The HEK-293 (human embryonic kidney) cell line, C2C12 (myoblasts) cell line, and MEF cell line were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). The HEK-293 cells, C2C12 cells and MEF cells were all immortalized and maintained in DMEM, supplemented with 10% fetal bovine serum (FBS) (Gibco, Grand Island, NY, USA), 100 U/ml penicillin and 100 µg/ml streptomycin. The cells were maintained at 37°C in a humidified 5% CO₂ atmosphere.

miR-155 mimic, miR-155 inhibitor, and shBMPR2 were purchased from GenePharma (Shanghai, China). The sequences for each oligonucleotide are listed in Table I. Entranster™ R4000 (Engreen Biosystems, Beijing, China) was utilized to transfect miR-155 mimic, miR-155 inhibitor and miR-155 negative control (NC) according to the manufacturer's instructions. To transfect shBMPR2 and it negative control shNC, Lipofectamine 2000 transfection reagent (Invitrogen, Shanghai, China) was used according to the manufacturer's instructions. In this study, C2C12 cells and MEF cells were transfected with with 100 nM of miR-155 mimic, or miR-155 inhibitor or NC in a 24-well plate or 500 nM of the above in a 6-well plate. In a 6-well plate, shNC or shBMPR2 were transfected into cells at 800 ng each well.

C2C12 cells and MEF cells were cultured in 24-well plates, each well was added with 500 µl DMEM supplemented with 10% FBS, and before using BMP9 and si-Runx2, we tested the titer of these recombinant adenoviruses. Each well contained 3 µl of polybrene to contribute to the adenovirus entry into cells, different volumes of adenovirus was added, until the infection efficiency reached 30%, the volume of the adenovirus was the titer for the corresponding cells, respectively. In a 6-well plate, 12 µl of polybrene and four-fold titer volume of adenovirus were added after that the cells attached.

Total RNA was extracted from C2C12 cells and MEF cells at the indicated time points using TRIzol reagent (Ambion, Aukland, New Zealand) according to the manufacturer's instructions for the analysis of mRNA and miRNA expression. The concentration and purity of RNA were quantified using spectrophotometer (NanoDrop 1000 Spectrophotometer; Thermo Fisher Scientific Inc., Waltham, MA, USA). The extracted total RNA (2,000 ng) was reverse transcribed using reverse transcription primers according to the manufacturer's instructions (Reverse Transcriptase M-MLV, Takara code: D2639A; Takara Bio Inc., Otsu, Japan). The primers used for reverse transcription PCR (RT-PCR) and RT-qPCR are listed in Table II. RT-qPCR were performed on a Bio-Rad iQ5 instrument (Bio-Rad, Hercules, CA, USA) used cDNA as templates to amplify target genes, mixing with SYBR^®-Green PCR Master Mix (Takara Bio, Inc.), Fold-changes in expression were calculated using the 2^-∆∆Ct method, and each experiment was performed in triplicate. Data were analyzed using Optical system software, version 2.0. The expression levels of β-actin and the small U6 RNA were used as internal controls for mRNA and miRNA, respectively.

ALP staining was performed with BCIP/NBT Alkaline Phosphatase Color Development kit (C3206; Beyotime, Shanghai, China) according to the manufacturer's instructions. Briefly, we seeded the C2C12 cells and MEF cells in 24-well plates followed by treatment with NC, miR-155 mimic or miR-155 inhibitor and BMP9 for 7 days. The supernatant was discarded, cells were washed three times using phosphate-buffered saline (PBS). Then 3 ml of ALP developing buffer was mixed with 10 µl BCIP, and 20 µl NBT to form BCIP/NBT working liquid and 200 µl BCIP/NBT was the working liquid in each well of 24-well plates, and the plates were placed in the dark to undergo ALP staining for 30 min. Finally, BCIP/NBT working liquid was removed by suction and the staining observed. Cells were scanned at both a lower (ScanMaker 3600; Shanghai Microtek Technology Co., Ltd., Shanghai, China) and higher magnification under a bright field microscope (T-DH; Nikon Corp., Tokyo, Japan). ALP activity was measured using an established enzymatic assay. ALP substrate was prepared from solution A (8 ng naphtol-AS-TR phosphate/300 ml N,N-dimethylformamide; both from Sigma-Aldrich, St. Louis, MO, USA) and solution B (24 mg fast blue BB salt/30 ml of 100 mmol/l Tris HCl, pH 9.6). The two solutions were combined, 10 mg of MgCl₂ was added, the resulting solution was used immediately. Enzymatic activity was determined in cell lysates that were solubilized with 0.1% Triton X-100. Aliquots (5 µl) of each sample were incubated with 15 µl of AP substrate buffer (100 mmol/l diethanolamine, 150 mmol/l NaCl, 2 mmol/l MgCl₂ p-nitrophenylphosphate at 2.5 mg/ml) and 5 µl of ALP substrate under the conditions of protection from light for 30 to 40 min at room temperature. ALP activity was tested using a luminometer (Promega, Madison, WI, USA).

C2C12 cells and MEF cells were seeded in 24-well plates treated with NC, miR-155 mimic or miR-155 inhibitor and BMP9 for 14 days. The supernatant was discarded, and cells washed three times with PBS. The cells were fixed with 0.05% ice-cold glutaraldehyde (Chongqing Chuandong Chemical Group) for 10 min and rinsed with double-distilled H₂O. Then stained with 40 mM (2%) Alizarin red S (Sigma-Aldrich) at pH 4.0, rocked gently for ~5 min. The cells were rinsed three times with double-distilled H₂O and then rinsed for 15 min with PBS and gently shaken. Finally cells were scanned at both a lower (ScanMaker 3600; Shanghai Microtek Technology Co., Ltd.) and higher magnification under a bright field microscope (T-DH; Nikon Corp.).

C2C12 cells and MEF cells were plated into a 10-cm Petri dishes. After treating cells as before, and when cells reached 80% confluence, cell lysate was collected and protein concentrations were determined using the Bio-Rad protein assay kit. The protein (50 µg) was boiled for 5 min in Laemmli buffer [62.5 mM Tris (pH 6.8), 1% sodium dodecyl sulfate (SDS), 20% glycerol, 0.01% bromophenol blue, and 100 mM DTT] before being loaded onto a 12% SDS-polyacrylamide gel and then transferred to PVDF membranes. Specific protein bands were detected with mouse anti-BMPR2 polyclonal antibody (1:200 dilution, cat. no. BM2821; BOSTER), rabbit anti-p-Smad1/5/8 monoclonal antibody (1:1,000 dilution, cat. no. sc-12353), rabbit anti-Smad1/5/8 monoclonal antibody (1:1,000 dilution, cat. no. sc-6031-R, lot no. D1911), rabbit anti-Runx2 monoclonal antibody (1:1,000 dilution, cat. no. sc-10758, lot no. A0810) (all from Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), rabbit anti-osteocalcin (OCN) polyclonal antibody (1:200 dilution, cat. no. bs-4917R; Beijing Biosynthesis Biotechnology Co., Ltd.), rabbit anti-osteopontin (OPN) polyclonal antibody (1:500 dilution, cat. no. wl00691; Wanleibio, Shenyang, China), or anti-β-actin mouse monoclonal antibody (1:1,000 dilution; Zhongshan Golden Bridge Biotechnology Co., Ltd., Beijing, China) overnight at 4°C, and then washed with TBST three times, each time for 15 min, the membranes were incubated with goat anti-rabbit or goat anti-mouse horseradish peroxidase conjugated secondary antibody (1:5,000 dilution; Zhongshan Golden Bridge Biotechnology Co., Ltd.), for 1 h at 37°C. After that, membranes were washed with TBST three times, each time for 15 min. The secondary antibodies were detected with western chemiluminescence reagent (Millipore Corp., Billerica, MA, USA). The results were recorded using the Bio-Rad electrophoresis documentation system (Gel Doc 1000) and Quantity One software, version 4.5.0.

TargetScan (http://www.targetscan.org) and PicTar (http://pictar.mdc-berlin.de/) were utilized to identify the possible target genes of miR-155 in BMP9-induced osteogenic differentiation of C2C12 cells and MEF cells. The 3′-UTR of Runx2 and BMPR2 containing miR-155-binding sequences, and the sequences contained binding sites with miR-155 were synthesised following the manufacturer's instructions (pMIR-REPORT™ system, miRNA expression reporter vector; Ambion, Carlsbad, CA, USA). The sequences contained binding sites with miR-155 are listed in Table III. The annealing fragments were then cloned into the SpeI/HindIII backbone of the pMIR-REPORT microRNA (miRNA) expression reporter (Ambion). Binding-region mutations were achieved using an oligo synthesis following the manufacturer's instructions (Ambion). Transient transfection into HEK-293 cells (1×10⁴ cells/well) was carried out in 24-well plates with Lipofectamine 2000 transfection reagents (Invitrogen) following the manufacturer's instructions. The cells were co-transfected with 800 ng of the luciferase construct plasmid and miR-155 mimic or NC, and luciferase assays were performed using the dual-luciferase reporter assay system (Ambion) according to the manufacturer's instructions. Luminescent signals were quantified using a luminometer (Promega), and each value of luciferase activity was normalized by the luciferase activity of cells that transfected empty vector only.

All animal experiments were approved by Institutional Animal Care and Use Committee (IACUC) of Chongqing Medical University. The nude mice were purchased from Huafukang (Beijing Hfk Bioscience Co., Ltd., Beijing, China). MEF cells were transfected as described above, and the long-lasting mimic (mmu-agomiR-155) and inhibitor (mmu-antagomiR-155) of miR-155 were used, and the sequences are fully consistent with mmu-miR-155 or mmu-anti-miR-155. The sequence of stable NC is the same as the NC used before. In short, after treating cells like before, cells were collected at 80% confluence and for subcutaneous injection (5×10⁶/injection) into the flanks of the athymic nude mice (5 animals/group, 4–6-week old, female) (44). At 4 weeks after implantation, heterotopic bones were harvested from mice after euthanasia, fixed in 4% paraformaldehyde and scanned using a µCT 40 system (Scanco Medical, Brüttisellen, Switzerland). Parameters computed from these data include bone mineral density (BMD), bone volume (BV/TV), trabecular thickness (Tb.Th), trabecular number (Tb.N) and trabecular separation (Tb.Sp). For histological analysis of bones, after scanning used µCT, decalcified in 10% EDTA for ~4 weeks, before being embedded in paraffin. Sections (5 µm) were cut and stained with hematoxylin and eosin (H&E) or with Masson's trichrome according to the manufacturer's instructions. Bone histomorphometry was performed using a microscope (T-DH; Nikon Corp.).

The quantitative experiments were performed in triplicate and/or repeated three times. Data are represented as the mean ± SD. Statistical significance between the control and treatment groups were determined by one-way analysis of variance and the Student's t-test. All statistical analyses were performed using GraphPad Prism 5 (GraphPad Software, Inc., La Jolla, CA, USA). P<0.05 was considered statistically significant.

Several differential expression miRNAs including miR-21, miR-23b and miR-155 have been previously identified during the early stage of mesenchymal stem cells (MSCs) osteogenesis induced by BMP9 through microarray data analysis (data not shown). Allowing us to consider that there are some associations between miR-155 and BMP9 in osteogenesis. During the early stage of osteogenic differentiation induced by BMP9, we detected the expression of miR-155 by using RT-qPCR on different days. It was found that compared with the expression level of 0 day, the expression level of miR-155 was upregulated in the early stage of BMP9-induced C2C12 cells and MEF cells osteogenic differentiation, and on the 5th day reached the maximum expression, then decreased on the 7th day (Fig. 1A). In our laboratory, BMP9 contains green fluorescence protein (GFP) tag, and so the results of fluorescence microscopy proved that BMP9 was transfected into C2C12 and MEF cells effectively (Fig. 1B). To further confirm the osteoinduction effect of BMP9, RT-qPCR was used to detect the mRNA expression level of osteoblast-specific genes over 7 days, including Runx2 and ALP. The results revealed that successful osteogenic differentiation of the C2C12 and MEF cells was induced (Fig. 1C).

Subsequently, to investigate the roles of miR-155 in the osteogenic differentiation of C2C12 cells and MEF cells induced by BMP9, the expression level of miR-155 following transfection were tested by RT-qPCR, in the whole study, we set control groups contain BMP9 group and BMP9 + miR-negative control group (BMP9 + NC) (Fig. 3A). After transfecting miR-155 or anti-miR-155 into C2C12 cells and MEF cells, respectively, induced osteogenesis by BMP9 for 7 days. ALP staining results showed that compared with the control groups, overexpressed miR-155 (BMP9 + miR-155) downregulated the early osteogenesis differentiation indicator-ALP activity, but inhibition of miR-155 (BMP9 + anti-miR-155) attenuated this inhibitive effect (Fig. 2A, top panel), quantitative determination of ALP activity, the results were in accordance with the ALP staining of C2C12 cells and MEF cells, respectively (Fig. 2A, bottom panel).

We further analyzed the effect of miR-155 on late osteogenic differentiation by Alizarin Red S staining of matrix mineralization. The results were consistent with ALP staining, overexpressed exogenous miR-155 repressed the BMP9-induced calcium deposition of C2C12 cells and MEF cells, and decreased the endogenous miR-155 adverse effect (Fig. 2B). In conclusion, the above results indicated that miR-155 impairs the BMP9-induced osteoblast lineage commitment of MSCs.

To detect the influence of miR-155 on the osteogenesis induced by BMP9 on mRNA level, firstly, we used RT-qPCR to measure the expression of miR-155 after transfection to verified that our transfection was successful. The data showed that during osteogenic differentiation induced by BMP9, compared with control group, the intracellular miR-155 level was significantly elevated by transfection with miR-155, whereas transfected with anti-miR-155 led to distinct reduction in miR-155 content (Fig. 3A). RT-qPCR was used to evaluate the expression level of Runx2 and osterix (OSX) on the 3rd day, the expression of ALP and OCN on the 7th day of BMP9-induced osteogenesis. Compared with the control groups, the expression of these osteogenesis markers were decreased by miR-155, conversely, anti-miR-155 increased their expression in the process of osteogenic differentiation induced by BMP9 in C2C12 cells, except the expression of OSX and ALP (Fig. 3B). The results of MEF cells were consis tent with C2C12 cells (Fig. 3C). Taken together, miR-155 could suppress the expression of Runx2, OSX, ALP and OCN, which are bone differentiation related markers.

Previous studies revealed that Smads, p38 and ERK1/2 are involved in BMP9-induced osteogenic differentiation (45). We hypothesized that miR-155 may through inhibiting canonical Smad-dependent signaling repress the activity of downstream transcription factors to achieve suppressed osteoblastogenesis in MSCs. The expression levels of p-Smad1/5/8 and Runx2 were measured by western blot analysis after treatment of cells as above. Compared with the control group, overexpressed miR-155 group (BMP9 + miR-155) diminished the expression of p-Smad1/5/8 and Runx2, however, downregulated miR-155 group (BMP9 + anti-miR-155) eliminated this inhibitory effect in C2C12 cells (Fig. 4A) and MEF cells (Fig. 5A). Based on these data, we concluded that the negative effect of miR-155 on osteogenic differentiation may be through suppressing canonical BMP (BMP/Smad) signaling.

To further examine the role of miR-155 in late stage of osteogenic differentiation, we tested the expression of OCN and OPN by western blot analysis after treating C2C12 cells and MEF cells indicated above for 7 days. The results showed that compared with NC group (BMP9 + NC), exogenous overexpression of miR-155 (BMP9 + miR-155) decreased the expression of OCN and OPN, and downregulated miR-155 group (BMP9 + anti-miR-155) eliminated this inhibitory effect in C2C12 cells (Fig. 4B) and MEF cells (Fig. 5B). The above data demonstrated that inhibition of miR-155 led to BMP signaling that coincided with increased bone formation, strongly suggested that miR-155 suppressed BMP9-induced osteogenic differentiation by repressing BMP signaling.

To further study the mechanisms of miR-155 inhibited osteogenic differentiation induced by BMP9, we used biological information analysis to find the potential targets of miR-155 in BMP signaling. From the target prediction tools (TargetScan and PicTar), we found the target binding sites of miR-155 on the 3′-UTR of Runx2 and BMPR2, which can complementary bind with miR-155 seed sequence. In order to evaluate whether Runx2 and BMPR2 are direct targets of miR-155, we measured the mRNA and protein expression levels following the regulation of miR-155 expression by transfecting the MSCs with NC, miR-155, or anti-miR-155. After determining the transfection efficiency (Fig. 6A), we discovered that in overexpressed-miR-155 C2C12 cells, the mRNA expression level of Runx2 and BMPR2 was not significantly altered compared with NC. In anti-miR-155 group, the expression levels had no significant changes as miR-155 group compared with NC (Fig. 6B). While, in MEF cells, the expression of Runx2 was increased in anti-miR-155 group compared with NC group, the expression level was not significantly altered in miR-155 group. The mRNA expression of BMPR2 had no markedly significant differences in different groups (Fig. 6D). However, compared with NC, the protein expression level of Runx2 and BMPR2 was decreased in miR-155 group in C2C12 cells and MEF cells, respectively (Fig. 6C and E). These results demonstrated that miR-155 significantly decreased the protein expression level of Runx2 and BMPR2, but did not markedly altered the mRNA expression levels, as compared with NC, respectively.

To test whether miR-155 directly binds to the 3′-UTR of Runx2 and BMPR2, luciferase reporter assays were performed in HEK-293 cells. We directly designed and synthesized sequences matching miR-155 target sites and cloned these wild-type and mutant-type into the multiple cloning sites of pMIR-REPORT luciferase, termed WT-Runx2, MT-Runx2, WT-BMPR2 and MT-BMPR2 (Fig. 6F). Luciferase levels were significantly reduced when cells were co-transfected with WT-Runx2 and miR-155 compared with the cells that co-transfected with WT-Runx2 and NC. However, this difference was abolished when the binding site was mutated suggesting a specific Runx2∷miR-155 direct interaction (Fig. 6G). For BMPR2, the result was the same as Runx2, when co-transfected with WT-BMPR2 and miR-155 into HEK-293 cells, luciferase activity was decreased remarkable, mutated binding site could offset this depressant effect demonstrating a specific BMPR2∷miR-155 direct interaction (Fig. 6G). In conclusion, the results suggested that Runx2 and BMPR2 are direct targets of miR-155.

To validate the role of Runx2 and BMPR2 in miR-155 inhibited osteogenic differentiation of MSCs induced by BMP9, we tested the expression changes of Runx2 and BMPR2 under the influence of miR-155 in osteogenesis. In C2C12 cells and MEF cells, in the process of differentiation, compared with NC, miR-155 repressed the expression of Runx2 at mRNA level (Fig. 3B and C) and protein level (Figs. 4A and 5A), and anti-miR-155 reduced the inhibitory effect of miR-155 on the expression of Runx2 and BMPR2. In C2C12 and MEF cells, miR-155 had no obvious influence at the expression of mRNA level, compared with NC during the differentiation (Fig. 7A). However, it reduced the protein expression level of BMPR2, and anti-miR-155 promoted the protein level in C2C12 cells compared with NC (Fig. 7B).

To further verify the association between miR-155 and its target genes, Runx2, and BMPR2, we blocked the expression of Runx2 by si-Runx2, and blocked the expression of BMPR2 by shBMPR2. Through blocking the target genes of miR-155, the effect of miR-155 on osteogenesis of MSCs was evaluated, to speculate on the function of the target genes of miR-155. After transfecting si-Runx2, shBMPR2 or their controls into C2C12 cells and MEF cells, respectively, used RT-qPCR to examine the interfering efficiency, data showed that compared with the control groups, the expression of Runx2 and BMPR2 was significantly decreased (Fig. 8A). The results of western blot analysis shown that the protein expression of Runx2 and BMPR2 was significantly reduced compared with their control group, respectively (Fig. 8B).

In C2C12 cells, the expression of Runx2, OSX, ALP and OCN were distinctly decreased when transfected with si-Runx2 (BMP9 + si-Runx2) compared with RFP (BMP9 + RFP); and in co-transfection with miR-155 and si-Runx2 group (BMP9 + miR-155 + si-Runx2), the expression of these four bone-related genes were significantly increased compared with co-transfection with miR-155 and RFP group (BMP9 + miR-155 + RFP) (Fig. 9A). In the process of osteogenesis, after transfecting shBMPR2 (BMP9 + shBMPR2), the expression of Runx2, OSX, ALP and OCN were markedly decreased compared with control group (BMP9 + shNC); in co-transfected miR-155 and sh-BMPR2 group (BMP9 + miR-155 + sh-BMPR2), the expression of these four bone differentiation related genes was notably increased compared with co-transfected with miR-155 and shNC group (BMP9 + miR-155 + shNC) (Fig. 9B). The results obtained from MEF cells coincided with the results of C2C12 cells (Fig. 10), when co-transfected with miR-155 and si-Runx2 or sh-BMPR2, the expression of Runx2, OSX, ALP and OCN was remarkably increased compared with co-transfected with miR-155 and RFP or shNC, respectively. These data indicated that knockdown of Runx2 and BMPR2, respectively, the effect of miR-155 on the osteogenic differentiation induced by BMP9 was reduced. In conclusion, miR-155 inhibited osteogenic differentiation induced by BMP9 may be through repression of the expression of its target genes-Runx2 and BMPR2, which are, at least partially, quite important components in BMP signaling.

To determine the consequences of miR-155 on osteogenic differentiation induced by BMP9 in vivo, we performed µCT analysis of heterotopic bones from 10-week-old female nude mice. The microarchitecture of bones were in accordance with data from cell culture experiments, µCT images of ectopic bones showed decreased trabecular bone volume in BMP9 + agomiR-155 group compared to BMP9 + NC group, while in BMP9 + antagomiR-155 group, the trabecular bone volume was increased compared with BMP9 + NC group (Fig. 11A). In µCT analyses, BMP9 + agomiR-155 group exhibited significant decrease in BMD, BV/TV and Tb.Th, but a significant increase in Tb.SP, there were no variances in Tb.N compared with BMP9 + NC group, respectively. On the contrary, antago-miR-155 elevated BV/TV and Tb.Th of the bones consequently, and had no effect on Tb.N, Tb.SP and BMD in the ectopic bone induction of BMP9 as well (Fig. 11B). After the µCT scan, all specimens were subjected to a histological study involving H&E staining and Masson's trichrome staining. In histological analyses, the results correspond to the results that we gained from cell culture in vitro. agomiR-155 inhibited the osteogenic differentiation of MEF cells induced by BMP9 in vivo, decreased trabecular bone mass, whereas antagomiR-155 reversed this phenomenon (Fig. 11C). The results of Masson's trichrome staining were very similar to the H&E staining in all the groups, showing that the chondrocytes were stained in BMP9 + antagomir-155 group more, but in BMP9 + agomiR-155 group, the staining was lighter compared with BMP9 + NC, respectively (Fig. 11D). Given the above, in vivo, agomiR-155 depressed the ectopic bone formation of MEF cells induced by BMP9, antagomiR-155 facilitated ectopic bone formation, conversely.