Department of Oral Cell Biology, Academic Centre for Dentistry Amsterdam (ACTA), University of Amsterdam and VU-University of Amsterdam, Research Institute MOVE, VU University, Amsterdam, the Netherlands
Department of Oral Implantology and Prosthetic Dentistry, Academic Centre for Dentistry Amsterdam (ACTA), University of Amsterdam and VU-University of Amsterdam, Research Institute MOVE, VU University, Amsterdam, the Netherlands
The osteogenesis of bone marrow stromal cells (BMSCs) is of paramount importance for the repair of large-size bone defects, which may be compromised by the dietary-accumulated all-trans retinoic acid (ATRA). We have shown that heterodimeric bone morphogenetic protein 2/7 (BMP2/7) could induce bone regeneration in a significantly higher dose-efficiency in comparison with homodimeric BMPs. In this study, we evaluated the effects of ATRA and BMP2/7 on the proliferation, differentiation, mineralization and osteogenic genes. ATRA and BMP2/7 exhibited both antagonistic and synergistic effects on the osteogenesis of BMSCs. ATRA significantly inhibited proliferation and expression of osteocalcin but enhanced the activity of alkaline phosphatase of BMSCs. On day 21, 50 ng/mL BMP2/7 could antagonize the inhibitive effects of ATRA and significantly enhance osteogenesis of BMSCs. These findings suggested a promising application potential of heterodimeric BMP2/7 in clinic to promote bone regeneration for the cases with dietary accumulated ATRA.
Maintenance of bone homeostasis relies on the bone development and remodeling process, which continually replaces old and damaged bone with new tissue (Parfitt 1994). In this process, bone marrow stromal cells (BMSCs) play a crucial role. BMSCs are also multipotent to differentiate into various mesenchymal lineages such as osteoblasts, chondrocytes, adipocytes, neurocytes, and fibroblasts. Furthermore, BMSCs are regarded as one of the important cell sources for bone tissue engineering. Bone remodeling and regeneration are controlled by the careful orchestration of a number of signaling components (Ducy et al. 2000), such as retinoid acid and bone morphogenetic proteins (BMPs).
All-trans retinoic acid (ATRA), a metabolite of vitamin A, can help cells to grow and develop, especially in the embryo. Bone is a target of retinoid acid. It has been shown that ATRA plays an important role in the regulation of bone cell functions (Skillington et al. 2002; Michaelsson et al. 2003; Jacobson et al. 2004). ATRA is likely to be one of the factors involved in bone metabolism by its ability to promote physiologic remodeling (Michaelsson et al. 2003). On the other hand, ATRA in high doses, that can be derived from hypervitaminosis A and alcoholism, can promote bone resorption and osteoporosis (Saneshige et al. 1995). The suppression of ATRA on bone metabolism can be mediated by both increasing osteoclastic bone resorption (Wang et al. 2008) and decreasing osteoblastic bone formation (Jacobson et al. 2004). It has been established that ATRA can slow the proliferation of preosteoblasts and prevent the formation of a mineralized matrix. This effect may be due to the induction and maintenance of a partially differentiated non-proliferating state (Kitching et al. 2002).
Bone morphogenetic proteins (BMPs) – a group of growth factors under the superfamily of transforming growth factor-β (TGF-β) (Yuan et al. 2011) – play an important role in osteogenesis and bone metabolism (Reddi 1997). Furthermore, homodimeric BMPs are the most important osteoinductive cytokines for bone tissue engineering (Vaccaro et al. 2002; Hayashi et al. 2010). Homodimeric BMP2 and BMP7 have already been approved for clinical application to promote bone formation. However, due to their high doses of clinical usage and an extremely high risk of side effects such as overstimulation of osteoclastic activity and ectopic bone formation, the use of homodimeric BMPs was limited (Kaneko et al. 2000; Boden et al. 2002). One approach to this dilemma is to adopt more potent forms of BMPs – heterodimeric BMPs (Guo & Wu 2012). We have already shown heterodimeric BMP2/7 induced cell differentiation with significantly lower threshold concentrations but similar maximum effects (Zheng et al. 2010). In an in vivo experiment, we showed that low-dose BMP2/7 heterodimer facilitated more rapid bone regeneration in better quality in peri-implant bone defects than the homodimeric BMPs (Wang et al. 2012).
Hitherto, the combinatory effect of BMPs and ATRA is still controversial in inducing osteogenic differentiation (Wang et al. 2008; Chen et al. 2010; Karakida et al. 2011). One opinion is that ATRA and the BMP signaling pathways cooperate to inhibit osteogenesis and promote adipogenesis of mouse embryonic palate mesenchymal cells, while the others found that retinoic acid cooperates with BMP2 to induce osteoblastic differentiation of C2C12 cells. Clarification of ATRA and BMPs' effects on BMSCs is of great importance for developing an efficient strategy for bone tissue engineering. In this study, we investigated the respective and the combinatory effects of ATRA and BMP2/7 on the in vitro osteogenesis of BMSCs.
Materials and methods
We tested the respective and the combinatory effects of BMP2/7 and ATRA on the osteogenesis of BMSCs. BMSCs were treated in the presence or absence of ATRA (1 μmol/L) and different concentrations of BMP2/7 (5 ng/mL, 50 ng/mL). Flow cytometry was used to characterize BMSCs. We examined the DNA content as proliferation index, the activity of alkaline phosphatase (ALP) as a marker for early differentiation, osteoclacium (OCN) secretion as a maker for terminal differentiation as well as calcium deposition. In addition, we investigated the expression of osteogenic genes, such as ALP, OCN and Runx2 (Runt-related transcription factor 2).
BMSC isolation and culture
Primary BMSC were harvested from a 4-week-old male Sprague–Dawley (SD) rats using previously established protocols (Wu et al. 2009; Ding et al. 2011; Wang et al. 2011). Briefly, femora and tibiae were dissected from the rat that had been euthanized by carbon dioxide gas. After the removal of soft tissue under sterile conditions, the bone tissues were incubated in culture medium (DMEM [Dulbecco's Modified Eagle Medium, Invitrogen, Grand Island, NY, USA] containing 10% FBS, 100 U/mL streptomycin and 100 U/mL penicillin). The marrow was flushed out from the bones with the culture medium by using a needle and syringe. Aspiration and blow of the dropper were repeated to obtain a suspension of single bone marrow cells. Then the cells were seeded into a cell culture Petri-dish (10 cm in diameter) and kept in a humidified incubator with 5% CO2 and 95% air at a temperature of 37°C. Hematopoietic and non-adherent cells were removed by subsequent medium changes. The culture medium was changed after 24-h culture, and thereafter routinely changed every 3 days. The unpassaged cells were defined as passage 0. As the culture reached almost 80–90% confluence, cells were detached with trypsin/ethylenediaminetetraacetic acid (EDTA) (0.25% w/v trypsin, 0.02% EDTA). Cell clusters were broken down by repeated pipetting. The cells of passage three were used for the following assays.
We adopted flow cytometry to identify the surface markers of BMSCs, such as CD11b, CD29, CD45, and CD90. The cells of passage three were collected and washed in 0.1 mol/L phosphate-buffered saline (PBS) twice. Thereafter, they were re-suspended in 2.5 mL pre-cooled PBS at the final concentration of 2 × 106 cells/mL and divided into five flow cytometry tubes. Then, the cells were incubated with fluoresceine isothiocyanate (FITC)-tagged anti-rat antibodies to the abovementioned surface markers: FITC-CD11, FITC-CD29, FITC-CD45, FITC-CD90 (Biolegend, CA, USA), or pre-cooled PBS (negative control) in the dark for 15 min at room temperature. After washing in PBS, the samples were then analyzed on the flow cytometer (Cytomics FC 500, Beckman, USA). Ten thousand cells were determined for each treatment. Additionally, the BMSC was identified through the following criteria: ≥95% of the BMSC population expressed CD29 and CD90, ≤5% expressed CD11b and CD45 (Vaquero et al. 2013).
Treatment of BMSC with BMP2/7 and ATRA
At sub-confluence, the cells of passage 3–5 that were used for research were detached with trypsin, pelleted and re-suspended in the culture medium. Thereafter, they were plated in 24-well plates at a density of 1 × 104 cells/well for cell proliferation assay, at a density of 5 × 104 cells/well in 6-well plates for ALP activity, OCN detection and PCR analysis. The cells were plated in 48-well plates for alizarin red staining at a density of 0.8 × 104 cells/well. After incubation for 24 h, cells were subjected to a low-serum medium (DMEM containing 1% FBS) for another 24 h and then followed by the treatments with BMP2/7 (R&D Systems, Minneapolis, MN, USA) or/and ATRA (Sigma-Aldrich, St. Louis, MO, USA). BMSCs were exposed to the following treatments: (i) control medium (culture medium with 1% FBS); (ii) control medium with 5 ng/mL BMP2/7; (iii) control medium with 50 ng/mL BMP2/7; (iv) control medium with 1 μmol/L ATRA; (v) control medium with 5 ng/mL BMP2/7 and 1 μmol/L ATRA; and (vi) control medium with 50 ng/mL BMP2/7 and 1 μmol/L ATRA. The concentrations of BMP2/7 and ATRA were determined based on previous studies. We have already shown that heterodimeric BMP2/7 was the most efficient in inducing proliferation of pre-osteoblasts at 5 ng/mL and in inducing ALP activity and OCN expression at 50 ng/mL (Zheng et al. 2010). In addition, 1 μmol/L ATRA has repeatedly been adopted to study the effects of ATRA on in vitro osteogenic differentiation (Wang et al. 2008; Chen et al. 2010; Sheng et al. 2010). The cells were maintained in their respective culture conditions with the culture medium and changed every 3 days. Additionally, the culture medium for alizarin red staining of each group was prepared in the same way as described previously and then treated with mineralizing medium (10% FBS, 50 μmol/L L-ascorbic acid, and 10 mmol/L β-glycerophosphate; Sigma-Aldrich) (Karakida et al. 2011).
Cell proliferation assay
To investigate cell proliferation of BMSCs in response to the different treatments, the number of cells was determined by the amount of cellular DNA after stimulation for 1, 4 and 7 days using Quant-i PicoGreen dsDNA Reagent and kits (Invitrogen, Molecular Probes, Eugene, OR, USA) as previously described (Horii et al. 2007). The fluorescent intensity of the solution mixed by the cell lysate and dye was measured using a fluorescence spectrometer (SpectraMax M5, Molecular Devices, Sunnyvale, CA, USA) with a setting of Ex 480 nm/Em 520 nm.
ALP activity assay
To determine the early osteogenic differentiation of BMSCs, ALP activity and protein content of these six groups were measured after 4- and 7-day treatments. The cells lysates were prepared, and then the ALP activity of the lysates was determined using LabAssay-ALP colorimetric assay kit (Wako Pure Chemicals, Osaka, Japan) according to the instructions. The total protein concentrations were determined by a commercial BCA Protein Assay kit (Beyotime, Shanghai, China). ALP activity was calculated as nmol p-NP released per hour and further normalized to the cell protein content.
OCN expression assay
At the end of the 4 and 7 days, the cell culture media of these six groups were collected to determine quantity of OCN secreted into the cell culture medium. The cell culture media were centrifuged at 12 857m/s2, 4°C for 5 min before detection. The OCN concentrations of the supernatants were determined by ELISA using a rat OCN EIA kit (Westang, Shanghai, China) (Zhu et al. 2004; Horii et al. 2007; Zheng et al. 2010).
Alizarin red staining
We compared the mineralization of BMSCs under the stimulation of BMP2/7 and ATRA. Quadruplicate cell cultures of each group were prepared and treated with mineralizing medium, which has been mentioned. After 14 and 21 days treatment, mineralized nodules were determined by alizarin red staining (Sigma-Aldrich) (Iwata et al. 2002; Karakida et al. 2011). Culture plates were photographed by NIS-Elements F2.20 (Nikon Eclipse 80i, Tokyo, Japan), and the calcified area was quantified using the software of Image-Pro Plus 6.0 analysis. After being photographed, 10% cetylpyridinium chloride (CPC, Sigma-Aldrich LLC, GER) was used to dissolve the mineralized nodules and release calcium-bound alizarin red S into solution. Then we detected the colorimetric absorbance at 562 nm.
Isolation of total RNA and real-time fluorescence qRT–PCR analysis
The total RNA was extracted from cells treated with BMP2/7 and ATRA using RNeasy Mini Kit and purified with RNase-Free DNase Set reagent (Qiagen, Germany) on day 4 and day 7, following the manufacturer's instructions. Total RNA was reverse transcribed to cDNA using the PrimeScript RT Master Mix Perfect Real Time reagent (Takara Bio, Shiga, Japan). Quantitative reverse transcription–polymerase chain reaction (qRT–PCR) was performed using PrimeScript RT reagent Kit Perfect Real Time (Takara Bio) according to the manufacturer's instruction. Specific primers used for detecting mRNA transcripts of the Runx2, ALP, OCN, and β-actin gene are as shown in Table 1. Transcripts were normalized to β-actin transcript levels. The n-fold upregulation was calculated for each gene of interest over the internal control gene (β-actin gene) according to the delt-delt-Ct method using the formula: 2−[(CT gene of interest−CTinternal control)sample-(CT gene of interest−CT internal control)control] (Wu et al. 2009).
Table 1. Primer sequences for real-time quantitative polymerase chain reaction analysis of the expression of Runx2, alkaline phosphatase (ALP) and osteocalcin (OCN) genes
Primers (F = forward; R = reverse)
F: 5′- TACAAGGTGGTGGACGGTGAAC -3′
R: 5′- GCCATGACGTGGGGGATGTA -3′
F: 5′- ACCCTCTCTCTGCTCACTCTGCT -3′
R: 5′- GCTGGGGCTCCAAGTCCATT -3′
F: 5′- AGCGGACGAGGCAAGAGTTT -3′
R: 5′- CCTAAATCACTGAGGCGGTCAG -3′
F: 5′- CCGTAAAGACCTCTATGCCAACA -3′
R: 5′- CTAGGAGCCAGGGCAGTAATCTC -3′
Each measurement within an individual experiment was repeated three times. Statistical comparisons among the results were made by one-way analysis of variance (anova). Post Hoc comparisons were made using Bonferroni corrections. The level of significance was set at P <0.05. SPSS software (version 20) for a Windows computer system was used for the statistical analysis.
Characterization of BMSCs
We used flow cytometry to characterize the biomarkers of BMSC (Fig. 1). More than 99% of expanded BMSCs of passage three were strongly positive for the surface markers CD29 and CD90 that were characteristics of BMSC. The BMSCs cultures contained little hematopoietic lineage cells, as indicated by the extremely low amount of CD11b- or CD45- expressing cells. These results indicated that the cells of passage three were high-purified BMSCs.
Cell proliferation assay
Without the presence of BMP2/7 and ATRA, cells significantly proliferated with time (Fig. 2). On day 1 and day 4, BMP2/7 alone did not significantly influence the proliferation of BMSCs irrespective of its concentrations. On day 7, 50 ng/mL BMP2/7 resulted in a significantly higher DNA amount than 5 ng/mL and 0 ng/mL. In contrast, ATRA alone could significantly inhibit the proliferation of BMSCs, resulting in only a significantly higher DNA amount from day 4 to day 7. Interestingly, BMP2/7 could antagonize the inhibitive effects of ATRA on cell proliferation. On day 4, BMP2/7 at both 5 ng/mL and 50 ng/mL resulted in a significantly higher DNA amount than ATRA alone. On day 7, on 50 ng/mL BMP2/7 was associated with a significantly higher DNA amount than ATRA alone.
In comparison with the contrary effects of BMP2/7 and ATRA on the proliferation, both cytokines significantly increased the ALP activity of BMSCs on day 4 and day 7 (Fig. 3). At the absence of ATRA, only 50 ng/mL BMP2/7 significantly promoted the cell ALP activity from day 4 to day 7. On day 4, the promoting effect of ATRA alone (about 1.71 folds) was similar with that of 50 ng/mL BMP2/7 alone (about 1.70 folds). While on day 7, the promoting effect of ATRA alone (about 3.10 folds) was significantly higher than that of 50 ng/mL BMP2/7 alone (about 2.24 folds). At the presence of ATRA, BMP2/7 at both 5 ng/mL and 50 ng/mL significantly promoted the cell proliferation from day 4 to day 7. ATRA and 50 ng/mL BMP2/7 synergistically promoted (about 7.01 folds) in the ALP activity of BMSCs on both day 4 and day 7.
BMP2/7 alone could significantly promote the late osteogenic differentiation marker ─ OCN in a dose-dependent manner on both day 4 and day 7 (Fig. 4). BMP2/7 of 50 ng/mL also promoted OCN from day 4 to day 7, while 5 ng/mL did not. In contrast, ATRA alone significantly inhibited OCN expression on both time points. In the presence of ATRA, the promoting effects of BMP2/7 were in dose-dependent and time-dependent manners. The ATRA-suppressed OCN could be completely restored by 50 ng/mL BMP2/7 on day 4 and 5 ng/mL on day 7. Furthermore, 50 ng/mL BMP2/7 could significantly enhance OCN (about 1.37 folds) on day 7 even in the presence of ATRA.
Cell matrix mineralization
After a 14-day treatment, mineralization of BMSCs in cell matrix was only found in the presence 50 ng/mL BMP2/7 (Fig. 5). Both calcification area and the absorbance without ATRA were significantly higher than that with ATRA.
On day 21, very mild calcium deposition could be detected in the two groups without BMP2/7. The influence of ATRA on the mineralization of BMSCs was very mild without the presence of BMP2/7. Regardless of ATRA, BMP2/7 could significantly promote mineralization in a dose-dependent manner. ATRA could significantly downregulate the osteogenesis induced by 5 ng/mL BMP2/7. However, the inhibitive effect of ATRA was insignificant for 50 ng/mL BMP2/7.
Expression of osteogenic genes
From day 4 to day 7, expression of Runx2 genes was significantly downregulated without the stimulation of either BMP2/7 or ATRA (Fig. 6A). 5 ng/mL BMP2/7 or ATRA only significantly enhanced the expression of Runx2 gene only on day 7. Without ATRA, 50 ng/mL BMP2/7 could significantly enhance the expression of Runx2 gene on both day 4 (1.20-fold and 1.22-fold) and day 7 (2.07-fold and 1.61-fold) in comparison with 0 ng/mL BMP2/7 and 5 ng/mL BMP2/7, respectively. Furthermore, ATRA and 50 ng/mL BMP2/7 significantly enhance the expression of Runx2 gene on both day 4 (2.06-fold and 2.23-fold) and day 7 (1.96-fold and 2.02-fold) in comparison with 0 ng/mL BMP2/7 and 5 ng/mL BMP2/7, respectively.
Different from ALP activity, ATRA didn't significantly influence the expression of ALP gene on day 4 (Fig. 6B). Furthermore, ATRA significantly suppressed the expression of ALP gene on day 7. Without ATRA, 5 ng/mL and 50 ng/mL BMP2/7 enhanced the expression of ALP gene only on day 7. On day 4, ATRA and BMP2/7 of both concentrations could significantly enhance the expression of ALP gene in comparison with the controls. On day 7, ATRA and 50 ng/mL BMP further significantly increased the expression of the ALP gene in comparison with on day 4. Without the presence of ATRA, a significant enhancement in the expression of OCN gene could be detected from day 4 to day 7. A significant suppression of OCN gene expression by ATRA was detected on day 7 irrespective of BMP2/7, while such a suppressive effect of ATRA was not seen on day 4 (Fig. 6C).
Bone marrow stromal cells are one of the main cells for the repair of large-size bone defects in vivo. Therefore, BMSCs are also commonly used for bone tissue engineering. During the differentiation of BMSCs towards osteoblastic lineage, various hormones and cytokines play important roles and regulate this process (Huang et al. 2004). It was suggested that the dietary overdosed ATRA may compromise the regenerative capacity of bone tissue (Jacobson et al. 2004). Recently, we showed that heterodimeric BMP2/7 could induce bone regeneration in a significantly higher dose-efficiency in comparison to the homodimeric BMPs (Wang et al. 2012). Consequently, in this study, we investigated the in vitro effects of ATRA and BMP2/7 on the proliferation, differentiation, and mineralization of BMSCs. We showed that 50 ng/mL BMP2/7 could antagonize the inhibitive effects of ATRA and significantly enhance bone regeneration. To our knowledge, this was the first study to report the combinatory effects of heterodimeric BMPs and ATRA on the osteogenesis of BMSCs.
Retinoic acid, an active metabolite of vitamin A, acts as an inducer of differentiation in a number of cell types, and it has become apparent that retinoic acid is involved in bone formation (Lohnes et al. 1993). Retinoic acid regulates the gene expression via its receptors. There are two families of retinoic acid receptors: RARs and RXRs, each of which has three isotypes (α, β and γ). RARs bind all-trans retinoic acid (ATRA) – an abundant form of retinoic acid – and form heterodimers with RXR. The RAR/RXR heterodimers bind DNA and directly regulate transcription of the target genes (Bastien & Rochette-Egly 2004). Although ATRA significantly enhanced ALP activity (Fig. 3), it significantly inhibited cell proliferation (Fig. 2), OCN expression (Fig. 4) and mineralization (Fig. 5). We also confirmed that the suppression of osteogenesis was not due to the promotion of adipogenesis (data not shown). These findings were in accordance with a previous publication that ATRA alone inhibited the osteogenic differentiation of rat bone marrow stromal cells (Wang et al. 2008). The mechanism for such an inhibitive effect of ATRA remains to be elucidated. It seems not to be mediated by the inhibition on the master gene of osteogenesis – Runx2, since ATRA alone could even enhance the transcription of Runx2 (Fig. 6A).
Bone morphogenetic proteins are important cytokines modulating osteogenesis. We have already shown that heterodimeric BMP2/7 exhibited significantly higher dose-efficiency in inducing both in vitro and in vivo osteogenesis than the corresponding homodimeric BMP2 and BMP7 (Zheng et al. 2010; Wang et al. 2012). In this study, we aimed to further investigate whether BMP2/7 could enhance the osteogenesis of BMSCs that was inhibited by ATRA. Although ATRA still significantly downregulated the mineralization induced by BMP2/7 on day 14, 50 ng/mL BMP2/7 could result in a significantly higher mineralization on day 21 irrespective of ATRA (Fig. 5). This finding suggested the promising application potential of heterodimeric BMP2/7 in clinic to promote bone formation even in the adverse microenvironment with dietary accumulated ATRA. Further studies are still needed to confirm the advantages of heterodimeric BMP2/7 over the respective homodimeric BMPs in this application.
The combinatory effects of ATRA and BMPs remain complicated. On one side, ATRA could significantly suppress BMP-induced proliferation (Fig. 2) and OCN expression (Fig. 4) of BMSCs. On the other side, ATRA could also significantly promote BMP-induced ALP activity (Fig. 3), as well as the expression of Runx2 gene and ALP gene (Fig. 6A,B). Furthermore, ATRA and BMPs could synergistically promote the osteogenic differentiation of non-osteogenically committed cells, such as myoblastic cells (Karakida et al. 2011) and preadipocytes (Skillington et al. 2002). This phenomenon may be due to the differential modulation of ATRA on BMP signaling. BMPs signaling can be mediated through classical Smad-dependent pathway and Smad-independent pathways (Derynck & Zhang 2003). BMPs bind to transmembrane serine/threonine kinase receptors on cell surfaces. Thereby, they trigger specific intracellular signaling pathways that activate and influence gene transcription. Activated BMP type I receptors phosphorylate Smad1, Smad5, and Smad8 (receptor-regulated Smads, R-Smads), which then assemble into a complex with Smad4 (common-partner Smad, Co-Smad) and translocate to the nucleus to regulate the transcription of target genes, such as Runx2 (Miyazono 1999). In addition, the activated BMP receptors can also initiate Smad-independent signaling pathways, resulting in the activation of ERK, p38, and JNK (Guicheux et al. 2003; Massague 2003; Hoffmann et al. 2005). RA could significantly promote the degradation of phosphorylated Smad1 through enhancing the interaction between pSmad1 and its ubiquitin E3 ligases (Sheng et al. 2010). This depended on the RA-increased Gadd45 expression and mitogen-activated protein kinase (MAPK) activation. On the other hand, ATRA could also cooperate with BMPs to enhance the ALP activity, which was mediated by RARγ (Karakida et al. 2011). Furthermore, ALP expression induced by BMPs was mediated by the activation of a Smad-independent signaling pathway p38 MAPK (Nohe et al. 2002). It may be concluded that ATRA simultaneously but respectively inhibits BMP-induced Smad-dependent signaling and promotes Smad-independent signaling. This may account for the phenomena: ATRA inhibits proliferation (Fig. 2), OCN expression (Fig. 4) but enhances ALP activity (Fig. 3). In addition, the inconsistent tendency between gene and protein of ALP and OCN suggested a modulation in the level of microRNA. Further study is still needed to elucidate this modulation.
The promoting effects of ATRA on BMP-induced ALP activity were positively correlated with the concentrations of BMPs (Karakida et al. 2011). In our study, we found that ATRA shows no or suppressive effects on the expression of ALP gene. However, ATRA and BMP2/7 could significantly enhance the expression of ALP gene in a BMP-dose-dependent manner (Fig. 6B). This finding suggested that ATRA took effects through amplifying BMP-induced Smad-independent signaling. Since ATRA inhibits Smad pathway, the significant upregulation of Runx2 gene could be directly stimulated by ATRA signaling and also probably mediated by Smad-independent MAPK signaling. 5 ng/mL BMP2/7 could enhance but not accelerate the osteogenesis of BMSCs (Fig. 5). In contrast, 50 ng/mL BMP2/7 could not only enhance but also accelerate the osteogenesis of BMSCs. Although a suppression of ATRA on the mineralization induced by BMP2/7 was detected on day 14, 50 ng/mL BMP2/7 resulted in a similar level of mineralization in the presence or absence of ATRA on day 21. This suggested that 50 ng/mL BMP2/7 could sufficiently antagonize the inhibitive effects of ATRA and enhance the osteogensis of BMSCs. In this process, the suppression of proliferation and OCN expression by ATRA may account for the delayed osteogenesis, while the synergistically amplified ALP gene and activity by BMP and ATRA may account for the comparable level of mineralization in comparison with that without ATRA.
In conclusion, ATRA and BMP2/7 showed both antagonistic and synergistic effects on the osteogenesis of BMSCs. This may be due to the differential modulation of ATRA on the BMP-induced Smad-dependent and Smad-independent signaling. 50 ng/mL BMP2/7 could significantly enhance the osteogenesis of BMSCs in the presence of ATRA. This suggested a promising application potential of heterodimeric BMP2/7 in clinic to promote bone formation even in the adverse microenvironment for the cases with dietary accumulated ATRA.
This study was supported by the grant (No. Y2111175 and No. Y2101440) from Natural Science Foundation of Zhejiang Province.
Wenjuan Bi: Concept, data collection, statistical analysis, drafting article and approval of the final version; Zhiyuan Gu: Design and securing of funding; Yuanna Zheng: Design and securing of funding; Limin Wang: Design and securing of funding; Jing Guo: Interpretation, revision, and approval of the final version. Gang Wu: Interpretation, drafting article. All authors approved the final version.