MicroRNA-705 regulates the differentiation of mouse mandible bone marrow mesenchymal stem cells

Animals

Sixty 6-week-old female C57BL/6J mice were randomly divided into two groups. Mice in the Sham group underwent sham surgery. Mice in the OVX group received a bilateral OVX under general anaesthesia. After surgery, all mice were housed under pathogen-free conditions (24 °C, 12-hour light/12-hour dark cycles and 50–55% humidity) for 3 months. All animal procedures were performed according to the guidelines of the Animal Care Committee of Zunyi Medical University, Zunyi, Guizhou, China.

Cell culture

Three months after the OVX and sham surgeries, MSCs were isolated from the mouse mandibles and cultured. Briefly, the mandibles were removed aseptically after euthanasia and dissected free of soft tissues. The periodontal tissue with the whole teeth was removed. Part of the alveolar bone surrounding the tooth socket was also removed to ensure clearance of dental pulp and odontoblast contamination. Then, the mandibles were smashed with scissors and flushed with α-MEM (Invitrogen, Carlsbad, CA, USA, supplemented with 10% FBS and 1% penicillin and streptomycin) to free the marrow cavities. The cell suspension was filtered, seeded into 5-cm culture dishes and grown in α-MEM in a humidified atmosphere with 5% CO₂ at 37 °C. The medium was changed every two days, and adherent cells at confluence were passaged with 0.25% trypsin/1 mM EDTA. Cells from passages 2–3 were used in the experiments.

Alizarin red staining

Mandibular MSCs at the 2nd passage were digested and seeded in osteogenesis-inducing medium (100 mg/ml ascorbic acid, 2 mM β-glycerophosphate and 10 nM dexamethasone) for two weeks to induce osteoblast differentiation. Alizarin red staining was performed as described previously (Yang et al., 2013). Quantification analysis was performed by extracting mineralized nodules with cetylpyridinium chloride and measuring the absorbance with a spectrophotometer at 570 nm.

Oil red O staining

Mandibular MSCs at the 2nd passage were digested and seeded in adipogenesis-inducing medium (0.5 mM isobutyl methylxanthine, 0.5 mM dexamethasone and 60 mM indomethacin; Sigma-Aldrich, St. Louis, MO, USA) for one week for adipogenic induction. Oil red O staining was performed to detect lipid droplets as described previously (Yang et al., 2013). Quantification analysis was performed by extracting lipid droplets with isopropanol and measuring the absorbance with a spectrophotometer at 520 nm.

Real-time qRT-PCR analysis of mRNAs and miRNAs

Total RNA was extracted using the TRIzol reagent (Invitrogen Life Technology, Carlsbad, CA, USA). Single-strand cDNA was synthesized from 2 mg of total RNA with the PrimeScript RT reagent kit (TaKaRa Bio, Dalian, China). For the qRT-PCR analysis, the miRNAs were reverse-transcribed with RT primers (RiboBio, Guangzhou, China). β-actin and U6 were used as loading controls for quantitation of the mRNAs and miRNAs respectively. The Bulge-Loop miRNA qRT-PCR Primer Set (RiboBio) was used for qRT-PCR of miR-705 (product ID: miRQ0003495-1-2) and U6 (Product ID: MQP-0202). All real-time PCR analyses were performed and analysed with the SYBR Premix Ex Taq II kit (TaKaRa) and the ABI Prism 7500 HT sequence detection system (Applied Biosystems, Foster City, CA, USA). The primer sequences are shown in Table S1.

Western blot

The western blot analyses were performed as previously described (Liao et al., 2013). The proteins were loaded on 10% sodium dodecyl sulphate polyacrylamide gels, transferred to polyvinylidene fluoride membranes (Millipore, Billerica, MA, USA) and blocked with 5% non-fat milk powder in PBST (containing 0.1% Tween). The membranes were probed overnight with the following primary antibodies: β-actin (Cell Signalling, Beverly, MA, USA), homeodomain-containing factor A10 (HOXA10) (Santa Cruz, Dallas, TX, USA) and Forkhead box O1 (FoxO1) (Cell Signalling). The membranes were incubated with peroxidase-conjugated secondary antibody (Boster, Wuhan, China). The blots were visualized through an enhanced chemiluminescence kit (Amersham Biosciences, Piscataway, NJ, USA) according to the recommended instructions.

Transfection of the miRNA mimics and inhibitor

The miR-705 mimics, inhibitor and negative control were purchased from RiboBio. The final concentration used for transfection was adjusted to 50 nM. The siPORTNeoFX transfection reagent (Ambion, Austin, TX, USA) was used according to the manufacturer’s instruction. Briefly, diluted miRNA mimics/inhibitor and diluted siPORTNeoFX transfection reagent were mixed and incubated for 10 min at room temperature. Cultured cells at 80% confluence were digested, and the suspensions were overlaid onto the transfection complexes and incubated at 37 °C for 48 h. Control group cells received the transfection reagent alone, and the other operations were the same.

Micro-CT analysis

The mice were scanned with the Inveon micro-CT system (Siemens AG, Germany). The scans were performed using an 80-kV and 500-mA micro-focus X-ray source. The cross-sectional volumetric bone mineral density (BMD) was measured at the angle of the mandible (the rectangle area near the joint area of the body and the ramus of the mandible posterior to the molars). Bone morphometric parameters and the bone volume relative to the tissue volume (BV/TV) were also assessed. Micro-CT scanning of femurs was performed as described previously (Yang et al., 2013).

Statistical analysis

The data are presented as the means ± SDs. Comparisons were made using one-way ANOVA followed by Tukey-Kramer test as a post-hoc. All experiments were repeated at least three times, and representative experiments were shown. Differences were considered significant at P < 0.05.

Cell lineage commitment of mandibular MSCs shifts to adipocytes during osteoporosis

The OVX models were constructed and confirmed as previously described (Liao et al., 2013). As expected, all mice manifested osteoporosis 3 months after OVX. In addition to the femur results, we also focused on the mandible (Fig. 1A). The BMD and BV/TV were significantly reduced in the femurs of OVX mouse (Figs. 1B, 1C). However, the BMD and BV/TV comparisons were barely significant between the OVX and Sham mandibles, which were quite different from the results obtained in the long bones (Figs. 1D, 1E).

MSCs were isolated from both the OVX and Sham mouse mandibles and cultured in vitro for further investigation. The flow cytometry analysis showed that these cells were all positive for CD29, CD105, CD106 and SCA-1 but were negative for CD34 and CD45. No significant difference was observed in the proliferation comparison.

We then compared the differentiation potentials of the OVX and Sham mandibular MSCs. The OVX mandibular MSCs formed much fewer mineralized nodules but more adipocytes than the Sham MSCs during induction, suggesting that weaker osteogenic but stronger adipogenic differentiation occurred in the OVX mandibular MSCs (Figs. 2A, 2B, 2E and 2F). This result was confirmed by qRT-PCR. Runt-related transcription factor 2 (RUNX2) and alkaline phosphatase (ALP) were analysed as two osteogenic marker genes. The OVX mandibular MSCs expressed significantly less RUNX2 and ALP than the Sham MSCs after 2 weeks of osteogenic induction (Figs. 2C, 2D). However, after 1 week of adipogenic induction, the expression of peroxisome proliferator-activated receptor- γ (PPAR-γ) and lipoprotein lipase (LPL), which were adipogenic marker genes, were significantly increased compared with the levels in the Sham MSCs (Figs. 2G, 2H).

Osteoporotic mandibular MSCs overexpress miR-705

In a previous study, we detected the overexpression of miR-705 in osteoporotic bone marrow using MSCs from OVX mouse femurs. Here, we confirmed that miR-705 was also overexpressed in osteoporotic mouse mandible BMMSCs. The qRT-PCR results showed that miR-705 was significantly upregulated in 3rd-passage OVX mouse mandibular MSCs compared to the expression level in the Sham MSCs (Fig. 3A).

Subsequently, to evaluate whether miR-705 affected osteogenic differentiation in mandibular MSCs, we examined miR-705 expression during osteogenic induction in vitro. According to the qRT-PCR results, miR-705 expression declined during osteogenic induction, with minimum expression witnessed on day 7 (Fig. 3B). Conversely, its expression increased during adipogenic induction and hit the maximum level on day 7 (Fig. 3C).

Inhibition of MSC osteoblast lineage commitment by miR-705

To investigate its role in cell lineage commitment, we knocked down and overexpressed miR-705 in mandibular MSCs with the si-miR-705 (inhibitor) and mimics respectively. The effects of the miRNA inhibitor and mimics were confirmed by qRT-PCR up to day 14 of induction (Fig. S1A). Alizarin red staining after 14 days of induction confirmed enhanced nodule formation by mandibular MSCs after miR-705 knockdown (Figs. 4A–4D). And qRT-PCR showed enhanced expression of both ALP and RUNX2 in the knockdown MSCs (Figs. 4E, 4F). However, overexpression of miR-705 by the mimics significantly decreased mineralization of the mandibular MSCs as well as the ALP and RUNX2 expression levels.

Promotion of MSC adipocyte lineage commitment by miR-705

To assess the function of miR-705 in adipocyte lineage commitment, we conducted a gain or loss of function assay on the mandibular MSCs. The effects of the miRNA inhibitor and mimics were confirmed by qRT-PCR up to day 7 of induction (Fig. S1B). Oil red O staining showed fewer fat droplets after 7 days of adipo-induction in the miR-705 knockdown MSCs (Figs. 5A–5D). Expression of the adipogenic marker genes LPL and PPAR-γ was also decreased on day 7 in the qRT-PCR assay (Figs. 5E, 5F). Conversely, promotion of both adipocyte formation and LPL and PPAR-γ expression was observed in the miR-705-overexpressing MSCs.

Knockdown of miR-705 rescues the cell lineage commitment disorder of MSCs from osteoporotic mandibles

Several miRNAs were shown to have functions in shifting MSC linage commitment. In this study, we tested whether knocking down miR-705 attenuated the lineage commitment disorder in MSCs from osteoporotic mandibles. Two groups of OVX mandibular MSCs were transfected either with the miR-705 inhibitor or the transfection reagent alone. And a group of Sham mandibular MSCs served as the control. As shown in Fig. 6, knocking down miR-705 partially rescued osteogenic differentiation of the OVX mandibular MSCs after osteo-induction. Significantly more mineralized nodules were observed in the knockdown group than those in the OVX group (Figs. 6A–6D). This result was confirmed by qRT-PCR, which showed increased RUNX2 and ALP mRNA expression in the OVX mandibular MSCs after miR-705 knockdown (Figs. 6E, 6F).

Next, we tested whether the knockdown of miR-705 reduced the excessive adipogenic differentiation of the OVX mandibular MSCs in the same way. The knockdown of miR-705 resulted in reduced lipid droplet accumulation in the induced OVX mandibular MSCs to a certain extent (Figs. 7A–7D). Consistently, it reduced the expression of the adipogenic marker PPAR-γ and LPL (Figs. 7E, 7F). These results suggested that the knockdown of miR-705 in OVX mandibular MSCs partially attenuated their adipogenic differentiation.

Knockdown of miR-705 increases the expression of target genes in MSCs from osteoporotic mandibles

To uncover details of the attenuation observed above, we tested the transcription levels of its predicted target mRNAs HOXA10 and FoxO1 at 48 h post-transfection. As shown in the qRT-PCR results, miR-705 expression was remarkably reduced at 48 h after inhibitor transfection (Fig. 8A). According to western blot results, HOXA10 and FoxO1 were suppressed in the MSCs from the OVX mandibles compared with the negative controls but were increased significantly 2 days after miR-705 inhibitor transfection (Fig. 8B).