pU6-M4 plasmid was kindly bestowed by Dolph L. Hatfield (National Institutes of Health, Maryland, United States). Other materials involved in the study were purchased from companies or corporations as below: competent Trans-5αE. coli (TransGen Biotech, Beijing, China), CNE-2 cells (China Center for Type Culture Collection), MEM medium and Opti-MEM (Gibco Corporation, Grand Island, NY), T4 DNA ligase and restriction endonucleases of BamHI, HindIII, Xho I, Mfe I and EcoR I (Thermo Fisher Scientific, Waltham, USA), rapid gel recovery kit and plasmid maxi kit and RNA extraction kit (QIAGEN, Hilden, Germany), lipofectamineTM2000 transfection Reagent Kit (Invitrogen Corporation , Grand Island, NY), BioTekesupermo III RT Kit and quantitative fluorescence PCR kit (Bio TeKe Corporation, Beijing, China). MTT, trypsin, rabbit anti-human Surviving, rabbit anti-human CDK1 and horseradish peroxidase-labeled goat anti-mouse antibodies were all purchased from Sigma Corporation, Shanghai, China, while rabbit anti-human β-actin antibody purchased from Masseachusettes, USA.

CNE-2 cells were cultured at 37˚C in MEM medium supplemented with 10% FBS (HyClone), 100 U/mL penicillin, 100 mg/mL streptomycin (Invitrogen) in an incubator with 5% CO₂. The cells were digested with 0.25% trypsin for passage when grown to 85% - 90% confluence.

Genbank accession number (survivinNM_001012271) and (CDK1 NM_001170406) were used for this study. Scramblesh RNA sequence without significant homology to mouse and human gene sequences, was used as negative control to validate the specific effects. Terminator code (TTTTT), restriction sites of BamHI (G^ATCC) and HindIII (A^AGCCT) were added to where to form BamHI-sense-loop-anti-sense-terminator-HindIII. A series of plasmids containing shRNA specific to Survivin and CDK1gene, either alone or in combination, were cloned into pU6-M4 vector by ligating the BamHI/HindIII-digested shRNA fragment to the vector after digestion by the same restriction endonucleases as described previously. The diagram of constructed pU6-survivin_shRNA-CDK1_shRNA was as shown in Figure 1.

Constructed plasmids were transfected into competent Trans-5αE.coli according to the instructions of Lipofectamine TM 2000 Transfection Reagent Kit. Seven groups were categorized according to the transfection: transfection with pU6-survivin; transfection with pU6-CDK1_shRNA; transfection with pU6-survivin_shRNA-CDK1_shRNA; transfection with pU6-survivin_shRNA-NC; transfection with pU6-CDK1_shRNA-NC; Lipofectamine 2000; control without adding anything. About 1 × 10⁵ CNE2 cells were seeded in each well of a six-well plate and cultured under proper condition. The cultured CNE2 cells were then used for plasmid transfection when cells covered 60% - 80%. Plasmids (5 μg) were transfected into CNE-2 cells with 12 μl Lipofectamine TM 2000 (Invitrogen) following the manufacturer’s protocol.

To evaluate the interference efficiency of constructed plasmids, the seven groups

of CNE-2 cells were harvested after 48 h of transfection and total RNAs were extracted with RNA extraction kit (QIAGEN, Hilden, Germany). RNA was reverse transcribed by BioTekesupermo III RT Kit. qPCR was performed using the SYBR green Chimeric fluorescence method (Bioteke Corporation, Beijing, China) following the manufacturer’s Protocol. Relative gene expression was quantified using the 2-ΔΔCT method with β-actin as internal reference and qualified with quantitative fluorescence PCR kit (Bio TeKe Corporation, Beijing, China). The 5’ primer of survivin gene was 5’-TCAAGGACCACCGCATCTCTA-3’, and the 3’ primer was 5’-TGAAGCAGAAGAAACACTGGGC-3’. The 5’ primer of CDK1 was 5’-CCTAGCATCCCATGTCAAAAACTTGG-3’ and the 3’ primer was 5’-TGATTCAGTGCCATTTTGCCAGA-3’. The 5’ primer of β-actin that acted as internal reference was 5’-GTGGTGGTGAAGCTGTAGCC-3’, and the 3’ primer was 5’-GAGACCTTCAACACCC-3’. RT-PCR was performed with the following parameters: predegeneration at 95˚C for 2 min, synthesis of cDNA with SuperScript^TM one-step RT-PCR kit on a PCR cycler by heating 60˚C for 1 min, followed by 40 cycles of amplification (denaturation at 95˚C for 15 s, anealing at 55˚C for 15 sec, extending at 72˚C for 45 s), and a final extension step at 72˚C for 5 min. The PCR cycle parameters were as following: predegeneration at 95˚C for 30 sec; 40 cycles of amplification (denaturation at 95˚C for 5 sec, and annealing at 60˚C for 20 sec).

CNE-2 cells were harvested and washed with PBS, then lysed with 1 × SDS loading buffer. Lysates from CNE-2 cells were added with protease inhibitorcocktail (Roche) and 1 mM PMSF (Calbiochem). The supernatant was acquired with centrifuging at 13,000× g for 10 min. The lysates were denatured and separated by 15% sodium dodecyl sulfate polyacrylamide gel electrophoresis, then transferred to PVDF membranes (Millipore, Bedford, MA, USA). 5% nonfast milk in Tris bufferedsaline (TBS) buffer with 0.05% Tween-20 was used to block at room temperature for 1 hour. Then the primary rabbit antibodies of anti-human CDK1 (1:1000) and survivin (1:1000) were added and incubated at 4˚C overnight. After five times of rinsing with TBST buffer, secondary goat anti-rabbit horseradish peroxidase conjugated antibody was added and incubated at room temperature for 2 h. Bands were visualized by ECL plus western blot detection reagent (GE, USA). The house-keeping gene β-actin was used as control and detected by using rabbit anti-β-actin antibody (1:4000, Abmart, China) plus HRP conjugated goat anti-rabbit secondary antibody (1:5000, Sigma).

CNE-2 cells were seeded into 96-well plates at a density of about 3 × 10⁵/well. CNE-2 cells were classed into five groups: transfection with pU6-survivin_shRNA; transfection with pU6-CDK1_shRNA; transfection with pU6-survivin_shRNA-CDK1_shRNA; transfection with pU6-survivin_shRNA-CDK1_shRNA; Control without adding anything. CNE-2 cells were transfected with 5 μg plasmids and 5 μl of Lipofectamine TM 2000 followed by incubating for 48 h. Then, 15 μl of MTT (5 mg/ml) was added and cultured 4 h, followed by adding 150 μl DMSO to each well. OD value of each well was analyzed at wavelength of 490 nm after shaking for 10 min, and the cell growth inhibition rate (IR) was calculated with the following formula: IR = (Acontrol group − A experimental group)/A control group × 100%. Each Experiment was repeated thrice.

The CNE-2 cells were categorized into the following five groups: transfection with pU6-survivin_shRNA; transfection with pU6-CDK1_shRNA; transfection withpU6-survivin_shRNA-CDK1_shRNA; transfection with pU6-survivin_shRNA NC-CDK1_shRNA NC; Mock transfection without adding anything. CNE-2 cells were harvested after 48 hour of post-transfection and digested with 0.25% trypsin to produce single-cell suspension, cells were adjusted to 1 × 10⁶/ml and then fixed with 70% cold ethanol at 4˚C overnight. CNE-2 cells were further treated with Annexin V and PI and incubated for 30 min at room temperature in the darkness. The percentage of CNE-2 apoptosis was determined by Flow cytometry (Flow cytometry cycle detection kit, FACSort flow cytometry, B & D companies; USA) operated by a specialist.

CNE-2 cells were trypsinized to produce suspension with DMEM medium, and about 5 × 10⁶ CNE-2 cells/200 μl were inject subcutaneously into BALB/c mice to generateCNE-2 xenografts. Thirty BALB/c nude mice were divided randomly into five groups for experimental tests (N = 6): pU6-survivin_shRNA-CDK1_shRNA group (0.1 ml solution containing 50 μg pU6-survivin_shRNA-CDK1_shRNA); pU6-survivin_shRNA (0.1 ml solution containing 50 μg pU6-survivin_shRNA); pU6-CDK1_shRNA (0.1 ml solution containing 50 μg pU6-CDK1_shRNA); pU6-survivin_shRNA NC-CDK1_shRNA NC (0.1 ml solution containing 50 μg pU6-survivin_shRNA NC-CDK1_shRNA NC); mock transfection (0.1 ml PBS alone). Once tumors were palpable (mean diameter, 5 mm), PBS or various plasmids were injected into multi-position of tumor of mice. Tumor volume (V) was evaluated as following:

V = L W 2 / 2

(L: tumor length; W: tumor width). Tumor volume was evaluated every 3 days.

Graphpad Prism 5 software was adopted for statistical analysis. Data were expressed as mean ± SEM. One-way ANOVA with Tukey’s test was applied for analysis data from multiple groups, while t test applied for data from two groups. The statistical significance was evaluated by p < 0.05 was considered statistically significant.

Down-regulations of survivin and CDK1 were observed in CNE-2 cells interfered by pU6-survivin_shRNA, pU6-CDK1_shRNA and pU6-survivin_shRNA-CDK1_shRNA. Of the the designed sequences of shRNAs of survivin and CDK1, No. 3 of survivin and CDK1 manifested the highest efficiency of knocking down targeted gene, which was used for further construction of pU6-survivin_shRNA-CDK1_shRNA and subsequent experiments. CNE-2 cells interfered by pU6-survivin_shRNA (Figure 2(a)) or pU6-CDK1_shRNA (Figure 2(b)) displayed that the designed plasmids were effective in knocking down expressions of targeted genes. pU6-survivin_shRNA-CDK1_shRNA were more efficient than pU6-survivin_shRNA and pU6-CDK1_shRNA in knocking down survivin and CDK1 in CNE-2 cells. The relative mRNA level of survivin in CEN-2 cells interfered by pU6-survivin_shRNA-CDK1_shRNA was about one third of that interfered by pU6-survivin_shRNA (0.18 vs 0.49, p < 0.01), and similar result observed in the relative mRNA level of CDK1 (0.15 vs 0.44, p < 0.01) (Figure 2(c)). Western blot analyses also supported pU6-survivin_shRNA-CDK1_shRNA was more efficient than pU6-survivin_shRNA and pU6-CDK1_shRNA in knocking down survivin and CDK1 in CNE-2 cells (Figure 2(d)).

The MTT colorimetric assay showed that CNE-2 cells transfected by pU6-survivin_shRNA, pU6-CDK1_shRNA and pU6-survivin_shRNA-CDK1_shRNA all resulted in growth inhibition. The rate of growth inhibition of CNE-2 cells transfected with pU6-survivin_shRNA-CDK1_shRNA was significantly higher than those transfected with both pU6-survivin_shRNA (41.21% vs 25.6%, p < 0.01) and pU6-CDK1_shRNA (41.21 vs 15.62%, p < 0.01) (Table 1). Flow cytometry also demonstrated that apoptosis in CNE-2 cells interfered by pU6-survivin_shRNA-CDK1_shRNA (15.2%) was higher than those interfered by pU6-survivin_shRNA (5.4%) and pU6-CDK1_shRNA (4.7%) (Figure 3).

The tumor growth speeds of the pU6-survivin_shRNA NC-CDK1_shRNA NC group and mock transfection group were the fastest, and both were higher than those of the other three groups. As shown in Figure 4, the tumor volume of CNE-2 xenografts in nude mice treated with pU6-survivin_shRNA-CDK1_shRNA at a dose of 50 μg three times for 3 weeks were significantly smaller than those of treated

with pU6-survivin_shRNA (513.7 vs 1091.2, p < 0.01) and pU6-CDK1_shRNA (513.7 vs 1500.2, p < 0.01). Moreover, both pU6-survivin_shRNA and pU6-CDK1_shRNA manifested the activity of antitumor in mice received CNE-2 xenografts. No death, significant systemic adverse reactions and body mass changes were observed in mice during three weeks, and no obvious interclass differences observed in the volume of the tumors.