BDNF siRNA was designed according to the siRNA design guidelines of Ambion Inc. (Carlsbad, CA, USA). General design guidelines are as follows. The selected 19–22 base siRNA sequences were designed with 30–50% guanine cytosine content to avoid inverted repeats (23). Four siRNAs were chosen based on the sequence of the human BDNF gene. They covered different regions of the BDNF sequence and showed no homology with other human genes. A scrambled siRNA was used as a negative control. The target sequences and corresponding hairpin siRNA sequences for the 5 siRNAs, designated siRNA1, siRNA2, siRNA3, siRNA4 and scramble siRNA, are shown in Fig. 1. The structure of hairpin siRNA was BamHI + sense + loop + antisense + termination signal + SalI + HindIII. pGenesil-1 [enhanced green fluorescent protein (EGFP) purchased from Genesil (Wuhan, China)] was introduced for the construction of recombinant eukaryotic human BDNF siRNA expression vectors. Five pairs of hairpin siRNA sense and antisense sequences were synthesized, annealed and cloned into the BamHI/HindIII cloning site of pGenesil-1, respectively. The products were transformed into DH5α-competent cells. Kanamycin-resistant colonies were chosen, identified by restriction digestion, and further confirmed by DNA sequencing. The synthesis of all DNA chains and DNA sequencing was performed by Genesil.

The cervical carcinoma cell line HeLa was cultured in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin, 100 U/ml streptomycin, and was routinely maintained at 37̊C in 5% CO₂. Medium was changed every 3 days.

HeLa cells were seeded in 6-well plates (5×10⁵/ml). Transient transfection was performed using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. For stable transfection, HeLa cells were cultured in the standard culture medium for 48 h after transfection. G418 selection at 800 µg/ml was applied and continued for 4 weeks until single colonies were formed. In parallel, non-transfected cells were also placed in standard culture medium to ensure the potency and selectivity of G418. Positive clones were maintained in the culture medium with G418 at 300 µg/ml. To determine the transfection efficiencies and expression effects of BDNF siRNA in HeLa cells, EGFP expression was examined by microscopy (magnification, ×400) at 24 h after transfection and after the positive clones were established.

Total RNA from HeLa cells was isolated using TRIzol extraction according to the manufacturer's protocol (Invitrogen). Complementary DNA (cDNA) was subsequently synthesized from total RNA (5 µg) using RevertAid First-Strand cDNA Synthesis kit (Fermentas, Burlington, Ontario, Canada). Then, 1 µg of cDNA was subjected to PCR for the selected genes. Primers used were as follows: 5′-GCAGCCTTCTTTTGTGTAACC-3′ and 5′-AGAGTGATGACCATCCTTTTC-3′ for BDNF (594 bp); as well as 5′-GAAGGTGAAGGTCGGAGTC-3′ and 5′-GAAGATGGTGATGGGATTC-3′ for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (226 bp) as control; 5′-GTTTCATAAGATCCCACTGGATGG-3′ and 5′-TGCTGCTTAGCTGCCTGAGAGTTA-3′ for TrkB (260 bp); as well as 5′-TGAGACCTTCAACACCCCAG-3′, and 5′-GCCATCTCTTGCTCGAAGTC-3′ for β-actin (312 bp) as control. PCR profiles consisted of denaturation at 94̊C for 1 min, annealing at 55̊C for 30 sec, and extension at 72̊C for 1 min. The samples were amplified for 32 cycles. PCR products were separated by electrophoresis on 1.5% agarose gels, stained with ethidium bromide and photographed.

Cells were washed twice with ice-cold phosphate-buffered saline (PBS), and then harvested on ice with NP40 lysis buffer, containing 50 mmol/l Tris-HCl, pH 7.4; 150 mmol/l NaCl; 1% NP40; 5 mmol/l EDTA; 5 mmol/l NaF; 2 mmol/l sodium vanadate; 1 mmol/l phenylmethylsulfonyl fluoride; 5 mg/ml leupeptin; and 5 mg/ml aprotinin. The protein content was quantitated by Bio-Rad protein assay (Bio-Rad, Hercules, CA, USA). The lysate was then boiled for 5 min for protein denaturation. Protein samples containing an equal amount (60 µg) of protein were separated by electrophoresis on 10% polyacrylamide-SDS gels and transferred onto nitrocellulose membranes. Non-specific binding of antibodies was blocked by incubation in Tris-buffered saline (TBS) containing 0.1% Tween-20 (TBS-T) and 5% non-fat milk for 1 h, followed by overnight incubation with rabbit anti-human BDNF (SC-546) (1:500) or rabbit anti-human TrkB (SC-12) (both from Santa Cruz Biotechnology, Santa Cruz, CA, USA) (1:500) or rabbit polyclonal IgG to GAPDH (1:1,000) at 4̊C. Rabbit anti-pAKT (1:500) and anti-AKT (1:1,000) were obtained from Cell Signaling Technology (Beverly, MA, USA). After washing, the membranes were incubated with horseradish peroxidase (HRP)-conjugated goat anti-rabbit secondary antibodies (1:5,000) at room temperature for 1 h. Immunoreactive bands were visualized using the ECL kit.

Cell proliferation was assessed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT; Sigma Chemical Co., St. Louis, MO, USA) assay. Cells were seeded in a 96-well plate at a density of 1.5×10⁴ cells/well and allowed to adhere overnight. Transient transfection was performed using Lipofectamine 2000 (Invitrogen), according to the manufacturer's instructions. After 12, 24, 48, 72 and 120 h of transfection, 20 µl of 5 mg/µl MTT solution was added into each well of the plate and cells were incubated at 37̊C for an additional 4 h. The culture supernatant was removed, 150 µl of dimethyl sulfoxide (DMSO; Sigma) was added to dissolve the crystals. Spectrophotometric absorbance of the samples was measured by a microtiter plate reader at 570 nm. Considering that the initial cell concentrations of each group may be not identical, data at 12 h was set as the control. The cell proliferation rate was calculated as follows: Proliferation rate (%) = (sample absorbance/control absorbance) × 100. Each value represents 6 replicates, and each experiment was repeated 3 times.

HeLa cells were divided into non-transfected (P_non), pGenesil-1-tranfected (P₀) and positive experimental groups (P_BDNF1). Cells in the P_non, P₀ and P_BDNF1 groups were cultured in 6-well plates. After 48 h of incubation, a total of 1×10⁶ cells were harvested, washed twice with PBS and fixed with cold 70% ethanol overnight at −20̊C. Fixed cells were centrifuged at 1,200 rpm and washed with PBS. Cells were stained in the dark with propidium iodide (PI; 50 µg/ml; Sigma) and 0.1% RNaseA (Invitrogen) at 4̊C for 1 h. Measurement of nuclear DNA was performed using flow cytometry (FCM). The results obtained reflected the percentage of cells in each phase of the cell cycle.

After 24 h of incubation in 6-well plates, the supernatant were removed. Then, the cells were fixed with cold methanol (0̊C) for 10 min and washed twice with PBS. Hoechst 33258 (1 mg/ml) was added and incubation for 30 min at 4̊C was carried out in a dark place. Cells were then washed with PBS and the apoptotic cells were observed using an Olympus BH-2 fluorescence microscope (magnification, ×200; Olympus, Tokyo, Japan). Cells stained bright blue were considered as apoptotic cells. Random cells (500) were counted and the apoptotic rate was calculated as follows: Hoechst 33258-stained cells/500 cells×100%.

Cell migration was quantified by the number of cells that directionally migrated through a 8-µm-pore polycarbonate filter (porosity, 8 µm; Costar, Appleton Woods, Birmingham, UK) in Boyden chambers. Briefly, the lower surface of the filter was coated with 10 mg of gelatin. HeLa cells were serum-starved overnight and resuspended in serum-free medium, and then 200 µl of the cell suspension was replaced onto the upper chamber of each well at a final concentration of 1×10⁶ cells/ml. FBS (10%)-containing medium was added to the bottom chamber and cells were allowed to migrate for 6 h at 37̊C. Non-migrated cells on the upper membrane surface were removed with a cotton swab. The migratory cells attached to the lower membrane surfaces were fixed with 4% paraformaldehyde in PBS and stained with Wright staining. Cells were counted at a magnification of ×400 using standard microscopy, and the mean number of cells/field in 5 random fields was recorded. Triplicate filters were used and the experiments were repeated 3 times. The invasion assay was performed as above except that the upper surface of the filters were coated with 25 µg Matrigel, the cell concentration was adjusted to 2×10⁵ cells/ml and the time was prolonged to 24 h.

SPSS for Windows (SPSS 13.0; SPSS, Inc., Chicago, IL, USA) was used to analyze the data. Data are expressed as the mean ± standard deviation (SD) from at least 3 separate experiments. Student's t-test was used to determine the significant difference between 2 groups. For comparison between 3 or more groups, one-way ANOVA test was used to determine statistical significance. The level of significance was set at p<0.05.

As shown in Fig. 2, RT-PCR and western blot analysis revealed that BDNF and TrkB were highly expressed in the HeLa cells at both the mRNA (Fig. 2A) and protein level (Fig. 2B). BDNF may provide autocrine support for TrkB-expressing HeLa cells as reported in the nervous system (24).

As shown in Fig. 3A-C, 4 experimental groups of plasmids pBDNF1, pBDNF2, pBDNF3 and pBDNF4, and the negative control group plasmid pScr were successfully generated. The multiple clone sites of pGenesil-1 are: HindIII, shRNA, BamHI, U6 promotor, EcoRI, SalI and XbaI. A specific SalI site was designed and synthesized into the shRNA sequences. Recombinant vectors were digested to a 400 bp fragment by SalI with the correct insertion. Most of the HeLa cells were effectively transfected with the recombinant BDNF siRNA expression vectors. The transient transfection efficiencies were ~60–80%. EGFP expression was not significantly reduced after the positive clones were generated (Fig. 3D-F).

HeLa cells were divided into non-transfected (P_non), pGenesil-1-tranfected (P₀), pScr (P_Scr) and positive experimental groups (P_BDNF1, P_BDNF2, P_BDNF3 and P_BDNF4, respectively). As shown in Fig. 4, semi-quantitative RT-PCR and western blot analysis were performed when the positive clones had been established. The results revealed that siRNA1 (p<0.01) and siRNA4 (p<0.05) both led to significant reductions in BDNF expression without marked changes in GAPDH (GAPDH, data not shown), while pGenesil-1, siRNA2, siRNA3 and scramble siRNA barely altered the expression of BDNF (p>0.05). Compared with the level of BDNF in the P_non group, siRNA1 led to nearly 99% reduction in the relative mRNA level, while a 58% decrease in the relative protein level was noted. siRNA1 was selected as the most efficient siRNA for use in the next experiments. Stable HeLa cell clones in the P_non, P₀ and P_BDNF1 groups were used as experimental objects.

FCM was performed to evaluate the cell cycle profile in the BDNF-siRNA-transfected HeLa cells. As shown in Fig. 5A-D, the percentage of cells in the G1 phase in the P_BDNF1 group (70.73±4.15%) was much higher than that observed in the P_non (55.33±5.64%) (p<0.01) and P₀ group (57.47±2.98%) (p<0.05). Meanwhile, the percentage of cells in the G2 phase correspondingly decreased in the P_BDNF1 group (11.15±2.88%) compared with the P_non (24.83±3.67%) (p<0.01) and P₀ group (21.28±5.38%) (p<0.05). The percentage of cells in the S phase had no significant change in the 3 groups (p>0.05) (Fig. 5A-D). Together, the results showed that BDNF-siRNA induced cell cycle arrest at the G1 phase and decreased the distribution of cells in the G2 phase. The growth curves of cells in the 3 groups as determined by the MTT assay showed significant growth inhibition of the BDNF-siRNA-transfected HeLa cells (Fig. 5E).

To analyze the involvement of BDNF in cell apoptosis, Hoechst 33258 staining was performed to investigate the apoptosis-related changes in cell morphology and to further evaluate the apoptotic rates in the P_non, P₀ and P_BDNF1 groups. Observation with fluorescence microscopy (magnification, ×200) revealed a significant increase in the number of cells in the P_BDNF1 group showing nuclear condensation and fragmentation which was not observed in the P_non and P₀ groups (Fig. 6A-C). As shown in Fig. 6D, the percentage of apoptotic/necrosis cells in the P_BDNF1 group (27.14±4.57%) was much higher than those observed in the P_non (0.84±0.39%) and the P₀ group (2.68±1.02%) (p<0.01).

To evaluate the role of BDNF in cell migration and invasion, Transwell assay was used as a tool to determine the migratory and invasive capabilities of HeLa cells in the P_non, P₀ and P_BDNF1 groups. As shown in the Fig. 7A-D, the number of cells in the P₀ group that had migrated to the underside of the filters was similar to that of the cells in the P_non group, whereas cells in the P_BDNF1 group showed a significantly reduced migratory capability compared with the other 2 groups (p<0.01). Migrated cells/field in the P_BDNF1 group (37±17), were significantly less than those in the P_non (105±31) and P₀ group (92±28). Migratory capability was impaired in the BDNF-knockdown cells compared with that noted in the non-transfected cells within at least a 2-fold reduction (p<0.01). As shown in Fig. 7E-H, similar to the migration assay, the number of invaded cells/field in the P_BDNF1 group (24±12), was significantly less than those in the P_non (85±26) and P₀ groups (75±20). Invasive capability was significantly impaired in the BDNF-knockdown cells compared with non-transfected cells with at least a 3-fold reduction (p<0.01). Using stable cell lines expressing shRNA against BDNF, these results indicate that endogenous BDNF is essential for the ability of HeLa cells to migrate and invade normally.

We next sought to study the effect of BDNF on the activation states of PI3K, a member of the signaling pathways known to be involved in mediating tumor cell proliferation and migration. AKT is a downstream target of PI3K-generated signals and becomes activated after phosphorylation of Ser473. The results showed that the level of phosphorylated AKT was significantly impaired in the BDNF-knockdown HeLa cells compared with the level noted in the non-transfected cells (Fig. 8).