A total of 58 primary osteosarcoma tissues and matched adjacent non-tumor tissues were collected from patients (35 males and 23 females; mean age, 31.4 years) in the Second Affiliated Hospital of Nanchang from April 2008 to September 2015. Prior to surgical resection, no patients received radio- or chemotherapy. Tissues were immediately snap-frozen in liquid nitrogen following surgical resection and stored in liquid nitrogen. The clinicopathological characteristics of patients included in the present study are summarized in Table I. The current study was approved by the Ethics Committee of the Second Affiliated Hospital of Nanchang University (Nanchang, China) and informed consent was obtained from all participants.

The human osteoblast cell line hFOB and the osteosarcoma cell lines U2OS, Saos-2, MG63 and SW1353 were purchased from the Cell bank of the Chinese Academy of Sciences (Shanghai, China). All cell lines were cultured in Dulbecco's Modified Eagle's medium (DMEM; Thermo Fisher Scientific, Inc., Waltham, MA, USA) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Inc.) in a 37°C humidified atmosphere of 5% CO₂.

U2OS cells were transfected with miR-92b inhibitor or negative control (NC) inhibitor; miR-92b mimic or scramble miR mimic (miR-NC); or pcDNA3.1-DKK3 expression plasmid using Lipofectamine^® 2000 (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. Following 48 h transfection, levels of miR-92b or DKK3 expression were determined.

Total RNA was extracted using TRIzol^® Reagent (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. To detect miR expression, RT-qPCR was performed using the All-in-One^™ miRNA qRT-PCR Detection kit (GeneCopoeia, Inc., Rockville, MD, USA) and an ABI 7500 thermocycler (Applied Biosystems; Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. The thermocycling conditions were as follows: 95°C for 10 min, and 45 cycles of denaturation at 95°C for 15 sec and annealing/elongation at 60°C for 15 sec. The U6 gene was used as an internal control. The primers for miR-92b (cat. no. HmiRQP0834) and U6 (cat. no. HmiRQP9001) were purchased from Guangzhou Fulengen Co., Ltd. (Guangzhou, China). To detect mRNA expression, total RNA was converted to cDNA using the PrimeScript 1st Strand cDNA Synthesis kit (Takara Bio, Inc., Tokyo, Japan) according to the manufacturer's instructions. A SYBR Green I Real-Time PCR kit (Biomics Biotechnologies, Co., Ltd., Nantong, China) was then used to perform qPCR according to the manufacturer's instructions. GAPDH was used as the internal reference for mRNA. The primers used were as follows: DKK3, forward, 5′-AGGACACGCAGCACAAATTG-3′ and reverse, 5′-CCAGTCTGGTTGTTGGTTATCTT-3′; GAPDH, forward, 5′-ACAACTTTGGTATCGTGGAAGG-3′ and reverse; 5′-GCCATCACGCCACAGTTTC-3′. The thermocycling conditions were 95°C for 3 min, followed by 40 cycles of 95°C for 15 sec and 60°C for 15 sec. Relative expression was analyzed using the 2^−ΔΔCq method (20).

Cells were solubilized in cold radioimmunoprecipitation assay lysis buffer (Beyotime Institute of Biotechnology, Shanghai, China). Protein concentration was determined using the BCA assay kit (Thermo Fisher Scientific, Inc.) and proteins (50 µg) were separated with 10% SDS-PAGE and transferred onto a polyvinylidene difluoride membrane (Thermo Fisher Scientific, Inc.). The membrane was blocked with PBS containing 5% milk (Yili Group, Beijing, China) for 3 h at room temperature. Following 3 washes with PBS (Beyotime Institute of Biotechnology), the membrane was incubated with rabbit polyclonal anti-DKK3 antibody (1:100; ab187532; Abcam, Cambridge, MA, USA) or rabbit polyclonal anti-GAPDH antibody (1:50; ab37168; Abcam) at room temperature for 3 h. Following 3 washes with PBS, the membrane was incubated with goat anti-rabbit immunoglobulin G (1:5,000; ab6721; Abcam) at room temperature for 40 min. An ECL kit (Thermo Fisher Scientific, Inc.) was then used to perform enhanced chemiluminent detection. Relative protein expression was presented as the density ratio vs. GAPDH using Image-Pro Plus software 6.0 (Media Cybernetics, Inc., Rockville, MD, USA).

U2OS cell suspension (5×10⁴ cells/well) was plated in a 96-well plate, and cultured for 0, 24, 48 or 72 h. Subsequently, MTT (10 µl, 5 mg/ml) was added to each well and then incubated at 37°C for 4 h. The supernatant was removed and 100 µl dimethyl sulfoxide (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) was added to each well to dissolve the purple formazan. The absorbance at 570 nm was measured using the Model 680 Microplate Reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA).

U2OS cell suspension (1×10⁶ cells/ml) was prepared in DMEM, 300 µl of which was added to the upper transwell chamber (BD Biosciences, Franklin Lakes, NJ, USA) pre-coated with Matrigel (BD Biosciences). Subsequently, 300 µl DMEM with 10% FBS was added to the lower chamber. Following 24 h culture at 37°C, cells that did not invade through the membrane in the filter were lightly wiped using a cotton-tipped swab (BD Biosciences). The filter was then fixed in 90% alcohol at room temperature for 10 min and cells were stained at room temperature for 10 min using 0.1% crystal violet (Beyotime Institute of Biotechnology). Invading cells were observed under an inverted microscope and images were captured.

Targetscan (http://www.targetscan.org) was used to predicate the potential targets of miR-92b, according to the manufacturer's instructions. ‘Human’ was selected as the species and ‘miR-92b’ was entered. The mutant type (MT) of DKK3 3′UTR lacking complementarity with the miR-92b seed sequence was generated using the QuickChange Site-Directed Mutagenesis kit (Agilent Technologies Inc., Santa Clara, CA, USA), according to the manufacturer's instructions. The wild-type (WT) or MT of DKK3 3′UTR was then cloned downstream of the firefly luciferase coding region of the pMIR-GLOTM Luciferase vector (Promega Corporation, Madison, WI, USA). U2OS cells were co-transfected with WT-DKK3-3′UTR or MUT-DKK3-3′UTR plasmid, and miR-92b mimic or miR-NC, respectively, using Lipofectamine 2000 (Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Following 48 h transfection, luciferase activity was determined using the dual-Luciferase Reporter assay system (Promega Corporation) according to the manufacturer's instruction and normalized to Renilla luciferase activity.

The results are expressed as the mean ± standard deviation of three independent experiments. Student's t test was used to analyze the difference between two groups. One-way analysis of variance with the Tukey post hoc test was used to analyze the differences between more than two groups, and Pearson's correlation analysis was used to look for associations between groups. SPSS 19 (IBM Corp., Armonk, NY, USA) was used to perform statistical analysis. P<0.05 was considered to indicate a statistically significant difference.

RT-qPCR was performed to measure miR-92b expression in a total of 58 osteosarcoma tissues as well as matched adjacent normal tissues from patients with osteosarcoma. The results indicated that miR-92b levels were significantly increased in osteosarcoma tissue compared with adjacent normal tissue (P<0.01; Fig. 1A). Furthermore, it was significantly upregulated in the osteosarcoma cell lines U2OS, Saos-2, MG63 and SW1353, compared with normal osteoblast hFOB cells (P<0.01; Fig. 1B).

The association between miR-92b expression and clinical characteristics in osteosarcoma was investigated. Mean miR-92b levels were used as the cutoff point and this cutoff point was used to divide patients with osteosarcoma into a high miR-92b expression group and a low miR-92b expression group. A total of 30 patients were in the high miR-92b expression group, whereas 28 patients were in the low miR-92b expression group (Table I). High miR-92b levels were significantly associated with lung metastasis and an advanced clinical stage of osteosarcoma (P<0.05; Table I). However, no significant associations were identified between miR-92b expression and age, sex, tumor size, location, serum lactate dehydrogenase or serum alkaline phosphatase in osteosarcoma (Table I). The results demonstrated that upregulation of miR-92b may be associated with osteosarcoma progression.

The regulatory role of miR-92b in osteosarcoma in vitro was determined using U2OS cells, as miR-92b expression was highest in U2OS cells compared with the other osteosarcoma cell lines. miR-92b expression was upregulated in U2OS cells, thus they were transfected with either miR-92b inhibitor or NC inhibitor. Transfection with miR-92b inhibitor significantly decreased miR-92b expression compared with the NC inhibitor group (P<0.01; Fig. 2A). MTT and Transwell assays were further conducted to examine cell proliferation and invasion, respectively. miR-92b knockdown significantly decreased U2OS cell proliferation and invasion compared with the NC inhibitor group (P<0.01; Fig. 2B and C). This suggests that miR-92b may promote the proliferation and invasion of osteosarcoma cells. To further confirm these results, U2OS cells were transfected with miR-92b mimic or miR-NC. Transfection with miR-92b mimic significantly upregulated miR-92b levels compared with miR-NC transfection (P<0.01; Fig. 2D). Overexpression of miR-92b significantly increased the proliferation and invasion of U2OS cells (P<0.01; Fig. 2E and F), indicating that miR-92b promotes the proliferation and invasion of osteosarcoma cells.

The potential target of miR-29b in osteosarcoma cells was investigated. Bioinformatics predication indicated that DKK3 was a putative target of miR-92b (Fig. 3A). Furthermore, overexpression of miR-92b significantly reduced DKK3 expression (P<0.01; Fig. 3B), whereas knockdown of miR-92b significantly increased DKK3 expression in U2OS cells (P<0.01; Fig. 3C), indicating that the expression of DKK3 is negatively regulated by miR-92b.

To further investigate the association between miR-92b and DKK3, WT-DKK3-3′UTR and MUT-DKK3-3′UTR luciferase reporter plasmids were generated (Fig. 3D and E). Subsequently, U2OS cells were co-transfected with WT-DKK3-3′UTR or MUT-DKK3-3′UTR luciferase reporter plasmid and miR-92b mimic or miR-NC, respectively. Luciferase activity was significantly decreased in U2OS cells co-transfected with miR-92b mimic and WT-DKK3-3′UTR luciferase reporter plasmid compared with the control group (P<0.01; Fig. 3F). However, luciferase activity was unchanged in cells co-transfected with miR-92b mimic and MUT-DKK3-3′UTR luciferase reporter plasmid compared with the control group (Fig. 3F). Taken together, the aforementioned results demonstrate that DKK3 is a target gene of miR-92b in U2OS cells.

It was speculated that DKK3 may be involved in the miR-92b-induced proliferation and invasion of U2OS cells. To clarify this, miR-92b-overexpressing osteosarcoma cells were transfected with pcDNA3.1-DKK3 expression plasmid to restore DKK3 expression. Following transfection, DKK3 expression was significantly higher in the miR-92b+DKK3 group compared with the miR-92b group (P<0.01; Fig. 4A). MTT and Transwell assays indicated that the proliferation and invasion of U2OS cells was significantly decreased in the miR-92b+DKK3 group compared with the miR-92b group (P<0.01; Fig. 4B and C). This suggests that restoration of DKK3 expression reverses the increased proliferation and invasion of osteosarcoma cells induced by miR-92b overexpression, suggesting that miR-92b stimulates the proliferation and invasion of osteosarcoma cells, at least partly, by targeting DKK3.

DKK3 expression was measured in osteosarcoma tissues and cell lines. Levels of DKK3 mRNA were significantly decreased in osteosarcoma tissues compared with adjacent normal tissues (P<0.01; Fig. 5A). DKK3 mRNA was also significantly downregulated in the osteosarcoma cell lines compared with normal osteoblast hFOB cells (P<0.01; Fig. 5B). Furthermore, an inverse correlation was detected between miR-92b expression and DKK3 mRNA levels in osteosarcoma tissues (P<0.01; Fig. 5C). These results suggest that the decreased expression of DKK3 may be caused by the upregulation of miR-92b in osteosarcoma.