MicroRNA‐92b‐5p modulates melatonin‐mediated osteogenic differentiation of bone marrow mesenchymal stem cells by targeting ICAM‐1

Abstract Osteoporosis is closely associated with the dysfunction of bone metabolism, which is caused by the imbalance between new bone formation and bone resorption. Osteogenic differentiation plays a vital role in maintaining the balance of bone microenvironment. The present study investigated whether melatonin participated in the osteogenic commitment of bone marrow mesenchymal stem cells (BMSCs) and further explored its underlying mechanisms. Our data showed that melatonin exhibited the capacity of regulating osteogenic differentiation of BMSCs, which was blocked by its membrane receptor inhibitor luzindole. Further study demonstrated that the expression of miR‐92b‐5p was up‐regulated in BMSCs after administration of melatonin, and transfection of miR‐92b‐5p accelerated osteogenesis of BMSCs. In contrast, silence of miR‐92b‐5p inhibited the osteogenesis of BMSCs. The increase in osteoblast differentiation of BMSCs caused by melatonin was attenuated by miR‐92b‐5p AMO as well. Luciferase reporter assay, real‐time qPCR analysis and western blot analysis confirmed that miR‐92b‐5p was involved in osteogenesis by directly targeting intracellular adhesion molecule‐1 (ICAM‐1). Melatonin improved the expression of miR‐92b‐5p, which could regulate the differentiation of BMSCs into osteoblasts by targeting ICAM‐1. This study provided novel methods for treating osteoporosis.


| INTRODUC TI ON
Millions of humans in the ageing populations, especially post-menopausal women suffer from osteoporosis throughout the world. 1 Osteoporosis is a progressive systemic skeletal disease which is characterized by low bone mass and microarchitecture deterioration of bone tissue, leading to reduced bone mineral density (BMD) and elevated risk of fractures. 2 Osteoporosis is attributed to the dysfunction of bone metabolism, which is caused by the imbalance between new bone formation by osteoblasts and old bone resorption by osteoclasts. Osteoblasts in the bone environment are mainly derived from bone marrow mesenchymal stem cells (BMSCs) which have the capacity to differentiate into various cell types, including osteoblasts, chondrocytes and adipocytes. 3,4 Thus, BMSCs, with inherent osteogenic differentiation potential, have been widely used for cell therapy and tissue regeneration based on their multilineage differentiation potential and easy access. 5 The dysfunction of BMSCs because of pathological stimuli or gene mutations is also involved in the occurrence and development of osteoporosis. Some studies uncovered that BMSCs engraftment may promote osteogenesis and produce therapeutic effects on bone fractures in animals. 6,7 Thus, how to effectively promote osteogenic differentiation of BMSCs has been regarded as an important strategy for osteoporosis treatment.
Melatonin (N-acetyl-5-methoxytryptamine) is a neurohormone produced and secreted by the pineal gland, and pineal synthesis is primarily regulated by the light and dark environment working via suprachiasmatic nucleus. 8 Melatonin plays a critical role in a wide variety of physiological functions, such as circadian rhythms, renal function, hormone secretion, antioxidant defence, immune responses, reproduction control, lipid metabolism, tooth development, anti-tumour capacity and etc. [9][10][11][12] In addition, accumulating evidence has revealed that melatonin is also involved in various diseases, including periodontal disease, cancer, diabetes, pulmonary fibrosis, spinal cord injury, Alzheimer's disease and hepatic fibrosis. 13,14 Recently, the association between melatonin and bone development has received increased attentions. For example, melatonin accelerates osteoblast differentiation of pre-osteoblastic MC3T3 cell lines and improves the bone-forming capacity under hypoxic condition. 15 Furthermore, treatment with melatonin stimulates matrix mineralization and promotes new bone formation of mice. 16 Our previous study suggested that melatonin may reverse iron overload-induced dysfunction of BMSCs. 17 However, little is known regarding the mechanisms of melatonin promoting the osteogenic differentiation of BMSCs, which need further investigations.
MicroRNAs (miRNAs) are a family of abundant, endogenously expressed, double-stranded small non-coding RNAs of about 20-25 nucleotides. 18,19 MiRNAs function as key regulators of gene expression by binding to the 3'-UTR of their target mRNAs and play a critical role in a wide variety of biological processes, including cell proliferation, cell apoptosis, cell fate and organ development. 20,21 In addition, miRNAs participate in multiple diseases, such as cancers, diabetes mellitus, heart failure, liver injury, cerebrovascular disease and glaucoma. [22][23][24] Furthermore, recent studies have reported that miRNAs are related to bone metabolism and bone regeneration. Research has shown that 10-11 translocation (Tet) plays a key role in maintaining BMSCs and bone homeostasis by controlling miR-297a-5p, miR-297b-5p, and miR-297c-5p release. 25 Besides, many studies have revealed that miRNAs play an important role in the melatonin-mediated biological functions. [26][27][28][29] For example, it has been reported that melatonin successfully accelerates the chondrogenic by increasing the expression of miR-526b-3p and miR-590-5p by regulating SMAD7. 30 Nevertheless, the information about the roles of miR-NAs in the protection of melatonin on osteogenesis of BMSCs has not been fully explored. 9 Herein, we demonstrated that melatonin has a role in osteogenesis of BMSCs, and melatonin regulates osteoblast differentiation by increasing the expression of miR-92b-5p which directly targets intracellular adhesion molecule-1 (ICAM-1). The findings suggest that melatonin may be an effective therapeutic treatment for osteoporosis.

| BMSC isolation and culture
injected with pentobarbital sodium (Sigma, USA) and anaesthetized, and then killed. The femur and tibias were separated and BMSCs were flushed from bone marrow cavity using 4 mL volume of normal culture medium (Cyagen, USA). And these cells were regarded as passage 0 (P0). BMSCs were cultured and maintained in 25 cm 2 culture flasks and maintained in a cell culture incubator containing 5% CO 2 and 95% humidity at 37°C (Thermo, USA).

| Transfection
To identify the role of miRNAs in osteogenic differentiation of BMSCs, miR-92b-5p mimics and miR-92b-5p AMO were designed and synthesized by GenePharma, China. Their corresponding negative controls (NC) were also provided by the manufacturer.

| Osteogenesis induction
To induce osteogenic differentiation of BMSCs, the cells were cultured in 6-well or 24-well plates under standard culture conditions of 37°C and 5% CO 2 . The cells were maintained in normal culture medium (Cyagen, USA) until 80% confluence and then they were differentiated into osteoblasts using osteogenic differentiation-inducing medium (Cyagen, USA). Osteogenic differentiation-inducing medium was composed of 175 mL culture medium, 10% FBS, 1% glutamine, 1% penicillin-streptomycin, 0.2% ascorbic acid, 1% β-glycerophosphate and 0.01% dexamethasone. The cells were induced into osteoblasts for 14 days and the medium was changed every 3 days.

| Melatonin treatment
Melatonin was purchased from Sigma (USA) and 23.228 mg melatonin was totally dissolved in 10 mL double distilled water. The concentration of melatonin at 10 mmol/L was regarded as the working solution used for further experiments. For the analysis evaluating the role of melatonin, the cells were grown in the presence of differentiation-inducing medium combined with different concentration of melatonin for 14 days. The cells without melatonin treatment were regarded as the controls.

| Alizarin red S staining
To examine the activity of osteoblasts, the BMSCs were placed in 24well plates and were induced into osteoblasts. Mineralization of cells was assessed on day 14 by performing alizarin red S (ARS) staining and the cells were washed with PBS gently three times. Then the cells were embedded in 4% PFA (Solarbio, China) for 30 minutes at room temperature. Then the cells were incubated using 1 mL ARS staining solution which was purchased from Cyagen (USA) for 30 minutes at room temperature. ARS staining solution chelated with calcium of cells and then formed an alizarin red-calcium complex which exhibited a bright red colour. After three washes with PBS to remove needless unbound stains, the degree of osteogenic differentiation was observed under an optimal microscope (Nikon, Japan). The experiment was performed at least three times.

| Alkaline phosphatase staining
To assess the calcium deposition by osteoblasts, BMSCs were cultured in osteogenic differentiation-inducing medium for 14 days. After osteogenic differentiation, the cells were stained with alkaline phosphatase

| RNA isolation, cDNA synthesis and realtime qPCR
To evaluate the expression level of osteogenic marker genes and miR-92b-5p, we performed real-time qPCR analysis according to the protocol. In brief, total RNA was extracted from BMSCs after treatment with melatonin or miRNAs by using TRIzol reagent (Invitrogen, USA). cDNA was converted from 500 ng of total RNA using reverse  Table 1.

| Western blot
Western blot was conducted to examine the protein expression of target genes, including ICAM-1. Briefly, BCA protein assay kit (Beyotime, China) was used to quantify the protein samples. BMSCs were grown in 6-well plates at an initial density of 1 × 10 5 cells/cm 2 and transfected with miRNAs. Then, after osteoblastic differentiation induced by OM for additional 14 days, the cells were harvested. Cell pellets were obtained after centrifugation at 15000 rpm, washed with PBS and mixed with ice-cold lysis buffer as described in a previous study. 34,35 Then cellular extracts were shifted to SDS-PAGE and transferred to a PVDF membrane (Millipore, USA). The membranes were blocked with 5% non-fat milk with gentle shaking for 2 hours. After washing with TBST for three times, the membranes were incubated with specific primary antibodies overnight with a gentle shaking at 4°C. Subsequently, the membranes were incubated with the secondary antibody for 1 hour at room temperature. After washing, strength of each membrane was detected by an ImageJ software (National Institutes of Health, USA) using Odyssey Infrared Imager. Anti-ICAM-1 (CST, USA) was purchased from Omnimabs. The expression level of β-actin was used as the reference.

| Establishment of OVX-induced osteoporotic mice
To analyse the potential of osteogenesis of BMSCs in osteoporotic mice, 8-week-old female C57BL/6J mice were purchased to construct an ovariectomy-induced osteoporotic mouse model based on established protocol. 36 All the work was performed in compliance with the guidelines of the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals. The mice were randomly divided into Sham group or OVX group. After an intraperitoneal injection of pentobarbital sodium, the mice underwent bilateral ovariectomy (OVX) or sham operation (Sham) respectively. Successful OVX models were defined as osteoporotic mice in contrast to the Sham group. After additional 4 weeks of feeding, the mice were killed and BMSCs were isolated from tibias and femurs as mentioned above. The cells from the two groups were cultured for osteogenic potential analysis.

| Immunofluorescence staining
To identify the osteoblast activity in response to miRNA treatments, the protein expression of Runx2 was measured by immunofluorescence staining. In short, BMSCs were seeded in glass slides

| Luciferase reporter assay
The candidate targets of miR-92b-5p were predicted by TargetScan

| Statistical analysis
All experiments were repeated three times independently. All statistical analysis was performed by GraphPad Prism 5.0 software (GraphPad Software, USA). P < 0.05 was regarded statistically significant.

| Melatonin promotes the osteogenic differentiation of BMSCs
In our previous study, we found that melatonin protected BMSCs against iron overload-induced inhibition of osteogenic differentiation, but the underlying mechanism was not characterized. 17 In  Figure 1F). Importantly, 10 μmol/L melatonin-treated BMSCs achieved a peak in the expression levels of master genes during osteogenesis ( Figure 1F). Melatonin at 10 μmol/L therefore was regarded as the effective concentration and used for further analysis.

| Beneficial effect of melatonin on the osteogenesis was blocked by luzindole
To determine if melatonin promoted the osteogenic differentiation in a melatonin receptor-dependent manner, we further tested the effects of luzindole (Luz), a known receptor antagonist of melatonin on melatonin-induced osteogenic effect. We treated BMSCs with NM, OM, OM + 10 μmol/L melatonin, OM + 10 μmol/L melatonin + 10 μmol/L luzindole respectively.
After differentiation of 14 days, ARS staining and ALP staining demonstrated that OM significantly enhanced the capacity of osteogenic differentiation of BMSCs, which was further promoted by 10 μmol/L melatonin treatment (Figure 2A,B).
However, as shown in Figure 2A,B, the increase in the osteogenesis caused by melatonin was blocked by 10 μmol/L luzindole, suggesting melatonin receptor involvement (Figure 2A,B).
In addition, our data suggested that BMSCs in the presence of 10 μmol/L melatonin exhibited a marked increase in the formation of calcium nodules, which sharply declined in luzindole-treated cells (Figure 2A,B). Moreover, the obtained results uncovered that the expression level of markers for osteoblast was obviously increased by melatonin at a concentration of 10 μmol/L compared with control group, which was visibly reduced in the presence of luzindole ( Figure 2C). Thus, the osteoblast formation potential of BMSCs caused by melatonin was blocked by luzindole, suggesting that the melatonin receptor was involved in the process. its negative control ( Figure 3E). Also, immunofluorescent staining revealed that the stimulatory effect of miR-92b-5p mimics on osteoblast differentiation was consistent with the results above ( Figure 3F). Therefore, the data suggested that miR-92b-5p promoted the differentiation of BMSCs into mature osteoblasts.

| Silencing of miR-92b-5p inhibits the osteogenesis of BMSCs
It was interesting to note that miR-92b-5p played a vital role in the differentiation of BMSCs into osteoblasts. Thus, we further investigated the role of miR-92b-5p AMO in the committed differentiation of BMSCs. MiR-92b-5p AMO and its NC were transfected into BMSCs respectively. As expected, BMSCs transfected with miR-92b-5p AMO exhibited fewer calcium nodules and a reduced number of osteoblasts in contrast with NC after osteogenic differentiation of 14 days ( Figure 4A,B). Consistently, the decline in the expression levels of osteogenic-related genes in BMSCs after exposure to miR-92b-5p AMO was quantified by real-time qPCR analysis ( Figure 4C). Immunofluorescence staining also revealed that knockdown of miR-92b-5p obviously reduced the expression of Runx2, a  Figure 4D). Together, the above results showed that miR-92b-5p AMO treatment led to decreased osteoblast differentiation of BMSCs.

| Osteoblast differentiation elevated by melatonin was abrogated by miR-92b-5p AMO
To clarify the mechanisms as to how melatonin regulated osteogenesis by increasing the expression of miR-92b-5p, we treated BMSCs with 10 μmol/L melatonin and miR-92b-5p AMO. After a 14-day osteogenic induction, the size and intensity of calcium deposition were markedly higher in the OM-induced group than that in the NM-treated group ( Figure 5A,B). The osteoblast differentiation was increased after treatment with 10 μmol/L melatonin, while knockdown of miR-92b-5p delayed the osteogenesis as revealed by ARS staining ( Figure 5A). Meanwhile, we found that miR-92b-5p AMO resulted in decreased number and area of mineralized nodules which were elevated by 10 μmol/L melatonin as observed in ALP staining ( Figure 5B). According to real-time qPCR analysis, inhibition of miR-92b-5p led to the down-regulation of  Figure 5C). These results clearly documented that melatonin accelerated osteoblast differentiation, which was reversed by miR-92b-5p AMO in BMSCs.
These results suggested that a close association existed between the expression of miR-92b-5p and the role of melatonin in the osteogenesis of BMSCs. Thus, the above data confirmed that melatonin regulated the osteogenesis through enhancing the expression of miR-92b-5p.

| miR-92b-5p restored osteogenesis of BMSCs from OVX-induced osteoporotic mice
To detect the role of miR-92b-5p in BMSCs from osteoporotic mice, we established Sham and OVX-induced osteoporotic mice. The osteoporotic models were successfully conducted ( Figure 6A,B). Based on the above results, we further determined the role of miR-92b-5p in the osteoblast differentiation of BMSCs isolated from OVX-induced osteoporotic mice. First, we identified that miR-92b-5p was differentially expressed in mice between Sham group and OVX group. The expression of miR-92b-5p was distinctly decreased in BMSCs and bone tissues from osteoporotic mice ( Figure 6C). Therefore, we supposed that miR-92b-5p might be involved in the development of osteoporosis. ARS staining and ALP staining were performed to assess the effect of miR-92b-5p on the potential of osteogenic differentiation in BMSCs of mice from the Sham and OVX groups. ARS staining indicated that BMSCs from osteoporotic mice displayed fewer calcium deposits than the control group ( Figure 6D). However, the decrease in the osteogenic differentiation caused by OVX surgery was elevated by miR-92b-5p transfection ( Figure 6D). Moreover, as shown in ALP staining, osteogenesis was thoroughly impaired in OVX-induced osteoporotic mice compared with Sham group ( Figure 6E).
However, the sharp decline in the osteogenic differentiation of BMSCs from OVX mice was restored by miR-92b-5p mimics ( Figure 6E). Comparisons in the protein expression of Runx2 between these two groups were conducted using immunofluorescence analysis. As shown, the number of the Runx2-positive BMSCs was visibly decreased in the OVX-induced group compared with Sham group ( Figure 6F). In addition, the percentage of Runx2-positive cells was elevated by miR-92b-5p transfection ( Figure 6F). These results revealed that miR-92b-5p restored the attenuation in osteogenesis of BMSCs which was caused by osteoporosis, and the therapeutic increase in miR-92b-5p in BMSCs

| miR-92b-5p was involved in the osteogenesis by directly targeting ICAM-1
To explore the underlying mechanism related to the function of miR-92b-5p in the osteogenesis of BMSCs, we predicted the possible target genes of miR-92b-5p by TargetScan online. ICAM-1 was suggested as the candidate gene for miR-92b-5p because of the highpotential binding sites between miR-92b-5p and ICAM-1 ( Figure 7A).
Meanwhile, it has been uncovered that ICAM-1 plays a critical role in osteogenesis. As shown in Figure 7B, the luciferase reporter assay revealed that the relative luciferase activity was down-regulated in cells treated with miR-92b-5p mimics and ICAM-1, while the association was completely abrogated when the 3′-UTR of ICAM-1 was mutated.
Furthermore, real-time qPCR and Western blot analysis documented that miR-92b-5p mimics effectively reduced the expression levels of ICAM-1, while miR-92b-5p AMO elevated the expression of ICAM-1 ( Figure 7C,D). The above results suggested that miR-92b-5p participated in the osteogenesis of BMSCs by directly targeting ICAM-1.

| D ISCUSS I ON
In this study, we revealed that melatonin regulates the osteogenic differentiation through a melatonin receptor pathway. Furthermore, we found that melatonin accelerates the osteogenesis by improving the expression of miR-92b-5p and, miR-92b-5p enhances the differentiation into mature osteoblasts by targeting ICAM-1. This study provides more insights into the role and mechanism of miRNAs in regulating osteogenesis of BMSCs.
Osteoporosis is a severe progressive systemic skeletal disease which causes humpback, osteodynia, thoracocyllosis and especially fragility fractures. 37 Osteoporosis is caused by the imbalance between osteoclast-mediated bone resorption and osteoblast-induced bone formation of osseous tissues. The present attempts to treat osteoporosis have focused on exercising, calcium supplement and F I G U R E 5 MiR-92b-5p AMO reversed melatonin-induced promotion of the differentiation into osteoblasts.
(A and B) The increased in osteogenesis by 10 μmol/L Mel was inhibited by miR-92b-5p AMO transfection. Scale bar = 100 μm. (C) The expression of osteoblast marker genes was up-regulated by melatonin treatment, but reversed by the knockdown of AMO. *P < 0.05, **P < 0.01 and ***P < 0.001 compared with NC vitamin D administration. 38 Besides these existing treatment methods, new strategies to effectively prevent the development of osteoporosis by improving bone formation are always urgently required.
Melatonin is a major indoleamine produced and secreted by pineal gland; melatonin exhibits a variety of biological actions. 39,40 Recently, melatonin has been reported to participate in the differentiation of BMSCs. 41  cells, miR-34a in neurons were proved to be regulated by melatonin treatment. 27,35,44 There are, however, no reports demonstrating an association between melatonin-induced osteogenesis and miRNAs as an underlying mechanism. Therefore, we further tested whether melatonin regulates osteoblastic differentiation of BMSCs by modulating the expression of miRNAs. In the present study, we treated BMSCs with different concentrations of melatonin and found that 10 μmol/L was the most effective concentration for enhancing the capacity of osteoblast differentiation of BMSCs. We found that the expression of miR-92b-5p was significantly increased after treatment with 10 μmol/L melatonin. Previous studies have showed that miR-92b-5p was related to acute heart failure, infant respiratory virus infection, coronary microembolization and early-onset atrial fibrillation, but there are no reports related to the role of miR-92b-5p in the BMSC fate. [45][46][47] Our data showed that miR-92b-5p facilitates osteoblast differentiation of BMSCs and reverses the osteoporosis-induced impaired osteogenic potential of BMSCs isolated from OVX-induced mice. Further investigation revealed that miR-92b-5p exhibited a positive role in the osteogenic differentiation of BMSCs via modulating ICAM-1 which been demonstrated to play a role in the osteogenic differentiation and bone regeneration in the previous studies. 48 ICAM-1 has been reported to inhibit the osteogenesis of BMSCs and serve as a new molecular target to accelerate bone regeneration and repair in inflammatory microenvironments. 49 According to a randomized controlled trial, melatonin treatment increases BMD and integrity of the femoral neck in postmenopausal women with osteopenia. 50 In that report, 1-3 mg melatonin per day increases BMD of femoral neck in a dose-dependent manner compared with placebo treatment, as BMD increased by 0.5% in the 1 mg/d group and by 2.3% in the 3 mg/d group. Besides, it has been reported that trabecular thickness in tibia was improved by 2.2% and BMD in the spine was increased To the best of our knowledge, the present study is the first to provide evidence about the correlation between melatonin and miRNAs during osteoblast differentiation in BMSCs. Melatonin may regulate osteoblastic differentiation of BMSCs in a melatonin receptor-dependent manner. Furthermore, our findings suggest that miR-92b-5p restores the osteoporosis-induced decline in the osteogenic potential of BMSCs. This study helps to clarify the effects of melatonin in the osteogenic differentiation and its mechanism. These findings open up new possibilities for melatonin use as an effective strategy to maintain bone metabolism in patients with osteoporosis.
In conclusion, we found that melatonin stimulates osteogenesis of BMSCs by increasing the expression of miR-92b-5p which directly targets at ICAM-1. Thus, the present study suggests a possible application of melatonin as a therapeutic agent for osteoporosis.