miR-486 inhibited osteosarcoma cells invasion and epithelial-mesenchymal transition by targeting PIM1

2.1 Patients and clinical samples

A collection of osteosarcoma patients were obtained from Qingdao West Coast New Area Central Hospital with 51 osteosarcoma tissues and corresponding paracancerous tissues. All the patents had not been treated before surgery, and the fresh tissue samples were immediately frozen in liquid nitrogen and saved at - 80 ∘ C. All the procedures were got the informed consent from patients and approved by the Ethical Committee of the participating Qingdao West Coast New Area Central Hospital.

2.2 Cell lines and culture condition

Two osteosarcoma cell lines (MG63 and SaOS2) and human osteoblastic cell line NHOst were purchased from the American Type Culture Collection (ATCC, Rockville, MD, USA). Both osteosarcoma and osteoblastic cells were cultured in RPMI-1640 medium (Gibco, Carlsbad, CA, USA) with 10% fetal bovine serum (FBS; Gibco) added in and incubated in a 37 ∘ C incubator with 5% CO 2 .

2.3 RNA isolation and qRT-PCR

TRIzol reagent (Takara Bio, Shiga, Japan) and MIRcute and Separation of miRNAs Kit (Tiangen, Beijing, China) were applied to extract total RNA or miRNAs from samples and cells. Followed, miRNA cDNA Synthesis kit (Abm Canada Inc., Milton, ON, Canada) and RT Master Mix (Abm Canada) were employed to synthesize cDNA or miRNA. In addition, miR-486 or PIM1, which were normalized by U6 small RNA and β -actin, was quantified by SYBR Prime Script miRNA RT-PCR Kit and SYBR Premix Kit (both purchased from Takara, Otsu, Japan). Relative expression of miR-486 and PIM1 were calculated used the 2 - Δ ⁢ Δ ⁢ Ct method.

2.4 Protein extraction and Western blotting

RIPA Lysis Buffer (Beyotime, Shanghai, China) supplemented with phenlymethanesulfonyl fluoride (PMSF) was utilized to assess the protein level of tissues and cells. After quantified the concentration by BCA Reagent Kit (Solarbio, Beijing, China), the proteins were separated by 12% SDS-PAGE through electrophoresis and then transferred onto 0.45 μ m PVDF membranes (Bod-Rad, Hercules, CA, USA). After being blocked in 5% nonfat dried milk for 1 h, incubated the membranes with specific antibodies: mouse monoclonal PIM1 antibody (1:1000), rabbit monoclonal antibody against E-Cadherin (1:1000) and rabbit polyclonal antibody against N-Cadherin (1:1000) and β -actin (1:2000, Abcam, Cambridge, MA, USA) at 4 ∘ C overnight respectively. The membranes was then incubated with rabbit (1:5000) or mouse (1:4000) secondary antibodies which attached with HRP-conjugated. Visualized the protein bands in enhanced chemiluminescence system (Pierce Biotech, IL, USA).

OPEN IN VIEWER

The invasive ability of osteosarcoma cells was evaluated by transwell assay. The transwell were put into 24-well plate, which appeared upper and lower chambers and the upper chamber covered with Matrigel (Clontech, Mountain View, CA, USA). Suspense the cells with medium without FBS and added 200 μ l cell suspension into the upper chamber, while 500 μ l medium containing 15% FBS were added into the lower chamber. After culture for 48 h, the cells still on the upper surface were wiped out with cotton swab. For the invasive cells, fix and followed stained by methanol and crystal violet respectively.

2.6 Transfection

For miRNA analysis, miR-486 mimic and inhibitor (GenePharma, Shanghai, China) were employed to overexpression or knockdown miR-486 in MG63 Cells. For PIM analysis, siRNA-PIM and pcDNA3.1-PIM as well as their negative control were purchased from GenePharma (Shanghai, China) to knockdown or re-express PIM. MG63 cells were seeded into 6-well plate and cultured for overnight at 37 ∘ C. Before transfection, we replaced the medium of the destined cells. Followed, mixed 10 μ L specific vectors and 8 μ L Lipofectamin 2000 reagent (Invitrogen, Carlsbad, CA, USA) and added into each well of the 6-well plate. For the cell transfected with miR-486 mimic, inhibitor or siRNA-PIM, harvest the cells after trnsfected 48 h for other study. However, for the cells transfected with pcDNA3.1-PIM, we plant to construct a cell, which stably expressed PIM protein. Thus, after transfected for 24 h, screened the cells stably expressing PIM with medium containing Geneticin (G418, Solarbio, Beijing, China).

OPEN IN VIEWER

TargetScan predicted that PIM1 was a target gene of miR-486 and the binding sequences were 5’-UCUGAG UGAAAUACUGUACAGGG-3’ on 3’-UTR. We mutated the binding sequences from GUACAGG to CAUGUCC and inserted the nucleotide fragments contain wide type and the mutant sequences into pmirGlo luciferase vector, which named pmirGlo-PIM1-WT (WT) and pmirGlo-PIM1-MUT (MUT) respectively. Whereafter, we co-transfected WT or MUT and miR-486 mimic into MG63 cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). The luciferase ability of firefly and Renilla were assayed used dual luciferase reporter assay system (Promega, Shanghai, China) after transfected 48 h, in which Renilla luciferase was the normalization for the firefly luciferase activity.

2.8 Statistical analysis

Statistical analysis was performed by SPSS version 16.0 software. The comparisons between two or multiple groups were analyzed by Student’s t -test or one-way ANOVA. The correlation between miR-486 and PIM1 was analyzed utilized by spearman’s rank correlation coefficient. Chi-square test and Kaplan-Meier method were performed to analyze the connection between miR-486 and clinicopathological features and the survival respectively. 0.05 was considered to be statistical significance.

3.1 The correlation of miR-486 and PIM1 in osteosarcoma tissues

To analyze the relationship of miR-486 and PIM1 in osteosarcoma, we calculated the miR-486 and PIM1 level by qRT-PCR. As expected, miR-486 was low expressed ( 0.0001) while PIM1 was over expressed ( 0.0001) in osteosarcoma tissues versus corresponding paracancerous tissues (Fig. 1A and B). Therefore, the paired correlation of miR-486 and PIM1 levels were evaluated either, and we discovered that miR-486 had negative connection with PIM1 ( 0.0001, r = - 0.5445) (Fig. 1C). On the other hand, the expression of miR-486 was explored in osteosarcoma and normal osteoblastic cells. Similar to the results in the tissues, miR-486 was low expressed in osteosarcoma cells MG63 ( P = 0.0004) and SaOS2 ( P = 0.0041) versus osteoblastic cell NHOst (Fig. 1D).

OPEN IN VIEWER

Transwell assay and western blot were performed to evaluate the migration, invasion and EMT abilities in osteosarcoma. We employed miR-486 mimic and inhibitor to over ( P = 0.0003) or knockdown ( P = 0.0044) miR-486, and the effect of transfection was shown in Fig. 2A. The invasive ability were reduced ( P = 0.0010) after transfected with miR-486 mimic, while promoted ( P = 0.0047) when knockdown miR-486 by miR-486 inhibitor (Fig. 2B). Meanwhile, western blot showed the epithelial marker E-cadherin was increased while mesenchymal marker N-cadherin was reduced when overexpressed miR-486, and opposite results were appeared when transfected with miR-486 inhibitor, which suggesting miR-486 could inhibit EMT of osteosarcoma cell (Fig. 2C). We checked morphologic changes with a microscope every other day, and we observed a gradual change from an epithelial to a mesenchymal morphology as shown in Fig. 2D. We observed the typical morphologic changes of EMT in MG63 cells.

3.3 PIM1 was upregulated in osteosarcoma cells and knockdown PIM1 inhibited cell invasion and EMT

The expression of PIM1 was also evaluated to be upregulated in osteosarcoma cells MG63 ( P = 0.0003) and SaOS2 ( P = 0.0014) versus osteoblastic cell NHOst (Fig. 3A). Thus, we utilized siRNA-PIM1 transfected into MG63 cells to knockdown PIM1, the knockout efficiency ( P = 0.0044) was shown in Fig. 3B. Followed, invasive ability ( P = 0.0077) was reduced after transfected siRNA-PIM1, revealed by transwell assays (Fig. 3C). On the other hand, epithelial marker E-cadherin was increased while mesenchymal markers N-cadherin decreased demonstrated by western blot (Fig. 3D). In short, our findings illuminated that knockdown PIM1 could inhibit osteosarcoma cell invasion and EMT.

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

TargetScan was utilized to predict the binding sequences between miR-486 and PIM1, and we discovered that miR-486 binding to PIM1 at 1264-1270 on 3’-UTR of mRNA, which were shown in Fig. 4A. Followed, binding sequences were mutated from GUAC AGG to CAUGUCC to certify miR-486 direct binding to PIM1 (Fig. 4A). Therewith, the wide type (WT) or mutant (MUT) sequences of PIM1 3’-UTR were both inserted into pmirGlo luciferase vector and then calculated the luciferase activity. As expected, the luciferase activity was reduced ( P = 0.0027) in cells with PIM1 3’-UTR wide type, while not ( P = 0.7465) in mutant cells (Fig. 4B). What’s more, PIM1 mRNA levels were evaluated in MG63 cells, after transfected with miR-486 mimic or miR-486 inhibitor. PIM1 mRNA level was decreased ( P = 0.0025) when transfected with miR-486 mimic, while inhibition of miR-486 increased ( P = 0.0016) the expression of PIM1 (Fig. 4C).

We have evaluated PIM1 was downregulated by miR-486 before, and we next evaluated the roles of PIM1 in miR-486-mediated MG63 cells. Followed, we discovered that PIM1 could re-expressed ( P = 0.0142) through overexpressed PIM1 both in mRNA and protein level (Fig. 5A). The invasive ability, which suppressed by miR-486 was rescued ( P = 0.0011) by re-expressed PIM1 in MG63 cells (Fig. 5B), which demonstrated that PIM1 could reverse partial role of miR-486 on cell invasion.

3.6 Identification of miR-486 and PIM1 associated with poor survival in osteosarcoma

To identify the connection between miR-486 and the clinicopathological features in osteosarcoma, 51 patients were separated into high and low two groups based on miR-486 expression, gender, age, tumor size, TNM stage, lymph-node metastasis, metastasis and PIM1 respectively. Chi-square test demonstrated that low level of miR-486 had collection with tumor size ( P = 0.040), TNM stage ( P = 0.020), lymph-node metastasis ( P = 0.031), metastasis ( P = 0.019) and PIM1 ( P = 0.020) (Table 1).

Kaplan-Meier analysis was applied to evaluate the 5-year overall survival (OS) depend on miR-486 and PIM1 expression. The OS in miR-486( - ) group was lower (Log-rank P = 0.0163) than that in miR-486( + ) group (Fig. 6A). On contaray, the OS ( P = 0.0095) was higher in PIM1 low expression group than in PIM1 over-expression group (Fig. 6B).