Four pancreatic cancer cell lines (PANC-1, BxPC-3, AsPC-1 and Capan-1), the immortalized pancreatic ductal epithelial cells (HPDE6-c7) and the bacterial liquid pGenesil-1 were obtained from Chongqing Key Laboratory of Molecular Oncology and Epigenetics (The First Affiliated Hospital of Chongqing Medical University, Chongqing, China). The cells were maintained in high-glucose Dulbecco's modified Eagle's medium (DMEM, Hyclone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS, Gibco, Carlsbad, CA, USA) and antibiotics, including 100 U/ml penicillin and 100 mg/ml streptomycin (Beyotime Biotechnology, Jiangsu, China). Cells were incubated at 37°C in a humidified atmosphere with 5 % CO₂ and passaged 1:3 once 80% confluence was reached approximately every 2–3 days. Cells were passaged twice, seeded onto a six-well plate and cultured overnight. Then 250 μl serum-free DMEM was separately mixed with 4 μg plasmids and 5 μl Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) for 5 min at room temperature, subsequently, the diluted plasmid DNA was combined gently with diluted transfected agent Lipo for further 30 min. Then the mixtures without serum were added in each well for cell starvation at 37° with 5% CO₂ incubator, after 4–6 h starvation, the medium was changed with the complete medium containing 10% FBS.

Total RNA was separately isolated from cell lines by using TRIzol reagent (Takara, Dalian, China), reverse transcribed using Go Script Reverse Transcription System (Promega, Madison, WI, USA), and then semi-quantitative RT-PCR was performed with Go Taq^® Flexi DNA polymerase (Promega). The temperature profile used for PCR was 95°C initial denaturation for 2 min, followed by 36 cycles of 95°C denaturation for 30 sec, 55°C primer annealing for 30 sec and 72°C extension for 30 sec, ultimately with a 3 min extension step at 72°C. β-actin was amplified as a positive control with 23 cycles as well. The amplification products were electrophoresed into a 2% agarose gel (Biowest, Barcelona, Spain), then images were captured by Bio-Rad gel imaging instrument and analyzed with Quantity One software. The primer sets for DACH1 gene were: forward, 5′-TTGCATACGGTCTACACCAAG-3′; reverse, 5′-TGAGACCAGGGACAGAATGC-3′. β-actin primers: forward, 5′-CTCCATCCTGGCCTCGCTGT-3′; reverse, 5′-GCTGTCACCTTCACCGTTCC-3′.

The Homo sapien DACH1 gene sequence (GenBank Accession no. NM_080759.5) was identified in the NCBI gene bank. According to the principles of siRNA design, three small interfering RNA (siRNA) targeting human DACH1 (site at 1325th, 1543th and 1744th base) were designed by primer design software Primer Premier 5 and Oligo 6, which has no sequences homologous with any other gene. Moreover, a 21-base non-specific sequence was designed as a negative control. Then, the short hairpin structure was constructed with a stem-loop TTCG, connecting the DACH1-target siRNA sequence and its reverse complementary sequence. By BLAST with the sequences in the international nucleotide database, the specificities of the three DACH1-target shRNAs were confirmed. These four oligonucleotides were synthesized by Invitrogen (Shanghai, China).

The four groups of oligonucleotides were separately dissolved in distilled water, making the final concentration at 50 μmol/l, and were then mixed with 5X annealing buffer for DNA oligos according to the manual (Nobleryder, Beijing, China). The mixture was incubated in 95°C for 2 min, and then dropped to 4°C by gradient cooling every 90 sec in the polymerase chain reaction (PCR) machine. The empty vector pGenesil-1 was extracted from bacterial liquid by E.Z.N.A. Plasmid Mini kit I (Omega Bio-Tek, Norcross, GA, USA) and double digested with BamHI and HindIII enzyme (Takara) in a 37°C water bath for 6 h. Afterward, E.Z.N.A Gel Extraction kit (Omega Bio-Tek) was used to recycle the restriction fragment after endonuclease reaction. Subsequently, T4 DNA ligase (Takara) was chosen to connect the shRNA with the purified pGenesil-1 plasmid fragment at 16°C in the metal bath overnight. The ligation products were transformed into the competent Escherichia coli cells DH5α (Tiangen, Beijing, China). Recombinant plasmids were extracted by E.Z.N.A. Plasmid Mini kit I for analysis of enzyme identification and sequencing by Sangon Biotech Co. (Shanghai, China). The correct plasmids were named as pshRNA1-DACH1, pshRNA2-DACH1, pshRNA3-DACH1, pGenesil-1-shRNA-negative control (pshRNA-NC).

Endotoxin-free plasmids were extracted using the E.Z.N.A. Endo-Free Plasmid Mini kit I (Omega Bio-Tek). Pancreatic cancer cell line Capan-1, which expresses high level of endogenous DACH1, was transfected with the recombinant plasmids via Lipofectamine 2000 according to the protocol mentioned above. Cells were divided into six groups: pshRNA1/2/3-DACH1, pshRNA-NC, pGenesil-1 and untreated cell. Forty-eight hours after transfection, the expression of DACH1 was tested by semi-quantitative RT-PCR and western blotting. The band intensity values were then analyzed with Quantity One software to choose the most effective interference plasmid. Besides, the expression of GFP was observed under an inverted fluorescence microscope (Leica DM IRB).

Total protein was isolated from cells from each group by using RIPA lysis buffer and PMSF at a ratio of 100:1 and the protein concentration was detected by BCA protein assay kit (Beyotime Biotechnology). A total of 40 μg of protein lysate for each sample was separated by using sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE), followed by electrophoretic transfer to PVDF membranes (Millipore, Billerica, MA, USA) at a constant current of 250 mA and the membranes were blocked with 5% nonfat milk for 1–2 h. Then membranes were incubated with the first antibody at 4°C overnight, followed with HRP-conjugated goat anti-rabbit/mouse IgG (Cell Signaling Technology, Danvers, MA, USA). The proteins of interest were detected using BeyoECL plus reagent (Beyotime Biotechnology). The antibodies for western blot analyses were as follows: rabbit anti-DACH1 (Proteintech, Wuhan, China), mouse anti-β-actin, rabbit anti-Bcl-2, anti-caspase-3 and anti-PARP (Cell Signaling Technology), rabbit anti-E-cadherin, mouse anti-N-cadherin and rabbit anti-Vimentin (Abcam, Cambridge, MA, USA). The band intensity values were analyzed with Quantity One software. Results were recorded as fold changes compared to an internal reference standard.

To assess the influence of the interference plasmid transfections on cell growth rates, Capan-1 cells were seeded onto 96-well plates in triplicate at a density of 5000 cells/well and cultured overnight. Cells were transfected with three groups of plasmids respectively: pshRNA1-DACH1, which identified the highest inhibitory rate, the negative control pshRNA-NC and the empty vector pGenesil-1. At 24, 48, 72 h incubation after treated, 10 μl CCK8 solution was added to each well and wells containing medium without cells were used as blank control. The plates were incubated for further 2 h. Cell viability was measured with the Cell Counting Kit8 (CCK8, Dojindo Molecular Technologies, Japan). The optical density (OD) values were measured at 450 nm and data are shown as means ± SD.

Capan-1 cells were cultured in a 24-well plate with a cell density of 1×10⁴ each well. The screening concentrations of G418 were 400, 500, 600, 700, 800, 900, 1000, and 1100 μg/ml, and three replicates were made for each concentration. Then, a corresponding volume of G418 solution (Amresco, Solon, OH, USA) was added to each well combined with complement medium and cells, and changed with new culture medium every 2–3 days. After selection for two weeks, the minimum lethal concentration of G418 to Capan-1 cells was regarded as the final selection concentration.

Capan-1 cells were monolayer cultured in petri dishes until adherence confluence of ~70%, then transfected with pshRNA1-DACH1, pshRNA-NC and pGenesil-1 plasmid via Lipofectamine 2000. After 48 h, cells with different treatments were separately collected, the suspension was inoculated into a new 6-well plate, which was selected for 2 weeks using G418 with a mini selection concentration of 800 μg/ml. Culture medium and G418 were changed every 2–3 days. After 14 days of selection, cells were fixed with 4% paraformaldehyde for 30 min, followed by 0.5% crystal violet staining for further 30 min. Colonies with >50 cells were counted.

Cell mobility was evaluated by using a 24-well Transwell chamber with an 8 μm pore-size PET membrane (Corning Inc., Corning, NY, USA). Briefly, cells were divided into three groups transfected with appropriate plasmids as mentioned above. After 48 h transfection, cells were digested with Trypsin/EDTA and collected from culture flasks. Capan-1 cells were then suspended in serum-free medium at a density of 2.5×10⁵ cells/ml and 200 μl cell suspension was added into the upper well, and 700 μl of the complete medium containing 10% FBS was placed into the lower chamber. The 24-well plate was cultured at 37°C, in 5% CO₂ incubator for 24 h. Cells that migrated from the upper chamber to the lower surface of the PET membrane were fixed with 40 g/l paraformaldehyde for 30 min and then stained with 1 g/l crystal violet for another 30 min. Finally, the Transwell chamber was washed with PBS buffer and five random visual fields were counted to calculate the average number under a light microscope (×200).

To detect the effect of knockdown of DACH1 on cell invasion ability, 24-well Transwell chambers (8 μm pore-size) were used. Matrigel (BD Biosciences, San Jose, CA, USA) was mixed with cold DMEM without fetal bovine serum at a ratio of 1:7, which simulates the extracellular matrix structure. Then the mixtures were loaded onto the upper chamber and incubated at 37°C for 4 h. Capan-1 cells, transfected with the pshRNA-DACH1, pshRNA-NC and vector plasmids, were collected and suspended in serum-free medium. When the mixtures converted to solid state, approximately 3×10⁴ cells from each group were planted into the upper compartment and allowed to penetrate to the lower well containing complete medium with 10% FBS. After incubation for 36 h at 37°C humidified atmosphere with 5% CO₂, cells on the upper side of the filter were wiped off with cotton swabs, subsequently, the invaded cells on the bottom of the PET membrane were fixed with paraformaldehyde then stained with crystal violet as above. The numbers of the invasion cell were counted at five different areas.

The pshRNA-DACH1, pshRNA-NC and vector plasmids were separately transfected into Capan-1 cells as described above. For cell cycle analysis, cells were treated for 48 h, later digested by Trypsin without EDTA, suspended and washed twice by pre-cold PBS buffer, then fixed in ice-cold 70% ethanol at 4°C overnight and analyzed by flow cytometer in Academy of Life Science, Chongqing Medical University. For assessing cell apoptosis, three groups of cells were cultured for 48 h, digested with Trypsin/EDTA-free, suspended in PBS buffer and flow cytometry analysis was immediately performed. Each experiment was repeated three times.

Statistical analysis was performed using SPSS version 18.0 software. Measurement data were presented as mean ± SD. The expression difference of DACH1 between pancreatic cancer cell lines and normal cells was calculated by the Student's t-test, while comparison of three different treated groups was assessed using one-way analysis of variance. P<0.05 was considered to indicate a statistically significant difference.

DACH1 expression was altered in different cancers as previously described (7,9,10,12). To explore the expression difference of DACH1 between pancreatic normal epithelial cell HPDE6-c7 cells and cancer cell lines (PANC-1, BxPC-3, AsPC-1, Capan-1), semi-quantitative RT-PCR and western blotting were performed. As shown in Fig. 1A, the mRNA level of DACH1 was significantly higher in Capan-1 cell line, compared with that in the immortalized epithelial cells. Moreover, DACH1 protein expression was further confirmed abundant in Capan-1 cancer cells (Fig. 1B), which correspond to the RT-PCR analysis. These results indicated that DACH1 was upregulated in Capan-1 pancreatic cancer cells, we assumed DACH1 might act as a cancer-promoting gene in pancreas, which may play an opposite role in tumor proliferation and in the growth of pancreatic cells as compared to cancer cells originating from other organs such as breast, prostate and stomach.

Three DACH1-shRNA were designed and synthesized while a nonspecific shRNA as negative control (Table I), which were then inserted into pGenesil-1 vector plasmid. The recombinant plasmids and pGenesil-1 were identified by using EcoRI single enzyme digestion and EcoRI/HindIII double enzyme digestion. Our data demonstrated that with restriction endonuclease digestion, the DNA fragments of the four pshRNA-DACH1 were greater than pGenesil-1 (Fig. 2). Simultaneously, sequencing analysis showed that the inserted fragment completely matched with the expected sequence (Fig. 3). All these results confirmed that the interference plasmids were constructed successfully.

To analyze the knockdown efficiency of the interference plasmid targeting DACH1, samples were divided into six groups with different treatments: three interference groups (pshRNA1/2/3-DACH1), pshRNA-NC, the empty plasmid pGenesil-1 and the untreated cell group. After 48 h transfection, the mRNA and protein levels of treated and untreated groups were, respectively, detected by RT-PCR and western blotting. As shown in Fig. 4A, the RNA expression of the six groups was lowest in the pshRNA1-DACH1, which was consistent with the protein level (Fig. 4B). Based on these statistics, we chose pshRNA1-DACH1 as the most efficient interference plasmid to complete the following experiment.

The effect of shRNA targeting DACH1 on cancer cell proliferation and growth was then evaluated by using CCK8 assay and colony formation test. As shown in Fig. 5, the survival colony number of the interference group compared to the negative control was 56±14 versus 144±24 (P<0.01), indicating ~60% reduction in RNAi-mediated transfected cells. To further confirm the cell viability, CCK-8 kit was used in accordance with the instruction steps. The OD values were measured with enzyme-link meters at 450 nm and recorded in Table II. The statistics showed that cell proliferation rates were all significantly reduced after 24 h, 48 or 72 h transfection with pshRNA1-DACH1 in Capan-1 cells (Fig. 6). Therefore, the results demonstrated that knockdown of DACH1 could inhibit the proliferation and growth of Capan-1 cells.

Cell cycle and apoptosis were analyzed by flow cytometer with Annexin V-FITC/PI staining. Results of cell cycle distribution among pshRNA-, control- and vector-transfected groups have no significant difference (data not shown). However, representative results of apoptosis in different treatment groups are shown in Fig. 7, and revealed a significant increase of apoptotic cells in the pshRNA-transfected group in comparison with the control one. To further explore the significance of DACH1 on cell apoptosis and the involved mechanisms, we investigated the expression of apoptotic markers using western blotting. Our data showed that downregulation of the DACH1 activity inhibited Bcl2 accumulation in Capan-1 cancer cells, inducing its downstream target genes the cleaved-caspase 3 and cleaved poly (ADP-ribose) polymerase PARP expression, compared to the control groups (Fig. 8A and B). Taken together, these results indicated that knockdown of DACH1 expression could inhibit tumor cell proliferation and growth, induced cell apoptosis, which may function as a tumor-promoting gene in pancreatic cancer.

To determine whether DACH1 could influence tumor cellular mobility, Transwell assay was performed. In Transwell migration or invasion experiments, the invaded cell numbers were significantly less in the pshRNA1-DACH1 interference cells than the control cells (P<0.05) (Fig. 9A and B). Our results demonstrated DACH1 indeed possess the ability to promote metastasis in Capan-1 pancreatic cancer cells.

Tumor cell metastasis is a complex process with multiple factors, which are the biological properties of malignant carcinomas. Epithelial-mesenchymal transition (EMT) is known as a dynamic cellular process and often occurs during cancer cell migration and invasion (13). To extend our analysis, we further examined whether downregulated expression of DACH1 in Capan-1 cancer cells was able to trigger a shift from mesenchyme to epithelia. As shown in Fig. 10A and B, a remarkable increase of E-cadherin and reduction of N-cadherin and Vimentin expression were detected in Capan-1 cells which were transfected with pshRNA-DACH1 plasmid. These findings suggested that knockdown the expression of DACH1 in Capan-1 cells could result in preventing EMT.