Moderate SMFs attenuate bone loss in mice by promoting directional osteogenic differentiation of BMSCs

Background Osteoporosis is a common metabolic bone disease without effective treatment. Bone marrow-derived mesenchymal stem cells (BMSCs) have the potential to differentiate into multiple cell types. Increased adipogenic differentiation or reduced osteogenic differentiation of BMSCs might lead to osteoporosis. Whether static magnetic fields (SMFs) might influence the adipo-osteogenic differentiation balance of BMSCs remains unknown. Methods The effects of SMFs on lineage differentiation of BMSCs and development of osteoporosis were determined by various biochemical (RT-PCR and Western blot), morphological (staining and optical microscopy), and micro-CT assays. Bioinformatics analysis was also used to explore the signaling pathways. Results In this study, we found that SMFs (0.2–0.6 T) inhibited the adipogenic differentiation of BMSCs but promoted their osteoblastic differentiation in an intensity-dependent manner. Whole genomic RNA-seq and bioinformatics analysis revealed that SMF (0.6 T) decreased the PPARγ-mediated gene expression but increased the RUNX2-mediated gene transcription in BMSCs. Moreover, SMFs markedly alleviated bone mass loss induced by either dexamethasone or all-trans retinoic acid in mice. Conclusions Taken together, our results suggested that SMF-based magnetotherapy might serve as an adjunctive therapeutic option for patients with osteoporosis. Supplementary information Supplementary information accompanies this paper at 10.1186/s13287-020-02004-y.


Background
Osteoporosis is a common bone and metabolic disease, characterized by reduced bone formation and increased fat accumulation in the bone marrow space, leading to decreased bone mineral density (BMD) and increased bone fractures [1]. As such, it is reported to be responsible for more than 1.5 million fractures worldwide annually [2]. Both adipose-derived stem cells (ASCs) and bone marrow-derived mesenchymal stem cells (BMSCs) are multipotent progenitor cells with the potential to differentiate into various mature cell types, including osteoblasts, chondrocytes, and adipocytes [3,4], which have been utilized extensively in the research field of bone tissue regeneration. Despite higher proliferative capacity for ASCs, BMSCs display more potential for osteogenic and chondrogenic differentiation [5,6]. Disrupting the dynamic balance between adipogenic and osteogenic differentiation of BMSCs, such as increased differentiation toward adipocytes or reduced differentiation toward osteoblasts, has been shown to cause osteoporosis and other bone and metabolic diseases [7,8]. As pleiotropic drugs, metformin and rapamycin have been reported to promote bone generation and improve bone density by increasing osteogenic differentiation potential of BMSCs [9,10]. Moreover, physical activities have also been found to induce mobilization of stem cells and promote bone regeneration [11,12]. However, the mechanisms behind the fine-tuned regulation of the commitment of the BMSC lineage to differentiate to osteoblasts versus adipocytes remain elusive. Humans are exposed to the naturally occurring magnetic fields of the earth. Lack of exposure to natural magnetic fields, such as staying in space, has been reported to cause insomnia, fatigue, depression, and increased predisposition for osteoporosis in humans [13], suggesting that the magnetic field might be beneficial for physiological function and human health. Accumulated evidence has shown that static magnetic fields (SMFs) might act on a variety of potential targeting organs and tissues, exerting favorable physiological effects on many biological systems, including suppression of inflammatory reactions [14], reduction of edema formation [15], improvement of microcirculation and blood flow [16,17], relief of osteoarthritis-induced pain [18], and facilitation of wound healing [19]. Therefore, magnetotherapy has been officially approved by the US Food and Drug Administration (FDA) for the treatment of pain and edema in superficial soft tissues for orthopedic applications [20]. Moreover, moderate-intensity SMF has been shown to enhance repair after cartilage damage and accelerate the formation of new bone tissue in rat models, by promoting extracellular matrix deposition [21,22]. SMF treatment has been reported to increase BMD of osteoporotic lumbar vertebrae in ovariectomized rats [23] and prevent the architectural deterioration and strength reduction of bones in streptozotocin-treated diabetic rats by enhancing osteogenic differentiation [24]. These observations strongly suggest that SMF might help prevent and treat osteoporosis. However, whether SMFs might participate in the reciprocal regulation between adipocyte and osteoblast differentiation of BMSCs and the subsequent control of the adipoosteogenic balance remains to be determined.
In the present study, we stimulated BMSCs during adipogenic and osteogenic differentiation with different intensities of SMFs (0 T, 0.2 T, 0.4 T, and 0.6 T), and found SMFs promoted osteoblastic differentiation, but inhibited adipogenic differentiation of BMSCs in an intensity-dependent manner. Whole genomic RNA-seq analysis demonstrated SMFs decreased PPARγ-driven gene expression while increased RUNX2-driven gene transcription in BMSCs. Moreover, SMFs markedly alleviated bone mass loss induced by either dexamethasone (Dex) or all-trans retinoic acid (ATRA) in mice. Thus, our study reveals moderate intensity SMFs may serve as an adjunctive therapeutic option for osteoporosis.

Animals
Wild-type C57BL/6J mice were purchased from Gem-Pharmatech Co. Ltd. and were housed in specific pathogen-free animal facilities of the Tianjin Medical University. Accordingly, 8-10-week-old male mice were used in this study. All animal experiments were performed in accordance with the Guidelines of the Institutional Animal Care and Use Committee of Tianjin Medical University.

Moderate SMF exposure system
For the creation of the SMF exposure system, gradient permanent neodymium magnets (130 × 110 × 60 mm) were assembled beneath the cell plates to expose the cultures to north fields (N; Fig. 1a). The distribution of the representative SMF (0.6 T), measured using a digital Tesla meter (HT20), was shown in Fig. 1b, c. For mouse exposure, the magnetic or nonmagnetic sandwich plates (230 × 130 × 15 mm), in which 24 magnets (10 mm diameter and 15 mm thickness) were inserted with alternating magnetic poles (north or south; N or S) facing up, were placed beneath the mouse cages and separated by spaced holes (Fig. 1d) as previously reported [19,25]. The SMF intensity distribution of control and 0.6 T magnetic sandwich plates were shown in Fig. 1e and Fig. 1f, respectively, as measured by a space magnetic field measuring instrument (FE-2100RD).

Cell culture
BMSCs were isolated from femoral bones of male mice (4-6 week-old), as previously reported [26,27]. At passage 3, BMSCs were subjected to purification by flow cytometry through the identification of CD29, CD44, CD90, CD105, and Sca-1 positive surface markers and CD11b, CD34, CD45, and c-kit negative surface markers, as shown in sFig. 1. Then, BMSCs were subjected to adipogenic and osteogenic differentiation using the respective induction media.

Osteogenic differentiation assay
BMSCs were grown in basal medium until appropriate cell confluence (approximately 70%). For osteogenic differentiation, BMSCs were cultured in osteogenic induction medium containing 50 mM ascorbate-2phosphate, 0.1 mM dexamethasone, and 10 mM βglycerol phosphate for 21 days. The culture medium was changed every third day. After osteogenic differentiation for 3 days, cultured BMSCs were fixed with 70% (v/v) ethanol and incubated in 0.25% (w/v) naphthol AS-BI phosphate solution and 0.75% (w/v) Fast Blue BB in 0.1 M Tris buffer before alkaline phosphatase (ALP) staining. A commercial ALP activity kit (Sigma-Aldrich) was used to assess the ALP activity of differentiated cells according to the manufacturer's instructions, following normalization by total cell protein. At the end of differentiation, alizarin red S (ALS) staining was performed to evaluate the mineralization of the cell matrix.

Histological analysis
Paraffin-embedded femurs were sectioned (5 μm), deparaffinized, and then stained with hematoxylin and eosin (H&E) for morphological analysis. For immunohistochemistry staining, deparaffinized and dehydrated tissue sections were re-hydrated before being subjected to antigen retrieval, and then blocked with diluted normal serum for at least 1-1.5 h at 25°C to eliminate nonspecific binding. The slides were incubated with primary antibodies against osteocalcin (OCN) (1:300; Abcam) overnight at 4°C. After careful washing, the sections were incubated with horseradish peroxidase conjugates to detect positive signals, followed by counterstaining with hematoxylin (Sigma-Aldrich). Slides incubated with polyclonal rabbit IgG (Abcam) served as negative controls. Pictures were captured and monitored using a Leica Microsystems microscope (Leica Microsystems Ltd.), while the ImageJ software (National Institutes of Health) was used to analyze the number or area of adipocytes and osteoblasts.

Real-time quantitative PCR
Total mRNA was extracted from BMSCs grown in adipogenic or osteogenic medium using TRIzol reagent (Invitrogen) and reverse-transcribed to cDNA using a Reverse Transcription Reagent kit (Takara Bio Inc.) according to the manufacturer's protocol. The resulting cDNA was amplified using the Real-Time PCR system (LightCycler 480 II, Roche) with 40 cycles. Sequences of primers for target genes are presented in Supplementary  Table 1. The mRNA levels of the specific genes were normalized to that of a reference gene (β-actin) within the samples.

RNA-seq
BMSCs plated in 10 cm dishes were exposed on a control iron plate (0 T) or a magnetic plate (0.6 T) for 48 h respectively. Then, BMSCs were washed by cold PBS for 3 times and then lysed by using Trizol at room temperature. Samples were sent to Novogene on dry ice for library preparation and sequencing. A total amount of 3 mg RNA per sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext UltraTM RNA Library Prep it for Illumina (NEB, USA) following manufacturer's recommendations and index codes were added to attribute sequences to each sample. The clustering of the indexcoded samples was performed on a cBot Cluster Generation System using TruSeq PE Cluster Kit v3-cBot-HS (Illumia) according to the manufacturer's instructions. After cluster generation, the library preparations were sequenced on an Illumina platform and 150 bp pairedend reads were generated. R software version GSEA and package fgsea were used for cellular pathway analysis including Hallmark gene sets and GO gene sets.

Microcomputed tomography (micro-CT) scan
Femurs of the same side were dissected from mice treated with ATRA for 21 days [28] or treated with Dex for 3 months [29] and fixed with 4% paraformaldehyde for more than 24 h. Then, the samples were scanned and analyzed using a micro-CT System (Skyscan 1172, Bruker) with a high resolution (voltage 50 kV; current 201 μA; resolution 12 mm/pixel), as reported [30]. The NRecon image reconstruction software (Bruker), CTVol 3D model visualization software (Bruker), and CTAn data analysis software (Bruker) were used to analyze the parameters of the trabecular bones. For distal femurs, regions of interest were selected for the analysis of BMD, trabecular bone volume per tissue volume (Tb.BV/TV), trabecular separation (Tb.Sp), trabecular number (Tb.N), and trabecular thickness (Tb.Th).

Statistical analysis
Data were analyzed using GraphPad Prism version 6.0 and are presented as the mean ± standard error of the mean. Two-tailed unpaired Student's t test or one-way ANOVA with Bonferroni post hoc analyses were applied when appropriate for multiple comparisons. A value of P < 0.05 was considered statistically significant.

Moderate SMFs inhibited adipogenic differentiation of murine BMSCs
To investigate the effect of SMFs on the adipogenic differentiation of BMSCs, BMSCs grown in adipogenic medium were exposed to different intensities of moderate SMFs (0, 0.2, 0.4, and 0.6 T). ORO staining showed that SMFs significantly inhibited lipid-droplet formation in BMSCs in adipogenic medium in an intensitydependent manner (Fig. 2a, b). The expression of adipogenic transcription factors, such as CCAAT/enhancer-binding protein α, β, and δ (Cebpα, Cebpβ, and Cebpδ), was shown to be lower in SMF-stimulated BMSCs than those in control cells (Fig. 2c-e). Moreover, the mRNA levels of Pparγ adipogenic marker and those of its downstream targets, adiponectin, cluster of differentiation 36 (Cd36), and fatty acid binding protein 4 (Fabp4) (Fig. 2f-i), were demonstrated to be markedly decreased in SMFs-treated BMSCs compared with iron control (0 T). In contrast, high intensity SMF (0.6 T) displayed a more apparent inhibitory effect on BMSC differentiation toward adipocytes than lower intensities (0.2 and 0.4 T; Fig. 2a-f).
Moderate SMFs suppressed PPARγ-mediated signaling in BMSCs upon adipogenic differentiation RNA profile analysis also revealed that some adipogenic transcription factors (Cebpα, Cebpβ, Cebpδ, and Pparγ) and adipocyte marker genes (CD36, Fabp4, Igfbp2, Scd1, Fasn, Lpi, Mgst3, and Lep) were downregulated in 0.6 T SMF-stimulated BMSCs compared with those in unchallenged BMSCs (Fig. 3a). The downregulation of the Pparγ and Fabp4 genes was further validated in SMFstimulated BMSCs by western blot analysis (Fig. 3b). To correlate the differentially expressed genes with biological functions, we analyzed the functional bias of the differentially expressed genes according to Gene Ontology (GO) enrichment. Our results revealed that SMF altered many metabolic processes in BMSCs, including peptide biosynthetic process, translation, ATP metabolic process, purine ribonucleoside metabolic process, and nucleoside metabolic process (Fig. 3c). Gene-set enrichment analysis (GSEA) showed that PPARγ-positively correlated genes were remarkably enriched in iron control-treated BMSCs, compared with those in SMFexposed cells (false discovery rate (FDR) q value = 0.009) (Fig. 3d).

Moderate SMFs promoted osteogenic differentiation of murine BMSCs
To investigate whether exposure to SMF might affect the osteogenic differentiation of BMSCs, we examined the phenotypical changes of SMFs-challenged BMSCs cultured in osteogenesis induction medium. After osteogenic induction, we found that moderate SMFs enhanced the production of ALP, an early marker of osteogenesis in BMSCs in a dose-dependent manner (Fig. 4a, b). Consistently, ALS staining and quantification of ALS absorption revealed a significant enhancement of osteogenesis in SMF-stimulated BMSCs compared with controls (Fig. 4c, d). Moreover, the mRNA levels of the key osteogenic transcription factors Runx2 and osterix (Osx) (Fig. 4e, f) and those of col1α1, col1α2, Alp, and Spp1 osteogenic marker genes ( Fig. 4g-j) were demonstrated to be increased in SMF-exposed BMSCs in a dose-dependent fashion. These results suggested that SMFs promoted osteoblast differentiation of BMSCs.

Moderate SMFs facilitated BMSC differentiation to osteoblasts via activating the Runx2 signaling pathway
RNA-seq analysis uncovered that the osteogenic transcription factor Runx2 and the osteoblast markers (Col1α1, Col1α2, Alp, Ogn, Omd, Dcn, Mgp, Ank, Sparc, Smpd3, Bgn, and Matn4) were also notably upregulated in SMF-exposed BMSCs cultured in osteogenesis induction medium, as compared with those in control BMSCs (Fig. 5a). Consistent with the mRNA expression results obtained by qPCR and RNA-seq, western blot analysis also showed an increase in the protein expression of the two key osteogenic markers, RUNX2 and ALP, in SMFexposed BMSCs (Fig. 5b). Moreover, GO analysis showed that, upon osteogenic stimulation, SMF altered the gene expression associated with the development of the skeletal system, metabolic process of collagen, catabolic process of lipids, and others (Fig. 5c). Finally, GSEA analysis revealed that RUNX2 positively correlated genes were notably enriched in SMF-exposed BMSCs (0.6 T), compared with control cells (0 T) (FDR q value = 0.034) (Fig. 5d).

Moderate SMFs reduced ATRA-or Dex-induced bone loss in mice
Chronic administration of the active metabolite of vitamin A, ATRA, has been reported to lead to significant bone loss in mice [28]. To explore whether SMFs might exert a preventive role in osteoporosis, ATRAchallenged mice were exposed to SMFs of varying intensities (0, 0.2, 0.4, and 0.6 T). ATRA significantly induced bone loss in mice by decreasing bone density (Fig. 6a-c) and Tb.BV/TV and Tb.N (Fig. 6d, e), thus increasing Tb.Sp (Fig. 6f) without markedly influencing trabecular thickness (Fig. 6g). Consistent with the micro-CT images (Fig. 6a, b), histological analysis showed that ATRA promoted the accumulation of adipocytes in the bone marrow (Fig. 6h-j) and reduced OCN-positive osteoblasts on both the trabecular and endosteal bone surfaces (Fig. 6k, l) in mice. However, SMFs were shown to notably attenuate the ATRA-induced bone loss in mice by increasing bone density and the total trabecular ratio, reducing the deposition of adipocytes in the bone marrow, and inducing osteoblast regeneration in mice (Fig. 6a-l). Moreover, this effect was noted to be occurring in an SMF intensity-dependent fashion.
We also tested the effects of SMF intervention on Dex-induced bone loss in mice. As shown in Fig. 7a-l, SMFs significantly prevented Dex-induced bone loss in mice by increasing bone density and the total trabecular ratio, reducing the deposition of adipocytes in the bone marrow and inducing osteoblast regeneration in mice in an intensity-dependent manner.

Discussion
In this study, we stimulated BMSCs during adipogenic and osteogenic differentiation with different intensities of SMFs (0, 0.2, 0.4, and 0.6 T) and found that SMFs promoted osteoblastic differentiation but inhibited adipogenic differentiation of BMSCs in an intensity- b Western blot analysis of PPARγ and FABP4 in SMFexposed BMSCs cultured in adipogenic medium. c Gene Ontology (GO) pathway enrichment analyses of significantly downregulated genes in SMF-exposed BMSCs cultured in adipogenic medium. Top 20 enriched pathways are shown.*P<0.05. d Gene set enrichment analysis (GSEA) indicates that PPARγ-positively correlated genes are significantly reduced in SMF-exposed BMSCs compared with those in control cells dependent manner. Whole genomic RNA-seq analysis demonstrated that SMFs decreased PPARγ-driven gene expression but increased RUNX2-driven gene transcription in BMSCs. Moreover, SMFs markedly alleviated bone mass loss induced by either Dex or ATRA in mice. Thus, our study reveals the therapeutic potential of moderate-intensity SMFs in osteoporosis.
Magnetic fields are known to modulate the behavior of stem cells through multiple routes [31]. They are known to reduce inflammation, facilitate wound healing, and increase blood circulation [32]. In addition, magnetic field therapy has been approved for the management of pain and edema in superficial tissue inflammation by FDA [20]. Extremely low-frequency magnetic fields have been shown to influence cell proliferation by activation of Na + /K + [33] and Ca 2+ [34] channels. SMF has been commonly used in clinical practice as a tool, such as in magnetic resonance imaging (MRI). Moderate-intensity SMFs (1 mT to 1 T) were reported to enhance proliferation, migration, and dentinogenesis of dental pulp stem cells by activating the p38 mitogen-activated protein kinase pathway [35][36][37] and to induce osteo/odontogenesis Fig. 4 SMFs promote osteogenic differentiation of BMSCs. a Representative images of alkaline phosphatase (ALP) staining of BMSCs cultured in osteogenic medium and treated with gradient SMF for 3 days. Scale bar, 100 μm. b Detection of ALP activity. *P < 0.05, **P < 0.01 vs control or 0 T group. n = 4 per group. c, d Representative images of alizarin red S (ALS) staining (c) and quantification of matrix mineralization by ALS absorption (d) in BMSCs grown in osteogenesis induction medium and stimulated with gradient SMF for 21 days. Scale bar, 100 μm. *P < 0.05, **P < 0.01, as indicated. n = 4 per group. e-j Expression of key osteogenic transcription factors (Runx2 and Ost) (e and f) and osteoblast marker genes (Col1α1, Col1α2, Alp, and Spp1) (g-j) in BMSCs grown in osteogenesis induction medium and treated with gradient SMF. *P < 0.05, **P < 0.01, as indicated. n = 8-10 per group. ns=not significant and mineralization in dental pulp stem cells [37,38]. Similarly, moderate-intensity SMF also promoted neuronal differentiation in fetal rat brain neural progenitor cells [39] and induced the proliferation and osteoblastic differentiation of BMSCs [40,41].
BMSCs are known to differentiate into adipocytes instead of osteoblasts with aging, leading to the occurrence of osteoporosis, which is characterized by bone loss and progressive accumulation of fat [42,43]. Here, we demonstrated that SMFs repressed adipogenic differentiation but promoted osteogenic differentiation of BMSCs in an intensity-dependent manner by suppressing the PPARγ signaling pathway and activating the RUNX2 signaling pathway. Their capacity to differentiate into osteoblastic and adipogenic lineages is dependent on various signaling pathways and key transcription factors [44]. Activation of Runx2 and Osx has been found to promote osteogenic differentiation and inhibit adipogenic differentiation of BMSCs [45]. Conversely, Cebpα, Cebpβ, and Pparγ have been reported to drive adipogenesis and disrupt b Western blot analysis of RUNX2 and ALP in SMF-exposed BMSCs cultured in osteogenesis differentiation medium. c GO pathway enrichment analyses of significantly upregulated genes in SMF-exposed BMSCs grown in osteogenesis differentiation medium. Top 20 enriched pathways are shown.*P<0.05. d GSEA indicates that osteogenesis-positively correlated genes are significantly enriched in the SMF-exposed BMSCs compared with those in control cells osteogenic differentiation of BMSCs [46,47]. We demonstrated that moderate-intensity SMFs notably inhibited the expression of the Cebpβ and Pparγ adipogenic transcription factors in BMSCs cultured in adipogenesis induction medium but upregulated the Runx2 osteogenic transcription factor and its downstream targeted genes in BMSCs upon osteogenic stimulation. Bioinformatics analysis showed that SMFs altered the gene expression associated with the metabolism of fat in adipogenic medium-cultured BMSCs and promoted the expression of genes associated with skeletal development and osteoblast differentiation in osteogenic medium-cultured BMSCs.
Because SMFs are time-independent fields whose intensity can be spatially dependent, they are able to freely penetrate the biological tissues [48]. The biological effects of SMFs are known to be related to the dosing Sp) (f), and trabecular thickness (Tb.Th) (g) in SMF-exposed mice after ATRA injection. *P < 0.05, **P < 0.01, as indicated. n = 4-5 per group. h-j Representative images of H&E staining (h) with quantification of number (i) and area (j) of adipocytes in distal femurs. Scale bar, 100 μm. n = 5 per group. k-l Representative images of osteocalcin (OCN) immunohistochemical staining (k) with quantification of number (l) of osteoblasts in distal femurs. Scale bar, 100 μm. *P < 0.05, **P < 0.01, as indicated. n = 5 per group. ns=not significant regimen, target tissue, magnet characteristics, and the magnet support device [49]. Mechanistically, SMF has been shown to be able to directly regulate the cell shape and plasma membrane structure [50], interact with magnetic materials found in tissues [51], and modulate intracellular levels of reactive oxygen/nitrogen species [52]. Additionally, SMFs have also been reported to exert their effects by enhancing synthesis and secretion of membrane-derived microvesicles (MVs), mediating drug delivery and perhaps by inducing mitophagy [31,53,54]. Recently, an iron-sulfur cluster protein, called ironsulfur cluster assembly protein 1, was identified and validated in mammalian cells as a magnetic sensor [55,56]. However, further studies are required to investigate how SMFs regulate the RUNX2/PPARγ-mediated signaling pathways during BMSC differentiation.
Glucocorticoids are widely prescribed for the treatment of autoimmune and inflammatory diseases; Fig. 7 SMFs reduces dexamethasone (Dex)-stimulated osteoporosis in mice. a, b Representative 2D (a) and 3D (b) micro-CT images of femurs from control and Dex-treated mice with gradient SMF stimulation for 3 months. c-g Quantitative micro-CT analysis of BMD (c), BV/TV (d), Tb.N (e), Tb.Sp (f), and Tb.Th (g) in SMF-exposed mice after Dex injection. *P < 0.05, **P < 0.01, as indicated. n = 4-5 per group. h-j H&E staining (h) and quantification of number (i) and area (j) of adipocytes in distal femurs. Scale bar, 100 μm. n = 6 per group. k-l Representative images of OCN immunohistochemical staining (k) and quantification of number (l) of osteoblasts in distal femurs. Scale bar, 100 μm. *P < 0.05, **P < 0.01, as indicated. n = 5 per group. ns=not significant however, long-term treatment in the clinical setting has been shown to frequently cause osteoporosis. We also showed here that moderate SMFs promoted BMSC differentiation to osteoblasts and prevented ATRA-and Dex-induced bone loss in mice. Moreover, others have also reported that SMFs markedly accelerate the healing of osteotomized femur and promote bone regeneration in rats [23,57]. Therefore, SMF might be helpful for the prevention and treatment of osteoporosis.
Both SMF and electromagnetic fields (EMF) including pulsed electromagnetic fields (PEMF) can substantially facilitate bone healing. For instance, EMFs positively regulates osteogenic lineage commitment of BMSCs [58]. Similarly, low-frequency PEMFs facilitate bone repair by promoting osteoblast proliferation and osteogenic differentiation of BMSCs [59]. Thus, as a noninvasive, effective, and clinical safety treatment, EMF therapies have been commercially used to promote bone un-united fracture healing [59][60][61][62] and other skeletal disorders [63]. Compared to EMFs, SMFs appear more convenient without additional electrical devices, which avoids electric or heat hazard to surrounding tissue [64]. Importantly, SMF stimulation is suitable for long-term topical treatment [40,65]. In this study, we demonstrate the moderateintensity SMFs ranging from 0.2 T to 0.6 T induce bone repair in mice in a dose-dependent manner.

Conclusions
In summary, SMFs protected against ATRA-or Dexinduced osteoporosis in mice by promoting the RUNX2mediated osteogenic differentiation and suppressing the PPARγ-mediated lipogenic differentiation of BMSCs. These results suggested that SMF-based magnetotherapy might be beneficial for patients with osteoporosis.

Highlights
Moderate-intensity SMF promotes osteogenic differentiation of BMSCs. Moderate-intensity SMF suppresses lipogenic differentiation of BMSCs. Moderate-intensity SMF reduces bone mass loss induced by either dexamethasone or all-trans retinoic acid in mice.