LncRNA PVT1 promotes proliferation and invasion through enhancing Smad3 expression by sponging miR-140-5p in cervical cancer

Human normal cervical epithelial cell line (End1/ E6E7) and human cervical cell lines (HeLa and SiHa) were obtained from the Cell Bank of the Chinese Academy of Sciences (Shanghai, China). The cells were maintained in Dulbecco’s minimal Eagle’s medium (DMEM; Invitrogen, Carlsbad, CA, USA) containing 6% fetal bovine serum (FBS; Gibico, Grand Island, NY, USA) and 1% penicillin/ streptomycin at 37°C in a humidified atmosphere containing 5% CO².

A bioinformatics website (http://www.rna-society.org/raid2/index.html) was used to identify the binding sites for miR-140-5p, lncRNA PVT1 and Smad3, and obtain the fragment sequences containing action sites. The bioinformatics software was then applied to predict the binding sites between miR-140-5p and PVT1 as well as between miR-140-5p and Smad3. Then the wild-type PVT1 (PVT1-WT) , mutant PVT1 (PVT1-MUT), wild-type Smad3 (Smad3-WT) and mutant Smad3 (Smad3-MUT) containing predicted miR-140-5 binding sites were synthesized and cloned into a pGL3 vector. Then, the above constructed vectors were transfected into cervical cancer cells in the presence of miR-NC or miR-140-5p inhibitor. Luciferase activity was detected using the Dual-Luciferase Reporter Assay System (Promega, Madison, USA).

Sh-PVT1, negative control (sh-NC), miR-140-5p mimics and miR-140-5p inhibitor were synthesized by GenePharma (Shanghai, China). After reaching 60%-70% confluence, the cervical cancer cells were transfected by using Lipofectamine 2000 (Life Technologies, Carlsbad, CA) according to the manufacturer’s instructions.

After transfection, cells were seeded in 6-well plates at a density of 1×10³ cells/well. The cells were then fixed and stained with 0.1% crystal violet after incubation for 14 days. The numbers of colonies were counted under microscope. This assay was performed at least three times.

Cells were incubated in 6-well plates at a density of 1 × 10⁶ cells/well. After transfection, the cells were cultured with serum-free medium for 24 h. Wounds were created by using a sterile plastic pipette tip to scratch the cell layer. Then the cells were incubated with medium containing 10% FBS for 48 h. Images were acquired by microscope. This assay was performed at least three times.

The transwell assays were determined using transwell chambers (Corning, NY, USA) as previously described.22 Briefly, transfected cells were suspended in serum-free medium and seeded into the upper chamber, while medium containing 10% FBS was added in the bottom chamber. After incubation at 37°C for 24 h, the cells that invaded through matrix membrane were fixed with 4% paraform-aldehyde and stained with crystal violet. Then, stained cells were photographed and counted from five random fields per filter for analysis. This assay was performed at least three times.

Total RNA was extracted from cell samples using Trizol reagents (Thermo Fisher Scientific, Waltham, MA, USA) from cell lines, and cDNA was synthesized using PrimeScript^TM cDNA Kit (Takara, Dalian, China) according to the manufacture’s protocols. qRT-PCR was determined using an ABI 7000 Prism Step One plus detection system (Life Technologies, USA). The relative expression was normalized using GAPDH as an internal reference gene, and U6 was used as the endogenous control of miR-140-5p. Fold changes were calculated using the formula 2^-ΔΔCt. All qRT-PCR reactions were performed three times independently. The primer sequences used for qRT-PCR as follow.

Cell samples were harvested and lysed using cell lysis buffer on ice. Total proteins were extracted and concentrations were detected using a BCA protein assay kit (Beyotime, China). Same amounts of protein samples were isolated by 12% SDS-PAGE gels and then transferred onto polyvinylidene fluoride (PVDF) membranes. The membranes were then blocked with TBS-T containing 5% nonfat milk powder for 2 h at RT and incubated with primary antibodies at 4°C overnight. The membranes were then incubated with secondary antibody for 1 h at RT. The images were visualized using an enhanced chemiluminescence system. Signals were quantified and analyzed using the Image-Pro Plus 6.0 software (Media Cybernetics, Sarasota, USA). The following primary antibodies were applied: anti-Smad3 (1:1000; #9513), anti-E-cadherin (1:1000; #14472), anti-vimentin (1:1000; #3932), anti-N-cadherin (1:1000; #4061) anti-Snail (1:1000; #3879) (All purchased from Cell signaling Technology). The corresponding HRP-conjugated antibody (1: 5000; Abcam, UK) was used as the secondary antibody.

Statistical analysis was performed using SPSS 22.0 software (IBM Corporation, NY, USA). All data were expressed as the mean ± standard deviation (SD). All experiments were performed triplicate independently. Comparisons were determined using one-way analysis of variance (ANOVA) or Student’s t-test. P < 0.05 was considered significantly different.

PVT1 forward 5’-AAAACGGCAGCAGGAAATGT-3’ and reverse 5’-GGAGTCATGGGTGTCAGACA-3’.

miR-140-5p forward 5’-GGGCCAGTGGTTTTACCCTA-3’

and reverse 5’-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGAC CTACCA-3’.

Smad3 forward 5’-CTCCAAACCTATCCCCGAAT-3’

and reverse 5’-CCTGTTGACATTGGAGAGCA-3’. U6 forward 5′-AAAGCAAATCATCGGACGACC-3′

and reverse 5′- GTACAACACATTGTTTCCTCGGA-3′.

GAPDH forward 5′-AGAAGGCTGGGGCTCATTTG -3′

and reverse 5′-AGGGGCCATCCACAGTCTTC-3’.

To determine the roles of lncRNA PVT1, miR-140-5p and Smad3 in cervical cancer metastasis, expressions of lncRNA PVT1, miR-140-5p and Smad3 in cervical cancer cell lines (HeLa and SiHa) and human normal cervical cell line (End1/E6E7) were determined by qRT-PCR and western blotting. The results in Figure 1A and 1B show that lncRNA PVT1 expression was significantly upregulated in cervical cancer cell lines compared with that in the normal cervical cell line by over 1.6-fold (P < 0.01) and 1.3-fold (P < 0.05) respectively. While, the expression of miR-140-5p was remarkably downregulated in cervical cancer cell lines compared with the normal cervical epithelial cells as their respective reductions were 51.8% and 36.7% (P < 0.01). Furthermore, the mRNA and protein expressions of Smad3 were significantly increased in the cervical cancer cell lines compared with that in the normal cervical epithelial cells with 1.5-fold, 1.3-fold increase in mRNA levels and 3.8-fold, 3.3-fold increase in protein levels, respectively (Figure 1C and 1D, P < 0.01).

Figure 1

Bioinformatics analysis showed that miR-140-5p might be a target of PVT1 (Figure 2A). Dual luciferase assay was further conducted to determine whether PVT1 could regulate miR-140-5p expression by acting as a molecular sponge, showing that miR-140-5p mimics could significantly inhibit the luciferase activity of PVT1-WT in both cancer cell lines, but had no significant effect on that of PVT1-MUT (Figure 2C and 2D). Furthermore, bio-informatics analysis revealed miR-140-5p can also bind directly to Smad3 (Figure 2B). The results of Figure 2E and 2F showed that miR-140-5p mimics could remarkably suppress the luciferase activity of Smad3-WT in both cervical cancer cell lines, and had no effect on that of Smad3-MUT.

Figure 2

Since PVT1 was highly expressed in cervical cancer cells, revealing that lncRNA PVT1 could be a potential oncogene during the progression of cervical cancer. To figure out the function of lncRNA PVT1 during the progression of cervical cancer cells, PVT1 specific sh-RNA was applied to knockdown the expression of PVT1. As shown in Figure 3A and 3B, the expressions of PVT1 were significantly downregulated in sh-PVT1 transfected cervical cancer cells compared with that in cells transfected with sh-NC and the control group as their respective reductions were 39.8% and 43.7% (P < 0.01). While, the expressions of miR-140-5p in sh-PVT1 transfected cervical cancer cells were significantly upregulated compared with cells transfected with sh-NC and the control group by over 1.4-fold (P < 0.05, Figure 3C) and 1.9-fold respectively (P < 0.01, Figure 3D). Meantime, the mRNA levels (Figure 3E and 3F) and protein levels (Figure 3G and 3H) of Smad3 were significantly decreased after transfecting with sh-PVT1 in cervical cancer cells with 25.3%, 38.5% reductions in mRNA levels and 57.2%, 62.3%

Figure 3

reductions in protein levels, respectively (P < 0.01). To explore the role of PVT1 in the proliferation of cervical cancer cells, colony formation assay was performed. As shown in Figure 3I and 3J, PVT1 knockdown significantly inhibit the proliferation of cervical cancer cells compared with the sh-NC group. Next, the migration activities of cervical cancer cells were detected by wound healing assay.

The results showed that PVT1 knockdown could significantly decrease the migration activities of cervical cancer cells compared with the sh-NC groups, with an inhibition rate of 51.2% and 61.3% in both cells lines respectively. (Figure 4A-4D, P < 0.01). Moreover, invasion capacity of cervical cancer cells was further determined by transwell assays. The results showed that PVT1 suppression could remarkably inhibit the invasion capacities of cervical cancer cells compared with the sh-NC groups, as their respective reductions were 59.2% and 61.2%, respectively (Figure 4E-4H, P < 0.01). Besides, expressions of E-cadherin, N-cadherin, vimentin and Snail were determined. The results showed that downregulation of PVT1 lead to significant increase in the expressions of E-cadherin (1.9-fold and 2.1-fold) and significant decrease in the expressions of N-cadherin (41.2% and 37.1%), vimentin (30.1% and 13.6%) and Snail (39.2% and 45%) compared with the sh-NC group (Figure 4I and 4J, P< 0.05). These data indicated that PVT1 could serve as an oncogenic lncRNA in the progression of cervical cancer.

Figure 4

To determine whether miR-140-5p was implicated in the effect of lncRNA PVT1 on the progression of cervical cancer cells, sh-PVT1 transfected cervical cancer cells were transfected with or without miR-140-5p inhibitor. The results showed that miR-140-5p inhibitor could decline the PVT1 inhibition-mediated decreasing effect on the Smad3 expression in cervical cancer cells (Figure 5A-5D). Colony formation assay indicated that inhibition of miR-140-5p could reserve the inhibition effects of PVT1

Figure 5

downregulation on proliferation of cervical cancer cells (Figure 5E and 5F). Similar results could also be observed in wound healing and transwell assays, that decreased migration and invasion abilities induced by PVT1 downregulation could be rescued by miR-140-5p inhibitor (Figure 6A-6H). Besides, treatment with miR-140-5p inhibitor could suppress sh-PVT1-induced upregulation of E-cadherin expression and reverse sh-PVT1-induced down-regulation of N-cadherin, vimentin and Snail expressions (Figure 6I-6J). These data indicated that PVT1 knockdown could suppress the progression of cervical cancer cells via miR-140-5p.

Figure 6