Three human lung cancer cell lines (A549, H157 and H460) were cultured in a DMEM/F12 medium supplemented with 10% fetal calf serum (FCS) (both from HyClone, Beijing, China). Recombinant Wnt3a (5036-WNP-010; R&D Systems, Minneapolis, MN, USA) was reconstituted at 200 μg/ml in sterile PBS containing at least 0.1% bovine serum albumin and stored at −20°C. The cells were placed in a fresh medium plus 50 or 100 ng/ml Wnt3a and were cultivated for 24 h before cell analysis. PBS was used as control.

The plasmid encoding the shRNA targeting the Notch3 gene (sc-37136-SH; Santa Cruz Biotechnology, USA) and the non-target shRNA control plasmid were transfected into cells with TranSmarter (Abmart, Shanghai, China). After 48 h of transfection, the cells were selected with puromycin (Sigma-Aldrich, St. Louis, MO, USA) at a final concentration of 1 μg/ml for 2 weeks. Puromycin-resistant clones were isolated and subsequently cultured for the following experiment. Validation of the Notch3 silencing in the transfected cells was carried out by western blotting.

The cells seeded in 35-mm dishes were stained according to the following protocol. The medium mixture was discarded and 4% paraformaldehyde was added to fix the cells at room temperature (RT) for 10 min, and cells were washed with PBS for 5 min in a shaker and 0.2% Triton X-100 was added for 10 min. The cells were washed with PBS twice and TRITC-conjugated phalloidin (Sigma-Aldrich) was added at RT for 40 min, and the cells were washed with PBS containing 0.02% Triton X-100 for three times. Nuclei were counterstained with Hoechst 33342 (5 μg/ml) for 30 min at RT, cells were washed with PBS containing 0.02% Triton X-100 for three times and finally, 200 μl PBS was added. The staining was examined under a laser scanning confocal microscope (FV1000; Olympus, Tokyo, Japan).

The total RNA was prepared from the cultured cells using TRIzol reagent (Invitrogen Inc., Carlsbad, CA, USA) according to the manufacturer’s instructions. The cDNA was synthesized using oligo(dT) as a primer. Primers (Table I) were designed with Primer3 on line. The PCR was carried out using a PCR kit (RT-PCR AMV 3.0 kit; Takara, Tokyo, Japan). The PCR products were electrophoresed on a 2% agarose gel and visualized by staining with GeneFinder (Baiweixin, Xiamen, China). The bands were quantified by densitometry to obtain the integrated density values (IDV). The relative amounts of each protein to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) are represented as the ratio of their IDV in the histograms.

The cells in the medium with 1% FCS were seeded into the upper chambers of a 24-well Transwell plate (3422; Corning, NY, USA) with BioCoat Matrigel (1:8; BD Bioscience, San Jose, CA, USA). The medium with 10% FCS was added to the lower chambers as a chemoattractant. After 24 h of incubation, cells that invaded through the membrane filter were fixed with 75% ethanol and stained with 0.1% crystal violet. The number of invading cells was counted under an inverted microscope with a digital CCD imaging system [IX70/SPOT RT-KE (color), Olympus/DI, Japan/USA] with a ×10 objective in five random fields.

Soft agar plates were prepared using a Gene Med soft-agar kit (GMS10024; Gene Med, Shanghai, China) according to the manufacturer’s instructions. To prepare the base layer, reagent A mixed with reagent B in equal volumes was added into the 12-well plate and incubated at 50°C for 2 h to allow the agar to solidify. Top layer agar was prepared by mixing reagent C with reagent A at a 3:1 ratio for semi-solidity, and the cells suspended in this top layer of agar were plated over the base layer. Then, the plates were incubated at 37°C overnight. The next day the liquid reagent D was added over the top layer of agar. Moreover, the cultures were fed with 0.25 ml of reagent D at 2-day intervals for 2 weeks. In the Wnt3a treatment groups, Wnt3a was added to the semi-solidity agar reagent and the reagent D was replenished to the concentration of 100 ng/ml. Finally, colonies with diameter >120 μm were scored under a light microscope at low magnification [IX70/SPOT RT-KE (color), Olympus/DI, Japan/USA]. The numbers of the colonies in the treatment groups were converted to a percentage of the control (by considering the control as 100%).

Cells were lysed in a phospho-lysis buffer (50 mM Tris-Cl, pH 7.5, 150 mM NaCl, 1 mM MgCl₂, 0.5% NP-40, 1 mg/ml BSA, 0.1 mM PMSF). Samples were analyzed by 10% SDS-polyacrylamide gel electrophoresis, followed by western blotting using rabbit monoclonal anti-Notch3 (D11B8; Cell Signaling Technology, Beverly, MA, USA), rabbit polyclonal N-cadherin (W745), rabbit polyclonal E-cadherin (R868), (both from Bioworld Technology, Nanjing, China), and vimentin (ab8069; Abcam, Cambridge, MA, USA), respectively, and the secondary antibody of goat anti-rabbit IgG conjugated with horseradish peroxidase (HRP; HuaAn Biotech, Hangzhou, China). The membrane was developed using chemiluminescent reagents (SuperSignal West Pico Chemiluminescent Substrate; Pierce, Rockford, IL, USA). The bands were quantified by densitometry to obtain the IDV. The relative amounts of each protein to actin are represented as the ratio of their IDV in the histograms.

Data are presented as the mean ± SE of at least three independent experiments. The one-way analysis of variance (ANOVA) was used for statistical analysis. Statistical significance was accepted at P<0.05.

Wnt3a, a member of the secreted Wnt ligands, is consistently used to mimic the biochemical effects of the canonical Wnt signaling pathway (13,14). Notch3 is one of the four Notch receptors identified in mammals. Previous studies concerned with the pathogenesis of lung cancer have identified that Notch3 plays an essential role in NSCLC (15,16). Compared with the control, Wnt3a treatment increased the mRNA expression of Notch3 and its downstream gene, HES1, in all the three cell lines in a dose-dependent manner. As for HEYL, another downstream gene of Notch3, Wnt3a treatment dose-dependently increased its mRNA expression in the H157 and A549 cells. In the H460 cells, only 100 ng/ml of Wnt3a increased the mRNA expression of HEYL (Fig. 1A and B). To further confirm the effects of Wnt3a on Notch3 expression, we assessed the protein level of Notch3, using western blotting in the absence or presence of 100 ng/ml of Wnt3a. Consistent with the data at the mRNA level, Wnt3a treatment significantly increased the expression of Notch3 at the protein level (Fig. 1C and D). Altogether, these data strongly indicate that Wnt3a may upregulate Notch3.

Notch3 is one of the four Notch receptors identified in mammals and its overexpression is frequently noted in NSCLC. In the present study, the expression of Notch3 was confirmed by western blotting. Three lung cancer cell lines transfected with Notch3 shRNA showed a significant reduction in Notch3 protein expression (Fig. 2A and B).

As in vitro invasion is one of the characteristics of metastasis, the Transwell invasion assay was carried out to assess the effects of Wnt3a and Notch3 on cell invasion. Cell invasion was increased in the three cell lines treated with 100 ng/ml Wnt3a. Notch3 reduction led to decreased cell invasion in the three cell lines. Moreover, Notch3 reduction significantly attenuated the effects of Wnt3a treatment on the invasive ability of cells (Fig. 3A and B). The data demonstrated that Wnt3a and Notch3 promoted in vitro invasion of the NSCLC cells and indicate that the Wnt pathway may promote cell invasion partially via Notch3.

For cells to undergo metastasis, they must have the ability to overcome anoikis and survive without cell-substratum interaction (17). The anchorage-independent growth in soft agar has been shown to be one of the independent factors of the metastatic potential in cancer (18,19). To further analyze the effects of Wnt3a and Notch3 on the characteristics of metastatic outgrowth, a soft-agar colony assay was used in the present study. When cultured in the reagent with 100 ng/ml Wnt3a, the cells formed compact spherical colonies that grew larger in size and in number, when compared to the control cells. Meanwhile, Notch3 reduction decreased the size and the number of the colonies. Consistent with the results of the cell invasion assay, Notch3 reduction also weakened the effects of Wnt3a on the size and number of colonies, suggesting that Notch3 may be involved in the colony outgrowth prompted by the Wnt signaling pathway in the NSCLC cells (Fig. 4).

Accumulating evidence suggests that the EMT is a key event in cancer metastasis. During EMT, cell morphological changes and dynamic remodeling of the actin cytoskeleton consistently occur (20,21). In the control cells, the majority of cells maintained typical epithelial morphology, while some of the Wnt3a-treated cells underwent elongation to become fibroblast-like spindle-shaped cells (Fig. 5A, black arrowhead). The F-actin staining also showed that some of the Wnt3a-treated cells became longer (Fig. 5B, white arrowhead). Moreover, we observed that the F-actin filaments of the control cells were mainly organized in cortical thin bundles whereas the actin filaments of the Wnt3a-treated cells appeared to be assembled into thick bundles at the cell surface in the A549 cells (Fig. 5B, white arrowhead). However, compared to the Wnt3a treatment, Notch3 shRNA appeared to have opposite effects on the cell morphology and actin reorganization (Fig. 5A and B). These results indicate that Wnt3a and Notch3 may induce the EMT process in the cultured cells.

To further explore the roles of Wnt3a and Notch3 on EMT, we measured the protein expression levels of E-cadherin, N-cadherin and vimentin, key protein markers involved in EMT. The downregulation of the epithelial marker E-cadherin and the upregulation of the mesenchymal markers N-cadherin and vimentin are all hallmarks of EMT (20). Wnt3a treatment significantly induced the downregulation of E-cadherin and the upregulation of N-cadherin and vimentin at the protein level. In contrast, Notch3 shRNA significantly induced the downregulation of N-cadherin and vimentin and the upregulation of E-cadherin. Moreover, Notch3 shRNA partly inhibited the regulation of Wnt3a on the expression of three protein factors (Fig. 5C and D). Taken together, these findings indicate that Wnt3a treatment is sufficient to induce EMT in the NSCLC cells, in which Notch3 is involved.