miR-182 is a negative regulator of osteoblast proliferation, differentiation, and skeletogenesis through targeting FoxO1

Authors

  • Kyoung Min Kim,

    1. Division of Endocrinology, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    2. Endocrine Research Institute, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
    • Group authorship/contribution: KMK and SJP contributed equally to this work.

  • Su Jin Park,

    1. Division of Rheumatology, Department of Internal Medicine, Hallym University Sacred Heart Hospital, Kyunggi, Korea
    Search for more papers by this author
    • Group authorship/contribution: KMK and SJP contributed equally to this work.

  • Seung-Hyun Jung,

    1. Department of Biology and Graduate School of Analytical Science and Technology, Chungnam National University, Daejeon, Korea
    2. Diabetic Complications Research Center, Division of Traditional Korean Medicine (TKM) Integrated Research, Korea Institute of Oriental Medicine (KIOM), Daejeon, Korea
    Search for more papers by this author
  • Eun Jin Kim,

    1. Department of Biomedical Science, College of Medicine, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Gadi Jogeswar,

    1. Endocrine Research Institute, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Jami Ajita,

    1. Brain Korea 21 Project for Medical Science, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Yumie Rhee,

    1. Division of Endocrinology, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    2. Endocrine Research Institute, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    Search for more papers by this author
  • Cheol-Hee Kim,

    1. Department of Biology and Graduate School of Analytical Science and Technology, Chungnam National University, Daejeon, Korea
    Search for more papers by this author
  • Sung-Kil Lim

    Corresponding author
    1. Division of Endocrinology, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    2. Endocrine Research Institute, Department of Internal Medicine, Yonsei University College of Medicine, Seoul, Korea
    3. Department of Biomedical Science, College of Medicine, Yonsei University College of Medicine, Seoul, Korea
    4. Brain Korea 21 Project for Medical Science, Yonsei University College of Medicine, Seoul, Korea
    • Division of Endocrinology, Department of Internal Medicine, Endocrine Research Institute, Department of Internal Medicine, Brain Korea 21 Project for Medical Science, and Department of Biomedical Science, College of Medicine, Yonsei University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, Seoul 120-752, Korea.
    Search for more papers by this author

Abstract

Uncontrolled oxidative stress impairs bone formation and induces age-related bone loss in humans. The FoxO family is widely accepted to play an important role in protecting diverse cells from reactive oxygen species (ROS). Activation of FoxO1, the main FoxO in bone, stimulates proliferation and differentiation as well as inhibits apoptosis of osteoblast lineage cells. Despite the important role of FoxO1, little is known about how FoxO1 expression in bone is regulated. Meanwhile, several recent studies reported that microRNAs (miRNAs) could play a role in osteoblast differentiation and bone formation by targeting various transcriptional factors. Here, we identified one additional crucial miRNA, miR-182, which regulates osteoblastogenesis by repressing FoxO1 and thereby negatively affecting osteogenesis. Overexpression of miR-182 in osteoblast lineage cells increased cell apoptosis and inhibited osteoblast differentiation, whereas in vivo overexpression of miR-182 in zebrafish impaired bone formation. From in silico analysis and validation experiments, FoxO1 was identified as the target of miR-182, and restoration of FoxO1 expression in miR-182–overexpressing osteoblasts rescued them from the inhibitory effects of miR-182. These results indicate that miR-182 functions as a FoxO1 inhibitor to antagonize osteoblast proliferation and differentiation, with a subsequent negative effect on osteogenesis. To treat bone aging, an antisense approach targeting miR-182 could be of therapeutic value. © 2012 American Society for Bone and Mineral Research.

Introduction

Osteoporosis is a representative age-associated human disease characterized by decreased density and quality of the skeleton with aging that leads to an increased risk of fracture in the elderly.1 The mechanism of aging is not yet fully understood, but several factors, such as oxidative stress, nuclear pore leakiness, and autophagy failure are thought to contribute based on pathophysiology studies.2–4 Among these, reactive oxygen species (ROS) are thought to play significant roles in the pathogenesis of age-associated human diseases and may also be important in bone loss with aging.5–7 To counteract the adverse effects of oxidative stress, cells possess intricate antioxidant defense mechanisms. In bone, FoxOs, a large family of forkhead proteins characterized by the presence of a winged-helix DNA binding domain called a forkhead box, are involved in the antioxidant process and play crucial roles in redox balance and osteogenesis.8, 9 Among the three predominant members of the FoxO family (FoxO1, FoxO3a, and FoxO4), FoxO1 is the main regulator of redox balance and function in osteoblasts and bone mass; depletion of FoxO1 specifically from osteoblasts results in decreased proliferation of osteoblasts, a decreased rate of bone formation, decreased bone volume, and increased oxidative stress levels.9

MicroRNAs (miRNAs) are endogenous noncoding single-stranded RNAs containing 21 to 23 nucleotides that regulate posttranscriptional gene expression by binding to 3′-untranslated regions (3′UTRs) of target messenger RNAs (mRNAs).10, 11 Recently, miRNAs have emerged as important regulators of various biological processes, including cellular differentiation, proliferation, apoptosis, and tissue development.12, 13 Moreover, alterations in miRNA expression are closely related to human disease.14, 15 Numerous miRNAs, eg, miR-2861, miR-206, miR-29, and miR 204, have been reported to be crucial regulators in bone by targeting various genes and can alter the phenotype of bone.15–18

In this study, we found that one additional crucial miRNA, miR-182, which regulates osteoblastogenesis by repressing FoxO1, plays an important role in bone homeostasis by an antioxidant mechanism and has a negative effect on osteogenesis. The expression of this miRNA in zebrafish was induced to explore the in vivo effects of miR-182 on bone formation. Our findings suggest that miR-182 works as a FoxO1 inhibitor to antagonize osteoblast proliferation and differentiation, and consequently has negative effects on osteogenesis. This is the first report that informs the regulation of FoxO1 by miRNA in bone.

Materials and Methods

Cell culture and alkaline phosphatase staining

We cultured C3H10T1/2 mesenchymal stem cells and MC3T3-E1 preosteoblasts as described.19 Primary mouse calvaria cells were isolated by sequential collagenase digestion and cultured as described.20 For differentiation studies, C3H10T1/2 mesenchymal stem cells and MC3T3E1 preosteoblasts were seeded in 12-well plates and were fed at confluency with the medium containing the differentiation supplements of 10 mM of β-glycerol phosphate (Sigma Chemical Co., St. Louis, MO, USA) and 50 ug/mL of ascorbic acid (Sigma). The osteogenic medium was changed every 2 days. Alkaline phosphatase staining (Sigma) was performed as previously described.19

Transfection of miRNAs

We seeded cells in 12- or 24-well plates on the day before transfection. The mature type of mmu-miR-182 (miR-182; Dharmacon, Lafayette, CO, USA; 5′-UUUGGCAAUGGUAGAACUCACACCG-3′) and an antisense inhibitor (anti-miR-182; Dharmacon) that was designed to bind to endogenous miR-182 when introduced into cells and inhibit its activity, were transfected at a final concentration of 50 nM, using cationic liposomes (Lipofectamine 2000; Invitrogen, Carlsbad, CA, USA) according to the manufacturer's Lipofectamine protocol. Nonspecific control miR (miR-Control; Dharmacon) was used as a control for nonspecific effects.

Cell proliferation and viability assays

We cultured MC3T3-E1 preosteoblasts in 12- or 24-well plates and transfected at 60% to 70% confluency using Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer's instructions. After 48 hours, the number of viable cells was calculated by direct counting using a hemocytometer in the presence of 0.05% trypan blue. Cell survival was determined using a Cell Counting Kit-8 (Dojindo Molecular Technologies, Kumamoto, Japan) following the manufacturer's instructions.

Cell death assay and flow cytometry

To detect apoptotic osteoblasts, we transfected cells at 60% to 70% confluency with miRNAs using Lipofectamine 2000 reagent (Invitrogen) according to the manufacturer's instructions. Forty-eight hours after transfection, cells were stained with 4,6-diamidino-2-phenylindole (DAPI) or propidium iodide (PI) (Sigma; 0.05 mg/mL) in phosphate buffered saline (PBS; Life Technologies, Gaithersburg, MD, USA) for 15 minutes at room temperature and visualized using a Leica DMR microscope (Bensheim, Germany). For flow cytometry, cells were prepared as described.19 The percentage of apoptotic cells was assessed using a FACS Calibur (Becton Dickinson, San Jose, CA, USA). Apoptotic cells were identified on the basis of fragmented DNA stained with PI (DNA content b2 N).

Real-time quantitative PCR for miRNA

We extracted total RNA using the mirVana microRNA extraction kit (Ambion, Austin, TX, USA). Quantitative real-time PCR (qRT-PCR) analysis of mir-182 was performed by using miRNA-specific TaqMan MicroRNA Assay Kit (Applied Biosystems, Foster City, CA, USA); 12.5 ng of total RNA was reverse transcribed using the corresponding RT primer and the TaqMan MicroRNA Reverse Transcription Kit (Applied Biosystems). PCR was performed on 1.33 µL of RT products in a reaction mixture containing TaqMan PCR probes and the TaqMan Universal PCR Master Mix (Applied Biosystems). U6 small RNAs were used to normalize input RNA/cDNA levels.

RT-PCR

We extracted total RNA using a standard protocol with Trizol reagent and then reverse-transcribed into cDNA using 2U Moloney murine leukemia virus (MMLV) reverse transcriptase (Promega, WI, USA). One microliter (1 µL) of the RT product was used as a template for PCR amplification of Runx2, type I collagen (COL1), alkaline phosphatase (ALP), osterix and β-actin using the following cycles with 10 pmol primers and Taq DNA polymerase (Promega): Runx2, 95°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds for 27 cycles; type I collagen, 95°C for 30 seconds, 56°C for 30 seconds, and 72°C for 30 seconds for 34 cycles; ALP, 95°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds for 34 cycles; osterix, 95°C for 30 seconds, 56°C for 30 seconds, and 72°C for 30 seconds for 34 cycles; β-actin, 95°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds for 27 cycles. Primer pairs were as follows: Runx2, Forward primer(FP) 5′-CCGCACGA-CAACCGCACCAT-3′ and reverse primer (RP) 5′-CGCTCCGGCCCACAAATCTC-3′; Type I collagen, FP 5′- GAGGCATAAAGGGTCGTGG-3′ and RP 5′-CATTAGGCGCAGGAAGGTCAGC-3′; ALP, FP 5′-GGGACTGGTACTCGGATAACG-3′ and RP 5′-CTGATATGCGATGTCCTTGCA-3′; β-actin, FP 5′-TTCAACACCCCAGCCATGT-3′ and RP 5′-TGTGGTACGACCAGAGGCATAC-3′; osterix FP 5′-CACATCCCTGGTGCGGCAA-3′ and RP 5′-CCGGGTGTGAGTGCGCACAT-3′. PCR products were separated by electrophoresis on a 1.5% agarose gel and visualized by ethidium bromide staining.

Western blot analysis

We performed Western blot analysis according to a standard protocol.19 The following antibodies were used for Western blot: p–protein kinase B (p-Akt), Akt, p–extracellular signal-regulated kinase 1/2 (p-ERK1/2), ERK, cleaved caspase-3, and FoxO1 (all from Santa Cruz Biotechnology, Santa Cruz, CA, USA). β-actin was purchased from Sigma.

Plasmid constructs and overexpression of plasmid

We obtained the expression vectors of FoxO1 in pcDNA3.1 (Invitrogen), including FoxO1 wild type, as a gift from Prof. EJ Lee (Endocrinology, Yonsei University College of Medicine, Seoul, Korea).21 pcDNA 3.1 was used as a control. Cells were transfected at 60% to 70% confluency using Lipofectamine 2000 reagent according to the manufacturer's instructions. Forty-eight hours after transfection, cells were processed for cell viability or cell death assay using the methods described above.

Luciferase constructs and reporter assay

We obtained the psiCHECK2 dual luciferase reporter (Promega) as a gift from Prof. Park (Seoul National University, Seoul, Korea). Wild type (WT) and mutant (Mu) constructs, four sequences in miR-182 binding lesion were replaced, with the upper section of the FoxO1 3′UTR (143–323) including the predicted and also already proven binding site of miR-182 (as described in Fig. 5A),22 were cloned into the psiCHECK2 vector downstream of the Firefly Luciferase gene using the following gene-specific primer pairs: WT FP 5′-CTCGAGTTTCCTCAGACTTGGCAACAGCGGCAGCAC-3′; WT RP 5′-GCGGCCGCACTTGAAAGCACTCCAGGATCTGAAAGTC-3′; Mu FP 5′-GCCATTGGGAATCATTACAATGAAGTGACCATCACTACACC-3′; Mu RP 5′-GGTGTAGTGAGTGGTTCACTTCATTGTAATGATTCCCAATGGC-3′. Cells were seeded into 24-well plates and cotransfected with 0.1 µg of psiCHECK2 parental (without the FoxO1 3′UTR), psiCHECK2-WT FoxO1 3′UTR, or psiCHECK2-Mu FoxO1 3′UTR vector and 50 pmol/L of miR-Control or miR-182 using Lipofectamine 2000. Cells were serum-starved for 4 hours and then added back the full media. Cells were harvested 48 hours posttransfection by treatment with trypsin, lysed in 1× Passive Lysis buffer (Promega), and centrifuged at 13,000 rpm (16,200g) for 10 minutes. The supernatant was assayed for firefly and Renilla luciferase activities following the Promega protocol using Dual-Glo Luciferase Assay system. Values were reported as firefly luciferase activity relative to Renilla luciferase activity.

Zebrafish experiments

We raised zebrafish under standard conditions at 28.5°C with a light-dark cycle of 14/10 hours. Embryos were obtained from crosses between AB wild-type adults and raised in embryonic water with methylene blue added to inhibit fungal growth. We injected one- or two-cell stage zebrafish embryos with approximately 1.5 µL of miRNA following a published protocol23; 25 µM mature dre-miR-182 or dre-miR-Control and 49 zebrafish embryos were finally microinjected, 23 as negative controls and 26 with miR-182. Microinjection was carried out under a dissection microscope (Leica, CA, Switzerland) using a WPI microinjector and picopump controller. After 5.2 days of embryogenesis, embryo and larvae were fixed in 4% paraformaldehyde in PBS, stained with alizarin red, and cleared as described. These experiments were repeated three times. Sequences for dre-miR-182 published in a previous report were synthesized by Bioneer Co., Ltd (Seoul, Korea): dre-miR-182, 5′-UUUGGCAAUGGUAGAACUCACA-3′; dre-miR-Control, 5′-UCACAACCUCCUAGAAAGAGUAGA-3′.24 To test whether mir-182 negatively regulates the expression of FoxO1 in zebrafish model, we amplified a full-length zebrafish foxo1a gene from the cDNA library and cloned it into the pGEM T-easy vector (Promega). To fuse the green fluorescent protein gene (GFP) with the foxo1a gene, Sma I and Nco I restriction enzymes were used to isolate the 5′-region (708 bp) of the foxo1a gene including the target region of miR-182 from the pGEM T-easy vector, and then we inserted the foxo1a fragments into the pCS2+ EGFP vector between the Sma I and Nco I sites, making the pCS2+ foxo1a-GFP construct. We injected the pCS2+ foxo1a-GFP DNA in one-cell–stage embryos and then injected these with a 1-nL dose of 25 µM dre-miR-182.

Statistical analysis

All data are presented as means ± SDs. The significance of differences between groups were analyzed using Student's t test (two-sided) or one-way analysis of variance for experiments with more than two subgroups at a significance level of p < 0.05. Statistical analyses were carried out using PASW statistics software (version 18; SPSS Inc., Chicago, IL, USA). All experiments were repeated at least three times and representative experiments are shown.

Results

miR-182 is expressed in bone and MC3T3E1 preosteoblasts and its expression increases with osteoblast differentiation

We first selected several miRNAs that were reported to be expressed in osteoblasts and also reported to be changed in relation to osteoblast differentiation in previous reports.15, 16, 25, 26 Next, we selected a few candidate miRNAs that were predicted to target important transcriptional factors in osteoblasts and as binding to 3′UTRs of conserved lesions of those genes based on bioinformatic searches. Finally, five candidate miRNAs were selected and their effects on proliferation and differentiation of osteoblasts were analyzed. Among these miRNAs, miR-182 showed a strong negative effect on osteogenic differentiation in ALP staining similar to that of miR-206, a known negative regulator of osteoblastogenesis (Fig. 1A).16 We examined the tissue distribution of miR-182 in normal adult mouse tissue. We detected high expression of miR-182 in the lung, calvaria, and fat, but only low-level expression of miR-182 in long bone (Fig. 1B). We also determined the expression pattern of miR-182 during osteoblast differentiation. C3H10T1/2 mesenchymal stem cells and MC3T3E1 preosteoblasts were cultured in osteogenic medium for 8 days and the expression level of miR-182 was measured at different time points during differentiation. The expression of miR-182 increased with osteoblast differentiation in MC3T3E1 preosteoblasts starting 2 days after induction (Fig. 1C). In C3H10T1/2 mesenchymal stem cells, expression of miR-182 increased after 4days and finally decreased at day 8 (Fig. 1D).

Figure 1.

miR-182 expression in tissues and during osteoblast differentiation. (A) Alkaline phosphatase (ALP) staining of MC3T3E1 preosteoblasts after transfection of miR-182 with two published miRNAs. (B) Quantitative RT-PCR analysis of miR-182 in mouse tissue. Total RNA was isolated from tissue. (C, D) Quantitative RT-PCR analysis of miR-182 expression during osteoblast differentiation. C3H10T1/2 mesenchymal stem cells and MC3T3E1 preosteoblasts were cultured in osteogenic differentiation medium for 8 days, and then total RNA from cells was isolated from the cells at the day indicated. The expression levels of miR-182 in tissues and cells were examined by real-time PCR using the TaqMan microRNA assay. RNU6 RNA was used as a loading control. The relative ratios are shown; the value for the liver and day 0 are designated as 1 in each of the corresponding experiments. *p < 0.05 compared to the 0 time point by Student's t test.

Effect of miR-182 on proliferation of C3H10T1/2 mesenchymal stem cells, MC3T3E1 preosteoblasts, and primary calvaria cells

To investigate the effects of miR-182 on the proliferation of C3H10T1/2 mesenchymal stem cells, MC3T3E1 preosteoblasts and primary calvaria cells, we measured the level of proliferation of these cells after transfection with miR-182 using the cholecystokinin (CCK) assay. As shown in Fig. 2A, levels of proliferation were significantly decreased in miR-182–transfected cells compared with miR-Control–transfected cells, and these effects were commonly observed in C3H10T1/2 mesenchymal stem cells, MC3T3E1 preosteoblasts, and primary calvaria cells. Transfection of all three cell types with miR-182 also decreased the number of viable cells compared to the control (Fig. 2B). To confirm that miR-182 was the cause of the negative proliferative effects observed, antisense inhibitor for miR-182 was cotransfected with miR-Control or miR-182 in these three cell lines and the antisense inhibitor of miR-182 was able to reverse the antiproliferative effect of miR-182 (Fig. 2C).

Figure 2.

miR-182 inhibits cell viability. (A, B) Level of proliferation and cell numbers after transfection of miR-182. *p < 0.05, **p < 0.01, ***p < 0.001 compared to miR-Control–transfected cells. (C) Rescue of the antiproliferative effects of miR-182 by transfecting cells with an antisense inhibitor of miR-182 (anti-miR-182). *p < 0.05, **p < 0.01, ***p < 0.001 compared to the counterparts shown.

Effect of miR-182 on the apoptosis of C3H10T1/2 mesenchymal stem cells, MC3T3E1 preosteoblasts, and primary calvaria cells

We also investigated whether miR-182 affected cell apoptosis. The reduced proliferation of miR-182 transfected cells was strongly correlated with an increased proportion of apoptotic cells. As shown in Fig. 3A, lactate dehydrogenase (LDH) release was dramatically increased by over twofold at 48 hours after transfection with miR-182 compared to transfection with miR-Control. Cell viability assessed by PI staining also showed significant increases of cell death in miR-182-transfected cells compared with those of miR-Control–transfected cells in both cell lines (Fig. 3B). To confirm the negative role of miR-182 on cell survival and identify the signaling molecules involved in the miR-182–induced enhancement of osteoblast apoptosis, we assessed the relative phosphorylation levels of Akt and ERK, which are known to be involved in the cell survival pathway. MC3T3E1 cells were cultured after transfection with control miRNA or miR-182. Western blot analysis showed remarkably decreased phosphorylation of AKT in miR-182–overexpressing MC3T3E1 cells, whereas the levels of pERK1/2 were not changed compared to miR-Control–transfected cells (Fig. 3C). We further studied whether the activation of caspase-3 was regulated by miR-182 in MC3T3E1 cells. Caspase-3 is primarily responsible for many of the signature events of apoptosis, including nuclear condensation, cell shrinkage, membrane blebbing, and DNA fragmentation.27, 28 As expected, the activity of caspase-3 was clearly higher in miR-182–transfected cells than control cells (Fig. 3C). Our results suggest that miR-182 enhances the osteoblast apoptosis related to AKT signaling, not to ERK signaling.

Figure 3.

miR-182 promotes cell apoptosis. (A) LDH assay of C3H10T1/2 mesenchymal stem cells, MC3T3E1 preosteoblasts and primary calvaria cells transfected with miR-Control or miR-182. *p < 0.05, **p < 0.01, ***p < 0.001 compared to miR-Control–transfected cells. (B) Detection of apoptosis in C3H10T1/2 mesenchymal stem cells and MC3T3E1 preosteoblasts transfected with miR-Control or miR-182 using propidium iodide (PI) staining and FACS analysis. Nuclei were stained with DAPI. The graph shows the percentage of PI-stained cells, indicating cells with condensed or fragmented nuclei. Values in A are representative of six independent experiments and values in B are representative of four independent experiments. (C) Molecules from different survival pathways were analyzed in cells transfected with miR-Control or miR-182 by Western blot 60 hours after transfection. Protein content was normalized by β-actin. The data shown are representative of three similar experiments.

Effect of miR-182 on the differentiation of C3H10T1/2 mesenchymal stem cells and MC3T3E1 preosteoblasts

We next performed experiments to determine the impact of miR-182 on osteoblast differentiation. miR-182–transfected C3H10T1/2 mesenchymal stem cells and MC3T3E1 preosteoblasts were induced to differentiate to an osteogenic lineage in osteogenic medium containing ascorbic acid and β-glycerophosphate. After 7 days of treatment, cells were subjected to alkaline phosphatase (ALP) staining. ALP staining of miR-182–transfected cells clearly showed downregulation of osteogenic differentiation (Fig. 4A). These results suggest that miR-182 acts as an osteogenesis inhibitor in mesenchymal cells and preosteoblasts. To further confirm these findings, we performed semiquantitative RT-PCR assays to determine if osteoblast marker genes, such as Runx2, Osterix, Type I collagen, and ALP, are also altered in a manner consistent with the effects of miR-182 on ALP staining. Overexpression of miR-182 inhibited the expression of type I collagen, Osterix, and ALP, and slightly inhibited the expression of Runx2 in MC3T3E1 cells (Fig. 4B). Taken together, our results suggest that miR-182 inhibits osteoblast differentiation.

Figure 4.

miR-182 inhibits osteoblast differentiation. (A) Alkaline phosphatase (ALP) activity after transfection of cells with miR-182. C3H10T1/2 mesenchymal stem cells or MC3T3E1 preosteoblasts were transfected with miR-Control or miR-182 and treated with osteoblast differentiation medium (ascorbic acid and β-glycerophosphate). ALP staining was performed 7 days after treatment. (B) The effects of miR-182 on osteoblast differentiation markers. MC3T3E1 preosteoblasts were transfected with miR-Control or miR-182, and the mRNA expression levels of Runx2, Osterix, Type I collagen, and ALP were measured at 48 hours after transfection by semiquantitative RT-PCR. Relative levels of mRNA for the indicated genes in the transfected cells were normalized to β-actin.

FoxO1 is a target of miR-182 in C3H10T1/2 mesenchymal stem cells and MC3T3E1 preosteoblasts

To understand the molecular mechanism that underlies the effects of miR-182 on C3H10T1/2 and MC3T3E1 cell viability and differentiation, we searched for potential targets of miR-182 using the miRNA target prediction algorithms, Targetscan, PicTar and miRBase.29 Interestingly, among the predicted target, we identified FoxO1, a transcriptional factor having crucial roles in osteogenesis-related proliferation and differentiation, as a strong candidate (Fig. 5A). The overexpression of miR-182 resulted in downregulation of FoxO1 in C3H10T1/2 cells and MC3T3E1 cells based on Western blot analysis (Fig. 5B). We next examined whether forced expression of FoxO1 eliminated the antiproliferative and proapoptotic effects of miR-182 in C3H10T1/2 cells, MC3T3E1 cells, and primary calvaria cells. Overexpression of miR-182 in these three cell lines decreased proliferation levels significantly and increased LDH release, but this inhibition was rescued by FoxO1 overexpression, even in the presence of miR-182 (Fig. 5C, D). Also in osteoblast differentiation, overexpression of FoxO1 showed a recovery effect to the diminished differentiation induced by miR-182 (Fig. 5E). Furthermore, we used a luciferase reporter assay to show that miR-182 directly interferes with FoxO1 expression. Transfection of miR-182–infected MC3T3E1 cells with a parental luciferase construct (without the FoxO1 3′UTR) or mutant FoxO1 3′UTR did not significantly alter the expression of the reporter. However, transfection of a luciferase construct in which wild type FoxO1 3′UTR was inserted showed significantly lower luciferase activity in miR-182–infected cells than in miR-Control–infected cells (Fig. 5F). The expression of FoxO1 was increased during osteoblast differentiation.30 miR-182, a negative regulator of FoxO1, was also increased along with osteogenesis in the present study and we speculated that it might be a feedback response to prevent further increases in expression of the target gene.

Figure 5.

miR-182 targets the FoxO1 3′UTR. (A) Schematic of the miR-182 putative target site in the mouse FoxO1 3′ untranslated region (UTR) and alignment of miR-182 with the FoxO1 3′UTR showing complementary pairing. *sequences changed to create a mutant FoxO1 3′UTR. (B) FoxO1 expression in C3H10T1/2 mesenchymal stem cells and MC3T3E1 preosteoblasts after transfection with miR-182. Western blots were analyzed using β-actin expression as a loading control. (C, D) Biological rescue of the effects of miR-182 on proliferation (C) and apoptosis (D) by overexpressing FoxO1 in C3H10T1/2 mesenchymal stem cells, MC3T3E1 preosteoblasts and primary calvaria cells. miR-182 or miR-Control transfected cells were also transfected with FoxO1 plasmid or mock plasmid. The values of CCK and LDH were assessed 48 hours after transfection. (E) Alkaline phosphatase activity after transfection of cells with miR-182 and/or FoxO1. MC3T3E1 preosteoblasts transfected with miR-182 were co-transfected with FoxO1 and treated with osteoblast differentiation medium (ascorbic acid and β-glycerophosphate). Alkaline phosphatase (ALP) staining was performed 7 days after treatment. (F) The luciferase reporter assay for the FoxO1 3′UTR in the presence of miR-182. Cells were co-transfected with miR-Control or miR-182 and parental (without FoxO1 3′UTR), wild type FoxO1 3′UTR, or mutant FoxO1 3′UTR luciferase constructs, and luciferase activities was measured. Values are reported as firefly luciferase activity to Renilla luciferase activity. *p < 0.05, **p < 0.01, ***p < 0.001 compared to cells transfected with miR-Control and the parental luciferase construct.

In vivo effect of miR-182 in bone in a zebrafish model

We investigated the effect of manipulating miR-182 levels on bone formation in vivo in zebrafish using miR-Control and miR-182. In zebrafish embryos, miR-Control and miR-182 were overexpressed by the injection of synthetic double-stranded miRNAs. The skeletal developments of the zebrafish embryos were analyzed by alizarin red staining. A total of 23 embryos were injected with miR-Control and 26 embryos with miR-182. In miR-182–injected embryos, three broad classes of phenotype were found (Fig. 6). Class I embryos (Fig. 6B) showed similar developmental morphology with miR-Control–injected embryos (Fig. 6A). Class II embryos exhibited little defect in development with shortened body lengths (Fig. 6C). Class III embryos had severe malformations such as an enlarged heart chamber and abnormal eyes, as shown in another reported experiment (Fig. 6D).24 In alizarin red staining of whole zebrafish embryos injected with control-miR (Fig. 6A') or miR-182 (Fig. 6B'–D'), the results also showed little different effects in each class. Class I embryos showed subtle decreases in bone formation, but not much differed with miR-Control injected embryos. However, Class II embryos exhibited significantly diminished bone formation, especially in the notochord lesion, compared to miR-Control–injected embryos (Fig. 6C') Moreover, in Class III embryos, which showed malformed embryos, skeletal development almost failed (Fig. 6D'). Figure 6E shows the percentage of embryos in each class type. To directly test whether the negative effect of miR-182 on osteogenesis in zebrafish is due to direct silencing of FoxO1, embryos were injected with both miRNAs and pCS2+ foxo1a-GFP. Embryos injected with both miR-182 and pCS2+ foxo1a-GFP lost GFP fluorescence, compared to controls (Fig. 6F). With the zebrafish model, we could also observe the negative role of miR-182 for skeletal development in vivo and confirm the inhibitory role of miR-182 in skeletogenesis.

Figure 6.

Effects of miR-182 on in vivo skeletogenesis using a zebrafish model. (A-D) Morphological phenotype of zebrafish embryos injected with dre-miR-Control (A) or dre-miR-182 (B-D). (A'-D') Alizarin Red staining of whole zebrafish embryos injected with dre-miR-Control (A') or dre-miR-182 (B'-D'). Embryos at one- or two-cell stages were microinjected with dre-miR-Control (A, A') or dre-miR-182 (B-D, B'-D'). Black arrow indicates diminished development of the notochord (C') and malformed embryos that almost failed the skeletal development (D'). (E) The percentage of embryos in each class type for each injection experiment. (F) Embryos injected with foxo1a-GFP alone (a,a′) or with foxo1a-GFP and miR-182 (b,b′). At the six-somite stage, embryos injected with foxo1a-GFP and miR-182 showed reduced GFP expression compared to foxo1a-GFP injected embryos. The figures shown are representative of three similar experiments. Nc = notochord; ps = parasphenoid; ot = otolith.

Discussion

miRNAs have recently emerged as important regulators of bone homeostasis as they have been shown to control the proliferation and differentiation of osteoblast lineage cells. miR-2861 promotes osteoblast differentiation by repressing HDAC5 expression.15 miR-206 inhibits osteoblast differentiation by suppressing the expression of connexin 43.16 miR-29a was identified as a negative regulator of the key Wnt antagonists, Dkk1, Kremen2, and sFRP2.17 miR-196a,31 miR-208,32 miR-29b,26 miR-20418 and miR-133,13525 play a role in osteoblast differentiation by targeting the various important transcriptional factors in bone. Moreover, genetic variations in miRNA and miRNA target sites have also been reported to be strongly associated with bone loss.14, 15 These observations suggest that miRNAs play crucial regulatory roles in bone and may be therapeutic targets for the treatment of various bone disease in humans.

In a previous study, several miRNAs were shown to be expressed in primary cultured calvaria cells.15 Bone morphogenic protein 2 (BMP-2) strongly affected the expression of several miRNAs, either positively or negatively in bone.16, 25 Based on these previous reports, we selected miR-182 as a potential regulator of osteoblast differentiation along with several other miRNAs. In our screening study, miR-182 revealed strong inhibitory effects both on the viability and osteogenic differentiation of MC3T3E1 preosteoblasts with a similar inhibitory potency as the positive control miR-206.15, 16 miR-182 was expressed in calvaria as well as white adipose tissue and the heart. Like many other regulatory miRNAs in bone, expression of miR-182 increased gradually following osteoblast differentiation.25, 26 It increased significantly at day 4 of differentiation, but slightly decreased at day 8 in C3H10T1/2 cells (Fig. 1C). In MC3T3E1 preosteoblasts, miR-182 expression was observed at day 2 after treatment of cells with osteogenic induction media, after which levels of the miRNA increased gradually (Fig. 1D). Based on the relatively high expression of miR-182 in bone tissue and the gradual increase in the expression of this miRNA according to osteoblast differentiation, we hypothesized that miR-182 plays an important role in the proliferation and differentiation of osteoblasts as well as skeletogenesis.

Overexpression of miR-182 in C3H10T1/2 mesenchymal stem cells, MC3T3E1 preosteoblasts, and primary calvaria cells resulted in a significant decrease in cell viability and an increase in cell apoptosis (Fig. 2A, B and Fig. 3A, B). Phosphorylation of pAKT was reduced and caspase 3 activity was enhanced 60 hours after overexpression of miR-182 (Fig. 3C). Akt1 is an established regulator of osteoblast survival; Akt1 deficiency or reduction of the phosphorylation of pAKT therefore enhances the susceptibility of osteoblasts to apoptosis and cause a subsequent decrease in bone mass and bone formation.33 However, introduction of anti-miR-182 was able to reverse the miR-182–promoted apoptosis in these three cells lines. Taken together, these results suggest a negative role of miR-182 in cell viability through inhibition of the PI3K/AKT pathways. miR-182 also affected osteoblast differentiation as well as had a significant negative impact on cell viability. ALP staining in both C3H10T1/2 cells and MC3T3E1 cells revealed that overexpression of miR-182 inhibited osteogenic differentiation. Semiquantitative RT-PCR revealed decreased expression of the early osteoblast markers, Runx2 and Osterix, 48 hours after treatment of cells with miR-182. Runx2 is also considered to be also an essential transcriptional factor for osteoblast differentiation. The other transcriptional factor, Osterix, is absolutely required for osteoblast commitment and osteoblast differentiation.34 However, we did not find any consensus target sequence for miR-182 in either the Runx2 or Osterix genes. Using an in vivo zebrafish model, we also confirmed that overexpression of miR-182 decreased osteogenic potential. All the above results indicate that miR-182 has negative effects on cell viability and osteogenic differentiation, thereby impairing osteogenesis.

The FoxO family is widely accepted to play an important role in protecting various cells from ROS, and members of this family are key antiapoptotic signals against oxidative stress stimuli.35 Among the FoxO family, FoxO1 is the main member of this family expressed in bone. It promotes osteoblast proliferation by maintaining protein synthesis and the redox balance, and also stimulates osteoblast differentiation.9, 30 An osteoblast-specific FoxO1 knockout mice model generated using α1(1) Collagen-Cre showed decreased bone mineral density (BMD), a decreased bone formation rate, and a decrease in viable osteoblast numbers.9 Another specific FoxO1 mouse model that was generated using Mx-Cre showed increased apoptosis and decreased osteogenic differentiation.8 Moreover, FoxO1 silencing in C3H10T1/2 cells suppressed osteogenic differentiation by repressing the expressions of Runx2 and Osterix, and impaired bone formation drastically with some increase in apoptosis.30 Although the roles and effects of FoxO1 differ somewhat depending on the stage of differentiation of development, it is clear that FoxO1 is an important positive regulator of bone formation. However, prior to this study, little was known about the regulators of FoxO1 in bone.

To study the molecular mechanism by which miR-182 regulates proliferation and the differentiation of osteoblasts, we searched for potential target genes with an established function in promoting osteogenesis by way of differentiation and proliferation using the programs, Target scan, Pictar, and miRBase. Among many candidate genes, we selected FoxO1 as a potential target gene of miR-182 based on two previous observations. First, FoxO1 inhibited apoptosis and promoted differentiation of osteoblast lineage cells, in direct contrast to the effects of miR-182 overexpression in osteo-lineage cells.8, 9, 30 The expression of Runx2 and Osterix has also been shown to be suppressed in FoxO1 knockout mice.8, 9 Second, miR-182 has been demonstrated to regulate FoxO1 in a breast cancer cell line.22 The putative binding sequence of miR-182 in FoxO1 is located at approximately 260 bp below the stop codon of the FoxO1 gene. Overexpression of miR-182 in C3H10T1/2 cells and MC3T3E1 cells resulted in obvious downregulation of FoxO1 at the protein level. Using FoxO1 3′UTR luciferase reporter assays, we confirmed that FoxO1 is a direct target of miR-182. Furthermore, overexpression of FoxO1 overcame the apoptosis-promoting, proliferation-inhibiting, and differentiation-inhibiting effects of miR-182. Taken together, all the above results suggest that miR-182, by targeting FoxO1, is a strong negative regulator of skeletogenesis. However, the possibility that miR-182 also targets other genes could not be completely ruled out, because miRNAs usually target several related genes for regulating specific pathways.

Among the many pathways implicated in aging, oxidative stress is one of the most intensely studied in relation to aging and age-associated human disease.36 Osteoporosis is a representative age-related disease, and oxidative stress is one of the critical components of age-related bone loss.6, 37 FoxOs have also been proposed to counteract the adverse effects of oxidative stress.35 FoxO1 prevents cell apoptosis by maintaining the redox balance in cells and promotes cell osteogenic differentiation. FoxO1 has therefore been suggested to be crucial for preserving bone homeostasis.9 A regulatory connection between miR-182 and FoxO1 has already been reported in breast cancer cells and immune cells.22, 38 However, no previous study has examined the regulation of FoxO1 in bone cells. In the present study, we identified miR-182 as a major regulator of FoxO1. miR-182 may therefore be a potential therapeutic target for the prevention and treatment of osteoporosis.

In conclusion, miR-182 acts as a negative regulator of osteogenesis by repressing the expression of FoxO1, and antisense targeting of miR-182 could be of therapeutic value in bone aging.

Disclosure

All the authors state that they have no conflicts of interest.

Acknowledgements

The Plasmid construct for FoxO1 was a kind gift of Eun Jig Lee. This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MEST) (No. 20110001024). This research was also supported by a faculty research grant from Yonsei University College of Medicine in 2010 (6-2010-0058).

Authors' roles: KMK, SJP, and S-KL designed research; KMK, SJP, EJK, JA, and GJ performed research; SHJ and C-HK performed zebrafish research; KMK and YR analyzed data; and KMK, and S-KL wrote the article.

Ancillary