MC3T3-E1 cells were purchased from the Shanghai Cell Bank of Type Culture Collection of Chinese Academy of Sciences (Shanghai, China). Cells were cultured in 25 cm² flasks in α-modified Eagle's medium (α-MEM; Wisent Inc., St. Bruno, QC, Canada) supplemented with 10% fetal bovine serum (FBS, Gibco; Thermo Fisher Scientific, Inc., Waltham, MA, USA), 100 U/ml penicillin and 100 µg/ml streptomycin, maintained at 37°C in a humidified 5% CO₂ incubator (Sanyo Electric Co., Ltd., Moriguchi, Japan).

Briefly, 5×10³ cells/well were seeded in 96-well culture plates (Corning Incorporated, Corning, NY, USA). The cell culture medium was discarded after 24 h, and 180 µl different final concentrations (0.01, 0.1 and 1 mg/ml) puerarin were added to the experimental wells, while an equal volume of serum-free medium was added to the control wells. Following culturing for 24, 48 and 72 h, the proliferation of osteoblasts was measured by adding 20 µl MTT (5 mg/ml, Amresco LLC, Solon, OH, USA) in each well and incubating for another 4 h. A total of 150 µl dimethyl sulfoxide was added in each well. Following this, the supernatant was removed and absorbance was measured at 490 nm after oscillation for 10 min using a microplate reader (BioTek Instruments, Inc., Winooski, VT, USA). The viability of osteoblasts after transfection with miR-204 mimics, inhibitor or miR negative control was also examined using MTT assay.

Osteoblasts were cultured in a 24-well culture plate. Following treatment with different final concentrations (0.01, 0.1 and 1 mg/ml) puerarin for 48 h, cells were washed three times with PBS. The ALP positive cells were stained and ALP activity was assayed using a commercial kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) according to the manufacturer's instructions. The ALP activities were also examined in osteoblasts following transfection with miR-204 mimics, inhibitor or miR negative control.

Osteoblasts were seeded in a 24-well plate in α-MEM (Wisent, Inc.), supplemented with 10% FBS (Gibco; Thermo Fisher Scientific, Inc.), at the density of 1×10⁵ cells/well. The concentration of 0.01, 0.1 and 1 mg/ml puerarin was added to the experimental wells, and cultured in a humidified 5% CO₂ incubator at 37°C for 48 h. Total RNA was prepared using TRIzol reagent (Invitrogen; Thermo Fisher Scientific, Inc.). Purity and concentration of the isolated RNA were detected by a nucleic acid protein analyzer (BioDrop µLite; Biodrop, Cambridge, UK). The RNA was then reversed-transcribed to cDNA with a reverse transcription kit (Takara Bio, Inc., Otsu, Japan), following the manufacturer's protocol. The final reaction solution (25 µl) contained 2 µl cDNA product, 12.5 µl 2X Realtime PCR SYBR premix (Takara Bio, Inc.), 0.5 µl each of forward and reverse primers and 9.5 µl distilled water. The two-step PCR amplification conditions were: Pre-denaturation of 95°C for 30 sec, followed by 40 cycles of 95°C for 5 sec, 60°C for 34 sec. The fluorescence signal detection was measured by Mx3000P real-time PCR system (Stratagene; Agilent Technologies, Inc., Santa Clara, CA, USA). Target genes included transforming growth factor (TGF)-β1, Smad2, Smad3 and Runx2, while β-actin served as an internal control. Primers were designed and synthesized by GenScript (Piscataway NJ, USA) and are presented in Table I.

Protein was extracted from the cells with total protein extraction kit (cat. no. BC3710; Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) after treatment with 0.01, 0.1 and 1 mg/ml puerarin for 48 h, and then the protein content was measured by the bicinchoninic acid assay method, according to the manufacturer's instructions (cat. no. PC0020; Beijing Solarbio Science & Technology Co., Ltd.). For western blot analysis, 25 µg of samples were subjected to SDS-PAGE gel and electro-transferred onto a polyvinylidene difluoride membrane. The membrane was then blocked with 5% skim milk in TBS containing 0.1% Tween-20 (TBST) for 1 h at room temperature. Following washing three times in TBST for 5 min, the protein samples were incubated with primary antibodies (Cell Signaling Technology, Inc., Danvers, MA, USA) for TGF-β1 (cat. no. 709), Smad2 (cat. no. 3122), Smad3 (cat. no. 9523), Runx2 (cat. no. 12556) and β-actin (cat. no. 4970; all 1:1,000) at 4°C overnight. Following washing another three times with TBST, the samples were incubated with horseradish peroxidase-conjugated anti-rabbit IgG secondary antibodies (cat. no. 7074; 1:3,000; Cell Signaling Technology, Inc.) for 1 h at room temperature. Following a final wash of TBST, the protein bands of interest were visualized by the enhanced chemiluminescence detection system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The intensity of the bands was measured using the ImageJ2× v2.1.4.7 software (National Institutes of Health, Bethesda, MD, USA) to assess the relative protein levels.

Expression profiles analysis of miRNAs was performed by KangChen Biology Engineering Co., Ltd. (Shanghai, China) following treatment of osteoblasts with 0.1 mg/ml puerarin for 48 h.

The Runx2-targeting miRNAs were predicted using TargetScan (http://genes.mit.edu/targetscan), PicTar (http://pictar.bio.nyu.edu), miRBase (http://mirbase.org/) and miRDB (http://mirdb.org/miRDB/) programs, and then compared with the results of expression profile analysis to determine the Runx2-targeting miRNAs.

Following incubating osteoblasts with puerarin (0.1 mg/ml) for 48 h, total RNA was extracted using TRIzol reagent (Takara Bio, Inc.). The reverse-transcription reaction solution (20 µl) contained 10.0 µl 5X Reverse Transcription Mix (Takara Bio, Inc.), 1.0 µl Stem-loop RT primers (GenScript), 500 ng miRNA, 2.0 µl HiScript Enzyme Mix (cat. no. R223; Vazyme, Piscataway, NJ, USA) and RNase-free water. The PCR amplification conditions were: 25°C for 5 min, 45°C for 50 min, 85°C for 50 min. Specific miRNA stem-loop RT primers for mouse miR-204 and internal control U6 were designed and synthesized by Shanghai Sangon Pharmaceutical Co. Ltd. (Shanghai, China). Primers used in the stem-loop RT-qPCR are as following: RT primer, 5′-GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACAGGCAT-3′; PCR upstream primer, 5′-GCGGCGGTTCCCTTTGTCATCC-3′; downstream primer 5′-ATCCAGTGCAGGGTCCGAGG-3′ for miR-204; PCR upstream primer, 5′-CTCGCTTCGGCAGCACA-3′; downstream primer, 5′-AACGCTTCACGAATTTGCGT-3′ for U6.

Empty vectors were prepared for constructing Runx2 expression vectors, sequences of the wild-type Runx2 3′-UTR (ttctatgcacgtattgtacaaattgtgct ttgtgccacaggtcatgatcgtggatgagtttactctgaacttcaaagggactatttgtatt gtatgttgcaactgtaaattgaattatttggcatttccccctctcatgattgtaatatt) and mutant Runx2 3′-UTR (ttctatgcacgtattgtacaaattgtgctttgtgcc acaggtcatgatcgtggatgagtttactctgaacttcaaatccactatttgtattgtatgttg caactgtaaattgaattatttggcatttccccctctcatgattgtaatatt) were inserted into the empty vectors. The recombinant plasmids were synthesized by Shanghai GenePharma Co, Ltd. (Shanghai, China).

Osteoblasts were seeded in 24-well plates at a density of 2×10⁴ per well. At 24 h incubation, cells in each well were transfected with the recombinations of wildtype Runx2 3′-UTR or mutant Runx2 3′-UTR plasmids (100 ng) and miR-204 mimics (2 µM) or miR control (2 µM) using Lipofectamine 2000 (1 ml; Invitrogen; Thermo Fisher Scientific, Inc.). Cells were collected following 48 h transfection, then firefly and Renilla luciferase activities were analyzed using Dual-Luciferase Reporter Assay System (Promega Corporation, Madison, WI, USA). The ratio of firefly to Renilla luciferase activity was used as relative luciferase activity.

The mouse miR-204 mimics, miR-204 inhibitor and negative control were obtained from Shanghai GenePharma Co., Ltd. Cells were seeded in six-well plates at a density of 1×10⁵ per well and transfected with 16 µl RNAs using 4 µl Lipofectamine 2000 transfection reagent (Invitrogen; Thermo Fisher Scientific, Inc.) when the cells reached 90% confluence. Control cells were only transfected with 4 µl transfection reagent. Then cells were cultured in a humidified 5% CO₂ incubator at 37°C for 48 h. Cells were ready for RT-qPCR and western blot analysis of Runx2. Viability and ALP activity were also examined.

The results were presented as mean ± standard deviation. Data comparisons were analyzed by Student's t-test or one-way analysis of variance with SPSS software (version, 17.0; SPSS Inc., Chicago, IL, USA). Comparisons between groups were analyzed using the Bonferroni post hoc test. P<0.05 was considered to indicate a statistically significant difference.

The authors first examined whether puerarin could affect osteoblast viability. Osteoblasts were cultured with puerarin at various concentrations. The cells treated with puerarin had significantly higher viability compared with that of the control cells (P<0.05), except those treated with puerarin at a lower concentration (0.01 mg/ml) for 24 h and 48 h (Fig. 1A). The osteoblasts treated with puerarin for 48 h at various concentrations (0.01, 0.1 and 1 mg/ml) had significantly higher ALP activity than that of the control cells (P<0.05; Fig. 1B). These data suggested that puerarin may promote osteoblast viability and differentiation.

In order to explore the proliferation effect of puerarin on osteoblasts, the authors studied the expression of osteoblast-related regulatory factors (TGF-β1, Smad2, Smad3 and Runx2). Following treating cells with puerarin for 48 h, we measured TGF-β1, Smad2, Smad3 and Runx2 gene expression with RT-qPCR and also the protein expression with western blotting. The transcription levels and the translation levels of the genes were all significantly increased in the cells treated with puerarin (P<0.05; Fig. 2). The data imply that these factors may be involved in the promotion effects of puerarin on osteoblasts. Studies have indicated that Runx2 is a master regulatory factor that serves a crucial role on the regulation of transcription in osteoblasts. Afterwards, the authors focused on the expression of Runx2-targeting miRNAs (17).

Analysis of miRNA expression profiles demonstrated that the expression of miR-204 and other 118 miRNAs changed following puerarin treatment for 48 h (18). In addition, miRNA target analysis programs (TargetScan, PicTar, miRBase and miRDB) predicted that miR-204 may directly target the osteogenic crucial regulator Runx2. Then, the effect of puerarin on miR-204 expression was examined by stem-loop RT-qPCR. The expression level of miR-204 was significantly decreased following culturing the cells with 0.1 mg/ml puerarin for 48 h, compared with that of the cells in absence of puerarin (P<0.05; Fig. 3A). The results suggested that puerarin downregulated the expression of miR-204.

Combined with the previous validation that puerarin promoted osteoblast viability and differentiation through upregulation of Runx2, however, the authors demonstrated that miR-204 was downregulated. Moreover, according to the bioinformatics analysis, miR-204 was suggested to bind to the 3′-UTR of Runx2 through specific binding sites (Fig. 3B). Taken together, future studies are prompted to determine whether Runx2 is a direct downstream target of miR-204.

To further investigate whether Runx2 can be directly targeted by miR-204, luciferase reporters containing wildtype or mutant 3′-UTR of Runx2 were engineering and transfected with miR-204 mimics or miR negative control. As presented in Fig. 4A, transfection with miR-204 mimics of wildtype-Runx2-3′-UTR recombinant plasmid had lower luciferase activity compared with that of the miR negative control, while transfection with miR-204 mimics of mutant-Runx2-3′-UTR recombinant plasmid had no statistically significant difference compared with that of the miR negative control. These data indicated that miR-204 dramatically repressed the luciferase activity of the reporter containing the wild-type full-length 3′-UTR of Runx2, but not the mutant full-length 3′-UTR of Runx2, implying that miR-204 may directly target Runx2.

The effects of overexpression and inhibition of miR-204 on Runx2 were then investigated. The expression of Runx2 was analyzed using RT-qPCR and western blotting methods in osteoblasts at 48 h following transfection with miR-204 mimics, inhibitor or miR negative control. The results of RT-qPCR (Fig. 4B) did not show significant difference, but western blotting analysis (Figs. 4C and D) indicated that overexpression of miR-204 in osteoblasts significantly decreased the expression level of Runx2, while inhibition of miR-204 had the opposite effect. The effects of alternation on protein level, rather than on mRNA level, suggesting that a role of miR-204 for post-transcriptional regulation. Collectively, these results indicated that Runx2 is a direct target of miR-204.

Finally, the authors investigated the effect of miR-204 on osteoblasts. Following transfection with miR-204 mimics, inhibitor or miR negative control for 48 h, the viability and ALP activity of osteoblasts were tested. The cells that were transfected with miR-204 mimics had significantly lower viability, but cells transfected with miR-204 inhibitor had higher viability compared with that of the control cells (Fig. 5A). Similarly, transfection with miR-204 mimics significantly reduced ALP activity, whereas transfection with miR-204 inhibitor had the opposite effect (Fig. 5B). These results suggested that miR-204 may inhibit osteoblast viability and differentiation.