study. The efficiency of silencing and the expression level of eIF4E protein was measured with expression of green fluorescent proteins. In order to analyze the downregulation of eIF4E expression in cells, western blot assay was performed. As shown in Fig. 2A, the levels of eIF4E protein expression in A2780 and C200 cells were both significantly reduced by pGIP-4e1 and pGIP-4e2 in comparison with its negative control cells, and the inhibitory rates were ~65% and 55% (p<0.05) in C200 cells, and 75 and 55% (p<0.05) in A2780 cells, respectively. Moreover, pGIP-4e1 showed stronger inhibition effect in both cell lines. Thus, we focused our further analyses on the pGIP-4e1 plasmid at the next biological assay. By the time-course analysis, we observed that downregulation of eIF4E protein expression level occurred at 12 h after transfection. The eIF4E reduction peak was at 24 h in C200 cells and 36 h in A2780 cells, respectively. The eIF4E reduction in both A2780 and C200 cells was sustained up to 72 h (Fig. 2B). The results indicated successful knockdown of eIF4E by pGIP-4e1 in ovarian cancer cells. In addition, no effects of shRNA plasmid were observed on the expression of β-action used as an internal control gene in this study.

We examined the anti-proliferate response of pGIP-4e1 in ovarian cancer cell lines C200 and A2780 with BrdU proliferation assay. The results showed that the growth of both A2780 and C200 cells were inhibited by pGIP-4e1 plasmid after transfection for 24 h (Fig. 3A). BrdU-FITC immunofluorescence assay indicated BrdU incorporation was significantly decreased in cells that were transfected with pGIP-4e1 and displayed less density than that of control cells (Fig. 3B).

To investigate the potential of knockdown of eIF4E in reducing colony formation of tumor cells in vitro, we examined the stably transfected C200 and A2780 cells. First, we generated stable cell lines expressing shRNA specific to eIF4E, C200-4e1 and A2780-4e1, C200-NS and A2780-NS (negative plasmid control). The expression of eIF4E protein was analyzed by western blot analysis to test the inhibitive effect of pGIP-4e1 in these stable cell lines. Results of western blot analysis showed that eIF4E protein level was significantly decreased by pGIP-4e1shRNA in ovarian cancer cells (Fig. 4A). Colony formation assay was analyzed two weeks after plating. All cells were fixed in paraformaldehyde and stained with crystal violet (Fig. 4B). The mean values from plates are shown in Fig. 4B. In our results, pGIP-4e1 reduced the colony formation, the number of colonies of both cancer cells was much lower than those of control cell lines, respectively (p<0.05). It further indicated that inhibition of eIF4E expression suppresses the growth of cancer cells.

Some research has shown that eIF4E might be involved in regulation of cancer metastasis. Therefore, we evaluated whether the role of eIF4E knockdown affected the migration and invasion of ovarian cancer cells. The C200 and A2780 cells stably transfected pGIP-4e1 shRNA were analyzed by Transwell assays. Western blot analysis showed that eIF4E protein level was significantly decreased by pGIP-4e1 shRNA in C200 and A2780 cells (Fig. 4A). The results of transwell assays showed that knockdown of eIF4E significantly reduced the number of migration and invasion in both cell lines compared with the control cells (Fig. 5). It suggested that elevated eIF4E expression was associated with positive regulation of ovarian cancer cell migration and invasion.

To elucidate the mechanisms underlying the siRNA mediated growth inhibition, further analysis was performed to test the effects of eIF4E-shRNA on the cell cycle progression of C200 and A2780 cells and each assay was performed three times. The mean values of triplicate experiments are shown in Fig. 6. Flow cytometric cell cycle analysis of logarithmically growing C200 cells revealed a distribution of continuously proliferating cells in G₁ phase, S phase and the mitotic G₂/M phase as evaluated in Fig. 6A. Compared with that of transfected C200-NS cells, the percentage of C200-4e cells in the G₁ phase increased by 17.2±2.1%, while the percentage of C200-4e cells in S phase significantly reduced by 18.5±1.2% (p<0.05). As shown in Fig. 6A, cell cycle analysis showed a significantly greater distribution in subG₁ phase indicating more apoptotic cells in C200-4e cells than in C200-NS cells (p<0.05). However, a significant difference was not found in G₂/M phase between C200-4e and C200-NS cells, suggesting that knockdown of eIF4E induces G₁ cell arrest and apoptosis but does not increase G₂/M phase in which mitotic arrest occurs. Using cleaved PARP as readout of apoptosis, we further determined whether knockdown of eIF4E induces apoptosis in the tested cell lines. As presented in Fig. 6B, we detected cleaved form of PARP in C200-4e or A2780-4e cells. Interestingly, we found the stronger expression of cleaved-PARP in C200-4e cells. As a positive control, tumor necrosis factor-related apoptosis-inducing ligand (TRAIL) induced strong cleavage of PARP in the both cell lines.

The lower level of 4E-BP_1/2 and elevated eIF4E proteins indicated that cisplatin-resistant C200 cells showed a translational activated state, and it was involved in development of acquired resistance to cisplatin. We speculated that inhibition of eIF4E would overcome cisplatin resistance and enhance the sensitivity of ovarian cancer cells to cisplatin. To test this hypothesis, an MTT assay was used to examine the impact on cell response to cisplatin by knockdown of eIF4E in C200 and A2780 cells. Generally, the approximate cisplatin IC₅₀ value of A2780 cell line was 10–20 μM, and the equivalent IC₅₀ value for C200 was ~200 μM. The C200 cell lines were ~20-fold resistant to cisplatin compared with the A2780 (23). As presented in Fig. 7A, we tested the cisplatin IC₅₀ of stable transfected C200-4e1 cells compared with C200-NS cells, the cisplatin IC₅₀ value of C200-4e1 cells was significantly decreased by 28.2% (p<0.05). Accordingly, the cisplatin IC₅₀ also significantly decreased by 18.2% in A2780 cells (Fig. 7B). Thus, it was clear that knockdown of eIF4E expression was able to enhance the effect of cisplatin in resistant ovarian cancer cells.

As a small molecule inhibitor of eIF4E/eIF4G interaction, 4egi-1 behaves as 4ebp1 mimetic to inhibit cap-dependent translation initiation. Thus, we further analyzed whether 4egi-1 could enhance the cytotoxic effects of cisplatin on C200 cells. Our results showed that the combination of 4egi-1 and cisplatin was more potent than either agent alone in inhibiting the growth of C200 cells. The combination indexes were <1 for combination treatments (Fig. 7C), indicating synergy between 4egi-1 and cisplatin in cytotoxicity. This result further suggested that inhibition of cap-dependent translation indeed overcame cisplatin resistance and enhanced the sensitivity of ovarian cancer cells to cisplatin.