The present study was approved by the Ethics Committee of Jishou University (Jishou, China). Osteosarcoma tissue (n=20) and their matched normal adjacent tissue was obtained from the Department of Orthopedics, People's Hospital of Xiangxi Autonomous Prefecture, Jishou University (Jishou, China). Informed consent was obtained from all patients. The tissues were stored in liquid nitrogen shortly following surgical resection.

The human osteosarcoma cell lines Saos-2, MG-63 and U2OS, and the human osteoblast cell line hFOB 1.19 were purchased from the American Type Culture Collection (Manassas, VA, USA). The cells were cultured in RPMI-1640, Dulbecco's modified Eagle's medium (DMEM) or DMEM/F12 medium (all from Invitrogen Life Technologies, Carlsbad, CA, USA), respectively, supplemented with 10% fetal bovine serum (FBS; Invitrogen Life Technologies) at 37°C in a humidified atmosphere containing 5% CO₂.

The total RNA was isolated using TRIzol reagent (Invitrogen Life Technologies), according to the manufacturer's instructions. Reverse transcription was performed using the Takara RNA PCR kit (Takara Bio Inc., Shiga, Japan) according to the manufacturer's instructions. RT-qPCR was performed using a Hairpin-it™ miRNA qPCR Quantitation kit (Gene Pharma, Shanghai, China) for miRNA detection, or the SYBR green qPCR assay kit (Invitrogen Life Technologies) for mRNA detection, on a Light Cycler 480 (Roche, Basel, Switzerland). GAPDH and U6 small nuclear RNA were used as internal controls. The primers used were as follows: ROCK1 forward, 5′-AGGAAGGCGGACATATTAGTCCCT-3′ and reverse, 5′-AGACGATAGTTGGGTCCCGGC-3′; GAPDH forward, 5′-GGAGCGAGATCCCTCCAAAAT-3′ and reverse, 5′-GGCTGTTGTCATACTTCTCATGG-3′. The PCR cycling conditions were as follows: 95°C for 10 min, 40 cycles of denaturation at 95°C for 15 sec then annealing/elongation at 60°C for 60 sec. The 2-^ΔΔCT method was used for relative quantification of the differences in the expression levels of each target.

ROCK1 small interfering (si)RNA and control siRNA were purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA). miR-144 mimics and scrambled small RNA oligonucleotide were synthesized by Gene Pharma. The cells were cultured to 80% confluence and were transfected with these small RNA oligonucleotides using Lipofectamine 2000 (Invitrogen Life Technologies), according to the manufacturer's instructions.

Targetscan software (release 6.2) was used to predicate the putative target genes of miR-144 (13). As described previously (12), the full length 3′-UTR of ROCK1 was amplified by PCR from genomic DNA and cloned at the SacI and XhoI sites into the pmirGLO vector (Promega, Madison, WI, USA). The primers for the 3′-UTR of ROCK1 mRNA were as follows: Forward, 5′-GCACTAGTGTTCCATCTTCGGACGTTGA-3′ and reverse, 5′-ATACGCGTCTTCAACAGACCATGCTCCC-3′. The mutant 3′-UTR of ROCK1 mRNA was generated using a Quick Change mutagenesis kit (Stratagene, Heidelberg, Germany). Lipofectamine 2000 was used to co-transfect the cells with reporter vectors and miR-144 mimics or negative control. Following co-transfection for 48 h, the luciferase activity was determined using a dual luciferase reporter assay system (Promega) according to the manufacturer's instructions.

An MTT assay was used to quantify cell proliferation according to the manufacturer's instructions. Briefly, 10⁴ cells/well were plated in a 96-well plate and incubated for 0, 12, 24 or 48 h at 37°C and 5% CO₂. To assess cell proliferation, 10 µl MTT (5 mg/ml; Sigma-Aldrich, St. Louis, MO, USA) was added to each well and the cells were subsequently incubated for 4 h at 37°C and 5% CO₂. The supernatant was removed and 100 µl dimethylsulfoxide (Sigma-Aldrich) was added to dissolve the precipitate. The absorbance was measured at 492 nm using a microplate reader (3550; Bio-Rad Laboratories, Inc., Hercules, CA, USA).

A cell migration assay was performed using Transwell chambers. A cell suspension, containing 5×10⁵ cells/ml without serum, was added into the upper chamber of the Transwells (BD Biosciences, Franklin Lakes, NJ, USA) and medium containing 10% FBS was added into the lower chamber. Following incubation for 24 h, the cells which had not migrated through the pores were removed using a cotton bud. The filters were subsequently fixed with 90% alcohol and stained with crystal violet (Sigma-Aldrich). Five random fields were counted under an inverted microscope (IX71; Olympus, Tokyo, Japan).

A cell suspension, containing 5×10⁵ cells/ml without serum, was added into the upper chamber of transwells pre-coated with Matrigel (BD Biosciences), and medium containing 10% FBS was added into the lower chamber. Following incubation for 24 h, the cells which had not migrated through the pores were removed using a cotton bud. The filters were subsequently fixed with 90% alcohol and stained with crystal violet. Five random fields were counted under an inverted microscope (Olympus).

The proteins in tissues or cells were isolated using radioimmunoprecipitation assay buffer (Sigma-Aldrich) and were quantified using a bicinchoninic acid protein assay kit (Pierce Biotechnology, Inc., Rockford, IL, USA). The proteins (60 µg) were separated using 10% SDS-PAGE (Sigma-Aldrich) and were transferred onto a polyvinylidene difluoride (PVDF) membrane, which was subsequently incubated with Tris-buffered saline with Tween-20 (Sigma-Aldrich) containing 5% milk at room temperature for 3 h. The PVDF membrane was incubated with mouse anti-ROCK1 (1:100; cat. no. ab45171; Abcam, Cambridge, UK), and mouse anti-GAPDH (1:100; cat. no. ab181602; Abcam) primary antibodies at room temperature for 3 h. Following washing three times with phosphate-buffered saline (Sigma-Aldrich) with Tween-20, the membrane was incubated with the goat anti-mouse secondary antibodies (1:5,000; ab175773; Abcam) at room temperature for 40 min. Chemiluminescent detection was performed using an enhanced chemiluminescence kit (Pierce Biotechnology, Inc.). The relative protein expression was analyzed using ImagePro plus software 6.0 (Media Cybernetics, Rockville, MD, USA) and was represented as the density ratio normalized to GAPDH.

SPSS 17.0 software (SPSS, Inc., Chicago, IL, USA) was used to perform statistical analysis. Values are expressed as the mean ± standard deviation. Differences were assessed using two-tailed Student's t-test. The association between the expression levels of miR-144 and ROCK1 was assessed using two-tailed Pearson's correlation. P<0.05 was considered to indicate a statistically significant difference.

To investigate the function of miR-144 in osteosarcoma, the expression levels in human osteosarcoma tissues and their matched adjacent normal tissues were assessed using RT-qPCR. As shown in Fig. 1A, miR-144 was markedly downregulated in osteosarcoma tissues compared with that in adjacent normal tissues. In addition, it was demonstrated that miR-144 was downregulated in the three osteosarcoma cell lines Saos-2, MG-63 and U2OS compared with that in the normal osteoblast cell line hFOB 1.19 (Fig. 1B). Accordingly, the present study suggested that downregulation of miR-144 may be important in osteosarcoma.

The present study further investigated whether miR-144 affects osteosarcoma cell migration and invasion. MG-63 cells were transfected with miR-144 mimics or negative control, and cell migration and invasion assays were performed. As expected, transfection with miR-144 led to an increase in the expression of miR-144 in MG-63 cells (Fig. 2A). Furthermore, as shown in Fig. 2B–D, restoration of miR-144 caused a decrease in cell proliferation, migration and invasion of MG-63 cells. These findings indicated that miR-144 exerts a suppressive role in the regulation of osteosarcoma cell proliferation, migration and invasion.

Potential targets of miR-144 were further investigated using Targetscan software. ROCK1 was predicated to be a putative target gene for miR-144, which has been demonstrated to regulate cell proliferation, migration and invasion (14). The effect of miR-144 on the protein expression of ROCK1 was assessed and it was demonstrated that the protein levels of ROCK1 were markedly reduced following the overexpression of miR-144 in MG-63 cells (Fig. 3A). Whether ROCK1 was a direct target gene of miR-144 was determined by performing a luciferase reporter assay in MG-63 cells (Fig. 3B–D). As shown in Fig. 3D, co-transfection with wild-type ROCK1-3′-UTR plasmid and the miR-144 mimics resulted in a significant decrease in luciferase activity. However, co-transfection with the mutant ROCK1-3′-UTR plasmid and miR-144 mimics revealed no change in luciferase activity. These results indicated that miR-144 negatively regulates the protein expression of ROCK1 by directly binding to the 3′-UTR of ROCK1, which can further cause translational inhibition in MG-63 cells.

Based on findings from the present study, it was speculated that ROCK1 may be involved in the regulatory role of miR-144 in osteosarcoma cell migration and invasion. To confirm this speculation, MG-63 cells were transfected with ROCK1 siRNA to knockdown the expression of ROCK1. As shown in Fig. 4A, the protein levels of ROCK1 were significantly reduced following transfection of MG-63 cells with ROCK1 siRNA. It was demonstrated that the downregulation of ROCK1 significantly inhibited MG-63 cell proliferation, migration and invasion (Fig. 4B–D). These results indicated that knockdown of ROCK1 inhibited osteosarcoma cell proliferation, migration and invasion, which resembled the suppressive effect of miR-144.

The mRNA expression levels of ROCK1 were determined in osteosarcoma tissues and their matched adjacent normal tissues, and the miR-1441 – ROCK1 correlation was investigated in osteosarcoma. As shown in Fig. 5A, the mRNA expression levels of ROCK1 were significantly increased in osteosarcoma tissues compared with those in their matched adjacent normal tissues. In addition, a significant inverse correlation was observed between the expression levels of ROCK1 and miR-144 (Fig. 5B). These findings suggested that the upregulation of ROCK1 is possibly due to the inhibition of miR-144, which may contribute to osteosarcoma cell migration and invasion.