Abstract
Background
MicroRNAs (miRNAs) are crucial regulators of osteoclast differentiation. Icariin (ICA), the primary active constituent of the traditional Chinese herb Epimedium, has been shown to inhibit osteoclastogenesis. However, whether ICA suppresses osteoclast differentiation via miRNA regulation remains to be elucidated.
Methods
We employed bioinformatic analysis coupled with in vivo validation to screen for and identify potential targets that mediate the effects of ICA on osteoclast formation. Subsequently, we carried out in vitro experiments involving the overexpression and knockdown of miR-376c-3p in bone marrow-derived macrophages (BMMs). Additionally, luciferase reporter assays and western blot analysis were utilized to ascertain the direct targets of miR-376c-3p.
Results
It was confirmed that miR-376c-3p plays a regulatory role in osteoclast differentiation within bone marrow-derived macrophages (BMMs). Notably, the knockdown of miR-376c-3p led to a partial reversal of the inhibitory effect that ICA exerts on osteoclastogenesis. Furthermore, matrix metalloproteinase-3 (Mmp3) was validated as a direct target of miR-376c-3p, providing evidence to suggest that ICA likely mediates its effects through the miR-376c-3p/Mmp3 signaling axis.
Conclusion
Our findings demonstrate that ICA exerts its anti-osteoporotic effect by downregulating the miR-376c-3p-mediated Mmp3 expression. This study provides insights that potentially facilitate ICA’s clinical application.
Graphical abstract
Supplementary Information
The online version contains supplementary material available at 10.1186/s13018-025-06589-z.
Keywords: Postmenopausal osteoporosis, Osteoclastogenesis, miR-376c-3p, Icariin, Matrix metalloproteinase-3
Introduction
Postmenopausal osteoporosis (PMOP) is a chronic systemic bone disorder that is primarily characterized by reduced bone mass and deterioration of bone microstructure. With the increase in the aging population, osteoporosis has become a growing medical, social, and economic concern [1]. Current therapeutic strategies for PMOP include estrogen replacement therapy, calcium and vitamin D supplementation, and bisphosphonates, which primarily function by inhibiting bone resorption [2]. Bone resorption, a key process in PMOP, is mediated by osteoclasts derived from differentiated bone marrow mononuclear macrophages (BMMs). Enhanced osteoclast differentiation leads to excessive bone resorption, thereby exacerbating postmenopausal osteoporosis [3]. Therefore, suppressing osteoclast formation may serve as an effective strategy in alleviating postmenopausal osteoporosis.
MicroRNAs (miRNAs) are small noncoding single-stranded RNAs, approximately 19–25 nucleotides in length, which regulate gene expression through the inhibition of protein translation [4]. They specifically target mRNAs via partial or complete base pairing with the 3’-untranslated region (3’-UTR), thereby participating in various pathophysiological processes [5–7]. Accumulating evidence suggests that miRNAs substantially influence many aspects in musculoskeletal diseases [8–11], including the formation process of osteoclasts in PMOP [12, 13]. miR-376c-3p is an miRNA that has widely been reported to be involved in cancer proliferation and migration [14, 15]. However, it was recently found that miR-376c-3p regulated bone formation [16], suggesting its potential role in mediating bone metabolism. However, the role of miR-376c-3p in osteoclastogenesis remains to be fully elucidated.
In recent times, traditional Chinese medicine has garnered growing interest due to its low risk of side effects and centuries of historical practice, especially for the treatment of challenging diseases such as PMOP. Icariin (ICA), the primary active constit-uent of the traditional Chinese herb Epimedium, exhibits remarkable anti-osteoporotic effects because of its dual pharmacological actions in promoting bone formation and inhibiting bone resorption [17, 18]. Several studies have demonstrated that ICA regulates osteogenic differentiation by modulating miRNAs, thereby mitigating osteoporosis [19–21]. However, experimental evidence regarding ICA’s inhibitory effects on osteoclastogenesis through miRNA regulation remains limited.
In this study, bioinformatics analysis and verification using an ovariectomized animal model identified miR-376c-3p as a potential target of ICA in treating PMOP. ICA was further validated to suppress osteoclast formation and alleviate PMOP by regulating the miR-376c-3p-mediated expression of matrix metalloproteinase-3 (Mmp3). Consequently, the current research findings suggest that ICA may have potential as a preventive measure and treatment for PMOP.
Materials and methods
Bioinformatics analysis
The RNA expression profiles were obtained from the GEO database (https://www.ncbi.nlm.nih.gov/geo/; accession number GSE230665). Postmenopausal osteoporosis-related genes were collected from GeneCards (https://www.genecards.orgGuideGettingStarted), OMIM (https://www.omim.org), and PharmGKB (https://www.pharmgkb.org). The potential target genes of ICA were retrieved from PubChem (https://pubchem.ncbi.nlm.nih.gov) and SwissTargetPrediction (http://swisstargetprediction.ch/). The intersection between the disease-related genes and ICA-predicted targets was identified. These candidate targets were subsequently validated using the GSE230665 dataset from GEO, with statistical significance set at P < 0.05. Visualization was performed using Venn diagrams and heatmaps, which were generated using the “pheatmap” package in R software. LASSO regression analysis was performed to identify the key differentially expressed genes using the “glmnet” package in R, with results visualized as boxplots with the “ggpubr” and “limma” packages. Potential mRNA-microRNA interactions were predicted based on the key genes using the TargetScan database, and the ceRNA network was visualized using Cytoscape 3.8.2.
Animals
Three-month-old female Sprague–Dawley rats (300 ± 20 g, n = 30) were obtained from Shanghai SLAC Laboratory Animal Co. Ltd. The animals were housed under controlled conditions (40%–70% humidity, 22 °C ± 2 °C) and randomly assigned to experimental groups. The OVX and OVX + ICA groups underwent bilateral ovariectomy, while the SHAM group received sham surgery. The OVX + ICA rats received ICA (20 mg·kg− 1·day− 1, Chengdu Herbpurify) daily for 12 weeks, whereas the SHAM and OVX groups received equal volumes of saline as described previously. All procedures complied with NIH guidelines for animal care and were reviewed and approved by the Animal Care Committee of Fujian University of Traditional Chinese Medicine (Approval No. 2023262).
Micro-computed tomography (micro-CT)
The tibiae were fixed in 4% paraformaldehyde and scanned using micro-CT (ZZKS-Micro-CT 4.1). Three-dimensional reconstruction and quantitative analysis of the BMD, BV/TV, Tb.N and Conn.Dn were performed using Medproject software.
Histological examination
The lumbar vertebrae’s samples were decalcified in EDTA for 6 weeks after fixation in 10% neutral buffered formalin for 24 h. Paraffin-embedded Section (4 μm) were stained with hematoxylin and eosin following standard protocols (Solarbio, China).
Quantitative real-time polymerase chain reaction(qRT-PCR)
Total RNA was extracted using Trizol reagent (Aidlab, China). cDNA synthesis and amplification were performed using PrimeScript™ RT and TB Green® kits (Takara). The relative expression was calculated using the comparative Ct (2−ΔΔCT) method and normalized to that of the reference genes Gapdh/β-actin (mRNA) or U6 (miRNA). The primer sequences are listed in Table 1.
Table 1.
The primers used in this study
| Gene | Sequence (5′-3′) | |
|---|---|---|
| Gapdh |
Forward: Reverse: |
ACGGCAAGTTCAACGGCACAG GAAGACGCCAGTAGACTCCACGAC |
| β-actin |
Forward: Reverse: |
CGCGAGTACAACCTTCTTGC CCTTCTGACCCATACCCACC |
| Mmp3 |
Forward: Reverse: |
ATGAACTTGGCCACTCCCTG GTGGGTACCACGAGGACATC |
| Mdm2 |
Forward: Reverse: |
CTTCGTGAGAACTGGCTTCC CCACTCAAGAACTGTCAAGTAACC |
| Gsk3b |
Forward: Reverse: |
CCACTCAAGAACTGTCAAGTAACC CCACGGTCTCCAGCATTAGTA |
| Il2 |
Forward: Reverse: |
CAAGCAGGCCACAGAATTGA GCTCATCATCGAATTGGCACTC |
| Esr1 |
Forward: Reverse: |
AACTTGCTCTTGGACAGGAATC AGAGACTTCAAGGTGCTGGATA |
| Crat |
Forward: Reverse: |
GATACCTCCTACGCCATTGC GGATTGTAGCAGATGCCGTAA |
| Tracp |
Forward: Reverse: |
CCATTGTTGGCTGCATACGG CACGGAAGGGTCCATGAAGT |
| Nfatc1 |
Forward: Reverse: |
GTCCTACGCCATCTGAGTCA GAGTGTTACTTGTTGCTCTTGAGA |
| Cstk |
Forward: Reverse: |
AACTGCGACCGTGATAATGTG CCGAGCCAAGAGAACATAGC |
| U6 |
Forward: Reverse: |
CTCGCTTCGGCAGCACATATACT ACGCTTCACGAATTTGCGTGTC |
| miR-376c-3p |
Forward: Reverse: RT: |
CCGCGCAACATAGAGGAAATT AGTGCAGGGTCCGAGGTATT GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACACGTGA |
| miR-497-3p |
Forward: Reverse: RT: |
TAAGCCAAACCACACTGTGGT AGTGCAGGGTCCGAGGTATT GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACTCTAAC |
| miR-935 |
Forward: Reverse: RT: |
TAATACCAGTTACCGCTTCCGC AGTGCAGGGTCCGAGGTATT GTCGTATCCAGTGCAGGGTCCGAGGTATTCGCACTGGATACGACGCGGTA |
Cell viability assessment
Bone marrow-derived macrophages (BMMs) were plated in a 96-well culture plate at a seeding density of 2 × 104 cells per well. The cells were then treated with varying concentrations of icariin (ICA) or left untreated, followed by incubation at 37 °C in a 5% CO₂ atmosphere for durations of 24, 48 and 72 h. Subsequently, 100 µL of CCK-8 reagent (APExBIO, USA) was introduced into each well, and the plates were further incubated at 37 °C for 1 h. Optical density (OD) was determined at a wavelength of 450 nm using a microplate reader.
Osteoclast differentiation and TRAP staining
BMMs were isolated from female Sprague–Dawley rats’ femurs and cultured in α-MEM (Procell, PM150421, China) containing 10% fetal bovine serum, antibiotics, and 20 ng/mL of macrophage colony-stimulating factor (M-CSF; PeproTech, USA) for 4 days. Osteoclast differentiation was induced with 50 ng/mL of RANKL (Abbkine, China) treatment for 5 days. TRAP staining was performed using a TRAP staining kit (G1492, Solarbio, China) according to the manufacturer’s protocols. Multinucleated (≥ 3 nuclei) TRAP-positive cells were identified as mature osteoclasts.
Transfection of miR-376c-3p mimics and inhibitors
BMMs were transfected with 100 nM of miR-376c-3p mimics/inhibitors or negative controls using Lipofectamine® RNAiMAX (Invitrogen, USA) according to the manufacturer’s instructions. The sequence of the miR-376c-3p mimics and inhibitors are listed in Table 2. The knockdown efficiency was verified by qPCR after 48 h.
Table 2.
The sequences were as follows
| Gene | Sequence (5′-3′) |
|---|---|
| miR-376c-3p mimics |
Sense: AACAUAGAGGAAAUUUCACGU antisense༚GUGAAAUUUCCUCUAUGUUUU |
| miR-376c-3p inhibitor | CAGUACUUUUGUGUAGUACAA |
| mimics NC |
Sense: UUCUCCGAACGUGUCACGUTT antisense༚ACGUGACACGUUCGGAGAATT |
| inhibitor NC | CAGUACUUUUGUGUAGUACAA |
Dual-luciferase reporter assay
Wild-type or mutant (MUT) 3’-UTR sequences of Mmp3 were cloned into GP-miRGLO vectors. Cotransfection with miR-376c-3p mimics was performed using Lipofectamine 2000, with the luciferase activity measured after 48 h, according to the manufacturer’s instructions (Yeasen, China).
Western blot analysis
Femur tissues and cultured cells were homogenized in RIPA buffer (Beyotime, China) containing a PMSF protease inhibitor (1:100 dilution). Following 4 h of lysis at 4 °C with constant shaking, the lysates were centrifuged at 14,000 rpm for 30 min to pellet the debris. The resulting supernatant was subjected to protein concentration measurement using a BCA protein assay kit. The protein lysates were separated by SDS-PAGE and transferred to polyvinylidene difluoride (PVDF) membranes. After blocking, the membranes were incubated with the primary antibodies (anti-Mmp3, 1:2000, Bioss; anti-Beta Actin, 1:20000, Proteintech), followed by horseradish peroxidase (HRP)-conjugated secondary antibodies. Signals were detected using an enhanced chemiluminescent (ECL) substrate (Abbkine, China). Image Lab software was used to quantify the density of each band.
Statistical analysis
Statistical analysis was performed using GraphPad Prism (GraphPad Software Inc., La Jolla, CA, USA). For comparison between two groups, t-test was applied. For comparisons among multiple groups, one-way ANOVA was applied. Data are expressed as the mean ± standard deviation (SD), with statistical significance defined as P < 0.05.
Results
Bioinformatics reveals ICA-specific target genes in PMOP alleviation
To identify key genes involved in the ICA-mediated mitigation of PMOP, we intersected PMOP-associated genes from GeneCards, OMIM and PharmGKB, differentially expressed genes from the GEO dataset GSE230665 (comparing gene expression between pre- and postmenopausal women) and drug-predicted targets from SwissTargetPrediction, which yielded 116 overlapping gene targets (Fig. 1A). Considering the statistic significance, 19 differentially expressed genes were statistically identified, which were visualized using a heatmap (Fig. 1B). LASSO regression analysis prioritized six candidate differentially expressed genes (MMP3, CRAT, ESR1, MDM2, GSK3B, and IL2) as potential therapeutic targets in ICA efficacy for PMOP treatment(Figure. 1 C, D).
Fig. 1.
Identification of therapeutic targets in ICA efficacy for PMOP treatment. (A) Venn diagram of ICA-regulated differentially expressed genes in PMOP. (B) The heatmaps of 19 ICA-regualted differentially expressed genes. (C) LASSO regression analysis. (D) The information of six potential genes in LASSO model
Validation of ICA-Specific target genes in PMOP
To verify whether the six differentially expressed genes were modulated by ICA treatment, we detected their expression in vivo. PMOP rat models were established by ovariectomy (OVX), and ICA was administered to establish an intervention group (OVX + ICA). Micro-CT and HE staining revealed that OVX rats exhibited thinner, sparser, and disorganized trabeculae compared to the control SHAM group, while ICA treatment significantly improved the bone microstructure (Fig. 2A, B). Moreover, the bone mineral density (BMD), bone volume fraction (BV/TV), trabecular number (Tb.N) and connectivity density (Conn.Dn) were lower in the OVX group than in the SHAM group. TRAP staining of bone tissue also revealed a greater number of osteoclasts in the OVX group compared to the SHAM group (Supplementary Fig. 1). These parameters were moderately recovered in the OVX + ICA group (Fig. 2C and F and Supplementary Fig. 1). All the bone morphometric parameters confirmed the efficacy of ICA for PMOP treatment.
Fig. 2.
Validation of hub genes in vivo. (A) Micro-CT scan of the tibiae in SHAM group, OVX group and OVX + ICA group. (B) The HE staining of lumbar vertebrae in SHAM group, OVX group and OVX + ICA group. (C) Change of BMD in SHAM group, OVX group and OVX + ICA group separately (n = 3). ****P < 0.0001, SHAM group versus OVX group and ****P < 0.0001, OVX + ICA group versus OVX group. (D) Change of BV/TV in SHAM group, OVX group and OVX + ICA group separately (n = 3). ***P = 0.0005 and **P = 0.0014. (E) Change of Tb.N in SHAM group, OVX group and OVX + ICA group separately (n = 3).****P < 0.0001, SHAM group versus OVX group and ****P < 0.0001, OVX + ICA group versus OVX group. (F) Change of Conn.Dn in SHAM group, OVX group and OVX + ICA group separately (n = 3). ***P = 0.0005 and **P = 0.0050. (G) mRNA level of Mdm2 was tested by qPCR in SHAM group, OVX group and OVX + ICA group (n = 3). Relative expression of Mdm2 was normalized against Gapdh (n = 3). **P = 0.0029, SHAM group versus OVX group and **P = 0.0077, OVX + ICA group versus OVX group. (H) mRNA level of Gsk3b was tested by qPCR in SHAM group, OVX group and OVX + ICA group(n = 3). Relative expression of Gsk3b was normalized against Gapdh (n = 3). ****P = < 0.0001 and ***P = 0.0002. (I) mRNA level of Mmp3 was tested by qPCR in SHAM group, OVX group and OVX + ICA group (n = 3). Relative expression of Mmp3 was normalized against Gapdh (n = 3). **P = 0.0027 and *P = 0.0102. (I) mRNA level of Il2 was tested by qPCR in SHAM group, OVX group and OVX + ICA group (n = 3). Relative expression of Il2 was normalized against Gapdh (n = 3). **P = 0.0047 and *P = 0.0327. All the data are shown as mean ± SD (n = 3) versus the OVX group in three independent experiments
The expression of the six candidate differentially expressed genes was measured by qPCR, and the results displayed upregulated expressions of Mdm2, Gsk3b, Mmp3, and Il2 in the OVX group, which was reversed by ICA treatment (Fig. 2G and J). No significant changes were observed for the expression levels of Crat or Esr1 (Supplementary Fig. 2), which suggests that Mdm2, Gsk3b, Mmp3, and Il2 are potential therapeutic targets of ICA in PMOP.
miR-376c-3p as a potential mediator of ICA efficacy in PMOP
To further explore the ICA-mediated miRNAs, we established a miRNA-mRNA regulatory network to identify the upstream miRNAs capable of coordinately regulating four differentially expressed genes (Mmp3, Mdm2, Gsk3b, and Il2). As visualized in Figs. 3A and 12 candidate miRNAs were identified. After detecting the expressions of these candidate miRNAs in vivo, it was revealed that miR-376c-3p was significantly downregulated in OVX rats and restored by ICA treatment (Fig. 3B), whereas the other miRNAs exhibited no detectable changes or did not exist in rats (Supplementary Fig. 3). Thus, miR-376c-3p emerged as a likely mediator of ICA’s therapeutic effects.
Fig. 3.
Verification of candidate miRNAs. (A) Construction of miRNA-mRNA regulatory network. The red nodes represent hub genes. The green nodes represent miRNA. (B) The expression of miR-376c-3p in SHAM group, OVX group and OVX + ICA group (n = 3). Relative expression of miR-376c-3p was all normalized against U6 (n = 3), ***P = 0.0002 and **P = 0.0079. All the data are shown as mean ± SD (n = 3) in three independent experiments
miR-376c-3p suppressed osteoclast differentiation
miR-376c-3p was recently reported to regulate bone remodeling by participating in osteogenesis [16]. However, its role in osteoclastogenesis remains unclear. Transfection with miR-376c-3p mimics and inhibitors significantly increased or decreased its expression, respectively, after 48 h (Supplementary Fig. 4). After a 5-day induction, miR-376c-3p mimics exhibited markedly weaker Tartrate-resistant acid phosphatase (TRAP) staining with fewer TRAP+ cell numbers as well as downregulated osteoclast markers, including Tracp, Ctsk, and Nfatc1 (Fig. 4A-E). Conversely, miR-376c-3p inhibition elevated the number of osteoclast TRAP+ cells and marker expressions (Fig. 4F-J). These results demonstrate that miR-376c-3p negatively regulates osteoclast differentiation.
Fig. 4.
The role of miR-376c-3p in osteoclast differentiation. (A-B) The representative images of TRAP staining in mimic NC group and miR-376c-3p mimic (A). The number of TRAP+ osteoclasts was observed and calculated (n = 3) (B), ****P < 0.0001. (C-E) mRNA levels of Tracp (C), Ctsk (D) and Nfatc1 (E), were tested as osteoclast markers by qPCR (n = 3), P values are indicated as follows: ***P = 0.0002; **P = 0.0011; ****P < 0.0001. (F-G) The representative images of TRAP staining in inhibitor NC group and miR-376c-3p inhibitor (F). The number of TRAP+ osteoclasts was observed and calculated (n = 3) (G), ***P = 0.0002. (H-J) mRNA levels of Tracp (H), Ctsk (I) and Nfatc1 (J), were tested as osteoclast markers by qPCR (n = 3). Relative expressions of Tracp, Ctsk and Nfatc1 were all normalized against β-actin (n = 3), P values are indicated as follows: *P = 0.0153; *P = 0.0368; ****P < 0.0001. All the data are shown as mean ± SD (n = 3) in three independent experiments
Knockdown of miR-376c-3p abolished the inhibitory effect of ICA on osteoclastogenesis
Given the anti-osteoclastogenic role of miR-376c-3p, we hypothesized that ICA might inhibit osteoclastogenesis through miR-376c-3p mediation. Cell viability assays confirmed that ICA was nontoxic at 0.02–2 µmol/L (Supplementary Fig. 5). ICA treatment generated a significant reduction of TRAP staining (Fig. 5A, B) as well as the downregulation of the expression of Trap, Ctsk, and Nfatc1 (Fig. 5C-E), suggesting its remarkable inhibition of osteoclast differentiation. Compared to the results in the ICA-administered group, the addition of an miR-376c-3p inhibitor during ICA-mediated osteoclast differentiation partially reversed ICA efficacy, as evidenced by greater TRAP+ cell numbers (Fig. 5A, B) and the higher expression of Tracp, Ctsk, and Nfatc1 (Fig. 5C-E). This indicates that ICA counteracts osteoclast formation by maintaining miR-376c-3p levels.
Fig. 5.
Silencing miR-376c-3p partially eliminates ICA’s suppressive impact on osteoclast formation. (A-B) The representative images of TRAP staining in inhibitor NC group, ICA + inihibitor NC group and ICA + miR-376c-3p inhibitor group (A). The number of TRAP+ osteoclasts was observed and calculated (n = 3) (B), ****P < 0.0001, inhibitor NC group versus ICA + inhibitor NC group and ****P < 0.0001, ICA + miR-376c-3p inihibitor group versus ICA + inhibitor NC group. (C) mRNA level of Tracp was tested as osteoclast markers by qPCR. Relative expression of Tracp was normalized against β-actin (n = 3). **P = 0.0016 and ***P = 0.0004. (D) mRNA level of Ctsk was tested as osteoclast markers by qPCR. Relative expression of Ctsk was normalized against β-actin (n = 3). **P = 0.0010, inhibitor NC group versus ICA + inhibitor NC group and **P = 0.0096, ICA + miR-376c-3p inihibitor group versus ICA + inhibitor NC group. (E) mRNA level of Nfatc1 was tested as osteoclast markers by qPCR. Relative expression of Nfatc1 was normalized against β-actin (n = 3). **P = 0.0024 and ***P = 0.0006. All the data are shown as mean ± SD (n = 3) in three independent experiments
ICA modulates bone loss via the miR-376c-3p/Mmp3 axis
To identify the direct target of miR-376c-3p, the expressions of four key differentially expressed genes were detected after transfection with miR-376c-3p mimics and inhibitors as well as treatment with ICA or ICA combined with a miR-376c-3p inhibitor. Among these, Mmp3 was the only gene that showed an downregulation upon miR-376c-3p mimic transfection and a upregulation with the inhibitor (Fig. 6A, B). Additionally, ICA treatment suppressed Mmp3 expression, which was partially restored in the presence of the miR-376c-3p inhibitor (Fig. 6C), while the remaining three genes exhibited no significant alteration (Supplementary Fig. 6). TargetScan predicted the binding sites between miR-376c-3p and Mmp3’s 3’UTR (Fig. 6D). Dual-luciferase assays revealed that miR-376c-3p mimics reduced the luciferase activity of the Mmp3 3’UTR construct. In contrast, the inhibitory effect was remarkably rescued after the putative binding sites were mutated (Fig. 6E). Furthermore, the expression of Mmp3 increased after BMMs were transfected with the miR-376c-3p inhibitor (Fig. 6F, G). Collectively, miR-376c-3p directly targets the 3’UTR of Mmp3 to regulate its expression.
Fig. 6.
ICA suppressed osteoclast formation via miR-376c-3p/Mmp3 axis. (A-B) The expression of Mmp3 after transfection with miR-376c-3p mimics (A), **P = 0.0055 and inhibitors (B), **P = 0.0014. (C) The expression of Mmp3 in inhibitor NC group, ICA + inhibitor NC group and ICA + miR-376c-3p inhibitor group. Relative expression of Mmp3 was all normalized against β-actin (n = 3), ***P = 0.0005 and **P = 0.0032. (D) Potential binding sites between miR-376c-3p and Mmp3’s 3’-UTR were predicted using TargetScan. The wild-type and mutated sequences of the Mmp3 3’-UTR were illustrated. (E) Relative luciferase activities were detected after cotransfection with miR-376c-3p mimics and wild-type or mutant (MUT) 3’-UTR sequences of Mmp3, **P = 0.0028 (F-G) Western blot analysis for Mmp3 was performed after transfection with miR-376c-3p inhibitor. Relative expression of Mmp3 was all normalized against β-actin (n = 3), *P = 0.0338. (H-I) Western blot analysis for Mmp3 in SHAM group, OVX group and OVX + ICA group. Relative expression of Mmp3 was all normalized against β-actin (n = 3), *P = 0.0236, SHAM group versus OVX group and *P = 0.0310, ICA + OVX versus OVX group. (J-K) Western blot analysis for Mmp3 in inhibitor NC group, ICA + inhibitor NC group and ICA + miR-376c-3p inhibitor group. Relative expression of Mmp3 was all normalized against β-actin (n = 3), ***P = 0.0003 and *P = 0.0315. All the data are shown as mean ± SD (n = 3) in three independent experiments
To verify whether ICA exerts its inhibitory effect on bone loss via the miR-376c-3p/Mmp3 axis, we performed immunoblotting to trace the expression of Mmp3 in vivo and in vitro. As shown in Fig. 6H and I, Mmp3 expression increased in the OVX group compared to that in the SHAM group, while it was reduced by ICA treatment. During osteoclast differentiation, the Mmp3 expression decreased in the ICA group and was partially restored in the ICA + miR-376c-3p inhibitor group (Fig. 6J, K). These findings establish that ICA may ameliorate PMOP through the miR-376c-3p/Mmp3 regulatory axis.
Discussion
Excessive bone resorption disrupts bone homeostasis and contributes to the development of PMOP. ICA has emerged as a promising therapeutic agent for the prevention and treatment of PMOP by suppressing osteoclastic activity [22]. In this study, we identified miR-376c-3p as a potential therapeutic target of ICA for the treatment of PMOP through bioinformatics analysis and validation in an OVX rat model. Functional experiments demonstrated that miR-376c-3p regulates osteoclast differentiation, as evidenced by the altered osteoclastogenesis upon the overexpression or inhibition of miR-376c-3p. Furthermore, the anti-osteoclastogenic effects of ICA were partially reversed upon miR-376c-3p inhibition, thus confirming its role as a pharmacological target of ICA. Mechanistically, dual-luciferase reporter assays and western blot analysis revealed that ICA alleviates PMOP by downregulating miR-376c-3p, thereby reducing the expression of Mmp3. These findings suggest that ICA exerts its therapeutic effects, at least in part, through the miR-376c-3p/Mmp3 axis.
Non-coding RNAs have been regarded as potential targets to regulate some musculoskeletal diseases [23–26]. miRNAs, a class of small non-coding RNAs, have been established as key regulators of osteoclastogenesis in PMOP, with notable examples including miR-378 and miR-24-3p [27, 28]. miR-376c-3p participates in regulating various pathological processes in vivo, including tumor proliferation, renal ischemia-reperfusion injury, and osteoarthritis [29–31]. Recently, miR-376c-3p inhibition was demonstrated to promote the osteogenic differentiation of bone marrow mesenchymal stromal cells [16], suggesting its potential involvement in bone remodeling during PMOP. In our study, we established a PMOP animal model, and a significant decrease in miR-376c-3p levels was observed, which contradicted the previous conclusion. Therefore, we speculated that miR-376c-3p might participate in PMOP bone metabolism by regulating osteoclast differentiation. Following the overexpression of miR-376c-3p in BMMs during osteoclast induction, the number of generated osteoclasts and expression of the osteoclast-specific genes were significantly reduced. Conversely, miR-376c-3p inhibition in BMMs during osteoclast induction led to significant increases in osteoclast formation and the expression of osteoclast-specific genes. These results collectively demonstrate that miR-376c-3p can suppress osteoclastogenesis, indicating miR-376c-3p as a new regulator in PMOP.
Metalloproteinase family has been proven to function in many musculoskeletal diseases [32, 33]. Mmp3 is a member of the metalloproteinase family but also belongs to the zinc-dependent proteolytic enzyme family that participates in extracellular matrix degradation and has been shown to play crucial roles in the remodeling of bone extracellular matrix [34, 35]. It has been reported that MMP3 cleaves membrane-bound FasL in osteoblasts via ERα signaling, increasing the concentration of the soluble C-terminal domain and subsequently promoting osteoclast apoptosis [36]. Furthermore, under the regulation of the STAT3 signaling pathway, MMP3 can be secreted by osteoblasts in response to mechanical stress and act to induce osteoclastic bone resorption [37]. Additionally, IL-3 differentially regulates membrane-bound and soluble RANKL in osteoblasts through metalloproteinases like MMP3 and the JAK2/STAT5 pathway, elevating the RANKL/OPG ratio in adult mice and potentially influencing osteoclastogenesis [38]. These findings highlight the multifaceted and context-dependent functions of MMP3 in bone resorption. Futhermore, MMP3 has been reported to be regulated by several miRNAs, such as miR-146a and miR-155 [39, 40]. In our study, dual-luciferase reporter assays further revealed that miR-376c-3p directly targets Mmp3 mRNA, thereby regulating its expression during osteoclastogenesis. MMP3 knockout in bone was found not to affect bone formation but rather inhibited bone resorption, which ultimately led to significantly increased bone density [41]. Therefore, we hypothesize that miR-376c-3p alleviates PMOP by suppressing osteoclast differentiation through the targeted regulation of Mmp3.
The treatment of PMOP is not limited to modern pharmaceuticals, as traditional Chinese medicine has demonstrated clinical efficacy in alleviating PMOP [42, 43]. Epimedium, a perennial understory herb, is widely used in the clinical treatment for bone-related disorders, such as PMOP and bone fractures, due to its properties of “nourishing the kidney and reinforcing the Yang“ [44]. ICA, an active flavonoid compound extracted from Epimedium, has been confirmed to exhibit therapeutic effects against PMOP; thus, it is regarded as an excellent complementary and alternative medicine for PMOP [45]. Therefore, investigating the mechanism of ICA is crucial for its clinical use for PMOP treatment. ICA has been demonstrated to exert anti-bone loss effects by regulating physiological processes such as cell death, oxidative stress, and inflammation [46–48]. However, the role of miRNAs in ICA-mediated inhibition of osteoclast differentiation remains poorly explored. In this study, the pharmacological effects of ICA were validated in vivo and in vitro. Through network pharmacology screening, we identified four potential differentially expressed genes. The subsequent construction of a miRNA-mRNA network and qPCR validation revealed miR-376c-3p as the common upstream miRNA. Notably, miR-376c-3p knockdown during ICA intervention partially attenuated its inhibitory effect on osteoclast differentiation, thereby confirming miR-376c-3p as a functional target of ICA. Furthermore, in vivo and in vitro experiments demonstrated significant alterations in the expression of miR-376c-3p-mediated Mmp3 following ICA intervention. Therefore, we propose that ICA alleviates PMOP by upregulating miR-376c-3p expression, thereby reducing Mmp3-mediated bone resorption. However, further studies involving direct in vivo modulation of miR-376c-3p are needed to fully validate its therapeutic role in ICA’s anti-osteoporotic effects.
Conclusions
In conclusion, we have confirmed miR-376c-3p as a new target to inhibit osteoclastogenesis. ICA exerts its anti-osteoporotic effects by downregulating miR-376c-3p-mediated expression of Mmp3. This study is conducive to the promotion of clinical use of ICA.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
The authors would like to thank Enago (www.enago.cn) for the English language review.
Author contributions
Hui Yan, Yunteng Xu and Dezun Ma identify the focus and overall direction of the review. Funded acquisitions Hui Yan, Yunteng Xu, and Dezun Ma; Sources, Hui Yan; Methodology, Hui Yan, Jingdi Li, Dingbang Xie, Yaoting Zou and Jiapo Zuo; Supervision, Yunteng Xu, Hui Yan and Dezun Ma; Written-original drafts, Hui Yan, Jingdi Li, Dingbang Xie, Yaoting Zou, Jiapo Zuo, Yunteng Xu and Dezun Ma; Written-review & edited by Hui Yan and Yunteng Xu. All authors contributed to this article and approved the submitted version.
Funding
This work was supported by Natural Science Foundation of Fujian Province Project (2023J05171), the Research Start-up Fund of Fujian University of Traditional Chinese Medicine (No. X2020009-Talents), Scientific Research Start-up Funds for High-level Talents of Fujian University of Traditional Chinese Medicine (X2024006-Talents) and Youth Science and Technology Innovation Talent Cultivation Program of FJTCM (XQC2024002).
Data availability
No datasets were generated or analysed during the current study.
Declarations
Ethics approval and consent to participate
All procedures complied with NIH guidelines for animal care and were reviewed and approved by the Animal Care Committee of Fujian University of Traditional Chinese Medicine (Approval No. 2023262).
Consent for publication
All authors approved the final manuscript and the submission to this journal.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Hui Yan and Jingdi Li are co-first authors
Contributor Information
Yunteng Xu, Email: XytDao@163.com.
Dezun Ma, Email: 2021025@fjtcm.edu.cn.
References
- 1.Walker Marcella D, Shane E. Postmenopausal osteoporosis. N Engl J Med. 2023;389(21):1979–91. [DOI] [PubMed] [Google Scholar]
- 2.Black Dennis M, Rosen Clifford J. Clinical Practice.Postmenopausal osteoporosis. N Engl J Med. 2016;374(3):254–62. [DOI] [PubMed] [Google Scholar]
- 3.Veis DJ, O’Brien CA, Osteoclasts. Master sculptors of bone. Annu Rev Pathol. 2023;18:257–81. [DOI] [PubMed] [Google Scholar]
- 4.Xu J, Liu S, Jin Y, Wang L, Gao J. MicroRNAs and lysosomal membrane proteins: critical interactions in tumor progression and therapy. Biochimica et biophysica acta (BBA) -. Reviews Cancer. 2025;1880(3):189303. [DOI] [PubMed] [Google Scholar]
- 5.Rupaimoole R, Slack FJ. MicroRNA therapeutics: towards a new era for the management of cancer and other diseases. Nat Rev Drug Discovery. 2017;16(3):203–22. [DOI] [PubMed] [Google Scholar]
- 6.Pritchard CC, Cheng HH, Tewari M. MicroRNA profiling: approaches and considerations. Nat Rev Genet. 2012;13(5):358–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bushati N, Cohen SM. microRNA Functions. Annual Review of Cell and Developmental Biology. 2007;23(Volume 23, 2007):175–205. [DOI] [PubMed]
- 8.Yang X, Yin P, Yao X, Zhang J. MicroRNAs in the diagnosis of osteoarthritis: a systematic review and meta-analysis of observational studies. J Orthop Surg Res. 2025;20(1):654. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zhao C, Li Q, Shen C. Therapeutic potential of miR-204-5p in intervertebral disc degeneration: targeting the SSRP1/NF-κB pathway to inhibit apoptosis. J Orthop Surg Res. 2025;20(1):586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Li F, Tan H, Zhang X, et al. LncRNA HCG18 regulates the progression of spinal tuberculosis by modulating the hsa-miR-146a-5p/TGF-β1/SMADs pathway. J Orthop Surg Res. 2025;20(1):484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Giordano L, Porta GD, Peretti GM, Maffulli N. Therapeutic potential of MicroRNA in tendon injuries. Br Med Bull. 2020;133(1):79–94. [DOI] [PubMed] [Google Scholar]
- 12.Yang X, Jin Q, Guo L. MiR-217 participates in the progression of postmenopausal osteoporosis by regulating the OPG/RANKL/RANK pathway. J Orthop Surg Res. 2025;20(1):600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Liu X, Zhang X, Cen M. Dysregulation of miR-106a-5p/PTEN axis associated with progression and diagnostic of postmenopausal osteoporosis. J Orthop Surg Res. 2025;20(1):456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ma Y, Cong X, Zhang Y, et al. CircPIP5K1A facilitates gastric cancer progression via miR-376c-3p/ZNF146 axis. Cancer Cell Int. 2020;20(1):81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wang K, Jin J, Ma T, Zhai H. MiR-376c-3p regulates the proliferation, invasion, migration, cell cycle and apoptosis of human oral squamous cancer cells by suppressing HOXB7. Biomed Pharmacother. 2017;91:517–25. [DOI] [PubMed] [Google Scholar]
- 16.Camp E, Pribadi C, Anderson PJ, Zannettino ACW, Gronthos S. miRNA-376c-3p mediates TWIST-1 Inhibition of bone Marrow-Derived stromal cell osteogenesis and can reduce aberrant bone formation of TWIST-1 haploinsufficient calvarial cells. Stem Cells Dev. 2018;27(23):1621–33. [DOI] [PubMed] [Google Scholar]
- 17.Si Y, Li Y, Gu K, Yin H, Ma Y. Icariin ameliorates osteoporosis in ovariectomized rats by targeting Cullin 3/Nrf2/OH pathway for osteoclast Inhibition. Biomed Pharmacother. 2024;173:116422. [DOI] [PubMed] [Google Scholar]
- 18.Bai L, Liu Y, Zhang X, et al. Osteoporosis remission via an anti-inflammaging effect by Icariin activated autophagy. Biomaterials. 2023;297:122125. [DOI] [PubMed] [Google Scholar]
- 19.Zhang Y, Zhang M, Li M, et al. Icariin-Enhanced osteoclast-Derived exosomes promote repair of infected bone defects by regulating osteoclast and osteoblast communication. Int J Nanomed. 2024;19:12389–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Teng J-W, Bian S-S, Kong P, Chen Y-G. Icariin triggers osteogenic differentiation of bone marrow stem cells by up-regulating miR-335–5p. Exp Cell Res. 2022;414(2):113085. [DOI] [PubMed] [Google Scholar]
- 21.Lian F, Zhao C, Qu J, et al. Icariin attenuates titanium particle-induced Inhibition of osteogenic differentiation and matrix mineralization via miR-21-5p. Cell Biol Int. 2018;42(8):931–9. [DOI] [PubMed] [Google Scholar]
- 22.Wang Z, Wang D, Yang D, et al. The effect of Icariin on bone metabolism and its potential clinical application. Osteoporos Int. 2018;29(3):535–44. [DOI] [PubMed] [Google Scholar]
- 23.Oliviero A, Della Porta G, Peretti GM, Maffulli N. MicroRNA in osteoarthritis: physiopathology, diagnosis and therapeutic challenge. Br Med Bull. 2019;130(1):137–47. [DOI] [PubMed] [Google Scholar]
- 24.Gargano G, Asparago G, Spiezia F, Oliva F, Maffulli N. Small interfering RNAs in the management of human osteoporosis. Br Med Bull. 2023;148(1):58–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Gargano G, Oliva F, Oliviero A, Maffulli N. Small interfering RNAs in the management of human rheumatoid arthritis. Br Med Bull. 2022;142(1):34–43. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gargano G, Pagano SM, Maffulli N. Circular RNAs in the management of human osteoporosis. Br Med Bull. 2025;153(1). [DOI] [PubMed]
- 27.Feng L, Yang Z, Hou N, et al. MiR-378 mediates the ovariectomy induced bone loss via exaggerating osteoclastogenesis and transforming growth factor beta impaired osteogenesis. Genes Dis. 2026;13(1):101754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Huang H, Chen J, Lin X, Lin Z. Morinda officinalis polysaccharides inhibit osteoclast differentiation by regulating miR-214-3p/NEDD4L in postmenopausal osteoporosis mice. Calcif Tissue Int. 2024;115(5):673–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Lichawska-Cieslar A, Szukala W, Ylla G, et al. MCPIP1 modulates the miRNA–mRNA landscape in keratinocyte carcinomas. J Experimental Clin Cancer Res. 2024;43(1):290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Li F, Xu Z, Xie Z, et al. Adipose mesenchymal stem cells-derived exosomes alleviate osteoarthritis by transporting MicroRNA – 376c-3p and targeting the WNT-beta-catenin signaling axis. Apoptosis. 2023;28(3):362–78. [DOI] [PubMed] [Google Scholar]
- 31.Xue J, Zhu K, Cao P, et al. Ischemic preconditioning-induced protective effect for promoting angiogenesis in renal ischemia-reperfusion injury by regulating miR-376c-3p/HIF-1α/VEGF axis in male rats. Life Sci. 2022;299:120357. [DOI] [PubMed] [Google Scholar]
- 32.Magra M, Maffulli N. Matrix metalloproteases: a role in overuse tendinopathies. Br J Sports Med. 2005;39(11):789–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Diniz-Fernandes T, Godoy-Santos AL, Santos MC, et al. Matrix metalloproteinase-1 (MMP-1) and (MMP-8) gene polymorphisms promote increase and remodeling of the collagen III and V in posterior tibial tendinopathy. Histol Histopathol. 2018;33(9):929–36. [DOI] [PubMed] [Google Scholar]
- 34.Jiang J, Wu Q, Rajasekaran S, Wu R. MMP3 at the crossroads: linking molecular pathways to disease diagnosis and therapy. Pharmacol Res. 2025;216:107750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Kulesza M, Kicman A, Motyka J, Guszczyn T, Ławicki S. Importance of metalloproteinase enzyme group in selected skeletal system diseases. Int J Mol Sci [Internet]. 2023; 24(24). [DOI] [PMC free article] [PubMed]
- 36.Garcia AJ, Tom C, Guemes M, et al. ERα signaling regulates MMP3 expression to induce FasL cleavage and osteoclast apoptosis. J Bone Min Res. 2013;28(2):283–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Gong X, Sun S, Yang Y, et al. Osteoblastic STAT3 is crucial for orthodontic force driving alveolar bone remodeling and tooth movement. J Bone Min Res. 2023;38(1):214–27. [DOI] [PubMed] [Google Scholar]
- 38.Singh K, Piprode V, Mhaske ST, Barhanpurkar-Naik A, Wani MR. IL-3 differentially regulates membrane and soluble RANKL in osteoblasts through metalloproteases and the JAK2/STAT5 pathway and improves the RANKL/OPG ratio in adult mice. J Immunol. 2018;200(2):595–606. [DOI] [PubMed] [Google Scholar]
- 39.Jia D, Li Y, Han R, et al. miR–146a–5p expression is upregulated by the CXCR4 antagonist TN14003 and attenuates SDF–1–induced cartilage degradation. Mol Med Rep. 2019;19(5):4388–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Korotkov A, Broekaart DWM, van Scheppingen J, et al. Increased expression of matrix metalloproteinase 3 can be attenuated by Inhibition of microRNA-155 in cultured human astrocytes. J Neuroinflamm. 2018;15(1):211. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Jehan F, Zarka M, de la Houssaye G, et al. New insights into the role of matrix metalloproteinase 3 (MMP3) in bone. FASEB Bioadv. 2022;4(8):524–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Pan M, Pan X, Zhou J, et al. Update on hormone therapy for the management of postmenopausal women. Biosci Trends. 2022;16(1):46–57. [DOI] [PubMed] [Google Scholar]
- 43.Li J, Fu SF, Yang Y, et al. Clinical practice of traditional Chinese medicine for the treatment of postmenopausal osteoporosis: a literature review. Climacteric. 2022;25(6):562–9. [DOI] [PubMed] [Google Scholar]
- 44.Zhuang W, Sun N, Gu C, et al. A literature review on Epimedium, a medicinal plant with promising slow aging properties. Heliyon. 2023;9(11):e21226. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Liu M, Zhong C, He R-x. Chen L-f. Icariin associated with exercise therapy is an effective treatment for postmenopausal osteoporosis. Chin Med J. 2012;125(10). [PubMed]
- 46.Li Z, Li D, Chen R, et al. Cell death regulation: A new way for natural products to treat osteoporosis. Pharmacol Res. 2023;187:106635. [DOI] [PubMed] [Google Scholar]
- 47.Indran IR, Liang RLZ, Min TE, Yong E-L. Preclinical studies and clinical evaluation of compounds from the genus epimedium for osteoporosis and bone health. Pharmacol Ther. 2016;162:188–205. [DOI] [PubMed] [Google Scholar]
- 48.Zhang X, Liu T, Huang Y, Wismeijer D, Liu Y. Icariin: does it have an osteoinductive potential for bone tissue engineering? Phytother Res. 2014;28(4):498–509. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
No datasets were generated or analysed during the current study.