The promoter sequence (1 kb) of the E-cadherin gene was retrieved and scanned for dsRNA targets based on the rational dsRNA design rules (17). A BLAST search was performed to exclude those targets that shared homology with nonspecific human sequences (blast.ncbi.nlm.nih.gov/Blast.cgi) and to reject those targets with single nucleotide polymorphisms (genome.ucsc.edu/cgi-bin/hgBlat). All the saRNA mimics were synthesized by RiboBio Co., Ltd. (Guangzhou, China). Control dsRNA (dsControl) was specifically designed to have no homology to any known human sequences (13). A small interfering RNA (siRNA) that specifically targets human E-cadherin mRNA (siEcad) was designed to silence E-cadherin expression (18). All the custom saRNA sequences are listed in Table I.

The human BCa 5637 and T24 cell lines were purchased from American Type Culture Collection (Manassas, VA, USA). The BCa cells were maintained in RPMI-1640 or α-MEM (HyClone; GE Healthcare Life Sciences, Logan, UT USA) supplemented with 10% foetal bovine serum (Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA) and 2 mM/l L-glutamine in a humidified atmosphere containing 5% CO₂ at 37°C. Cells were seeded in growth medium without antibiotics at a density of ~40% and then transfected with 50 nM dsRNAs or miR-373 mimics using Lipofectamine RNAiMax (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol.

Total cellular RNA was extracted from cells that had been transfected for 72 h using the TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). RNA (500 ng) was reverse transcribed by using the Takara RT kit (Takara Biotechnology Co., Ltd., Dalian, China). RT reaction conditions were 37°C for 15 min and 85°C for 5 sec, according to the manufacturer's recommendation. qPCR amplification was conducted using SYBR Premix Ex Taq™ II (Takara Biotechnology Co., Ltd.) according to the manufacturer's instructions on the ABI PRISM 7300 system (Applied Biosystems; Thermo Fisher Scientific, Inc.). All qPCR reactions consisted of hot start activated step (95°Cfor 30 sec), and 40 cycles of 5 sec at 95°Cand 30 sec at 60°C. Endogenous gene expression was normalized to GAPDH levels in the cells. Each measure was done in triplicate and the change in expression of the target genes was quantified using the 2^−∆∆Cq method (19). Specific primer sets were designed for qPCR and synthesized by Invitrogen (Thermo Fisher Scientific, Inc.; Table II).

Briefly, cells were harvested 72 h after dsRNA or miRNA treatment and lysed in ice-cold NP40 (Beyotime Institute of Biotechnology, Haimen, China) lysis buffer containing the protease inhibitor phenylmethanesulfonylfluoride (Roche Diagnostics, Basel, Switzerland) to obtain the total cellular protein. Nuclear and cytoplasmic β-catenin proteins were isolated from cells using a cell fractionation kit (Wuhan Boster Biological Technology, Ltd., Wuhan, China), according to the manufacturer's instructions. Protein concentration in lysates was quantified using an enhanced bicinchoninic acid assay kit (Beyotime Institute of Biotechnology) according to the manufacturer's instructions. Proteins (40 µg) were separated using 8% SDS-PAGE and transferred onto polyvinylidene fluoride membranes (EMD Millipore, Billerica, MA, USA). Following blocking for 1 h at room temperature with 5% bovine serum albumin (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany), they were incubated overnight at 4°C with primary antibodies against E-cadherin (1:1,000; cat. no. 610181; BD Biosciences, San Jose, CA, USA), cyclin D1 (1:1,000; cat. no. AF0931; Affinity Biosciences, Cincinnati, OH, USA), c-MYC (1:1,000; cat. no. AF0358; Affinity Biosciences), MMP2 (1:1,000; cat. no. BM0569; Wuhan Boster Biological Technology, Ltd.), and β-catenin (1:1,000; cat. no. AF6266; Affinity Biosciences). β-actin (cat. no. BM0627; Wuhan Boster Biological Technology, Ltd.) was used as an internal control at a 1:500 dilution. Subsequently, the membranes were incubated with horseradish peroxidase-labelled secondary antibodies (1:5,000; cat. nos. S0001 and S0002; Affinity Biosciences) at room temperature for 2 h. Antigen-antibody complexes were visualized using an enhanced chemiluminescence assay kit (Clarity Max Western ECL Substrate; Bio-Rad Laboratories, Inc., Hercules, CA, USA) for protein detection. The ECL signals were detected by chemiluminescence imaging system (GeneGnome XRQ; Syngene Europe, Cambridge, UK). The expression of β-catenin in the nuclear and cytoplasmic compartments was determined as described above.

To evaluate the effects of the overexpression of dsEcad-346 and miR-373 on the migratory ability of BCa cells, a scratch assay was performed. Cells were transfected with saRNAs or the negative control sequence for 72 h. Then, ~5×10⁵ cells were reseeded in a new 6-well plate. When the cells had grown to confluence, a micropipette tip was used to create a cell monolayer. The scratch widths at 0 h, 12 h and 24 h after wounding were calculated from photomicrographs captured using an inverted microscope (Olympus Corporation, Tokyo, Japan).

The motility of cells was evaluated with a 24-well Boyden chamber with an 8-µm pore size polycarbonate membrane (Corning Incorporated, Corning, NY, USA). For the Transwell invasion assays, the membrane was pre-coated with 50 µl 33.3% Matrigel (BD Bioscience) to form a matrix barrier. Following transfection of candidate saRNAs for 72 h, 200 µl serum-free medium containing 5×10⁴ cells were seeded in the upper chamber. A total of 500 µl of medium with 10% FBS was added to the lower chamber as a chemoattractant. After incubation for 48 h, the cells were fixed with 4% formaldehyde for 10 min and stained with 0.1% crystal violet at room temperature for 15 min. Non-migrated cells in the upper chamber were removed and the cells that had invaded the lower surface of the membrane were images using an inverted microscope (Olympus Corporation), and the number of cells was counted in four randomized visual fields. The Transwell migration assay was performed using the same method as the invasion assay, except the upper chambers were Matrigel-free.

To assess the distribution of cells in different phases of the cell cycle, a flow cytometric analysis of the cell cycle distribution was performed. After transfection of the corresponding miRNA or dsRNA for 72 h, the cells were harvested and fixed with 70% cold ethanol overnight. Then, the cells were incubated with RNase A at 37°C for 30 min and resuspended in prop-idium iodide (PI; Cell Cycle Detection kit; Nanjing KeyGen Biotech Co., Ltd., Nanjing, China) at 4°C for 30 min. Then, the samples were analysed for PI fluorescence intensity by flow cytometry to assess the relative cellular DNA content, and the data were interpreted using CellQuest software (version 3.3; BD Biosciences). Apoptosis was measured by flow cytometry following staining of cells with Annexin V-fluorescein isothiocyanate (FITC) and PI in accordance with manufacturer's recommendations [Multisciences (Lianke) Biotech Co., Ltd., Hangzhou, China]. The cell apoptosis rate was calculated as follows: Cell apoptosis rate (%) = (apoptotic cell number ⁄ total cell number) ×100.

The localization of E-cadherin and β-catenin in cells was detected by immunofluorescence. At 72 h post-transfection, cells were fixed in 4% CH3OH for 10 min at room temperature and then permeabilised with 0.5% Triton X-100 (Sigma-Aldrich; Merck KGaA) at room temperature for 20 min. Following blocking with 3% (w/v) bovine serum albumin for 1 h at 25°C, the cells were incubated with the E-cadherin antibody (1:100; cat. no. 610181; BD, Biosciences) and the β-catenin antibody (1:100; cat. no. AF6266; Affinity Biosciences) overnight. FITC-conjugated IgG (1:50; Wuhan Boster Biological Technology, Ltd.) was used as the secondary antibody. Nuclei were counterstained with DAPI (0.05 mg/ml) for 5 min at room temperature. Subsequently, the samples were analysed with an Olympus inverted fluorescence microscope (Olympus Corporation).

All data are presented as the mean ± standard deviation. Statistical analyses were performed by SPSS version 13.0 software (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 5.0 software (GraphPad Software, Inc., La Jolla, CA, USA). The differences between two groups were analysed using two-tail Student's t-test and differences between more than two groups were analysed using one-way analysis of variance (post hoc test, Tukey). P<0.05 was considered to indicate a statistically significant difference.

E-cadherin expression is reduced in BCa cells and that reduced expression is associated with cancer progression and poor overall survival of patients with BCa (7,11). To enhance the expression of E-cadherin in BCa cells, two 21-nucleotide dsRNAs targeting the E-cadherin promoter at sequence positions -238 (dsEcad-238) and -346 (dsEcad-346) were designed and synthesized (Fig. 1A). The candidate dsRNAs were transfected into human BCa cell lines for 72 h. The mRNA and protein expression levels of E-cadherin were detected by RT-qPCR and western blotting, respectively. E-cadherin mRNA levels in dsEcad-346-treated T24 and 5637 cells were significantly elevated by ~8.3- and ~3.7-fold, respectively, compared with the levels in mock and dsControl transfected cells. However, there was no significant change in the dsEcad-238-transfected group (Fig. 1B). Upregulation of the E-cadherin mRNA was strongly associated with an increase in E-cadherin protein expression (Fig. 1C). These results suggest that the novel designed dsEcad-346 induces E-cadherin expression by targeting the E-cadherin gene promoter in BCa cells.

To further detect the effect of saRNAs on the activation of E-cadherin, the expression of E-cadherin was examined using immunofluorescence. The expression of E-cadherin was increased on the membrane surfaces of T24 and 5637 cells transfected with dsEcad-346 (Fig. 2A). In addition, the results show that miR-373 can also upregulate the expression of E-cadherin on the cell membrane surface of T24 and 5637 cells, this is similar to the results of previous experiments in our experimental group that miRNA-373 can activate E-cadherin expression in bladder tumour cells (20). E-cadherin is known to inhibit tumourigenicity and tumour transmission. Previous research demonstrated that upregulation of the expression of E-cadherin induces apoptosis and inhibits the proliferation of breast carcinoma cells in vitro and in vivo (21). To further evaluate the physiological effects of dsEcad-346 and miR-373 on BCa cell growth, flow cytometry was performed to assess the distribution of cells in the cell cycle. Compared with the dsControl group, the dsEcad-346- and miR-373-transfected cells demonstrated a marked accumulation in the G0/G1 phase and a decrease in the S and M phases (Fig. 2B).

To analyse the cell apoptosis rate induced by the upregulation of E-cadherin expression, dsEcad-346 and miR-373 were transfected into T24 and 5637 cells for 72 h, and the cell apoptosis rates were analysed by flow cytometry assay. As shown in Fig. 2C and D, overexpression of dsEcad-346 and miR-373 increased apoptosis in T24 and 5637 cells. The results indicated that transfection of dsEcad-346 and miR-373 blocked cell cycle progression in the G0/G1 phase, which promoted the apoptosis of cells.

To determine the effects of saRNAs on the migration and invasion capacity of BCa cells, wound healing and Transwell migration assays were performed. In the wound healing assay, the migration distances of the cells transfected with dsEcad-346 and miR-373 were significantly reduced compared with the dsControl group after 12 h and 24 h of incubation (Fig. 3A and B). The results of the Transwell assay revealed that dsEcad-346 and miR-373 treatment resulted in a significant decrease in motility and invasion compared with the dsControl group (Fig. 3C and D).

To understand the mechanism of the inhibition of growth and migration by saRNAs through the upregulation of the expression of E-cadherin, the expression of β-catenin/TCF target genes were examined. To address this issue, RT-qPCR and western blotting were performed to determine the mRNA and protein expression levels of c-MYC, MMP2, and cyclin D1. RT-qPCR results demonstrated that transfection of dsEcad-346 and miR-373 mimics significantly decreased c-MYC, MMP2 and cyclinD1 mRNA levels in T24 and 5637 cells (Fig. 4A and B). Western blotting further confirmed significant suppression of the protein expression of these genes (Fig. 4C and D).

To explore whether E-cadherin activation by saRNA can influence β-catenin status in BCa cells, β-catenin expression was evaluated by immunofluorescence and western blotting. Compared with that of the dsControl group, the abundance of plasma membrane β-catenin in T24 and 5637 cells treated with dsEcad-346 and miR-373 was significantly increased, and nuclear abundance of β-catenin was reduced (Fig. 5A). However, immunoblot analysis suggested that the change in the total level of β-catenin was not marked (Fig. 5B). Cytosolic and nuclear fractions of the cell lysates were isolated, and the transfection of dsEcad-346 and miR-373 resulted in an increased level of cytosolic β-catenin and a decreased level of nuclear β-catenin (Fig. 5B). These results suggest that saRNA redistributes β-catenin from the nucleus to the cell membrane, resulting in the inhibition of β-catenin/TCF target genes.

To demonstrate that dsEcad-346 and miR-373 supress the expression of β-catenin/TCF target genes via elevated expression of E-cadherin, siEcad was used to knockdown E-cadherin expression in T24 and 5637 cells. Following co-transfection dsEcad-346 or miR-373 with siEcad in BCa cells for 72 h, the results demonstrated that the effect of E-cadherin activation mediated by the two saRNAs was markedly abrogated at the mRNA and protein levels (Fig. 6A and B). Then, the expression of c-MYC, MMP2 and cyclin D1 were examined in BCa cells. As illustrated, co-transfection with siEcad increased c-MYC, MMP2 and cyclin D1 expression dsEcad-346 or miR-373 (Fig. 6C and D). This result suggested that deEcad-346 and miR-373-induced inhibition of c-MYC, MMP2 and cyclin D1 expression is associated with expression of E-cadherin.

Whether activation of E-cadherin was involved in the inhibition of the growth and metastasis of BCa cells by dsEcad-346 or miR-373 was investigated. T24 and 5637 cells were co-transfected by dsEcad-346 or miR-373 with siEcad to silence the expression of E-cadherin. Flow cytom-etry revealed a marked reduction in the percentage of G0/G1 cells and corresponding increases in S and G2/M populations in treatment groups co-transfected with siEcad compared with dsEcad-346 or miR-373 (Fig. 7A). These results indicate that silencing the expression of E-cadherin significantly reduced the cycle arrest that was mediated by saRNAs. Correspondingly, cell apoptosis was also decreased following the knockdown of E-cadherin (Fig. 6B and C). Additionally, the ability to inhibit cell migration and invasion mediated by dsEcad-346 and miR-373 was significantly reduced by depleting E-cadherin (Fig. 6D and E). Taken together, the findings indicated that the antitumour effect of dsEcad-346 and miR-373 in BCa cells is predominantly mediated via upregulation of E-cadherin expression.