Tumor specimens and the adjacent non-cancer tissues were obtained from 86 osteosarcoma patients who underwent surgery in the Department of Orthopaedics, Children's Hospital of Zhejiang University School of Medicine between January 2003 and January 2010. Written informed consents were obtained from all participating patients. Clinical specimens were frozen and stored at −80°C. The patients did not receive any chemo- or radiotherapy before operation. The demographic and clinicopathological features of all included patients are presented in Table I. All protocols involving clinical samples were approved by the Ethics Review Committee of Children's Hospital of Zhejiang University School of Medicine.

Total RNA was isolated from the osteosarcoma tissues and cells using TRIzol (Life Technologies) according to the manufacturer's instructions. The TaqMan Human MiRNA Assay kit (Applied Biosystems, Foster City, CA, USA) and a SYBR^® Premix Ex Taq™ II (Perfect Real-time) kit (Takara Bio, Shiga, Japan) were employed to perform the PCR amplification. The primer of miR-130a (HmiRQP0156), PTEN (HQP015535), U6 (HmiRQP9001) and GAPDH (HQP006940) were obtained from Genecopoeia (Guangzhou, China). The relative expression of miR-130a is shown as fold difference relative to U6 while the relative expression of PTEN is shown as fold difference relative to GAPDH.

Human osteosarcoma cell lines HOS58, SaoS-2 and MG63, and the human normal osteoblasts (hFOB1.19) were grown in complete Dulbecco's modified Eagle's medium (DMEM; HyClone, Logan, UT, USA) supplemented with 10% fetal bovine serum (FBS; HyClone), 100 U/ml penicillin, and 100 µg/ml streptomycin. Cell cultures were incubated in a humidified incubator containing of 5% CO₂ at 37°C.

miR-130a expressing vector (HmiR0170-MR03), the control vector (CmiR0001-MR03), miR-130a inhibitor (HmiR-AN0156-AM03) and the negative control (CmiR-AN0001-AM03), were purchased from GeneCopoeia. Plasmids carrying PTEN expressing vector (#28298) and PTEN specific shRNA (#25638) were obtained from Addgene (Cambridge, MA, USA). The vectors mentioned above were transfected into the osteosarcoma cells using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) following the manufacturer's protocol.

The migration and invasion of osteosarcoma cells were evaluated in Transwell chambers with 8-µm inserts (Millipore Corp., Billerica, MA, USA) according to the manufacturer's instructions. Osteosarcoma cells were resuspended in serum-free medium and seeded into the top chamber of each insert. Serum-containing medium (10% FBS) was used in the lower chamber as the attractant. In the invasion assay, each upper chamber was coated with mixture of DMEM and Matrigel (Becton-Dickinson Labware, Bedford, MA, USA) at a ratio of 8:1, and 24 h after cell seeding, the cells in the upper surface of the filter were removed with a cotton swab, and the migrated or invaded cells were stained with 0.1% crystal violet. Cell numbers on the lower surface were counted. Three independent experiments were performed.

Cells were collected and washed twice with phosphate-buffered saline (PBS), and lysed with RIPA lysis buffer (BioMed, Beijing, China). Lysate was centrifuged at 14,000 rpm for 20 min and the supernatants containing protein were collected. Protein concentration was measured using the BCA kit (Pierce, Rockford, IL, USA). Protein (30–40 µg) of each sample was separated and transferred to PVDF membrane. The blots were incubated with the following primary antibodies: PTEN (1:1,000, Cell Signaling Technologies, Danvers, MA, USA), E-cadherin (1:1,000, Cell Signaling Technologies), vimentin (1:1,000, Cell Signaling Technologies) and GAPDH (1:1500, Santa Cruz Biotechnology, Santa Cruz, CA, USA). After washing the membranes with TBST, blots were incubated with anti-mouse or anti-rabbit secondary antibodies (1:10,000; Bio-Rad, Hercules, CA, USA), and the signals were detected using the Bio-Rad Gel imaging system.

Wild-type 3′-UTR sequence of PTEN (wt PTEN-3′-UTR) which was predicted to interact with miR-130a and the mutated sequence (mt PTEN-3′-UTR) were synthesized and inserted into the pGL3 control vector (Promega, Madison, WI, USA). For luciferase reporter assay, cells were transfected with the wild-type construct or mutant construct, and miR-130a expressing vector, miR-130a inhibitor, control vector or negative control. At 48 h after transfection, the luciferase activity was measured using the Dual-Luciferase Reporter Assay system (Promega, Shanghai, China) with a luminometer (Promega). Results were obtained from three independent experiments performed in duplicate.

SPSS statistical package for Windows version 13 (SPSS, Inc., Chicago, IL, USA) and GraphPad Prism 5 software (GraphPad Software, Inc., USA) were used to perform the statistical analysis, including the Pearson's Chi-square test, the Spearman's rank correlation coefficient, the Student's t-test, the Kaplan-Meier plot, and the log-rank test or ANOVA was performed in this study. P<0.05 was considered to indicate a statistically significant difference.

Quantitative RT-PCR assay was conducted to evaluate the expression level of miR-130a in osteosarcoma tissues and matched non-tumor tissues. miR-130a in osteosarcoma tissues was increased significantly compared with that in the adjacent non-tumor tissues (P<0.05, Fig. 1A). Moreover, in tissues with metastasis, the level of miR-130a was significantly higher than that in those without metastasis (P<0.05, Fig. 1B). Furthermore, the expression level of miR-130a in osteosarcoma cell lines was measured. Compared with normal human osteoblastic cell line, hFOB1.19, osteosarcoma cell lines, including HOS58, SaoS-2 and MG63, had significantly higher level of miR-130a (P<0.05, Fig. 1C). MG63 and SaoS-2 cells had obviously higher expression level of miR-130a than HOS58 cells (P<0.01, Fig. 1C). These results indicate that miR-130a is significantly upregulated in osteosarcoma and is associated with the metastasis of osteosarcoma.

Next, the clinical significance of miR-130a expression level in osteosarcoma tissues was examined. The expression of miR-130a was assessed either low (n=43) or high (n=43) based on the median level of miR-130a in the 86 patients cohort. As presented in Table I, high expression level of miR-130a in osteosarcoma patients was closely associated with metastasis (P=0.002) and advanced TNM stage (P=0.023). Importantly, Kaplan-Meier analysis further showed that patients with high miR-130a expression level had significantly shorter overall survival (P= 0.0056, Fig. 2A) and disease-free survival (P=0.0014, Fig. 2B). These results suggest miR-130a is a promising prognostic predictor for osteosarcoma patients.

The functional role of miR-130a in osteosarcoma cells was explored. HOS58 cells with miR-130a mimics, and the expression level of miR-130a was significantly increased after transfection (P<0.01, Fig. 3A). Functionally, as suggested by the Transwell assay, HOS58 cells overexpressing miR-130a (HOS58-miR-130a cells) showed increased migratory (P<0.05, upper part in Fig. 3B) and invasive (P<0.05, lower part in Fig. 3B) ability. In contrast, miR-130a inhibitor significantly downregulated the expression level of miR-130a in MG63 cells (P<0.01, Fig. 3C). The migration (P<0.05, upper part in Fig. 3D) and invasion (P<0.05, lower part in Fig. 3D) of MG63 cells was significantly decreased after the expression of miR-130a was inhibited. These data indicate that miR-130a potentiate the metastatic behavior of osteosarcoma cells.

The EMT process, which has been regarded as a fundamental process of the cancer metastasis (18–20), can increase the migratory and invasive ability of osteosarcoma cells (21). Therefore, we further explored whether miR-130a promotes the migration and invasion of osteosarcoma cells by promoting EMT. The results of western blot showed that overexpression of miR-130a in HOS58 cells resulted in decreased level of E-cadherin (P<0.05, Fig. 3E) and increased expression of Vimentin (P<0.01, Fig. 3E). On the contrary, downregulating miR-130a in MG63 cells led to increased level of E-cadherin (P<0.01, Fig. 3F) and decreased expression of Vimentin (P<0.01, Fig. 3F). These results suggest that miR-130a can promote EMT phenotype of osteosarcoma cells.

To further investigate the underlying mechanisms for the functional influence of miR-130a in osteosarcoma cells, two publicly available databases (TargetScan 6.2 and miRanda) were searched for the potential downstream target of miR-130a. PTEN, which is a well-recognized tumor suppressor and an important participator in the development and progression of osteosarcoma, was suggested to be a potential downstream target of miR-130a. As shown in Fig. 4A, the 3′-UTR of PTEN mRNA contained the complementary sequence of miR-130a. This suggests that miR-130a can potentially bind to the 3′-UTR of PTEN. To confirm this prediction, we performed dual-luciferase reporter assays to elucidate whether miR-130a could directly bind to the 3′-UTR of PTEN. Forced expression miR-130a significantly decreased the luciferase activity of PTEN with a wild-type (wt) 3′-UTR (P<0.01 Fig. 4B), but had no significant influence on that with a mutant (mt) 3′-UTR. On the contrary, when miR-130a inhibitor was transfected, the luciferase activity of wt PTEN 3′-UTR obviously increased (P<0.01, Fig. 4B) while that of mt PTEN 3′-UTR remained unchanged.

Moreover, we investigated whether alteration of miR-130a could affect the expression level of PTEN mRNA and protein. qRT-PCR and western blot were performed to examine the expression level of PTEN after forced expression of miR-130a in HOS58 cells. The result of qRT-PCR showed that PTEN mRAN level in HOS58 cells was significantly decreased after overexpression of miR-130a (P<0.01, Fig. 5A). Consistently, the protein level of PTEN was significantly decreased after overexpression of miR-130a (P<0.01, Fig. 5B). In contrast, inhibiting the expression of miR-130a in MG63 cells resulted in significantly increased level of PTEN mRNA (P<0.01, Fig. 5C) and protein (P<0.01, Fig. 5D). Taken together, these data indicate the PTEN is a direct downstream target of miR-130a, and miR-130a could regulate the expression of PTEN by interacting with the 3′-UTR of PTEN.

To further clarify whether PTEN can serve as the functional mediator of miR-130a in osteosarcoma cells, plasmids containing PTEN expressing vector and PTEN shRNA were transfected into HOS58 cells with overexpressed miR-130a (HOS58-miR-130a cells) and MG63 cells with downregulated miR-130a (MG63-anti-miR-130a cells), respectively. Functionally, overexpression of PTEN partly abrogated the promoting effects of miR-130a on the migration (P<0.05, upper part of Fig. 6A) and invasion (P<0.05, lower part of Fig. 6A) of HOS58 cells. Inhibiting PTEN expression in MG63-anti-miR-130a cells partly reversed the inhibitory effects of miR-130a on the migration (P<0.05, upper part of Fig. 6B) and invasion (P<0.05, upper part of Fig. 6B) of MG63 cells. Moreover, the results of western blot showed that transfection of PTEN expressing vector in HOS58-miR-130a cells resulted in significant increase of PTEN (P<0.01, Fig. 6C), and led to upregulation of E-cadherin (P<0.05, Fig. 6C) and downregulation of Vimentin (P<0.05, Fig. 6C). Transfection of PTEN shRNA in MG63-anti-miR-130a cells significantly inhibited the expression of PTEN (P<0.05, Fig. 6D), and resulted in the downregulation of E-cadherin (P<0.05, Fig. 6D) and increase of Vimentin (P<0.01, Fig. 6D).