Quercetin promotes bone marrow mesenchymal stem cell proliferation and osteogenic differentiation through the H19/miR-625-5p axis to activate the Wnt/β-catenin pathway

Background Quercetin and H19 can promote osteogenic differentiation of bone marrow mesenchymal stem cells (BMSCs). However, whether quercetin regulates H19 expression to promote osteogenic differentiation of BMSCs is unclear. Methods BMSC proliferation, matrix mineralization, and alkaline phosphatase (ALP) activity were assessed using the Cell Counting Kit-8, ALP assay kit, and alizarin red staining kit, respectively. Expression of H19, miR-625-5p, BMP-2, osteocalcin, and RUNX2 were measured by qRT-PCR; β-catenin protein level was measured by western blotting. Results Quercetin promoted BMSC proliferation, enhanced ALP activity, and upregulated the expression of BMP-2, osteocalcin, and RUNX2 mRNAs, suggesting that it promoted osteogenic differentiation of BMSCs. Moreover, quercetin increased H19 expression, while the effect of quercetin on BMSCs was reversed by silencing H19 expression. Additionally, miR-625-5p, interacted with H19, was downregulated during quercetin-induced BMSC osteogenic differentiation, which negatively correlated with H19 expression. Silencing miR-625-5p expression promoted BMSC proliferation and osteogenic differentiation, whereas miR-625-5p overexpression weakened the effect of quercetin on BMSCs. Finally, quercetin treatment or downregulation of miR-625-5p expression increased β-catenin protein level in BMSCs. Upregulation or downregulation of miR-625-5p or H19 expression, respectively, inhibited β-catenin protein level in quercetin treated-BMSCs. Conclusion H19 promotes, while miR-625-5p inhibits BMSC osteogenic differentiation. Quercetin activates the Wnt/β-catenin pathway and promotes BMSC osteogenic differentiation via the H19/miR-625-5p axis. Supplementary Information The online version contains supplementary material available at 10.1186/s12906-021-03418-8.

that can regulate BMSC osteogenic differentiation need to be developed for the management and treatment of these orthopedic diseases. Quercetin, a naturally available flavonoid and a wellknown phytoestrogen [5], exerts antioxidant and antiinflammatory properties. In vitro, quercetin promotes osteogenic differentiation of BMSCs via extracellular-signal-regulated protein kinases, adenosine 5'-monophosphate (AMP)-activated protein kinase/sirtuin 1, and estrogen receptor-mediated pathways [6][7][8]. In vivo, in rat models of postmenopausal osteoporosis, quercetin heightens BMSC osteogenic differentiation to increase the bone mineral density [9]. However, currently available scientific data has not yet elucidated the functional mechanism of quercetin. Therefore, further understanding of the mechanisms underlying quercetin effects will have immense clinical implications.
Therefore, in this study, the effect of quercetin and H19 expression on the proliferation and osteogenic differentiation of BMSCs was identified. Further, the mechanism was elucidated by which quercetin affects BMSC osteogenic differentiation.

Matrix mineralization
BMSCs were cultured in CM medium or osteogenic differentiation medium at 37 o C in a 5% CO 2 incubator. Every 3 days, fresh medium (preheated to 37 o C) was used to replace the old CM medium. After 21 days of quercetin treatment, the medium was removed, BMSCs were stained with Alizarin Red solution (Aladdin, Shanghai, China). Finally, the BMSCs were washed and images were acquired under a microscope (Beijing Cnmicro Instrument Co., Ltd., Beijing, China).
The information related to the primers used is listed in Table 1.

Bioinformatic database assay and luciferase reporter assay
LncBase Experimental v.2 and Starbase 3.0 were used to find the potential miRNAs binding to H19 [20,21]. Abnormally expressed miRNAs during BMSC osteogenic differentiation were analyzed using GEO2R in GSE148049 of the Gene Expression Omnibus (https:// www. ncbi. nlm. nih. gov/ geo/). The potential miRNAs at the intersection of three sets of results were selected. 293T cells (5×10 4 cells/well) were plated and cultured for 24 h. The binding of H19 to miRNAs was analyzed using the luciferase reporter assay. Briefly, the wild type H19 (H19-WT) and mutant H19 (H19-Mut, with mutated binding sites) sequences were cloned onto the luciferase reporter vector psi-CHECK2 and then transfected into 293T cells. Forty-eight hours later, the luciferase activity of renilla or firefly luciferase activity was evaluated by the dual luciferase reporter assay system (Promega). The renilla/firefly luciferase activity rate was lower in the co-transfected H19 and miR-625-5p mimic groups than that in co-transfected NC mimic and WT-H19 groups, suggesting that H19 can bind to miR-625-5p. The renilla/firefly luciferase rate did not change significantly between the co-transfected H19 and miR-625-5p mimic groups and the co-transfected NC mimic and WT-H19 groups, suggesting that H19 cannot bind to miR-625-5p.

Statistical analysis
Data, conform to normal distribution, are presented as the mean ± standard deviation. One-way analysis of variance was performed using SPSS 19.0 statistical software (IBM, Inc.) to analyze the statistical difference between more than three groups, followed by Tukey's post-hoc test. Statistical significance was set at P < 0.05.

Quercetin enhances BMSC osteogenic differentiation
The molecular structural formula of quercetin is shown in Fig. 1A. Cell proliferation increased significantly in quercetin treatment groups (both 5 and 10 μM) compared with that in the blank group from day 1 to day 7, whereas cell proliferation in the positive group increased significantly on day 1 and 2 but decreased on day 7 (Fig. 1B). ALP activity and the transcription of BMP-2, osteocalcin, and RUNX2 was significantly enhanced in the quercetin treatment and positive groups compared with those in the blank group ( Fig. 2A-D). Additionally, calcium nodules were observed in all groups after treatment for 21 days (Fig. 2E). Compared with that in the blank group, the number and area of calcium nodules notably increased in the quercetin treatment and positive groups (Fig. 2E). These results suggest that quercetin treatment significantly increased cell proliferation and osteogenic differentiation.

Quercetin promotes H19 expression
The H19 levels was measured by performing qRT-PCR after treating cells with quercetin and osteogenic differentiation medium (positive group). H19 expression increased significantly in the quercetin treatment and positive groups compared with that observed in the blank group (Fig. 3A). Furthermore, H19 expression had a positive relationship with ALP activity, and the

Silencing H19 expression reverses the effect of quercetin on BMSCs
To study the effect of H19, BMSCs were transfected with si-H19, followed by treatment with 10 μM quercetin; si-NC was used as control. H19 expression was successfully silenced by si-H19 (Fig. 4A). Following H19 silencing, the proliferation of BMSCs was inhibited on days 1, 2, and 7 (Fig. 4B). Moreover, on day 21 after H19 silencing, ALP activity and mRNA levels of BMP-2, RUNX2, and osteocalcin were significantly reduced ( Fig. 4C-D).
The number and area of calcium nodules were notably decreased following H19 silencing on day 21 (Fig. 4E).

H19 interacts with miR-625-5p
Firstly, we identified 254 miRNAs in the GSE148049 dataset that were differentially expressed during BMSC osteogenic differentiation. Additionally, 105 and 237 potential miRNAs binding to H19 were identified. miR-625-5p and miR-483-3p were selected through the intersection of the three sets of results (Fig. 5A).
Zhou et al reported that treatment with 2 μM quercetin promoted osteogenic differentiation of rat BMSCs, and this effect was even better than that observed following  [22]. Treatment with 0.1, 1, and 10 μM concentration quercetin enhanced osteogenic differentiation of rat BMSCs, particularly at 10 μM [23]. We observed that quercetin treatment induced osteogenic differentiation in BMSCs and found that 10 μM quercetin was the optimal concentration to achieve these effects. The inconsistent effects of quercetin treatment on rat, mouse, and human BMSCs may be attributed to the differential tolerance of quercetin in different species. Casado-Díaz et al reported that BMSC differentiation medium supplemented with 10 μM quercetin decreased BMSC proliferation and differentiation, while BMSC differentiation medium supplemented with 0.1 μM quercetin promoted BMSC proliferation and differentiation [24]. Casado-Díaz's study indirectly shows that quercetin promotes osteogenic differentiation. However, when present at a higher concentration (10 μM quercetin), in presence of another strong inducer of osteogenic differentiation (BMSC differentiation medium), it inhibits osteogenic differentiation. In our study, treatment with 1, 5, and 10 μM quercetin or BMSC differentiation medium treatment promoted BMSC proliferation and differentiation. Although based on our results we inferred that that 10 μM is the optimal quercetin concentration to induce osteogenic differentiation of BMSCs, the toxicity of high concentrations of quercetin (more than 2 μM) should be considered in animals and humans based on the above studies. Based on these observations, we propose that 1~2 μM quercetin may be the effective concentration to induce osteogenic differentiation without causing cytotoxicity. Further studies are warranted to validate these observations and inferences. Previous studies also reported that H19 heightened osteogenic differentiation of BMSCs [14,25,26]. Consistently, H19 expression was upregulated during BMSC osteogenic differentiation in this study. Additionally, silencing H19 expression in BMSCs reversed the osteogenic differentiation-inducing effects of quercetin. This result suggested that H19 participates in the regulation of quercetin-induced BMSCs osteogenic differentiation. Functionally, H19 heightened BMSCs osteogenic differentiation by inhibiting the expression of miR-140-5p and miR-149 [14,26]. In our study, H19 was found to interact with miR-625-5p during quercetin-induced BMSC osteogenic differentiation. The role of miR-625-5p on BMSC osteogenic differentiation was previously unclear. Our results revealed that silencing miR-625-5p promotes osteogenic differentiation of BMSCs, suggesting that miR-625-5p inhibits BMSCs osteogenic differentiation. Furthermore, miR-625-5p overexpression can reverse quercetin-induced osteogenic differentiation of BMSCs. Collectively, quercetin promotes BMSCs osteogenic differentiation by targeting the H19/miR-625-5p. In our study, H19 adsorbed miR-625-5p to regulate quercetininduced osteogenic differentiation. Consistently with our results H19 can adsorb miRNAs include miR-140-5p, miR-149, and miR-532-3p to regulate osteogenic differentiation induced by other factors or drugs [14,26,27].
These results indicate that H19 regulated osteogenic differentiation can be further modulated by different factors by adsorbing different miRNAs.
Clinically, osteogenic ability of BMSCs is weakened under some pathophysiological conditions, such as aging, menopause, trauma, and inflammation, all of which can lead to bone defects and osteoporosis [2][3][4]33]. We found that quercetin treatment induced BMSC proliferation and osteogenic differentiation, suggesting that quercetin can be used for the clinical treatment of osteoporosis and bone defects. H19 expression is inhibited in patients with osteoporosis and bone defects, suggesting H19 interacts with miR-625-5p and inhibits miR-625-5p expression. A miR-483-3p and miR-625-5p were selected from the intersection of the GSE148049 dataset, Starbase 3.0, and LncBase Experimental v.2 analysis results. Differentially expressed miRNAs during BMSC osteogenic differentiation were identified from by the GSE148049 dataset. Potential miRNAs binding to H19 were determined according to Starbase 3.0 and LncBase Experimental v.2 analysis. B miR-483-3p and miR-625-5p expression were analyzed in the GSE148049 dataset. ***P < 0.001 vs Day 0. C The binding sites between H19 and miR-625-5p were analyzed by Starbase 3.0. D The binding of H19 to miR-625-5p was measured by the luciferase reporter assay. ***P < 0.001. E miR-625-5p expression reduced during quercetin-induced BMSC osteogenic differentiation. **P < 0.01 and ***P < 0.001, vs Blank group. F miR-625-5p expression negatively correlated with H19 expression. G miR-625-5p expression increased at 21 days after transfection. ***P < 0.001 that H19 is a therapeutic target for these diseases [27,[34][35][36]. Since quercetin treatment increased H19 expression during quercetin-induced BMSCs osteogenic differentiation, quercetin may be used for the clinical treatment of osteoporosis and bone defects. Although our study presents some credible data, it had some limitations. First, the target genes of miR-625-3p remain unclear. Additionally, quercetin may also activate the ERK and p38 MAPK pathways to enhance osteogenic differentiation of BMSCs, which can also be activated by H19 in cardiomyoblasts [23,37], suggesting that the MAPK pathway may be another downstream pathway regulated by H19 in quercetin-induced osteogenic differentiation. Thus, to elucidate the relationship between quercetin, H19, and the ERK and p38 MAPK signaling pathways, further studies are warranted. Moreover, understanding the effect of quercetin on BMSC osteogenic differentiation in animal models and humans require further studies, and the nontoxic dose of quercetin remains to be confirmed.

Conclusions
Our study demonstrated that H19 promoted, while miR-625-5p inhibited osteogenic differentiation of BMSCs. Quercetin promoted BMSC proliferation and osteogenic differentiation via the H19/miR-625-5p axis to activate the Wnt/β-catenin pathway. Additionally, 1~2 μM quercetin may be the effective concentration to induce osteogenic differentiation without causing cytotoxicity. However, the concentration and mechanism by which quercetin facilitates the osteogenic differentiation of BMSCs in vivo requires further exploration.