A total of 18 pairs of fresh surgically resected OS tissue and adjacent bone tissue, 5 cm from the edge of tumor site, were obtained from OS patients (age range, 13-68 years; sex, 12 females and 6 males) after diagnosis by experienced pathologists between March 2015 and September 2017 at the Jingzhou Traditional Chinese Medicine Hospital (Hubei, China). All collected tissues were immediately frozen in liquid nitrogen. The present study was approved by the Ethics Committee of Jingzhou Traditional Chinese Medicine Hospital and all patients provided their written informed consent.

Human OS cell lines (Saos-2, MG-63, SW1353 and U2OS), osteoblast cell line hFOB 1.19 and embryonic kidney cell line 293T were purchased from the American Type Culture Collection. All cell lines were cultured in RPMI-1640 medium (Gibco; Thermo Fisher Scientific, Inc.) with 10% fetal bovine serum (Gibco; Thermo Fisher Scientific, Inc.), 100 U/ml penicillin and streptomycin (Invitrogen; Thermo Fisher Scientific, Inc.). Samples were maintained in a humidified atmosphere containing 5% CO₂ at 37˚C.

Oligonucleotides, including miR-106b-5p inhibitors (5'-ATCTGCACTGTCAGCACTTTA-3') and negative controls (miR-NC, 5'-TTCTCCGAACGTGTCACGT-3') were designed and synthesized by Shanghai GenePharma Co., Ltd. The open reading frame of CDKN1A, generated from RNA samples of Saos-2 cells (forward, 5'-CACCATGTCAGAACCGGCTGGGGATG-3'; reverse, 5'-TTAGGGCTTCCTCTTGGAGAAGATCAGC-3'), was inserted into the pcDNA3.1 expression vector to generate overexpressing recombinant vector pcDNA3.1-CDKN1A (Shanghai GenePharma Co., Ltd.). Small interfering (si)RNA for CDKN1A (si-CDKN1A) and its NC (si-NC) were synthesized by Shanghai GenePharma Co., Ltd. Saos-2 or U2OS cells (1x10⁴ cells per well) were seeded into six-well plates and transfected with 50 nM miRNA, 100 pmol siRNA and/or 4 µg plasmid using Lipofectamine® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) for 48 h according to the manufacturer's instructions.

Total RNA was extracted with TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. For miR-106b-5p detection, the temperature protocol for reverse transcription of miRNA was as follows: 37˚C for 60 min, 95˚C for 5 min and the samples were subsequently kept at 4˚C. RT-qPCR was performed in triplicate using a miRVana™ real-time RT-PCR microRNA detection kit (Thermo Fisher Scientific, Inc.) with U6 as an internal control. The thermocycling conditions were as follows: Initial denaturation of 95˚C for 2 min, followed by 40 cycles of 95˚C for 10 sec, 55˚C for 30 sec and 72˚C for 30 sec. For CDKN1A detection, cDNA was synthesized using a PrimeScript™ RT reagent kit (Takara Bio, Inc.). The temperature protocol for reverse transcription of RNA was as follows: 37˚C for 15 min and 85˚C for 5 sec. The expression of CDKN1A mRNA was determined using SYBR Premix Ex Taq™ II (Takara Bio, Inc.) using primers (forward, 5'-TCTGGGGTCTCACTTCTTGG-3' and reverse, 5'-ATGTGAGGAAGGCTCAGTGG-3') with GAPDH (forward, 5'-TGCACCACCAACTGCTTA-3' and reverse, 5'-GGATGCAGGGATGATGTTC-3') as an internal control. The thermocycling conditions were as follows: Initial denaturation at 95˚C for 10 min, followed by 40 cycles of 95˚C for 10 sec and 58˚C for 60 sec. Relative expression levels were calculated using 2^-ΔΔCq method (26).

Saos-2 or U2OS cells were seeded in 96-well plates at a density of 4x10³ cells per well and routinely cultured (37˚C; 5% CO₂) for 5 days. At the indicated time points, 20 µl MTT reagent was added to each well (Sigma-Aldrich; Merck KGaA) and incubated for another 2 h at 37˚C. The blue formazan crystals in each well were subsequently dissolved by adding 150 µl dimethylsulfoxide (Sigma-Aldrich; Merck KGaA). Cell proliferation was evaluated by recording the absorbance value at 595 nm using Model 680 microplate reader (Bio-Rad Laboratories, Inc.).

Flow cytometry analysis was performed to determine cell cycle distribution in OS cells. Saos-2 or U2OS cells at a density of 4x10⁵ cells per reaction, were washed with PBS and fixed with pre-cooled 70% ethanol at 4˚C for 30 min. Following centrifugation (450 x g at 4˚C and 5 min) and washing with PBS, cells were stained with 500 µl 1% propidium iodide (PI; Thermo Fisher Scientific, Inc.), followed by DNA content analyses using BD FACScan™ flow cytometer equipped with CellQuest Pro 4.0.2 software (BD Biosciences).

The putative target genes of miR-106b-5p were predicted using public available algorithms, including PicTar (https://pictar.mdc-berlin.de), TargetScan (http://targetscan.org) and miRDB (http://www.mirdb.org). The predicted miR-106b-5p binding sites in the wild-type (WT) 3'UTR of CDKN1A and the corresponding mutant type (MUT) miR-106b-5p binding sites were cloned into a pGL3 vector (Promega Corporation). For the luciferase reporter assay, 293T cells (1x10⁵ cells/well) were seeded in 96-well plates and co-transfected with 300 ng WT-CDKN1A or MUT-CDKN1A, and 100 nM of miR-106b-5p inhibitor or miR-NC using Lipofectamine® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) in accordance with the manufacturer's protocol. After 48 h transfection, the relative luciferase activities were measured using the dual-luciferase reporter assay system (Promega Corporation). Renilla luciferase activity was used for normalization.

Total protein was extracted using the RIPA protein extraction reagent (Beyotime Institute of Biotechnology) and the protein concentration was determined using the bicinchoninic acid Protein Assay kit (Beyotime Institute of Biotechnology). Approximately 30 µg protein was separated by 10% SDS-PAGE and then transferred to nitrocellulose membranes (Bio-Rad Laboratories, Inc.). The membranes were blocked in 5% non-fat milk containing 0.1% Tween-20 (TBST) for 2 h at room temperature and then probed with primary antibodies against p21 (1:1,000; cat. no. #2947; Cell Signaling Technology, Inc.) and GAPDH (1:5,000; cat. no. #2118; Cell Signaling Technology, Inc.) overnight at 4˚C. After washing with PBS, the membranes were incubated with a horseradish peroxidase-conjugated secondary antibody (1:20,000; cat. nos. ab205718; Abcam) for 2 h at room temperature. Protein bands were visualized with enhanced chemiluminescence reagents (Pierce Biotechnology; Thermo Fisher Scientific, Inc.) following the manufacturer's protocol.

All experiments were independently performed in triplicate. The data were analyzed with SPSS 19.0 software (SPSS, Inc.) and expressed as the mean ± standard deviation. Differences between two groups were assessed using Student's t-test. Differences amongst multiple groups were evaluated using one-way ANOVA followed by Tukey's post-hoc test. Correlations between the expression of miR-106b-5p and the expression of CDKN1A were analyzed by Spearman's rank correlation. P<0.05 was considered to indicate statistically significant difference.

RT-qPCR was used to detect the expression of miR-106b-5p in tumor tissue and adjacent tissue derived from OS patients. As presented in Fig. 1A, miR-106b-5p was significantly upregulated in OS tissue compared with adjacent tissue (P<0.001). In addition, the expression of miR-106b-5p in several human OS cells lines including MG-63, U2OS, SW1353 and Saos-2 was also analyzed. As presented in Fig. 1B, miR-106b-5p expression was significantly higher in OS cell lines compared with the osteoblast cell line hFOB1.19 (P<0.01 and P<0.001). These results suggested that increased miR-106b-5p may serve an important role in OS.

Saos-2 and U2OS cells demonstrated relatively higher levels of miR-106b-5p and were therefore selected for miR-106b-5p inhibitor transfection to knockdown miR-106b-5p. RT-qPCR was utilized to validate transfection efficiency. The results demonstrated that transfection with the miR-106b-5p inhibitor significantly reduced miR-106b-5p levels in Saos-2 and U2OS cells compared with the miR-NC group (Fig. 2A; P<0.001). The results of the MTT assay indicated that miR-106b-5p inhibition significantly suppressed the proliferation of Saos-2 and U2OS cells at day 3, 4 and 5 (Fig. 2B; P<0.001). As cell proliferation may be regulated by cell cycle progression, flow cytometry analysis was performed to examine the cell cycle distribution in Saos-2 and U2OS cells. As presented in Fig. 2C, downregulation of miR-106b-5p caused an increase in the cell population in the G0/G1 phase (P<0.001). Accordingly, the cell population in the S and G2/M phase was reduced in Saos-2 (P<0.01 and P<0.001) and U2OS cells (P<0.01 and P<0.001). These results indicated that miR-106b-5p accelerated cell proliferation which may be closely associated with cell cycle progression in OS cells.

Bioinformatics analysis was performed to predict the potential target genes of miR-106b-5p involved in cell cycle regulation. Among the predicted targeted genes, three miR-106b-5p targets associated with cell cycle regulation, including CDC37L1, CDKN1A and CCNG2 were selected as potential target genes of miR-106b-5p. Luciferase reporter assay showed that there was no significant interaction between miR-106b-5p and the 3'UTRs of CCNG2 and CDC37L1 (Fig. S1). It was determined that CDKN1A, a gene associated with G1-S transition, was a putative target gene of miR-106b-5p (Fig. 3A). As expected, the luciferase activity of the miR-106b-5p inhibitor + WT-CDKN1A was significantly increased compared with the miR-106b-5p inhibitor + MUT-CDK1A in 293T cells (Fig. 3B). In addition, to further confirm whether CDKN1A was regulated by miR-106b-5p, the expression of CDKN1A in transfected Saos-2 and U2OS cells was examined using RT-qPCR and western blotting. The results demonstrated that miR-106b-5p downregulation significantly upregulated the expression of CDKN1A mRNA (Fig. 3C; P<0.01) and protein levels (Fig. 3D and E) in both Saos-2 and U2OS cells. Furthermore, the expression of CDKN1A in OS tissues was determined. As depicted in Fig. 3F, CDKN1A had a significantly lower expression in OS tissues compared with adjacent tissues (P<0.001). Through a two-tailed Pearson's correlation analysis, it was determined that the expression of miR-106b-5p was inversely correlated with CDKN1A expression in OS tissues (Fig. 3G; P=0.0057). Taken together, the results suggested that CDKN1A was negatively regulated by miR-106b-5p in OS.

To illustrate whether CDKN1A acts as a downstream effector in miR-106b-5p-mediated OS cell proliferation and cell cycle progression, gain-of-function assays were performed in Saos-2 cells by transfecting pcDNA3.1-CDKN1A. Following transfection, the mRNA (Fig. 4A; P<0.001) and protein (Fig. 4B) levels of CDKN1A were markedly increased in the pcDNA3.1-CDKN1A group compared with the control group in Saos-2 cells. The MTT assay further indicated that overexpression of CDKN1A suppressed the proliferation of Saos-2 cells, which was identical to the effect exerted by miR-106b-5p knockdown (Fig. 4C; P<0.001). In addition, flow cytometry demonstrated that CDKN1A upregulation induced cell cycle arrest at the G0/G1 stage in Saos-2 cells, similar to the effect of miR-106b-5p knockdown (Fig. 4D; P<0.001). Collectively, these results suggested that CDKN1A might be a direct effector involved in the miR-106b-5p-mediated cell proliferation in OS.

To further confirm whether miR-106b-5p-mediated cell proliferation and cell cycle progression was dependent on its capacity to modulate CDKN1A expression, rescue experiments were performed by transfecting si-CDKN1A plasmid into Saos-2 cells treated with miR-106b-5p inhibitor. The expression of CDKN1A in Saos-2 cells was detected and it was determined that si-CDKN1A transfection markedly attenuated the increased CDKN1A mRNA (P<0.01; Fig. 5A) and protein (Fig. 5B) expression. As expected, the decreased cell proliferation and induced cell cycle G0/G1 phase arrest by miR-106b-5p inhibitor were partially abolished by CDKN1A knockdown in Saos-2 cells, as determined by an MTT assay (P<0.01 and P<0.001; Fig. 5C) and flow cytometry analysis (P<0.001; Fig. 5D). These results further demonstrated that CDKN1A might be a key regulator of miR-106b-5p-mediated cell proliferation and cell cycle progression in OS cells.