A total of 25 male BALB/c mice (5–6 weeks old; body weight, 15–21 g) and 2 pregnant female C57/BL mice (12–13 weeks old; body weight, 23–26 g) at 13 days post-coitum were used throughout the present study. Animals were obtained from the Laboratory Animal Centre, Jiangsu University (Zhenjiang, China). The animals were housed in cages in a temperature-controlled room (20-25°C and 40-0% humidity), with a 12-h light/dark cycle and free access to commercial food and water. All procedures involving animals and their care conformed to the USA National Institutes of Health guidelines (NIH Pub. No. 85-23, revised 1996). All animal experiments were reviewed and approved by the Institutional Animal Care and Use Committee of School of Animal Science and Technology, Yangzhou University (Yangzhou, China; approval no. YZUDWSY2017-0029). The procedures were performed in accordance with the Regulations of the Administration of Affairs Concerning Experimental Animals (China, 1988), and the Standards for the Administration of Experimental Practices (Jiangsu, China; 2008).

The isolation and culture of BMSCs was conducted as previously described (74). In brief, 5–6-week-old male BALB/c mice were sacrificed via the cervical dislocation method. The animals were then rinsed with 70% ethanol for few min. The hind limbs were excised from the trunk of the body, and the bones were kept in PBS for the subsequent steps under a sterile hood. The bones were placed on sterile gauze and were gently rubbed to remove any attached soft tissue. The epiphysis was cut, a 26-gauge syringe needle was inserted into the bone marrow cavity and the marrow was flushed out using Dulbecco's modified Eagle's medium (DMEM); flushing was continued until pale white bone was observed. The cell suspension was filtered through a 70-mm filter mesh to remove any bony spicules, muscle or cell clumps. The number of viable cells was counted via trypan blue staining. Cells were cultured in 95-mm dishes using 1 ml complete culture medium (CCM) at a density 25×10⁶ cells/ml, and then kept at 37°C in a humidified atmosphere containing 95% air and 5% CO₂. The CCM comprised DMEM containing 15% fetal bovine serum (FBS; HyClone; GE Healthcare Life Sciences, Logan, UT, USA), 2 mM L-glutamine and 1% penicillin-streptomycin (Beyotime Institute of Biotechnology, Haimen, China). Cells were cultured with frequent medium changes as described previously (74). When primary cultures became nearly confluent, the culture was treated with 0.5 ml 0.25% trypsin EDTA for 2 min at room temperature. The cells that were lifted within 2 min were harvested and cultured in a 60-mm plate. Once the culture reached 70–80% confluence, the cells were harvested for successive passages. The subsequent experiments were performed with cells from passages 3 and 4.

Pregnant female mice (C57/BL) at 13 days post-coitum was sacrificed via the cervical dislocation method. Uterine horns were removed, washed with PBS and opened. Embryos were harvested and the head and the visceral organs were removed. After washing with PBS, the embryos were minced, and then incubated at 37°C for 15 min in 0.25% trypsin EDTA with gentle shaking. Trypsin was neutralized with an equal amount of MEF medium and cells were collected by centrifugation at 500 × g for 7 min at room temperature (20-25°C). Cells were then resuspended and cultured on gelatin-coated dishes with MEF growth medium containing high glucose DMEM mixed with 10% FBS, 4 mM L-glutamine and 1:100 penicillin-streptomycin at 37°C with 5% CO₂. The cells were examined daily using an inverted microscope (Olympus TH4-200; Olympus Corporation, Tokyo, Japan) to assess their general appearance, and to identify any signs of microbial contamination. Once confluent, cells were passaged 1:3 and the second passage cells were trypsinized and frozen at −80°C. In the present experiments, MEFs within three passages were used to avoid replicative senescence.

The codon-optimized hLMP-3 cDNA sequence obtained from Pola et al (75), was synthesized, and cloned into T-Vector pMD19 (Takara Biotechnology, Co., Ltd., Dalian, China) by GenScript (Project ID. 7162905-1; GenScript, Jiangsu, China). cDNA sequencing and the primer design of hLMP-3 were also conducted by GenScript. The construction of the viral expression vector containing the transcription factors hLMP-3 was performed by Genomeditech (Shanghai, China; Project ID. GM-Lc-01147). The primers used for hLMP-3 (Accession number: AAK30569.1) were as follows: Forward, 5′-CCCTCGAGGGCTTGGCCATGGATAGTTTCAAGGTGGTC-3′ and reverse, 5′-GCGTCGACGTTCAGCCACTTGAGGCGGGCATCTG-3′ (restriction recognition sites are underlined).

BMSCs (passage 3) were used for MSC cell surface marker experiments. The MSC surface markers cluster of differentiation (CD)44 and CD90, and also the hematopoietic marker CD45 were examined by flow cytometry as previously described (76,77). Briefly, cells at passage 3 were harvested using trypsin/EDTA. Following cell counting, cell suspensions were stained with the following fluorescence-conjugated antibodies: Fluorescein isothiocyanate (FITC)-anti-mouse/human CD44 antibody (1:100; cat. no. 103021; BioLegend, Inc., San Diego, CA, USA), FITC anti-mouse CD45 monoclonal antibody (1:100; cat. no. 11-0451; eBioscience; Thermo Fisher Scientific, Inc.) and FITC anti-mouse CD90 monoclonal antibody (1:100; cat. no. 11-0903; eBioscience; Thermo Fisher Scientific, Inc.) for 1 h at 4°C. Cells stained with FITC-labeled rat anti-mouse immunoglobulin G (IgG; 1:100; cat. no. 406001; BioLegend, Inc.) served as controls. Cells were pelleted, washed twice with PBS and fixed with 1% paraformaldehyde in PBS. Cell surface antigens were detected with a flow cytometer using BD Accuri C6 Plus software (BD Biosciences, Franklin Lakes, NJ, USA).

BMSCs were seeded at a density of 50 cells/cm² and cultured until they reached 70% confluency. The CCM was then changed to OM (fresh complete medium with 15% FBS) supplemented with 50 µg/ml ascorbic acid (Sigma Aldrich; Merck KGaA, Darmstadt, Germany), 10 mM β-glycerol phosphate disodium salt (Sigma Aldrich; Merck KGaA), and 100 nM dexamethasone (Sigma Aldrich; Merck KGaA). Cells cultured in osteogenic medium only comprised the OM group; while in the curcumin group (CR group), 15 µM curcumin was added to the OM. This concentration was selected based on previous studies that used curcumin in cell culture to induce osteogenic differentiation with minimal cytotoxicity (53,78,79). Curcumin was prepared according to the method previously described by Gu et al (53). Briefly, curcumin (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) was dissolved in dimethyl sulfoxide and stored at −20°C. In the ATRA group, 1 µM ATRA (Sigma-Aldrich; Merck KGaA) was added to the OM. This concentration has repeatedly been used to study the effects of ATRA on in vitro osteogenic differentiation; since the normal physiological level of ATRA is ≤0.01 µM and the effective pharmacological concentration is >0.1 µM, the 1 µM concentration was selected to induce osteogenic differentiation (27,80–82).

The protocol used was a modified version of that described by Yamamoto et al (83); the preparation and dilution of the OM components were conducted according to the protocol provided by the supplier. MEFs were re-suspended in CCM, and seeded onto 35-mm dishes (Corning^®; 5×10⁴ cells/dish) or a 24-well plate (Corning^®; 1.2×10⁴ cells/well) on day 1. The next day, transduction of MEFs was performed using the supernatant containing the pGMLVPE1-hLMP-3 expression vector (multiplicity of infection=4), and supplemented with 4 µg/ml polybrene (Genomeditech). Following culture for 24 h, the virus-containing medium was replaced with an OM composed of fresh complete culture medium (10% FBS) supplemented with 50 µg/ml ascorbic acid, 10 mM β-glycerol phosphate disodium salt and 100 nM dexamethasone. In the curcumin-supplemented group, 15 µM curcumin was added to the medium.

Total RNA was extracted from the cell samples using TRIZOL^® reagent (cat. no. 15596-026; Invitrogen; Thermo Fisher Scientific, Inc.) according to the manufacturer's protocol. Following isolation, 1 µg RNA was reverse-transcribed into cDNA using FastQuant RT kits with gDNase (cat. no. KR106; Tiangen Biotech Co., Ltd., Beijing, China). The reverse transcription reaction was performed at 42°C for 15 min, followed by 95°C for 3 min. The cDNA samples were analyzed by RT-qPCR in a 7500 Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.) using Super Real Premix plus (SYBR Green; cat. no. FP205; Tiangen Biotech Co., Ltd.). The thermocycling conditions for qPCR were as follows: 95°C for 15 min, followed by 40 amplification cycles of 95°C for 10 sec and 60°C for 32 sec. The sequences of the RT-qPCR primers are listed in Table I. The primers were designed and manufactured Takara Biotechnology Co., Ltd. Relative quantification was calculated with 2^−ΔΔCq (84) and normalized to β-actin. Data are presented as levels relative to the expression level in the control cells.

Early osteogenic differentiation was evaluated by alkaline phosphatase staining (ALP). Matrix mineralization was evaluated by Alizarin red (ALZ) and von Kossa (VK) staining. The staining intensity was quantified using Fiji in ImageJ software, version 1 (National Institutes of Health, Bethesda, MD, USA) (85).

Histochemical staining was performed using 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium (BCIP/NBT) ALP color development kits (cat. no. C3206; Beyotime Institute of Biotechnology) according to the manufacturer's protocol. The cells were washed with PBS buffer, then fixed with 4% formalin at room temperature for 2 min. The cells were then washed 3 times with PBS. Following the last wash, an appropriate amount of BCIP/NBT staining solution was added, ensuring that the sample was fully covered. After the addition of the working solution, cells were incubated in the dark at 37°C for 5–30 min in an incubator until the desired color developed. The BCIP/NBT stain working solution was then removed, and cells were washed once or twice with distilled water to stop the color reaction.

The cells were washed with 1X PBS, and fixed in 10% formaldehyde in 1X PBS for 15 min. Following fixation, the cells washed with distilled water 2 or 3 times. Cells were incubated in 40 mM ALZ staining solution for 5 min in the dark. Finally, cells were washed with distilled water 4 times to remove the excess stain. The ALZ staining solution was prepared by diluting the ALZ staining powder (cat. no. A5533; Sigma-Aldrich; Merck KGaA) in distilled water. The pH was adjusted to 4.1–4.3 with 10% ammonium hydroxide using a pH Meter (Mettler Toledo, Columbus, OH, USA).

For VK staining, the cells were washed with PBS, fixed with 10% formalin and stained with freshly prepared 5% silver nitrate solution for 30 min with exposure to UV light. After washing with distilled water 3 times, the cells were treated with 5% sodium thiosulfate solution for 3 min to remove any remaining silver nitrate. Final washing with distilled water was repeated 3 times. The silver nitrate solution (5%) was prepared by diluting silver nitrate powder (cat. no. GB12595-90; Beijing HenGye Zhongyuan Chemical Co., Ltd., Beijing, China) in distilled water. The same diluent was used for the preparation of a solution of sodium thiosulfate (cat. no. 217263; Sigma-Aldrich; Merck KGaA).

The cells were rinsed briefly with PBS, and then fixed for 20 min with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) at room temperature. The cells were permeabilized for 10 min with 0.1% TritonX-100 in PBS, to allow for specific antibody entrance into the cells, and then blocked for 45–60 min with 4% bovine serum albumin (BSA) in PBS at room temperature. Cells were incubated overnight (16–18 h) at 4°C with mouse monoclonal OCN antibody (1:500; cat. no. sc-376726; Santa Cruz Biotechnology, Inc., Dallas, TX, USA). The cells were then washed 3 times with washing buffer. This was followed by 1 h incubation at room temperature with the secondary antibody Alexa Fluor 647-labeled goat anti-mouse IgG (1:500; cat. no. ab150115; Abcam, Cambridge, MA, USA). Following a final round of 3 washes with the wash buffer, nuclei were counterstained using DAPI at room temperature for 5 min (1 mg/ml PBS; Invitrogen; Thermo Fisher Scientific, Inc.). Cells were examined using inverted fluorescence microscopy (Olympus TH4-200).

All steps were conducted according to the manufacturer's protocol, using a Bio-Rad system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). Cell lysates were collected with RIPA buffer (cat. no. P0013B; Beyotime Institute of Biotechnology) supplemented with phenyl methane sulfonyl fluoride (Beyotime Institute of Biotechnology). Following centrifugation at 12,000 × g at 4°C for 10 min, the supernatant was collected and the protein concentration was determined using a BCA protein assay kit (Beyotime Institute of Biotechnology). Protein samples were heated for 10 min at 95°C, and then separated by 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (20 µg protein/lane). Samples were then blotted onto polyvinylidene fluoride membranes (Beijing Solarbio Science & Technology Co., Ltd.) using a Trans-Blot SD system (Bio-Rad Laboratories, Inc.). The membranes were blocked with BSA blocking buffer (cat. no. CW2143S; CWBIO, Beijing, China) at room temperature for 2 h, and then incubated with the primary mouse monoclonal osteocalcin (OCN) antibody (1:500; cat. no. sc-376726; Santa Cruz Biotechnology, Inc.) at 4°C overnight. After washing three times, the corresponding secondary antibody horseradish peroxidase-labeled goat anti-mouse IgG (1:2,000; cat. no. 665739; Merck KGaA) was added. The membranes were incubated at 37°C for 2 h and then washed with TBST (cat. no. CW0043S; CWBIO). Bands were visualized with enhanced chemiluminescence method western blot kits (cat. no. CW0048M; CWBIO) using the FluorChem Q system (ProteinSimple, San Jose, CA, USA). β-actin (1:1,000; cat. no. AA128; Beyotime Institute of Biotechnology) was used as a loading control with overnight incubation at 4°C. The protein band intensity was quantified using Fiji in ImageJ version 1 (85).

All quantitative data are expressed as the mean ± standard deviation (n=3). Statistical analyses were performed using SPSS software version 21 (IBM Corp., Armonk, NY, USA). Statistical significance was determined using one-way analysis of variance followed by Tukey's post hoc test. P<0.05 was considered to indicate a statistically significant difference. For some experiments the statistical significance was determined using paired sample t-tests (P<0.05 and P<0.01). GraphPad Prism software version 6 (GraphPad Software, Inc., La Jolla, CA, USA) was used to produce the graphs.

Bone marrow was harvested from BALB/c mice, and P3 cells were seeded into 35-mm culture dishes at a density of 25×10⁶ cells/ml. BMSCs are recognized as the adherent cells derived from bone marrow and are capable of extensive proliferation, producing a fibroblastic shape. Following 15 days of culture, almost homogeneous populations of fibroblast-like cells were observed (Fig. 1A). To confirm cell identity, immunofluorescent staining was used to identify the BMSCs with the surface markers CD90 and CD44, while the surface marker CD45 was used to detect hematopoietic cell contamination. The cultured cells were positive for the MSC markers CD44 and CD90, and negative for the hematopoietic cell marker CD45 (Fig. 1B).

BMSCs were induced to differentiate into an osteogenic lineage, and were able to proliferate in vitro. The morphology of BMSCs changed following the first week of culture in OM (OM group) from their spindle-like fibroblast shape to the characteristic cuboidal morphology of primary osteoblasts. Comparable results were observed in the CR group. By contrast, the BMSCs cultured in CCM did not differentiate into the osteoblast lineage, and cells in the ATRA group displayed a more elongated shape, and dendrites were clearly visible (Fig. 2).

The present study performed an in vitro mineralization assay to determine the effects of curcumin and ATRA on the onset of the osteogenic differentiation of BMSCs. ALP, ALZ and VK staining were conducted during the differentiation period. ALP activity was analyzed as an early indicator of osteogenic differentiation. The results revealed that OM induced ALP activity in the mouse BMSCs after 7 and 14 days. The ALP staining intensity was significantly increased in the CR group (P<0.01 vs. the OM and ATRA groups); while no significant difference was detected between the OM and ATRA groups. Matrix mineralization was detected by ALZ and VK staining. Mineralization was not clearly detected after 7 days, and was slowly developing after 14 days. However, after 21 and 28 days, the OM and CR groups exhibited positively stained mineralized nodules. Mineralization was significantly stronger in the CR group compared with the OM group. By contrast, the ATRA group did not show any signs of mineralization (Fig. 3).

In order to evaluate the molecular changes following the supplementation of the OM with curcumin and ATRA, the present study examined the expression levels of the main genes associated with osteoblastic differentiation, namely runt-related transcription factor 2 (Runx2), osterix and BMP2. These markers are well known for their roles in numerous osteoblast differentiation pathways. The expression levels of these bone markers were quantified at 7, 14 and 21 days (Fig. 4). The results revealed that curcumin-supplemented OM induced a significant upregulation of osteo-specific markers when compared with the ATRA and OM groups, and this upregulation appeared to be time-dependent. After 7 days, the CR group exhibited induced expression of osteo-specific markers when compared with the OM and ATRA groups. After 14 days, the CR group had significantly higher expression levels of the three tested markers when compared with the OM and ATRA groups (P<0.01). The expression levels of the markers had begun to decline in the ATRA group, and were not significantly different when compared with the OM group for Runx2 and BMP2. Notably, the OM group exhibited significantly higher (P<0.01) osterix expression when compared with the ATRA group. These results were consistent with the results of the mineralization assay, which revealed the inhibitory effect of ATRA during the differentiation process. Finally, after 21 days, the results revealed that the CR group had upregulated levels of the three markers when compared with the OM and ATRA groups (P<0.01). In addition, the expression levels of osteo-specific markers in the ATRA group appeared to decline further compared with those in the OM group. These RT-qPCR results demonstrated the positive effect of curcumin-supplemented medium on the osteogenic differentiation of BMSCs. By contrast, ATRA supplementation exhibited an inhibitory effect.

OCN, a marker of mature osteoblasts, was evaluated during the differentiation process using immunostaining and the protein levels of OCN were determined by western blot analysis. In line with the previous results, the CR group exhibited a distinct increase in the expression level of OCN when compared with the OM and ATRA groups. Furthermore, the addition of ATRA inhibited the induction of OCN expression (Fig. 5).

The aforementioned results reveal the positive role of curcumin-supplemented medium on the osteogenic differentiation of BMSCs. Therefore, the present study investigated whether this positive role could be applied in other cell types. In a previous study, the present research team studied the reprogramming of MEFs to osteoblast cells using hLMP-3 (86). The results revealed that the transduction of MEFs with a lentiviral vector expressing the osteogenic factor hLMP-3 induced osteoblast cell formation in vitro (86). The present study examined the effect of curcumin enrichment on the osteogenic differentiation of MEFs reprogrammed with the osteogenic factor hLMP-3. MEFs were successfully transduced with pGMLVPE1-hLMP-3, as previously reported (86). At 2 days post-transduction, the culture medium was changed to OM supplemented with curcumin. The effect of curcumin-supplemented medium was compared with that of the non-supplemented medium using RT-qPCR to detect the difference in the expression of bone markers between the two groups. The RT-qPCR results revealed that the addition of curcumin to the OM upregulated the expression of the bone markers Runx2, BMP and osterix at 7, 14 and 21 days post-transduction (Fig. 6).