Resveratrol promotes osteogenesis and alleviates osteoporosis by inhibiting p53

Although osteoporosis is one of the most common chronic age-related diseases, there is currently no gold standard for treatment. Evidence suggests resveratrol, a natural polyphenolic compound, may be helpful in the treatment of osteoporosis and other diseases. However, the molecular mechanisms underlying the anti-osteoporotic effects of resveratrol remain largely unknown. In the present study, KEGG pathway enrichment analysis of resveratrol-targeted genes identified 33 associated pathways, 12 of which were also involved in osteoporosis. In particular, the MDM2/p53 signaling pathway was identified as a potential key pathway among the shared pathways. In vitro experiments indicated that MDM2-mediated p53 degradation induced osteoblast differentiation, and resveratrol could partially reverse p53-dependent inhibition of osteogenic differentiation. These findings suggest resveratrol may alleviate osteoporosis at least in part by modulating the MDM2/p53 signaling pathway.


INTRODUCTION
Osteoporosis (OP) is a common metabolic bone disease in postmenopausal women and the elderly [1]. The pathological characteristics of osteoporosis include abnormal bone microarchitecture, decreased bone density, and increased bone fragility, which result in an increased risk of fractures [2]. An estimated 200 million people currently suffer from osteoporosis worldwide, and about 34 million patients in the USA have been diagnosed with osteoporosis or low bone mass [3]. Osteoporosis can cause pain, spinal deformities, and fragile fractures. Fragile fractures are usually caused by low-energy impacts, such as a fall from standing height, a slight collision, or other routine minor injuries. In a 50-year-old woman, the risk of developing osteoporotic fractures may be has high as 50% [4]. Vertebral and hip fractures are two common types of osteoporotic fractures. Vertebral fractures can cause long-term pain and severely affect the quality of life, while hip fractures can prevent patients from standing or walking and can even increase the risk of death [4]. Osteoporosis has thus become a major public health problem worldwide, placing heavy economic burdens on both patients and healthcare systems [3]. There is currently no gold standard treatment for osteoporosis.

AGING
Resveratrol (3,5,4′-trihydroxy-trans-stilbene) is a natural polyphenolic compound found in several plants, including grapes, berries, and peanuts [5,6]. Growing evidence shows that resveratrol has various biological effects [7][8][9], including protective effects in osteoporosis. For example, bone mass density (BMD) reduction and microarchitectural deterioration were reversed by resveratrol in a rodent model [8]. Resveratrol may impact bone homeostasis by acting as both an anabolic and antiresorptive agent [10]. However, the detailed mechanisms underlying the effects of resveratrol on bone health remain unclear.
Bioinformatics, which uses computational techniques to organize and integrate experimental data, has been used to great effect in many studies [11,12]. In this study, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways associated with both resveratrol-targeted genes and osteoporosis progression were analyzed using bioinformatic tools. The p53 signaling pathway was identified as a key KEGG pathway involved in the protective effects of resveratrol on osteoporosis. We therefore performed an in vitro study to further examine whether resveratrol could counteract the negative effects of p53 on MDM2-mediated osteogenic differentiation.

Resveratrol-targeted genes and the interaction network
In total, 30 resveratrol-targeted genes were identified in Search Tool for Interacting Chemicals (STITCH) using a three shell limit. Interaction networks among resveratrol-targeted genes were then constructed in STITCH (Figure 1A). TP53, SIRT1, PTGS1, SIRT3,  ESR1, PPARG, NOS3, AKT1, SIRT5, and PTGS2 were identified as members of the first shell, indicating that resveratrol might directly affect these genes . The second shell consisted of ATM, BRCA1,  FOXO1, MTOR, EP300, RICTOR, FOXO3,  CDKN1A, KAT2B, and MDM2, and the third shell  included HSP90AA1, HIPK2, NCOA3, CDKN2A,  MAPK8, SRC, USP7, RCHY1, CREBBP, and SP1, indicating that resveratrol might have secondary effects on these genes. A network visualization constructed based on interaction weights indicated that TP53 had the highest weight of all of these genes ( Figure 1B).

Identification of KEGG pathways associated with osteoporosis and resveratrol-targeted genes
Enrichment analysis of the resveratrol-targeted genes using Database for Annotation, Visualization, and Integrated Discovery (DAVID) identified 33 KEGG pathways with p <0.05. 110 KEGG pathways involved in human osteoporosis were retrieved using miRWalk2.0. Twelve KEGG pathways associated with both osteoporosis and resveratrol-targeted genes were identified and visualized using a Venn Diagram online tool ( Figure 2). Among them, the five KEGG pathways with smallest p values were prostate cancer pathway, pathway in cancer, glioma pathway, p53 signaling pathway, and cell cycle signaling pathway (Table 1).  Cell cycle signaling pathway CDKN1A, CDKN2A, EP300, CREBBP, TP53, MDM2, ATM 6.6E-6

Retrieval of KEGG pathways related to resveratroltargeted genes
The downstream cellular responses of the top five KEGG pathways associated with both osteoporosis and resveratrol-targeted genes are shown in Figure 5. They include G1/G2 arrest, cellular apoptosis, genomic instability, proliferation, G1/S progression, survival, and cell cycle arrest. These results indicate that resveratrol impacts a variety of cellular activities through the PI3K, p53, and cell cycle signaling pathways.

MDM2-mediated p53 degradation induces osteoblast differentiation in vitro
p53 gene and protein expression were measured in non-OP (n=10) and OP patients (n=10) using qRT-PCR and western blotting, respectively. p53 gene and protein expression were significantly higher in OP patients (p < 0.001, Figure 6A, 6B) than in non-OP patients. Transfection of human mesenchymal stem cells (hMSCs) with an MDM2 overexpression plasmid successfully increased MDM2 gene expression compared to untreated cells (Control) and empty vector transfection (Plasmid NC) ( Figure 6C). As expected, p53 gene and protein expression decreased significantly in hMSCs transfected with MDM2 plasmid (p < 0.001 Figure 6D, 6E). In addition, expression of the osteogenic genes ALP OCN and Runx2 also increased significantly in MDM2 plasmid-treated hMSCs compared to untreated control and empty vector-treated cells (p < 0.001 Figure 6F). Alizarin red-mediated calcium staining also showed enhanced mineral deposition in the MDM2 plasmid group (p < 0.001 Figure 6G, 6H). These results indicated that MDM2 induced p53 degradation and promoted osteogenesis in hMSCs.

Resveratrol partially reverses p53-induced inhibition of osteogenic differentiation
The effects of different concentrations of resveratrol on MDM2 expression were measured by qRT-PCR. 10 μM resveratrol induced both the largest increase in MDM2 gene expression ( Figure 7A) and the largest decrease in p53 expression ( Figure 7B). As expected, ALP, OCN, and Runx2 expression were significantly increased by resveratrol treatment (Figure 7C-7D). Moreover, MDM2 gene expression was significantly higher ( Figure 7E), while p53 expression was significantly lower AGING  AGING ( Figure 7F), in hMSCs treated with resveratrol than in the other groups. Furthermore, alizarin red-mediated calcium staining showed enhanced mineral deposition in the resveratrol group (p < 0.05 Figure 7G-7H), indicating that resveratrol might increase mineral bone mass. In summary, these results suggest that resveratrol promotes osteogenesis by inhibiting the p53 signaling pathway.

DISCUSSION
Although osteoporosis is a widespread disease [13] that has become a major challenge for public health systems worldwide [3], there is currently no gold standard treatment. While resveratrol can reverse both osteoporosis-associated BMD reduction and microarchitectural deterioration [8], the mechanisms responsible for these effects remain unknown. In this study, bioinformatics analyses and in vitro studies were performed to investigate the molecular mechanisms of resveratrol's effects on osteoporosis pathologies.
In this study, 12 KEGG pathways associated with both osteoporosis and resveratrol-targeted genes were identified using bioinformatics tools. Among these, the top five KEGG pathways with the smallest p values were prostate cancer pathway, pathway in cancer, glioma pathway, p53 pathway, and cell cycle pathway.
Resveratrol-targeted genes were associated with G1/G2 arrest, apoptosis, and genomic instability in the prostate cancer pathway (Figure 5A), apoptosis and proliferation in pathway in cancer ( Figure 5B), G1/S progression, proliferation, survival and genomic instability in the glioma pathway ( Figure 5C), cell cycle arrest and p53 feedback in the p53 pathway ( Figure 5D), and S-phase proteins and CycE DNA biosynthesis in the cell cycle pathway ( Figure 5E). Thus, resveratrol-targeted genes exerted biological effects primarily through the p53 signaling pathway. p53 inhibits cancer development and progression via several mechanisms, including apoptosis, regulation of DNA replication, cell division, and inhibition of angiogenesis [14,15]. The p53 protein is encoded by the TP53 gene, which was identified in this study as the hub gene with the highest degree of interaction in the network. TP53, CTNNB1, and SP1 modulate the expression of most of the differentially expressed genes that are upregulated and play important roles in primary osteoporosis [16]. Fu Jia et. al. demonstrated that pri-miR-34b/c rs4938723 and TP53 Arg72Pro polymorphisms may contribute to the risk of osteoporosis [17]. In this study, both qRT-PCR and western blots indicated that p53 was enriched in osteoporosis ( Figure 6A, 6B). In addition, qRT-PCR and Alizarin-red staining showed that MDM2-mediated inhibition of p53 induced osteoblast differentiation in vitro ( Figure 6C-6G), indicating that p53 promoted the pathological progression of osteoporosis.
Resveratrol (3,5,4′-trihydroxy-trans-stilbene) is a natural polyphenolic compound found in several plants [5,6]. Accumulating evidence shows that resveratrol has anti-inflammatory, antioxidant, and other protective effects in osteoporosis and in aging-induced cognitive impairment [7][8][9]. Ali Mobasheri and Mehdi Shakibaei AGING reported that resveratrol can modulate bone cell metabolism and bone turnover due to its osteogenic and osteoinductive properties [18,19]. In this study, resveratrol partially reversed p53-induced inhibition of osteogenic differentiation in in vitro experiments (Figure 7). These results indicate that resveratrol may protect against osteoporosis by inhibiting the p53 signaling pathway.
Some limitations in this study should be considered when interpreting the results. Firstly, the effects of different durations of resveratrol treatment were not investigated. Furthermore, potential differences in p53 enrichment in different osteoporosis subtypes were not examined. Finally, additional studies are needed to compare the effects of resveratrol on osteoporosis development and progression with those of other drugs.

AGING
In conclusion, bioinformatics analysis revealed that the protective effects of resveratrol against osteoporosis were associated with its interaction with the prostate cancer pathway, pathway in cancer, glioma pathway, p53 signaling pathway, and cell cycle signaling pathway. Our in vitro experiment further indicated that resveratrol exerts antiosteoporosis effects by inhibiting the p53 signaling pathway, and may thus serve as a promising osteoporosis treatment.

Construction of the resveratrol-targeted genes interaction network
The STITCH database integrates data on the effects of over 430000 chemicals on gene expression [20]. Resveratrol-targeted genes were identified using the default settings in STITCH, and the STITCH online tool was used to construct an interaction network for resveratrol and its targeted genes. Degree, betweenness, and closeness were analyzed for each gene in the network and visualized using Cytoscape 3.7.2. Degree represents the extent to which one node is associated with all the other nodes in the network, closeness represents how far one node is from other nodes in the network, and betweenness is the number of times a node acts as the shortest bridge between two other nodes. These measures for all resveratrol-targeted genes were then imported into Gephi software and a weighted interaction network was constructed.

Shared KEGG pathways involved both in osteoporosis and resveratrol-targeted genes
miRWalk2.0, a comprehensive archive containing the largest existing collection of predicted and experimentally verified miRNA-target gene interactions [21], was used to identify human osteoporosisassociated KEGG pathways. DAVID, an integrated biological knowledgebase with analytic tools for systematically extracting biological meaning from large gene/protein lists [22,23], was used to identify KEGG pathways enriched in resveratrol-targeted genes. KEGG pathways with p<0.05 were selected for subsequent analysis. KEGG pathways enriched in both osteoporosis and resveratrol-targeted genes were visualized using a Venn diagram (Venny 2.1, http://bioinfogp.cnb.csic.es/ tools/venny/index.html).

Identification of hub genes
Enrichment information for the top five KEGG pathways was presented with GOplot, an R package that visually combines expression data with functional analysis [24]. Genes involved in all top five shared KEGG pathways were considered hub genes. Centrality in the network and chromosome position for all resveratrol-targeted genes were visualized using the circlize package for R [25].

Retrieval of KEGG pathways related to resveratroltargeted genes
The top five shared KEGG pathways with the smallest p values were selected, and KEGG pathways related to resveratrol-targeted genes were established, using Pathway Builder Tool 2.0 (https://www.proteinlounge.com). Hub genes and their mechanisms of action are shown in the schematic diagrams.

Cell culture and transfection
BMSCs were kindly donated by the Huazhong University of Science and Technology, Wuhan, China. Cells were grown in a specific media designed for C57BL/6 mouse mesenchymal stem cells (#MUBMX-03011-440, Cyagen, Guangzhou, China) at 37°C in a 5% CO2 incubator. Cells were maintained for a maximum of 3 passages. Lipofectamine 3000 (#L3000001, ThermoFisher Scientific, USA) was used to transfect cells with siRNAs according to provided directions. MDM2 siRNA constructs (RIBOBIO, Guangzhou, China) were transfected at 50 nM. Plasmid NC and plasmid MDM2 were synthesized by GenePharma company (Shanghai, China). All in vitro experiments using cells were independently repeated three times.

qRT-PCR analysis
Total tissue/cell RNA was extracted using Trizol reagent (#15596018, Invitrogen, USA). RNA was reverse transcribed using the Verso TM cDNA Synthesis Kit (#AB-1054/A, ThermoFisher Scientific), and qRT-PCR reactions were performed using a Thermal Cycler C-1000 Touch system (#10021377, Bio-Rad CFX Manager, USA). GAPDH was used as a housekeeping gene. Target gene expression was quantified using the ΔΔCT method.
The primers used in this study are listed in Table 2.

Alizarin red staining
BMSCs were grown in 6-well plates in media containing 100 nM dexamethasone, 50 mM ascorbic AGING  Subsequently, 1 mL 0.5% alizarin red staining solution was used to stain the cells at room temperature for 15 minutes. After rinsing with distilled water for 5 minutes, red mineralized nodules were analyzed via a charge-coupled device microscope. Absorbance was measured at 570 nm. Experiments were repeated in triplicate.

Statistical analysis
Data are shown as means ± SD. GraphPad Prism 8.0 (GraphPad Software, CA, USA) was used to perform statistical analyses. One-way analysis of variance (ANOVA) with Tukey's post hoc test was used to compare three or more groups. Unpaired Student's t-test was used to compare data between two groups. P < 0.05 was considered statistically significant. The study schema is shown in Figure 8.

AUTHOR CONTRIBUTIONS
YY and YX conceived of and designed the study; GY, HZ, and BL supervised the study; TY, ZW, and XY performed bioinformatics analysis and experiments; HZ and WH analyzed the data; JX and YZ provided advice and technical assistance; TY, ZW, and XY wrote the manuscript. All authors approved the final manuscript.

CONFLICTS OF INTEREST
The authors declare no conflicts of interest.

FUNDING
This study was supported by the National Science Foundation of China (No.31600754, NO.81472144).