LncRNA PVT1 accelerates malignant phenotypes of bladder cancer cells by modulating miR-194-5p/BCLAF1 axis as a ceRNA

Up-regulated expression of PVT1 and down-regulated expression of miR-194-5p in BCs

The relative expression level of PVT1 and miR-194-5p was measured through RT-qPCR in BC samples. Compared to para-carcinoma tissues, the relative expression level of PVT1 was significantly increased about 2.28 times (43 of 70) of BC samples (P<0.001) (Figure 1A, 1B), and the relative expression level of miR-194-5p was significantly decreased about 42.92% (46 of 70) of BC samples (P=0.023) (Figure 1C, 1D). Compared to SV-HUC-1, the relative expression level of PVT1 was up-regulated in both BC cells, T24 about 3.76 times (P < 0.001) and UM-UC-3 about 2.84 times (P=0.004) (Figure 1E), and the relative expression level of miR-194-5p was decreased in both BC cells, T24 about 44.16% (P=0.003) and UM-UC-3 about 63.77% (P = 0.001) (Figure 1F). PVT1 expression levels were statistically positively correlated with miR-194-5p expression levels in BC samples and cells (Figure 1G, 1H). As following in Table 1, the up-regulated expression of PVT1 was closely associated with histological grade (P = 0.014), and TNM stage (P< 0.001) in bladder carcinomas, the low expression of miR-194-5p was significantly related to histological grade (P=0.006) and TNM stage (P<0.001). But sex, age, tumor size, and lymph node metastasis were no obvious correlation with the relative expression levels of PVT1 and miR-194-5p. Our results manifested that PVT1 could act as the oncogene and miR-194-5p could act as the antioncogene in BCs.

PVT1 acted as the oncogene

After 48 hours of transfection for siRNA, miRNA mimics or inhibitor and pcDNA3.1-PVT1, the relative expression levels of PVT1 and miR-194-5p were detected by qRT-PCR in T24, UM-UC-3 and SV-HUC-1. After transfection, the relative expression of PVT1 was down-regulated about 41.51% in T24 (p= 0.002) and was reduced about 44.32% in UM-UC-3 (P= 0.006) cells (Figure 2A). The relative expression of PVT1 was notably increased about 2.41 times in SV-HUC-1 (p<0.001) after 48 hours of transfection for pcDNA3.1-PVT1 (Figure 2B). The relative expression of miR-194-5p was reduced about 53.60% in SV-HUC-1 (P= 0.0185) by pcDNA3.1-PVT1 (Figure 2C).

CCK-8 was carried out to detect whether the knockdown of PVT1 could restrain the proliferation and pcDNA3.1-PVT1 could facilitate the proliferation in BC T24, UM-UC-3 and SV-HUC-1 cells. Our study demonstrated that si-PVT1 (Figure 2D, 2E) notably restrained both BC cells proliferation (p < 0.01). PcDNA3.1-PVT1 (Figure 2F) remarkably facilitated SV-HUC-1 cell proliferation (p < 0.01).

EdU was elucidated cell proliferation as well. Compared with control group, EdU positive T24 and UM-UC-3 cells in si-PVT1 group were reduced and pcDNA3.1-PVT1 group was reversed after transfection in SV-HUC-1 cell line.

EdU assay proved that the quantity of EdU positive cells was decreased about 47.98% in T24 (P=0.006) (Figure 2G) and about 40.53% in UM-UC-3 (P = 0.002) (Figure 2H) in si-PVT1 group. The quantity of EdU positive cells was increased about 2.31 times in SV-HUC-1 (P< 0.001) (Figure 2I) of pcDNA3.1-PVT1 group.

Our study manifested that the knockdown of PVT1 restrained bladder cell lines proliferation and over-expression PVT1 facilitated bladder cell lines proliferation.

Cell migration was detected after the transfection of siRNA and plasmids by scratch assay. Scratch assay that revealed that the ratio of the relative migration was decreased about 41.29% in T24 (P = 0.001) (Figure 3A) and decreased about 42.18% in UM-UC-3 (P = 0.004) (Figure 3B) in si-PVT1 group. The ratio of the relative migration in the pcDNA3.1-PVT1 group was up-regulated about 2.032 times in SV-HUC-1 (P<0.001) (Figure 3C).

Our results concluded that the knockdown of PVT1 restrained bladder cell lines migration. Overexpression PVT1 expression facilitated bladder cell lines migration.

Cell apoptosis was measured with flow cytometry assay. Compared with control groups, the ratios of apoptosis were up-regulated about 7.91 times in T24 (P<0.001) and increased about 4.99 times in UM-UC-3 (P=0.011) (Figure 3D, 3E) after transfection si-PVT1.

Compared with control groups, the ratios of apoptosis were dramatically reduced about 0.332 times in SV-HUC-1 (P=0.003) (Figure 3F) after the transfection of pcDNA3.1-PVT1.

In brief, the knockdown of PVT1 facilitated bladder cell lines apoptosis and overexpression PVT1 restrained bladder cell lines apoptosis.

MiR-194-5p acted as the antioncogene

The relative expression levels of miR-194-5p were remarkably reduced about 48.64% in T24 (p =0.044) and about 75.51% in UM-UC-3 (P<0.001) at 48 hours after the transfection of miR-194-5p inhibitor (Figure 4A). The relative expression levels of miR-194-5p were increased in about 2.460 times in T24 (p=0.002) and 3.259 times in UM-UC-3 (p< 0.001) at 48 hours after the transfection of miR-194-5p mimics (Figure 4A).

The results demonstrated that miR-194-5p mimics significantly restrained both BC cells proliferation remarkably (Figure 4B, 4C) (p < 0.01). MiR-194-5p inhibitor accelerated both BC cells proliferation (Figure 4B, 4C) (p < 0.01).

The quantity of Edu positive cells in miR-194-5p mimics group was notably decreased about 64.34% in T24 (P= 0.001) and dramatically reduced about 66.34% in UM-UC-3 (P< 0.001) (Figure 4D, 4E). The quantity of Edu positive cells in miR-194-5p inhibitor group was up-regulated about 1.70 times in T24 (P= 0.015) and elevated about 1.69 times in UM-UC-3 (P= 0.001) (Figure 4D, 4E). Our study illustrated that the up-regulation of miR-194-5p restrained BC cells proliferation and suppression of miR-194-5p accelerated BC cells proliferation.

The ratio of the relative migration was down-regulated about 52.10% in T24 (P< 0.001) and decreased about 42.53% in UM-UC-3 (P=0.0002) (Figure 4F, 4G) after the transfection of miR-194-5p mimics. The ratio of the relative migration was up-regulated about 2.121 times in T24 (P< 0.001) and increased about 2.002 times in UM-UC-3 (P=0.0003) (Figure 4F, 4G) after transfection miR-194-5p inhibitor. These data suggested that over-expression of miR-194-5p suppressed BC cells migration and knockdown of mi-194-5p accelerated BC cells migration.

Compared to NC groups, the ratios of apoptosis were gone up about 2.434 times in T24 (P=0.0001) and increased about 2.033 times in UM-UC-3 (P=0.0004) (Figure 4H, 4I) after transfection miR-194-5p mimics. Compared to NC groups, the ratios of apoptosis were lessened about 0.230 times in T24 (P= 0.0002) and about 0.228 times in UM-UC-3 (P=0.0005) (Figure 4H, 4I) after transfection miR-194-5p inhibitor. Ultimately, over-expression of miR-194-5p accelerated BC cells apoptosis and down-regulation of mi-194-5p restrained BC cells apoptosis.

PVT1 sponges miR-194-5p

Compared with si-NC groups, the miR-194-5p expressions were increased about 2.393 times in T24 (P=0.001) and about 3.223 times in UM-UC-3 (P= 0.004) (Figure 5A) in si-PVT1 groups. We used bioinformatics databases to predict some underlying binding sites of PVT1 with miR-194-5p. We verified the predictions by luciferase reporter assay. Our study proved that miR-194-5p mimics significantly suppressed PVT1 wild type reporter luciferase activity. Compared with the co-transfections NC+ pmirGLO-PVT1-Wt, the luciferase activity was notably reduced about 44.31% in T24 (P=0.0002) and about 45.88% in UM-UC-3 (P<0.001) in the co-transfections miR-194-5p mimics + pmirGLO-PVT1-Wt, nevertheless, miR-194-5p could not suppress the PVT1 mutant reporter vector luciferase activity (Figure 5B).

PVT1 sponging miR-194-5p mediated BC cell progression

The luciferase reporter assays have verified that PVT1 sponged miR-194-5p. Firstly, we must verify that miR-194-5p co-regulated the repressive progression induced by knockdown PVT1 in BC cells. That si-PVT1 co-transfected miR-194-5p mimics could manifest more powerfully suppressed effects on BC cells proliferation (Figure 5C, 5D) and migration (Figure 5E, 5F) than si-NC co-transfection with NC (si-NC+NC). Meanwhile, compared with si-NC+NC group, apoptosis was accelerated in si-PVT1 co-transfection miR-194-5p mimics (si-PVT1+miR-194-5p mimics) group (Figure 5G, 5H). Conversely, miR-194-5p inhibitor could partially reverse inhibited effects on BC cells progression induced by si-PVT1.

The CCK-8 assays have manifested that si-PVT1 co-transfected miR-194-5p mimics (Figure 5C, 5D) remarkably restrained both BC cells proliferation in (p<0.01 in T24 and UM-UC-3). At the same time, miR-194-5p inhibitor could partially reverse inhibited effects on BC cells proliferation induced by si-PVT1 (Figure 5C, 5D).

Compared with si-NC+NC, si-PVT1 co-transfected miR-194-5p mimics could decrease the ratio of the relative migration about 77.28% in T24 (P= 0.0003) and about 62.39% in UM-UC-3 (P< 0.001). Moreover, miR-194-5p inhibitor could partially reverse inhibited effects on BC cells migration induced by si-PVT1 and increased about 19.86% in T24 and about 23.55% in UM-UC-3 cell lines (Figure 5E, 5F).

Compared with si-NC+NC, si-PVT1 co-transfected miR-194-5p mimics could accelerate the ratio of the relative apoptosis about 10.44 times in T24 (P<0.001) and about 9.91 times in UM-UC-3 (P<0.001), what’s more, miR-194-5p inhibitor could partially reverse promoting apoptosis on BC cells migration induced by si-PVT1 and reduced about 255.36% in T24 and about 277.04% in UM-UC-3 (Figure 5G, 5H).

PVT1 sponging miR-194-5p closely promoted BCLAF1 expression

Compared with para-carcinoma tissues, the relative expression level of BCLAF1 was significantly increased about 1.82 times of BC samples (P=0.002) (Figure 6A). Compared with si-NC groups, the BCLAF1 expressions were decreased about 59.97% in T24 (P=0.001) and about 48.52% in UM-UC-3 (P=0.025) (Figure 6B) in si-PVT1 groups. Compared with NC groups, the BCLAF1 expressions were decreased in T24 and UM-UC-3 (P<0.01) after the transfection of miR-194-5p mimics and were up-regulated in T24 and UM-UC-3 (P<0.01) after the transfection of miR-194-5p inhibition (Figure 6C).

We used bioinformatics databases to predict BCLAF1 with miR-194-5p possible mutual binding sites, which was shown as Figure 6D. The predicted binding sites and binding effects were manifested through luciferase reporter assay.

Compared with the co-transfections with NC + pmirGLO-PVT1-Wt, our results have confirmed that miR-194-5p mimics dramatically restrained BCLAF1 wild type reporter luciferase activity, which decreased about 49.66% in T24 (p< 0.001) and about 49.47% in UM-UC-3 (p<0.001) in the co-transfection with miR-194-5p mimics + BCLAF1C-3’UTR-Wt. Inversely, miR-194-5p mimics could not restrain the BCLAF1 mutant binding sites reporter vector luciferase activity (Figure 6D).

We identified that protein is associated with BCLAF1 through immunoprecipitation (IP/co-IP) assays. CDC20 was detected to be a BCLAF1-binding protein in the bladder cell lines (Figure 6E). At the same time, IP pyrolysis products are used for proteomics detection (Figures 6Ka–f), which was furthermore proved that BCLAF1 bound multiple functional proteins in BC cells, and KEGG, GO and other analysis were performed (Also as shown as the GO.anno,KEGG.anno and Subcellular.localization_anno), and we further revealed that BCLAF1 could interact with CDC20.

Our study proved that the relative expression level of PVT1 was closely related to BCLAF1 expression and down-regulation of PVT1 could reduce BCLAF1 expression in BC cells. Our further experiments confirmed that restrained PVT1 referred to the WNT signaling in BC cells (Figure 6F). Western blotting was used to detect the WNT signaling that is associated downstream genes expression. Knockdown of PVT1 decreased CD44, MET, cyclinD1, CDC20, BCLAF1 and β-catenin expression in BC cells. Therefore, the down-regulation of PVT1 inhibited the progression of BC cells through inhibition of the Wnt/β-catenin signaling pathway because β-catenin was the vital protein in this signaling pathway.

We must confirm whether PVT1 increased BCLAF1 expression via miR-194-5p-dependent manner in BC cells. Our studies proved that over-expression of miR-194-5p reduced BCLAF1 expression and miR-194-5p inhibition up-regulated expression of BCLAF1 in BC cells (Figure 6G). Our experiments indicated that PVT1 could closely promote BCLAF1 expression via sponging miR-194-5p in BC cells.

The knockdown of PVT1 caused the reduction of MMP-2 and MMP-9 (Figure 6H). Thus, we concluded that the knockdown of PVT1 suppressed the migration of BC cell lines, which is in accordance with that after transfection miR-194-5p inhibition and mimics. Over-expression miR-194-5p decreased MPP2 and MPP9 expression and the knockdown of miR-194-5p increased MPP2 and MPP9 expression in BC cells (Figure 6I). Therefore, we concluded that PVT1 and miR-194-5p referred to the migration of BC cell lines.

PVT1 accelerated malignant phenotypes of BC cells via BCLAF1-dependent manner

We should confirm whether PVT1 accelerated malignant phenotypes via BCLAF1-dependent manner in BC cells. Further experiments proved that knockdown of BCLAF1 dramatically reversed the promotion of BCLAF1 expression induced by over-expression of PVT1 in BC cells (Figure 6J). Moreover, the knockdown of BCLAF1 reversed BC cells proliferation promotion (Figure 7A–7D) induced by over-expression of PVT1. And BCLAF1 knockdown could reverse BC cells migration (Figure 7E, 7F) induced by over-expression of PVT1. Meanwhile, knockdown of BCLAF1 could notably reverse BC cells apoptosis suppression (Figure 7G, 7H) induced by over-expression of PVT1. Our studies proved that PVT1 could accelerate malignant phenotypes of BC cells via BCLAF1-dependent manner.

The relative expression levels of miR-194-5p were significantly decreased in about 69.80% in SV-HUC-1 (p=0.0002) (Figure 8A). MiR-194-5p inhibition notably promoted bladder cells proliferation (Figure 8B) (p < 0.01). The ratio of the relative migration was up-regulated about 2.109 times in SV-HUC-1 (P= 0.026) (Figure 8C) after transfection miR-194-5p inhibitor.

These data suggested that the knockdown of mi-194-5p accelerated bladder cells migration. Compared with NC groups, the ratios of apoptosis were lessened about 0.264 times in SV-HUC-1 (P= 0.0008) (Figure 8D) after transfection miR-194-5p inhibitor.

MiR-194-5p co-regulated the repressive progression induced by knockdown BCLAF1 in BC cells. That si-BCLAF1 co-transfected miR-194-5p mimics could manifested more powerfully suppressed effects on BC cells proliferation (Figure 8E, 8F) and migration (Figure 8G, 8H) than si-NC co-transfection with NC (si-NC+NC), meanwhile, compared with si-NC+NC group, apoptosis was accelerated in si-BCLAF1 co-transfection miR-194-5p mimics (si-BCLAF1+miR-194-5p mimics) group (Figure 8I, 8J). Conversely, miR-194-5p inhibitor could partially reverse inhibited effects on BC cells progression induced by si-BCLAF1.

Knockdown of PVT1 suppressed BC cells tumorigenicity

The generation of xenograft was used to confirm whether PVT1 promoted the tumorigenicity of BC cells. Our experiments proved that the knockdown of PVT1 could restrain the tumorigenicity of BCs in vivo (Figure 9A–9F). Solid tumors that were obtained from mice were displayed in Figure 9A. It was confirmed that the relative expression of PVT1 was reduced in LV-shPVT1 groups compared with LV-shNC group of BC cells in vivo (Figure 9B). Tumor weight was lessened in LV-shPVT1 groups than LV-shNC groups in vivo (Figure 9C). Tumor growth was slower in LV-shPVT1 groups than LV-shNC groups in vivo (Figure 9D). Our experiments revealed that the knockdown of PVT1 could reduce CD44, MET, cyclinD1, CDC20, BCLAF1 and β-catenin expression of BC cells in vivo (Figure 9E). IHC experiments proved that the knockdown of PVT1 restrained Ki-67 and BCLAF1 expression (Figure 9F) of BC cells in vivo. Our experiments proved that PVT1 facilitated the tumorigenicity of BC via up-regulating BCLAF1.

As simulated diagram shown as Figure 10, PVT1 acts as a microRNA sponge that actively promotes the expression of BCLAF1 to sponge miR-194-5p and subsequently increases malignant phenotypes of BC cells. PVT1 is dramatically upregulated in BC cells and PVT1 could sponge miR-194-5p to closely promote BCLAF1 expression. Up-regulated BCLAF1 protein could facilitate transcription and translation of proteins operating through indispensably abnormal protein signaling pathways, and subsequently accelerate malignant phenotypes of BC cells.

Patient specimens

Our study included BC patients who received tumorectomy. We quick-freeze the BC specimens and paired normal peritumoral specimens in liquid nitrogen quickly after resection. Every one patient has signed the written informed consent. The first affiliated hospital of Soochow University research ethics committee approved the experiment.

Cell lines and cell culture

T24, UM-UC-3 and SV-HUC-1 were obtained from the Institute of Cell Biology, Chinese Academy of Sciences in Shanghai. T24 and UM-UC-3 cells were cultivated in DMEM (Gibco, USA). SV-HUC-1 cells were cultivated in F12 (Life technologies, USA). 1% antibiotics (100 U/ml penicillin and 100 μg /ml streptomycin sulfates) and 10% FBS were blended in DMEM and F12. The atmosphere of the incubator is at 37° C and 5% CO₂.

Cell transfection

Specific siRNA oligonucleotides were transiently transfected in T24, UM-UC-3 and SV-HUC-1 cells, si-PVT1 sense (5-CAGCTTCAACCCATTACGATT-3), si-BCLAF1 sense (5-GGTTCACTTCGTATCAGAA), si-NC and si-RNA (si-PVT1, si-BCLAF1) were obtained from Gene Pharma (Suzhou, China). Optimum density BC cells were cultivated and then transfected in 6-well plates. The plasmid vectors (pcDNA3.1-PVT1, negative control) were obtained from Gene Script (Nanjing, China). Cells were cultured for 24 hours before transfection. The cells were then transiently transfected with the corresponding vectors using Lipofectamine 3000 Transfection Reagent based on the product description. After 48 hours, cells transfected with the corresponding vector were collected.

RT-qPCR

Based on the product descriptions, the whole RNA was extracted from the specimens or the transfected cells using TRIzol reagent (Invitrogen, USA). The cDNA was synthesized from total RNA applying the Prime Script RT Reagent Kit with gDNA Eraser (Takara, Japan). The relative expression levels of PVT1 were measured by RT-qPCR with SYBR Premix Ex Taq II (Takara) on the CFX96 sequence detection system (Bio-Rad). The primer sequences were displayed in Supplementary Table 1. The endogenous controls were Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) and U6 small nuclear RNA. The relative quantification method (2^-ΔΔCt) was adopted to calculate the expressions that were normalized to endogenous controls.

Cell proliferation assays

CCK-8 (Beyotime, Shanghai) was applied to cell proliferation based on the product descriptions. Cells were incubated in a 96-well plate for 24 hours and then respectively transfected with siRNAs or plasmids in the CCK-8 assays. 0, 24, 48 and 72 h after transfection, the absorbance in each well was measured at by a microplate reader (Bio-Rad, USA).

EdU incorporation assay

EdU (Ethynyl-2-deoxyuridine) Apollo DNA in vitro kit (RIBOBIO, Guangzhou) was also used for cell proliferation that was detected by EdU incorporation assay based on the product descriptions. In a word, cells transfected siRNA or plasmid and were incubated with 100 μl of 50 μM EdU per well for 2 h at 37° C, respectively. Cells were visualized under a fluorescence digital camera system.

Cell migration assay

The cells were cultivated into the 6-well plates in the incubator. About 90% confluence was obtained before the transfection siRNA or plasma transfected cells. Use the sterilization 200ul pet tips to generate clean lines in 6-well plates. Use digital camera system to take photos in each well quickly. A day later, the picture was taken again. The travel distance is set at 0 and 24 hours.

Flow cytometry assay

SiRNAs or plasmid vectors were respectively transfected in bladder T24, UM-UC-3 and SV-HUC-1 cells. For 48 hours after transfection, cells were collected and resuspended in fixation fluid 5 μl of Annexin V-FIFC and 10 μl of propidium iodide were added to 195 μl of cell suspension. Flow cytometry (Beckman, USA) was applied to detect cell apoptosis.

Western blot analysis

Protein was separated by 10% SDS–PAGE and transferred to PVDF membranes. After blocking in the 5% non-fat milk and the membranes were incubated overnight for at least 16h in 4° C with the primary antibody. At room temperature for 1-2h, the membranes were then incubated with a secondary antibody and enhanced chemiluminescence ECL kit (Beyotime Biotechnology, China) was visualized. β-actin was the internal standard, and the antibodies were demonstrated in Supplementary Table 2.

Dual-luciferase assays

Based on the product descriptions, dual-Luciferase Reporter Assay System (Promega, USA) was used for the Dual-luciferase reporter assays. The binding and mutant sequences were respectively cloned into pmirGLO Dual-luciferase vectors (Fubio Biological Technology Co, Suzhou, China). PVT1, BCLAF1 WT or Mut constructed and co-transfected along with miR-194-5p mimics or NC, and then transfected with Lipofectamine 3000 and incubated for 48 h. Microplate reader was applied to measure the luciferase activities.

Mouse model experiments

Our experiment was approved by Institutional Ethics Review Board of Soochow University. 5-wk old male BALB/c nude mice were divided into two groups and each group included 5 mice. LV-PVT1 and LV-NC were purchased from Genechem (Shanghai, China). 2×10⁶ UM-UC-3 cells were injected into the mice dorsal flank regions. Every 5 days, tumor growth was measured. The formula, a*b²/2 (a represents the long diameter and b refers to the short diameter), was used to calculate tumor volume. Mice were executed after 30 days and the subcutaneous weight of each tumor was measured.

Statistical analysis

Assays were performed in triplicate at least and data were shown as mean ± standard deviation (SD) of those biological replicates or samples. SPSS 20.0 software (IBM, Chicago, IL, USA) etc. were used to analyze assays’ statistical analyses. Paired samples t-test was used to analyze the relative expression of PVT1 and miR-194-5p. ANOVA was used to analyze CCK-8 assay data. The independent samples t-test was used to analyze other data. P< 0.05 was regarded as the statistically significant one.