MiR-145 regulates osteogenic differentiation of human adipose-derived mesenchymal stem cells through targeting FoxO1

Cell cultivation and transfection

hASCs HMSC-ad cell line (Sciencell, USA) was cultured in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (Gibco), 100 units/ml penicillin, and 100 μg/ml streptomycin (Gibco) in 5% CO₂ in a humidified incubator at 37°C. pcDNA3.1 plasmids were used to construct FoxO1 overexpression plasmids. The synthesis of miR-145 mimics, control mimics, pcDNA3.1-FoxO1 and pcDNA3.1-control plasmids was purchased from Invitrogen (Carlsbad, CA, USA). The synthesis of FoxO1 siRNA was conducted by Shanghai Sangon Company. The synthetics were transfected using Lipofectamine, 2000 (Invitrogen, Carlsbad, CA, USA).

Osteoblast differentiation induction

To induce osteoblast differentiation, HMSC-ad cells were plated in six-well plates (1 × 10⁵ cells/well, 2 ml medium per well) and cultured normally. After five days cultivation (to 80%∼90% confluence), the media were replaced with osteoblasts-specific induction medium, the low glucose of Dulbecco's Modified Eagle's Medium with 10% FBS, 10 mM β-glycerophosphate, 0.1 uM dexamethasone, and 0.2 mM antiscorbic acid (Sigma–Aldrich, Louis, MO, USA) to induce osteoblast differentiation. Von Kossa staining was conducted after two weeks.

RT-qPCR

The total RNA was extracted from hASCs osteodifferentiated induction cells with TRIzol reagent (Life Technologies corporation, Gaithersburg, MD, USA) and reverse-transcribed into cDNA using PrimeScript II 1st Strand cDNA Synthesis Kit (TaKaRa, Tokyo, Japan). RT-qPCR was performed using SYBR Premix Ex Taq (TaKaRa, Tokyo, Japan). U6 was used as an endogenous normalization control for miR-145, while GAPDH was used as a FoxO1 mRNA control. The relative level of RNA was quantified using the 2^−ΔΔCt method. The primers were synthesized by Shanghai Sangon Company and are listed in Table 1.

OPEN IN VIEWER

Table 1.

RT-qPCR primer sequence.

Name	Primers
MiR-22	F: 5′-GCAGTTCTTCAGTGGCAA-3′
MiR-22	R: 5′-GAATACCTCGGACCCTGC-3′
U6	F: 5′-CTCGCTTCGGCAGCACA-3′
U6	R: 5′-AACGCTTCACGAATTTGCGT-3′
FoxO1	F: 5′-TTAGCCAGTCCAACTCGG-3′
FoxO1	R: 5′-TCTTGACCATCCACTCGTA-3′
GAPDH	F: 5′-TGGTCACCAGGGCTGCTT-3′
GAPDH	R: 5′-AGCTTCCCGTTCTCAGCC-3′

Western blot

Quantification of total protein was performed using Bradford protein assay (Bio-Rad, Hercules, CA, USA). Proteins were resolved by SDS-PAGE and transferred to a PVDF membrane using the iBlot Dry Blotting Transfer System (Life Technologies corporation, Gaithersburg, MD, USA). After blocking in PBS with 5% milk and 0.1% Tween for 2 h, the membrane was incubated with primary antibodies for goat anti-FoxO1, anti-Ocn, anti-BSP, anti-OPN (1:1000 dilution) (Santa Cruz Biotechnology, Santa Cruz, CA, USA) or goat anti-GAPDH polyclonal antibody (1:1000) (GenScript, Nanjing, China) at 4°C overnight and washed and incubated with secondary antibodies horseradish peroxidase labeled anti-goat (Abcam, Cambridge, MA, USA) for 1–2 h. After incubation, PVDF membrane was washed with PBS before placing in a darkroom. Moderate amount of DAB substrate was applied and the membranes were put in the darkroom for exposure. Images were obtained using a chemiluminescent image analyzer (LAS-4000 mini; Fujifilm) and cumulative optical density value of each target band was semi-quantified by ImageJ software (NIH). GAPDH was used as a loading control to quantify the relative protein level.

Alkaline phosphatase staining and alizarin red S staining

Seven or fourteen days after osteogenic induction, DMEM media were discarded. Cells were washed three times with PBS and fixed with 4% paraformaldehyde. The cell layer was subsequently washed and dried in air. Reaction solution was prepared using an alkaline phosphatase (ALP) Assay Kit (Beyotime Biotechnology, Shanghai, China). After reaction at 37°C for 1 h, purplish red granules formed were detected under an optical microscope; 7 or 14 days post osteogenic differentiation, cells were fixed with ethanol for 10 min. Lastly, the cells were treated with 0.1% alizarin red S (ARS)-Tris-Hcl (pH 8.3) at 37°C for 1 h. To quantify calcium deposition, cells were washed with distilled water and dried in air. Absorbance was measured under an optical microscope.

Dual-luciferase reporter assay

The FoxO1 3′UTR fragment containing the predicted potential miR-145-binding sites was amplified and pGL3-WT plasmids were constructed by PCR method. pGL3-Mut was generated by site-directed mutagenesis, replacing the first six ribonucleotides of the miR-145 complementary sequence. HEK293T cells were grown in 24-well plates and cotransfected with miR-145 mimics and pGL3-WT (miR-WT group), miR-145 mimics and pGL3-Mut (MiR-Mut group), control mimics and pGL3-WT (NC-WT group), control mimics and pGL3-MuT (NC-Mut group). The media were replaced after 6 h; 24 h after cotransfection, the Dual-Luciferase Reporter Gene Test Kit (Beyotime Biotechnology, Shanghai, China) was used to detect the luciferase activities in different groups. The firefly and the renal luciferase reagent were added to detect the luciferase activity of each group.

Statistical analysis

Affymetrix Data Mining Tool version 3.1 (DMT 3.1) and scripts called within R environment (http://www.r-project.org/) were utilized to screen differentially expressed mRNAs and miRNAs. The filtration criteria were: Log2 (Fold Change) >2 and P < 0.001. MiRNA analysis was conducted using GSE72429 and GPL16770. MRNA analysis was conducted using GSE63754 and GPL17077 platform. Three osteogenic differentiation hASCs and three normal hASCs without osteogenic differentiation were selected to screen out differentially expressed miRNAs and mRNAs, respectively. The results represented the mean of three independent experiments and data were presented as mean ± standard deviation (x ± s). Comparisons between two groups were analyzed by Student’s t test. For the testing among multiple groups, a one-way analysis of variance (ANOVA) was conducted. Statistical analysis was performed using GraphPad Prism 6.0 software. P-values < 0.05 were considered to be statistically significant.

FoxO1 was highly expressed after hASCs osteogenic differentiation

Three osteogenic HMSC-ad cell lines and three normal HMSC-ad without osteogenic differentiation cell lines were selected to screen out differentially expressed mRNAs. Among the 16 upregulated mRNAs with significant differences in osteoinductive HMSC-ad cell lines, there was FoxO1. The expression of FoxO1 in osteoinductive HMSC-ad cell lines was 2.29-fold higher in osteoinductive differentiation cell lines (Figure 1(a) and (b)). The expression of FoxO1 was gradually upregulated with the progression of osteogenic differentiation (Figure 1(c)). The protein expression of FoxO1 increased in hASCs after osteoinductive differentiation. Meanwhile, overexpression of FoxO1 (FoxO1 cDNA) and silence expression of FoxO1 (FoxO1 siRNA) was successfully constructed (Figure 1(d)).

OPEN IN VIEWER

Figure 1.

The expression of FoxO1 was up-regulated during hASCs osteogenesis. (a) Heat map revealed the FoxO1 was higher expressed after hASCs osteogenic differentiation. (b) Volcano plot showed the FoxO1 was highly expressed after hASCs osteogenic differentiation. (c) FoxO1 expression was up-regulated gradually along with the osteoinductive differentiation. (d) FoxO1 expression increased after osteoinductive differentiation in HMSC-ad cells. *P < 0.05, **P < 0.01, compared with day 0. (A color version of this figure is available in the online journal.)

FoxO1 promoted hASCs osteogenic differentiation

The result of RT-qPCR and Western blot was consistent. The expression of BSP, Ocn, and OPN in silence expression of FoxO1 group was downregulated since the fourth day and was more significant with the progression of osteogenic differentiation (P < 0.01). A remarkably opposite trend was observed in overexpression of FoxO1 group (P < 0.001) (Figure 2(a) and (b)).

OPEN IN VIEWER

Figure 2.

FoxO1 promoted hASCs osteogenesis. (a–b) RT-PCR and Western blot results showed that BSP, Ocn, and OPN mRNA expression in FoxO1 siRNA group were much lower than in NC group after osteoinductive differentiation, whereas those in FoxO1 cDNA group were much higher than in NC group after osteoinductive differentiation, especially four days after the differentiation induction. *P < 0.05, **P < 0.01, ***P < 0.001 compared with NC group. (c and d). ALP and ARS staining results showed that HMSC-ad osteogenesis differentiation after osteoinduction was much stronger in FoxO1 cDNA group yet much weaker in FoxO1 siRNA group at either time point (day 7 and day 14). *P < 0.05 compared with NC group. (A color version of this figure is available in the online journal.)

ALP staining and ARS staining showed that on days 7 and 14, ALP activity and calcification were lowered in FoxO1 siRNA group, but enhanced in FoxO1 cDNA group (P < 0.05) (Figures 2(c) and 3(d)). Collectively, these results revealed that FoxO1 could promote hASCs osteogenic differentiation.

MiR-145 was low expressed during hASCs osteogenic differentiation

Three osteogenic HMSC-ad cell lines and three normal HMSC-ad without osteogenic differentiation cell lines were selected to screen out differentially expressed miRNAs. Microarray analysis showed that miR-145 was low expressed in all three HMSC-ad osteogenic differentiation cell lines (Figure 3(a)). The expression of miR-145 in HMSC-ad was 2.09-fold lower after osteoinductive differentiation. MiR-145 expression was downregulated since the fourth day and was more significant along with the progression of osteogenic differentiation. Significant statistical difference was detected on day 14 (Figure 3(c)). MiR-145 and FoxO1 mRNA expression was detected in six cell lines on day 7 and 14 after osteo-differentiation induction, and the correlation analysis results showed that there was a significantly negative correlation between them (P < 0.01) (Figure 3(d)).

OPEN IN VIEWER

Figure 3.

MiR-145 expression was down-regulated during hASCs osteogenesis. (a) Heat map revealed the miR-145 was lower expressed after hASCs osteogenic differentiation. (b) Volcano plot showed the miR-145 was lower expressed after hASCs osteogenic differentiation. (c) MiR-145 expression was downregulated gradually along with osteoinductive differentiation. (d) There is a significantly negative correlation between miR-145 and FoxO1 expression. *P < 0.05, **P < 0.01 compared with day 0. (A color version of this figure is available in the online journal.)

MiR-145 targeted the 3′UTR region of FoxO1

According to the database TargetScan, the binding site of miR-145 was located between bases 770 and 776 of the FoxO1 3′-UTR, suggesting that FoxO1 might be the target of miR-145 (Figure 4(a)). The luciferase reporter gene assay was conducted in 293T cells to verify their binding relationship. The result revealed that the luciferase activity of cells cotransfected with miR-145 mimics and pGL3-WT was significantly lower, whereas miR-145 mimics had no such effect on the fluorescence of the Mut recombinant plasmid (Figure 4(b)). Taken together, these results revealed that miR-145 directly targeted FoxO1 3′UTR.

OPEN IN VIEWER

Figure 4.

MiR-145 targeted FoxO1 3′UTR. (a) The binding sites of miR-145 and FoxO1 as well as the putative binding relationship were illustrated. (b) Dual luciferase reporter gene assay results demonstrated that the luciferase activity in miR-145 mimics + wild type group was significantly weaker than in NC + wild type group. **P < 0.01 compared with NC group. (A color version of this figure is available in the online journal.)

MiR-145 prevented hASCs osteogenic differentiation via suppressing FoxO1 expression

The protein expression of the cells after osteoinductive differentiation was detected on day 7 and 14. The protein expression of BSP, Ocn and OPN showed identical tendencies on day 7 and 14. In FoxO1 cDNA cells, BSP, Ocn, and OPN levels were high, whereas in miR-145 mimics cells, their levels were lower than in NC cells (Figure 5(a)). APL activity and ARS activity (calcification) were optimal in FoxO1 cDNA group, while the worst in miR-145 mimics group (P < 0.05) (Figure 5(b) and (c)). The above results demonstrated that miR-145 inhibited expression of osteogenic differentiation protein markers BSP, Ocn, and OPN via suppressing FoxO1, preventing hASCs osteogenic differentiation.

OPEN IN VIEWER

Figure 5.

MiR-145 inhibited FoxO1 to suppress hASCs osteogenesis. (a) FoxO1, BSP, Ocn, and OPN expression in miR-145 mimics group was much lower, whereas those in FoxO1 cDNA group was much higher than in NC group. There was no significant difference between the NC group and the (b–c) ALP and ARS staining results demonstrated that miR-145 mimics significantly reduced HMSC-ad osteogenesis differentiation after osteoinduction and FoxO1 cDNA significantly increased HMSC-ad osteogenesis differentiation. The simultaneous transfection of FoxO1 cDNA and miR-145 mimics showed no significant difference from the NC group. *P < 0.05, compared with NC group; #P < 0.05, compared with miR-145 mimics group. (A color version of this figure is available in the online journal.)