Introduction
Cervical cancer remains the second most common cancer in women worldwide. In 2008, there were >529,800 new cases and ~275,000 deaths caused by this malignancy (1,2). Unfortunately, up to date, the molecular mechanisms involved in the initiation and progression of cervical cancer have not been elucidated. Understanding of the epigenetic modifications involved in cervical cancer, such as those mediated by microRNAs, might shed light on this issue (3,4).
microRNAs are a family of small, single-stranded RNAs, which can degrade or block the transcription of mRNA via directly binding to the 3′-untranslated region of their target genes (5,6). microRNAs have been demonstrated to regulate cell proliferation, division, differentiation, invasion, and migration (7,8). A number of microRNAs, such as microRNA-218 (miR-218) have been shown to be downregulated and play a role as tumor suppressors in human malignancies (9–13). miR-218 has been shown to suppress cellular proliferation and invasion and to induce apoptosis both in vitro and in vivo (13–15). In a previous study, we demonstrated that the circulating miR-218 level is significantly reduced in cervical cancer patients, and associates with the late tumor stage, lymphatic node metastasis and poor prognosis (16). Here, using a mouse model and primary cultured tumor tissues, we aimed to explore the effects of miR-218 restoration in human cervical cancer.
Materials and methods
Clinical specimens and primary cultures
A total of 25 fresh cervical cancer tissues were collected from patients who underwent surgery at the Affiliated Hospital of Jiangnan University (the Fourth People’s Hospital of Wuxi, Jiangsu Province, China). For RNA extraction, the tissues were instantly stored in RNAlater solution (Qiagen, Valencia, CA, USA) at −80°C.
For primary cultures, fresh tumor tissues were washed with Hank’s buffered salt solution (HBSS) and gently minced with a sterilized scalpel. Then, the tissues were digested with 0.1% collagenase IA (product no., C5894; Sigma-Aldrich, Minneapolis, MN, USA) for 2 h at 37°C. Following a wash with phosphate buffer saline, the mixture was resuspended to a density of ~1×105 cells/ml. Next, 2×104 cells/well were seeded into 96-well plates and incubated for 24 h prior to carboplatin treatment. Permissions for this study were obtained from the Ethical Committee of the Fourth People’s Hospital of Wuxi, and informed written consent was obtained from all the patients. Patient information was collected; this information is summarized in Table I.
Cervical cancer cell lines, culture and reagents
Four human cervical cancer cell lines were used in this study, HeLa, SiHa, C33A and CaSki, obtained from the American Type Culture Collection (ATCC; Manassas, VI, USA). The cells were cultured in the following media: RPMI-1640 supplemented with 10% fetal bovine serum (FBS) for CaSki, Dulbecco’s Modified Eagle’s medium/Nutrient Mixture F-12 (DMEM/F12) with 10% FBS for HeLa, SiHa and C33A. All cell lines were cultured at 37°C, with 5% CO2. Carboplatin was purchased from Sigma-Aldrich (product no., C2538) and was freshly prepared before each experiment at a 5 mM stock concentration. The miR-218 mimic and its negative control were purchased form GenePharma Co., Ltd. (Shanghai, China).
Transient and stable transfection with miR218
The pGenesil-1-miR-218 expression plasmid was obtained as previously described (15). Transient transfection was performed with the miR-218 mimic and negative control in all four cell lines. A total of 1×105 cells/well were plated into the 6-well plates and incubated overnight. Transfection with miR-218 mimic (or the negative control) was performed using Lipofectamine® 2000 (Thermo Fisher Scientific, Waltham, MA, USA), and G418 (0.4 mg/ml, Sigma-Aldrich) was used to select the stably transfected colonies.
MTT assay
Cells (2×103 or 2×104 cells/well for primary cultured cells) were seeded into 96-well plates and incubated overnight, then treated with different doses of carboplatin for 72 h. Next, 5 μl MTT (5 mg/ml; Sigma-Aldrich) and 100 μl dimethyl sulfoxide were sequentially added into each well, and the absorbance was measured at 570 nm. All procedures were performed in triplicate.
Cell-cycle analysis
Cells (3×106) were plated into a 60-mm dish and incubated overnight before treatment with the miR-218 mimic or carboplatin. Twenty-four hours later, the cells were harvested and suspended in 1 ml of staining solution, consisting of 50 μg/ml propidium iodide and 20 μg/ml RNase A (Thermo Fisher Scientific). Then, the cells were analyzed on a FACScalibur flow cytometer (BD Biosciences, San Diego, CA, USA). Flow cytometry data were analyzed using CellQuest™ software (BD Biosciences). The procedures were repeated in triplicate.
microRNA enrichment and reverse transcription-quantitative PCR (RT-qPCR)
Total RNA was extracted using the Invitrogen TRIzol reagent (Thermo Fisher Scientific). Then, the total RNA sample was enriched in microRNAs using the Applied Biosystems® mirVana miRNA Isolation kit (Thermo Fisher Scientific) according to the manufacturer’s instructions. The expression of miR-218 was measured by the Applied Biosystems Taqman miRNA assay kit that is specific to miRNA-218, on an Applied Biosystems StepOnePlus™ Real-Time PCR system (all from Thermo Fisher Scientific). U6 RNA was used as the endogenous control. The relative level of miR-218 was calculated with the 2−ΔΔCt method (17). All the procedures were repeated in triplicate.
Western blotting
Total protein was extracted using RIPA buffer (Beyotime Institute of Biotechnology, Shanghai, China) and the protein concentration was measured using the BCA Protein Assay kit (Bio-Rad Laboratories, Hercules, CA, USA). Equal amounts of protein were separated by 12% SDS-PAGE and transferred onto polyvinylidene fluoride membranes (Bio-Rad Laboratories). The membranes were incubated overnight at 4°C with the primary antibodies rabbit monoclonal anti-cyclin D1 and rabbit monoclonal anti-CDK4 (1:1,000 for anti-cyclin D1 and 1:500 for anti-CDK4; Cell Signaling Technology, Beverly, MA, USA), and then incubated for 2 h at room temperature with the secondary antibody horseradish peroxidase-conjugated anti-rabbit IgG (monoclonal; Cell Signaling Technology). The specific protein bands were detected using an Enhanced Chemiluminescence system (Beyotime Institute of Biotechnology).
Immunohistochemical staining (IHC)
IHC was performed as previously described (18). The primary antibodies used were: anti-Ki 67 (1:200; Boster Biological Technology, Ltd., Wuhan, China) and anti-cyclin D1 (1:300; Cell Signaling Technology). Staining was performed with the Histostain-Plus IHC kit (Mingrui Biotech Co., Ltd., Shanghai, China). The stained tissues were visualized on an optical microscope (Leica Microsystems GmbH, Wetzlar, Germany) and were separately evaluated by two pathologists.
The tumor xenograft growth assay
The experimental procedures using animals in this study were approved by the Animal Ethics Committee of the Jiangnan University. Six- to eight-week-old female BALB/c nude mice were purchased from SLAC Laboratory Animal Co., Ltd. (Shanghai, China). SiHa cells (1×106) stably expressing miR-218 were subcutaneously injected into the left posterior flanks of the nude mice. As a negative control, an equal amount of SiHa cells transfected with the empty vector was injected into the right flanks. To test the effects of miR-218 on cellular sensitivity to carboplatin, another group of nude mice bearing SiHa cells was established following the same procedure, and carboplatin (intraperitoneal injection, 30 mg/kg) was administrated at days 10, 17, 24 and 31. From 10 days after implantation, the tumor size was measured every five days. Tumor volume = (length × width2)/2. Four weeks later, all mice were sacrificed, and the tumor tissues were collected for further analysis.
Statistical analysis
The software SPSS 16.0 (SPSS Inc., Chicago, IL, USA) was used for statistical analysis. The χ2 and the t-test were used for categorical and quantitative data comparisons, respectively. P<0.05 was considered to indicate a statistically significant difference.
Results
miR-218 inhibits the growth of cervical cancer cells
Compared to the negative control, 24-h treatment with the miR-218 mimic increased the expression of miR-218 by >1,000-fold in all four cell lines (Fig. 1A), and significantly inhibited cellular growth (Fig. 1B–E, P=0.021 for CaSki, P=0.009 for HeLa, P=0.016 for SiHa, and P=0.029 for C33A, respectively).
miR-218 induces G1 phase cell-cycle arrest and inhibits cyclin D1 and CDK4
In HeLa and SiHa cells, overexpression of miR-218 arrested the majority of the cells at the G1 phase (Fig. 2A), but this arrest was not prominent in the C33A and CaSKi cell lines (Fig. 2C). In addition, the expression of cyclin D1 and CDK4 was reduced in miR-218-overexpressing HeLa and SiHa cells (Fig. 2B).
Stable overexpression of miR-218 inhibits tumor growth in vivo
By transfection of SiHa cells with the miR-218-expressing plasmid, we stably overexpressed miR-218, as much as 3,000-fold (Fig. 3A). The growth of the transfected SiHa cells was significantly reduced (Fig. 3B, P<0.01), similarly to the transient transfection experiments. In the mouse model established by injection of the miR-218-transfected SiHa cells, growth of the xenograft was also impaired (Fig. 3C, P<0.001); moreover, the tumor weight in these mice was reduced compared to the control group (Fig. 3D, P=0.01). Overexpression of miR-218 also inhibited the expression of Ki 67 (a marker of proliferation) and cyclin D1 (Fig. 3E).
miR-218 expression correlates to tumor sensitivity to carboplatin in primary cultured cervical cancer samples
As shown in Fig. 4A, reduced expression of miR-218 was associated with cervical cancer cell resistance to carboplatin, and as the level of miR-218 increased, the half maximal inhibitory concentration (IC50) of carboplatin (72-h treatment) gradually decreased. The correlation between miR-218 expression and IC50 was significant (R2=0.3319, P=0.0026).
Restoration of miR-218 sensitizes SiHa cells to carboplatin
In vitro, miR-218 overexpression increased cellular sensitivity to carboplatin and induced cell-cycle arrest (Fig. 4B and C). In vivo, tumor growth in the group overexpressing miR-218 was markedly lower compared to the control group, where miR-218 expression was lower (Fig. 4D, P<0.001), while tumor weight was also reduced (Fig. 4E, P=0.011). Carboplatin treatment further reduced the expression of Ki 67 and cyclin D1 when combined with overexpression of miR-218 (Fig. 4F).
Discussion
Uncontrolled tumor growth and chemo-/radio- resistance remain the leading causes of cancer-related death, especially in cervical cancer, where patients often experience tumor recurrence and poor outcome (19). Thus, valid approaches to suppress tumor growth and improve cellular sensitivity to routinely used therapeutic agents are urgently needed. The group of small, single-stranded nucleotide chains, microRNAs, has been demonstrated to be a potent tool to target the genes involved in tumor growth and reactions to drugs (20). For instance, SPC3649, a miR-122 antagonist, is currently being evaluated in a phase II clinical trial (NCT01872936) and may constitute the first commercial microRNA-based drug (21). Another example is MRX34, a drug designed to mimic the activity of human miR-34, now under phase I clinical trials (NCT01829971) (22).
We previously reported that loss of miR-218 relates to late tumor stage and lymphatic node metastasis, indicating the roles of miR-218 in the progression of cervical cancer. We now investigated the relationship between the miR-218 level and tumor sensitivity to carboplatin, as well as the effects of miR-218 restoration in cervical cancer progression.
miR-218 has been shown to participate in numerous signaling pathways through which it can suppress tumor growth, migration and invasion, such as IKK-β, NF-κB, Slits-Robo, survivin and PI3K/AKT (9,23,24). Using two transfection protocols, we upregulated miR-218 in four human cervical cancer cell lines, and in both series of experiments, cellular growth and cell-cycle progression were significantly inhibited. In vivo, overexpression of miR-218 also significantly suppressed the growth of SiHa xenografts. Our findings are supported by previous studies on nasopharyngeal cancer (14), gastric carcinoma (15)and other malignancies (12,13).
For cervical cancer patients at the late stage, platinum-based chemotherapy is commonly combined with surgery or radiotherapy. Since severe dose-dependent toxicity is frequently observed, oncologists nowadays opt for weekly chemotherapy, which has been demonstrated to be more efficient and have limited side-effects (25,26). In this study, we found that the low miR-218 level correlates to resistance to carboplatin in the primary cultured human cervical cancer samples, which was consistent with the data observed in the mouse model. Tumors where miR-218 expression was restored were more sensitive to carboplatin treatment than those with relatively reduced expression of miR-218, which indicates that upregulating miR-218 may be a promising strategy for patients who both show low expression of miR-218 and resistance to carboplatin. Apart for its suppressive effects on tumor proliferation (as shown by the Ki 67 IHC detection), we also found the miR-218 overexpression or miR-218 overexpression combined with carboplatin treatment significantly inhibits cell cycle progression, in agreement with a previous study (27).
In other malignancies, microRNAs such as miR-193b and miR-21, were proved to predict tumor sensitivity to chemotherapy (28,29). Here, we demonstrated that miR-218 expression correlates to carboplatin sensitivity. Moreover, we showed that restoration of miR-218 allows to overcome carboplatin resistance when miR-218 expression is low. In the near future, we will focus on development of miR-218 as a predictive marker for resistance to chemotherapy in cervical cancer.
In summary, based on our previous and the present findings, we conclude that restoration of the reduced miR-218 level in the cells can significantly suppress the growth of cervical cancer, potentially via the induction of cell-cycle arrest; furthermore, we found that both in vitro and in vivo, the high level of miR-218 can sensitize cervical cancer cells to carboplatin. Our findings provide strong arguments for the development of a novel therapy for cervical cancer based on miR-218, especially for patients with a reduced miR-218 level.