The double-stranded oligonucleotide siCE951 encoding the corresponding shRNA was synthesized by Shanghai Sangon Biological Engineering Co., Ltd. (China) according to the nucleotide fragment of the cyclin E mRNA sequence (NM 057182.1) 951–969 with BamHI and HindIII incision enzyme residues on both ends, respectively. The synthetic oligonucleotide chain was inserted into the pRNA-U6.1/Neo vector (GenScript, Piscataway, NJ, USA) between the BamHI and HindIII sites, and then transformed into E. coli DH5α for incubation. The recombinant plasmid was extracted followed by sequencing to confirm the insertion sequence. The recombinant vector pRNA-U6.1/Neo-siCE951 was obtained.

EC9706, Eca109 and KYSE30 esophageal cancer cell lines were placed in a 24-well plate at 2×10⁵ cells per well in order to be incubated in RPMI-1640 medium (Gibco, Carlsbad, CA, USA) containing 10% fetal bovine serum. When cell adhesion reached 90–95%, pRNA-U6.1/Neo-siCE951 and pRNA-U6.1/Neo-Con plasmids were respectively transfected into EC9706, Eca109 and KYSE30 cells using Lipofectamine™ 2000 (Invitrogen, Carlsbad, CA, USA). The study included experimental groups transfected with pRNA-U6.1/Neo-siCE951 plasmid, negative control groups transfected with pRNA-U6.1/Neo-Con plasmid and blank control groups transfected only with liposomes. Twelve hours after transfection, the medium was replaced by complete medium containing 600 μg/ml G418 for 4-week culture at 37°C in an atmosphere of 5% CO₂ in order to obtain stably transfected cells for future use.

Total RNA from each group was extracted using Qiagen RNeasy Mini kit (Qiagen, Hilden, Germany). The cyclin E mRNA in each group was amplified with a One Step SYBP PrimeScript^® RT-PCR kit (Takara Biotechnology Co., Ltd., Dalian, China). The amplification primers of cyclin E were 5′-CGGGTCCACAGGGATGCGAAGGA-3′ and 5′-CAG GTGTGGGGATCAGGGAGCA-3′. The amplification primers of internal control GAPDH were 5′-GCCTTCCGTGTC CCCACTGC-3′ and 5′-CAATGCCAGCCCCAGCGTCA-3′. Amplification reaction conditions were: pre-denaturing at 94°C for 20 sec and 60°C for 60 sec, for 40 cycles. Amplification was performed 5 times for each sample. The relative expression level of cyclin E mRNA was indicated with the ratio of cyclin E to internal control GAPDH.

Cell disruption was performed with RIPA for protein extraction. The protein concentration in the supernatant was determined using the Bradford method to adjust the protein load. The protein sample underwent SDS-PAGE, and was then transferred to a membrane. The membrane was sealed in 25 ml of fresh nonfat dry milk for 1 h followed by the addition of 1:800 mouse anti-cyclin E monoclonal antibody and mouse anti-GAPDH (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), for a 1- or 2-h incubation at room temperature. Horseradish peroxidase-conjugated goat anti-mouse IgG antibody (1:2000, Santa Cruz Biotechnology, Inc.) was added for 1-h incubation followed by coloration. The gray values of western blot bands were determined with Kodak Digital ID Image analysis software. The relative expression level of cyclin E protein was indicated with the gray value ratio of cyclin E to internal control GAPDH.

EC9706, Eca109 and KYSE30 esophageal cancer cell lines were placed in a 96-well plate at 1×10⁴ cells per well with 5 wells for each group. The absorbance value of each well was determined using a cell counting kit-8 (CCK-8, Dojindo, Shanghai, China), and then the average absorbance value was calculated. The cell growth curve in each group was drawn with incubation time as the horizontal axis and with OD value as the vertical axis.

Cells in the logarithmic growth phase were placed in a 6-well plate at a density of 1×10⁶ cells per well. After trypsinization, the cells were centrifuged at 800 rpm for 15 min for cell collection. The cells were resuspended in 0.4 ml PBS followed by addition of 0.7 ml absolute alcohol containing 3% serum for fixation at 4°C for 24 h. RNase-A was added to a final concentration of 50 μg/ml for digestion in a water bath for 30 min at 37°C. Propidium iodide (PI, Sigma, St. Louis, MO, USA) was added to a final concentration of 65 μg/ml for staining in an ice bath for 30 min away from light. After filtration with 300-screen nylon mesh, the cells were observed with a flow cytometer. The amount of cells in the G0/G1, S and G2/M phases were calculated. Testing was performed in triplicate in each group.

The cells were collected by centrifugation, and then washed three times with 1 ml cold PBS. After centrifugation, the cells were stained with Annexin V-FITC at room temperature in the dark for 20 min. After cell collection, the cells were resuspended. PI was added for staining in an ice bath for 5 min away from light. The cells were washed with PBS and then observed with a flow cytometer.

Agarose (1.2%) and 2X RPMI-1640 medium were mixed (at a ratio of 1:1) in a petri dish, and then 2X antibiotics and 20% of fetal calf serum were added. Agarose (0.7%) and 2X DMEM medium were mixed (1:1), followed by addition of 0.2 ml cells (5×10³/ml), which were added into the petri dish mentioned above for a 10- to 14-day incubation. Ten fields were selected to count the colony-forming units (number of cells >50 was counted as a colony-forming unit) under an inverted microscope in each group.

The cells were adjusted at 2×10⁵/ml in each group 48 h following transfection. The upper chamber of 24-well Transwell Permeable Supports with 8-μm pores (Corning Cat. no. 3422) was loaded with 200 μl cell suspension, and the lower chamber was loaded with 500 μl medium containing 10% serum for incubation in an atmosphere of 5% CO₂ at 37°C for 48 h. Five wells were set up for each group. The cells on the Matrigel and in the upper chamber were collected using cotton swabs. Ten fields were selected to be observed under a microscope and the mean was calculated.

Statistical treatment was performed with SPSS 16.0 software. All data are expressed as the means ± SD. One-way ANOVA was used for data analysis. Statistical significance was established at P<0.05.

Recombinant vectors were transformed and transfected bacteria were obtained. Sequencing for the target gene in the recombinant plasmid was performed. Results of sequencing were consistent with the hairpin single-strand DNA that had been designed, demonstrating that recombinant vectors were successfully constructed.

After siRNA vectors were transfected into EC9706, Eca109 and KYSE30 cells, the level of cyclin E mRNA expression was detected with fluorescence-quantitative PCR in each group. Results indicated that compared with the control groups, the level of cyclin E mRNA expression was significantly decreased in each experimental group (P<0.01, Fig. 1), demonstrating that cyclin E mRNA expression was significantly inhibited in EC9706, Eca109 and KYSE30 esophageal cancer cells.

Western blotting revealed the specific band at 54 kDa in the experimental groups and blank and negative control groups (Fig. 2). The western blot was significantly stronger in the blank and negative control groups than in the experimental groups, demonstrating that the designed siRNA vectors targeting cyclin E gene effectively interfered with cyclin E protein expression in EC9706, Eca109 and KYSE30 esophageal cancer cells.

Cell growth curves are shown in Fig. 3. There was no significant difference in absorbance values between the blank and negative control groups (P>0.05). Compared with the blank and negative control groups, the absorbance values on the third, fourth and fifth day were significantly decreased in the experimental groups (P<0.05), demonstrating that inhibition of cyclin E expression can decrease the growth velocity of EC9706, Eca109 and KYSE30 esophageal cancer cells.

The status of the cell cycle is shown in Fig. 4. Compared with the blank and negative control groups, the number of cells in the S and G2/M phases were reduced and the number of cells in the G0/G1 phase were increased in experimental groups (P<0.05), demonstrating that inhibiting cyclin E expression inhibits the cell cycle process, namely that the number of cells in the S and G2/M phases are reduced and the number of cells in the G0/G1 phase are increased in EC9706, Eca109 and KYSE30 esophageal cancer cells.

The early apoptosis rates were 10.67% in EC9706 cells, 11.7% in Eca109 cells and 13.83% in KYSE30 cells, and they were all significantly higher than that in the blank and negative control groups (P<0.01, Fig. 5).

The average number of colony-forming units in the experimental groups and the blank and negative control groups are shown in Fig. 6. There was no significant difference in the average number of colony-forming units between the blank and negative control groups (P>0.05). Compared with the blank and negative control groups, the average number of colony-forming units was significantly decreased in the experimental groups (P<0.05), demonstrating that inhibition of cyclin E expression decreases the number of colony-forming units in EC9706, Eca109 and KYSE30 esophageal cancer cells.

The average numbers of cells penetrating the transwell membrane in the experimental groups and the blank and negative control groups are shown in Fig. 7. There was no significant difference in the average number of cells penetrating the transwell membrane between the blank and negative control groups (P>0.05). Compared with the blank and negative control groups, the average number of cells penetrating the transwell membrane was significantly decreased in the experimental groups (P<0.05), demonstrating that inhibition of cyclin E expression decreases the invasive ability of EC9706, Eca109 and KYSE30 esophageal cancer cells.