Circular RNA GCLC promotes osteogenic differentiation of bone marrow mesenchymal stem cells via modulating miR-516b-5p/GNAS axis

Abstract

Objective

Osteoporosis (OP) is a systemic bone disease without effective treatment at present. The aim of this study is to explore the role of the circular RNA GCLC (circGCLC) in the osteogenic differentiation of bone marrow mesenchymal stem cells (BMSCs).

Methods

The expression of circGCLC, miR-516b-5p, and the G protein alpha subunit (GNAS) was determined during osteogenic differentiation of BMSCs. Plasmid or oligonucleotide transfection was used to modify the expression levels of circGCLC, miR-516b-5p, and GNAS for assessing their effects on osteogenic differentiation ability, osteogenesis-related proteins, and MAPK signalling pathway-related proteins in BMSCs. The interactions among circGCLC, miR-516b-5p, and GNAS were also investigated.

Results

During osteogenic differentiation of BMSCs, circGCLC and GNAS levels increased, whereas miR-516b-5p expression decreased. The knockdown of circGCLC inhibited osteogenic differentiation and MAPK signalling pathway activation in BMSCs. The increase in osteogenic differentiation that was induced by circGCLC upregulation was reversed by overexpressing miR-516b-5p. The inhibitory effects of circGCLC silencing on BMSC osteogenic differentiation were counteracted by GNAS upregulation. CircGCLC competitively bound to miR-516b-5p, mediating GNAS expression.

Conclusion

CircGCLC activates the MAPK signalling pathway by sponging miR-516b-5p, which mediates GNAS expression, thereby promoting the osteogenic differentiation of BMSCs.

Introduction

Osteoporosis (OP) is a chronic, progressive metabolic bone disease, which is primarily characterized by a reduction in the bone mass and the disruption of the bone microarchitecture. This imbalance in bone remodelling results in a decreased bone strength and an increased risk of fragility fractures [1–3]. The increasing incidence of OP, which is driven by the ageing demographics and sedentary lifestyles, poses a health challenge worldwide. Despite the availability of antiresorptive and anabolic agents, current treatment options are often associated with limited efficacy, potential side effects, and poor long-term compliance. Consequently, there is a compelling need to develop novel and effective therapeutic treatments to prevent or reverse OP progression [4]. In this context, bone marrow-derived mesenchymal stem cells (BMSCs), particularly osteoblasts, have garnered considerable attention because of their intrinsic self-renewal ability and multilineage differentiation capacity. Owing to their low immunogenicity, ease of isolation, and potential for genetic modification, BMSCs are considered candidates for bone regenerative therapies [5–7]. A deeper understanding of the mechanisms governing the osteogenic differentiation of BMSCs is critical for harnessing their full therapeutic potential for OP treatment.

Recent studies have highlighted pivotal roles of noncoding RNAs as key regulators of gene expression during tendon healing, osteoarthritis, rheumatoid arthritis and osteoporosis [8–13]. CircRNAs are endogenous noncoding RNAs that are characterized by covalently closed loop structures, which confer high resistance to exonucleases and enhanced stability to circRNAs within the cell. CircRNAs exhibit tissue-specific expression patterns and have been increasingly found to function as crucial regulators of bone metabolism [14, 15]. Several circRNAs have been implicated in modulating the osteogenic differentiation of BMSCs. For example, circYAP1 has been shown to mitigate osteoporotic features by promoting osteoblastogenesis [16], whereas circ0062582 and circ0006215 have been reported to increase osteogenic differentiation via distinct regulatory pathways [17]. Importantly, transcriptomic profiling of patients with OP revealed that the expression of circGCLC was significantly lower—by approximately 5.24-fold—than that in healthy individuals [18]. These findings suggest that circGCLC is involved in bone homeostasis; however, its precise functional role and underlying mechanisms in the osteogenic differentiation of BMSCs remain largely unknown.

This study aimed to elucidate the regulatory functions of circGCLC during the osteogenic differentiation of BMSCs. We hypothesized that circGCLC might act as a competing endogenous RNA (ceRNA) to sequester miR-516b-5p, thereby modulating the expression of GNAS, a gene that is known to influence osteogenesis. By revealing the circGCLC/miR-516b-5p/GNAS regulatory axis, this research offers novel perspectives on the molecular landscape of BMSC osteogenesis and identifies potential therapeutic targets for OP treatment.

Materials and methods

Ethics statement

This study was approved by the Ethics Committee of the Funing People’s Hospital. Written informed consent was obtained from all participants, as were their procedural details. Bone marrow samples were collected from fifteen healthy volunteers for the isolation of BMSCs.

Isolation and culture of BMSCs

BMSCs were isolated following previously described protocols [19]. Briefly, bone marrow aspirates were centrifuged, and the mononuclear cells were resuspended in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, MA, USA) supplemented with 10% foetal bovine serum (FBS; Sijiqing, Hangzhou, China). The cells were plated and incubated at 37 °C in a humidified atmosphere of 5% CO₂. Nonadherent cells were removed after 48 h. The medium was changed every three days. Upon reaching 80–90% confluency, adherent cells were detached with 0.25% trypsin containing 0.53 mM EDTA and subcultured. Cells from passages 3–5 were used. Each functional assay was performed with three biological replicates of BMSCs from three independent donors, with each donor’s cells assayed in technical triplicate.

Identification of BMSCs by flow cytometry

The immunophenotype of cultured BMSCs was verified using flow cytometry, as previously described [20]. The cells were incubated with fluorophore-conjugated monoclonal antibodies against CD14, CD45, CD44, and CD105 (BD Biosciences, CA, USA). Analyses were conducted by using a BD Influx cell sorter (BD Biosciences). BMSCs were characterized by low expression of haematopoietic markers (CD14⁻ and CD45⁻) and high expression of mesenchymal markers (CD44⁺ and CD105⁺).

Transfection of BMSCs

BMSCs were transfected with chemically synthesized oligonucleotides, including a miR-516b-5p mimic (5’-AUCUGGAGGUAAGAAGCACUUU-3’) and inhibitor (5’-AAAGUGCUUCUUACCUCCAGAU-3’) and their respective negative controls (GenePharma, Shanghai, China). In addition, a small interfering RNA targeting circGCLC (si-circGCLC), overexpression plasmids for circGCLC and GNAS (oe-circGCLC and oe-GNAS), and negative control constructs (si-NC and oe-NC) were designed and obtained from RiboBio (Guangzhou, China). Transfections were conducted with Lipofectamine™ 2000 (Invitrogen, USA).

Osteogenic induction and Alizarin red staining

For osteogenic differentiation, BMSCs were cultured in osteogenic induction medium (Sigma‒Aldrich, USA) containing DMEM, 10% FBS, 100 IU/mL penicillin, 100 µg/mL streptomycin, 0.1 µM dexamethasone, 10 mM β-glycerophosphate, and 50 µM ascorbic acid. After 14 days, the cells were fixed with 4% paraformaldehyde and stained with 1% Alizarin Red S (ARS; pH 4.3). To quantify mineralized nodules, the ARS dye was eluted with 10% cetylpyridinium chloride monohydrate (Sigma‒Aldrich), and the absorbance was measured at 562 nm by using a microplate reader.

Quantitative real-time PCR (qPCR)

Total RNA was isolated with a phenol–chloroform-based reagent (TRIzol; Gibco, Thermo Fisher Scientific) following the supplier’s instructions. First-strand complementary DNA (cDNA) was generated by using the PrimeScript™ reverse transcription master mix (Jiancheng Technology Co., Ltd.). The quantitative analysis of transcript levels was performed with a SYBR Green-based qPCR reagent (Toyobo Co., Ltd., Osaka, Japan) on the ABI 7300 quantitative PCR platform (Applied Biosystems, CA, USA). GAPDH or U6 small nuclear RNA was used as the endogenous normalization control. Relative mRNA expression was determined using the comparative Ct (2^−ΔΔCt) method [21]. The oligonucleotide primer sequences used in this assay are provided in Table 1.

Table 1.

Primer sequences

	Sequences (5′- 3’)
GAPDH	Forward: 5’- ATCTTCCAGGAGCGAGATCCC-3’
	Reverse: 5’- TGAGTCCTTCCACGATACCAA-3’
U6	Forward: 5’- CTCGCTTCGGCAGCACA-3’
	Reverse: 5’- AACGCTTCACGAATTTGCGT-3’
CircGCLC	Forward: 5’- GCCAATATGAGCCAG − 3’
	Reverse: 5’- CTTTCTTGGGAATCCAG − 3’
miR-516b-5p	Forward: 5’- GCCGAGATCTGGAGGTAAGA − 3’
	Reverse: 5’- CTCAACTGGTGTCGTGGA − 3’
ALP	Forward: 5’- TACTCGGACAATGAGATGCGCC − 3’
	Reverse: 5’-TTGTGCATTAGCTGATAGGCGA − 3’

Note: GAPDH, Glyceraldehyde-3-phosphate dehydrogenase; CircGCLC, Circular RNA GCLC; miR-516b-5p, microRNA-516b-5p; ALP, alkaline phosphatase

Dual-luciferase reporter assay

To evaluate the direct binding of miR-516b-5p to circGCLC or GNAS, a dual-luciferase reporter assay was conducted. Reporter plasmids containing either the wild-type (WT) or mutated (MUT) 3′ untranslated region (UTR) sequences of circGCLC and GNAS were inserted into a luciferase expression vector. Human embryonic kidney 293T (HEK-293T) cells were cotransfected with these constructs, along with either a synthetic miR-516b-5p analogue or a nontargeting negative control (NC), by using the Lipofectamine™ 2000 transfection reagent. Luminescence output was assessed with the Dual-Luciferase^® reporter assay kit (Promega, USA), with firefly luciferase activity normalized against Renilla luciferase activity as an internal control.

RNA Immunoprecipitation (RIP) assay

To confirm the association between circGCLC and miR-516b-5p, RNA‒protein complex immunoprecipitation was conducted with the Magna RIP kit for RNA-binding protein analysis (Millipore). BMSCs were lysed with RIP lysis buffer supplemented with a protease inhibitor cocktail and an RNase inhibitor. The resulting cell extracts were incubated with magnetic beads coupled to either an anti-Ago2 monoclonal antibody (ab57113; Abcam) or an isotype-matched IgG control (7076; Cell Signaling Technology). Following RNA extraction from the immunoprecipitates, the levels of circGCLC and miR-516b-5p were assessed using qRT‒PCR.

Fluorescence in situ hybridization (FISH)

The intracellular distributions of circGCLC and miR-516b-5p were examined through FISH. Bone marrow-derived mesenchymal stromal cells were fixed with paraformaldehyde, permeabilized with a detergent, and incubated with fluorescently labelled oligonucleotide probes specific for circGCLC (tagged with Cy3) and miR-516b-5p (tagged with FAM) (Servicebio, Wuhan, China). The cell nuclei were counterstained with DAPI. Fluorescence images were captured using a confocal laser scanning microscope (TCS SP8; Leica Microsystems) [22].

Western blotting

Proteins were extracted from BMSCs with RIPA lysis buffer (Beyotime, China) and quantified with a BCA protein assay kit. Equal amounts of protein were resolved via SDS‒PAGE and transferred onto nitrocellulose membranes. The membranes were blocked and probed with primary antibodies against RUNX2 (ab23981), OSX (ab22552), OCN (ab93876), p-JNK (9251), JNK (9252), p-p38 (9211), p38 (9212), GNAS (ab83735), and GAPDH (2118), all at 1:1000 dilutions. After incubation with an HRP-conjugated secondary antibody (ab6721, Abcam; 1:2000), the protein bands were visualized with enhanced chemiluminescence (ECL) reagents and quantified with the ImageJ software.

Statistical analysis

Each assay was independently conducted three times. The results are expressed as the mean ± standard deviation. Statistical computations were carried out using Prism version 9.0 (GraphPad Software, CA, USA). The data distribution was evaluated using the Kolmogorov–Smirnov normality test. For comparisons between two groups, an unpaired Student’s t test was applied, while multiple group differences were assessed via one-way ANOVA, followed by Tukey’s multiple comparisons test. Differences were deemed statistically significant when p values were less than 0.05.

Results

CircGCLC expression is upregulated during osteogenic differentiation of BMSCs

The cells expressed characteristic mesenchymal surface markers, with positive expression of CD105 and CD44 and negative expression of the haematopoietic markers CD14 and CD45 (Fig. 1A). Following osteogenic induction, enhanced mineralization was observed using Alizarin Red S (ARS) staining, along with ALP upregulation, indicating successful cell differentiation (Fig. 1B, C). Importantly, quantitative PCR analysis revealed that circGCLC expression was significantly elevated during osteogenic differentiation (Fig. 1D), suggesting a potential regulatory role for circGCLC in this process.

Fig. 1.

Increased expression of circGCLC during the osteogenic differentiation of BMSCs. A. Flow cytometry to identify the surface markers CD14, CD45, CD44 and CD105 of BMSCs; B. ARS staining to detect the osteogenic differentiation and calcification ability of BMSCs, scale bar = 50 μm; C. PCR to detect ALP in BMSCs; D. PCR to detect circGCLC expression during osteogenic differentiation; data were expressed as mean ± SD (N = 3). *P < 0.05

Knockdown of CircGCLC impairs osteogenic differentiation of BMSCs

To investigate the functional role of circGCLC, siRNA-mediated knockdown of circGCLC was performed in BMSCs. qPCR confirmed the effective silencing of circGCLC (Fig. 2A). Functionally, circGCLC knockdown led to decreased matrix mineralization and a reduced ALP mRNA level (Fig. 2B, C). Furthermore, western blot analysis revealed that the levels of RUNX2, OSX, and OCN were reduced in circGCLC-depleted BMSCs (Fig. 2D). Given the importance of the MAPK pathway in osteoblastogenesis [19], we evaluated the phosphorylation status of JNK and p38. Notably, p-JNK and p-p38 levels were decreased in circGCLC-silenced cells, whereas total JNK and p38 levels remained unchanged (Fig. 2E). These data collectively suggest that circGCLC facilitates the osteogenic differentiation of BMSCs, potentially through MAPK signalling activation.

Fig. 2.

CircGCLC knockdown prevents BMSCs from osteogenic differentiation. A. PCR to detect circGCLC in BMSCs after circGCLC knockdown; B. ARS staining to detect the osteogenic differentiation and calcification ability of BMSCs after circGCLC knockdown, scale bar = 50 μm; C. PCR to detect ALP expression in BMSCs after circGCLC knockdown; D-E. Western blot to measure RUNX2, OSX, OCN, p-JNK and p-p38 in BMSCs after circGCLC knockdown; data were presented as mean ± SD (N = 3)

CircGCLC functions as a CeRNA to inhibit miR-516b-5p expression

On the basis of the competitive endogenous RNA (ceRNA) hypothesis [23], we explored potential miRNA targets of circGCLC. miR-516b-5p was selected on the basis of bioinformatic predictions and its reported involvement in bone metabolism [24]. During osteogenic induction, miR-516b-5p expression decreased progressively (Fig. 3A), indicating an inverse relationship with circGCLC expression. In circGCLC-silenced BMSCs, miR-516b-5p was upregulated (Fig. 3B), suggesting that circGCLC negatively regulates miR-516b-5p expression.

Fig. 3.

CircGCLC imposes an inhibitory impact on miR-516b-5p expression. A. PCR to detect miR-516b-5p during osteogenic differentiation of BMSCs; B. PCR to detect miR-516b-5p in BMSCs after circGCLC knockdown; C. starbase to predict the potential binding sites of circGCLC and miR-516b-5p; D. FISH experiment to detect the co-localization of circGCLC and miR-516b-5p in BMSCs, scale bar = 50 μm; E-F. Dual-luciferase reporter assay and RIP assay to detect the targeting relationship between circGCLC and miR-516b-5p; data were presented as mean ± SD (N = 3)

The binding sites between circGCLC and miR-516b-5p were predicted using online bioinformatics tools (Fig. 3C). Colocalization of circGCLC and miR-516b-5p in the cytoplasm was confirmed via FISH, further supporting their interaction (Fig. 3D). The miR-516b-5p mimic repressed luciferase activity in cells transfected with wild-type circGCLC constructs but not in those transfected with mutant constructs, which lacked the miRNA binding site (Fig. 3E). RIP with an anti-Ago2 antibody revealed significant enrichment of both circGCLC and miR-516b-5p, indicating their coexistence in the RISC complex (Fig. 3F).

CircGCLC promotes osteogenesis via miR-516b-5p Inhibition

To verify that circGCLC promotes osteogenesis by inhibiting miR-516b-5p, rescue experiments were performed. In BMSCs overexpressing circGCLC, cotransfection with a miR-516b-5p inhibitor increased ALP expression, matrix mineralization (Fig. 4B, C), and the protein expression of RUNX2, OSX, and OCN (Fig. 4D). These effects were accompanied by increased activation of MAPK signalling, as shown by the elevated p-JNK and p-p38 levels (Fig. 4E). The restoration of miR-516b-5p expression in circGCLC-overexpressing cells abolished these effects, resulting in the suppression of osteogenic marker expression and MAPK pathway activation (Fig. 4B–E). These findings suggest that circGCLC exerts its proosteogenic function by sponging miR-516b-5p.

Fig. 4.

CircGCLC triggers osteogenic differentiation of BMSCs through miR-516b-5p. A. PCR to detect miR-516b-5p in BMSCs after co-transfection; B. ARS staining to detect the osteogenic differentiation and calcification ability of BMSCs after co-transfection; scale bar = 50 μm; white arrows indicate the areas of co-localization; C. PCR to detect ALP expression in BMSCs after co-transfection; D-E. Western blot to measure RUNX2, OSX, OCN, p-JNK and p-p38 in BMSCs after co-transfection; data were presented as mean ± SD (N = 3)

GNAS is a downstream target of miR-516b-5p

GNAS was found to be involved in bone formation and osteogenic differentiation of BMSCs [25, 26]. Our data revealed that GNAS mRNA expression increased during the osteogenic induction of BMSCs (Fig. 5A). Bioinformatic analyses revealed the binding sites between miR-516b-5p and the 3’ untranslated region (UTR) of GNAS (Fig. 5B). miR-516b-5p inhibition elevated GNAS expression (Fig. 5C). Dual-luciferase reporter assays demonstrated that the miR-516b-5p mimic reduced luciferase activity of the GNAS-WT construct but had no effect on the GNAS-MUT construct (Fig. 5D). RIP assays confirmed the interaction between miR-516b-5p and GNAS (Fig. 5E), indicating that GNAS is a bona fide downstream target of miR-516b-5p.

Fig. 5.

GNAS is a target of miR-516b-5p. A. Western blot to detect GNAS expression after osteogenic induction of BMSCs; B. starbase to predict potential binding sites of GNAS and miR-516b-5p; C. Western blot to detect GNAS in BMSCs after miR-516b-5p inhibition; D-E. Dual-luciferase reporter assay and RIP assay to detect the targeting relationship of GNAS and miR-516b-5p; data were presented as mean ± SD (N = 3)

The circGCLC/miR-516b-5p/GNAS axis facilitates osteogenic differentiation of BMSCs

To further verify the regulatory axis, we examined whether GNAS mediates the effects of circGCLC on osteogenesis. The knockdown of circGCLC significantly reduced GNAS expression, while circGCLC overexpression had the opposite effect (Fig. 6A). Importantly, the overexpression of GNAS in circGCLC-silenced BMSCs rescued the suppressed phenotype, restoring ALP expression, calcium deposition (Fig. 6C, D), and protein levels of osteogenic markers (Fig. 6E), as well as those of p-JNK and p-p38 (Fig. 6F).

Fig. 6.

CircGCLC/miR-516b-5p/GNAS axis promotes BMSCs differentiating into osteoblasts. A. Western blot to detect GNAS expression in BMSCs after circGCLC knockdown or overexpression; B. Western blot to detect GNAS expression in BMSCs after co-transfection; C. ARS staining to detect the osteogenic differentiation and calcification ability of BMSCs after co-transfection; scale bar = 50 μm; D. PCR to detect ALP expression in BMSCs after co-transfection; E-F. Western blot to measure RUNX2, OSX, OCN, p-JNK and p-p38 in BMSCs after co-transfection; data were presented as mean ± SD (N = 3)

Discussion

OP is a complex skeletal metabolic condition that is characterized by reduced bone density, disruption of trabecular and cortical bone structures, and increased susceptibility to fractures [27, 28]. With the global rise in ageing populations, the incidence of OP continues to increase annually, indicating that OP is a significant public health concern [29]. In addition to the risk of fractures, OP is frequently associated with comorbidities such as intervertebral disc degeneration, particularly in middle-aged and older individuals [30]. This degenerative synergy is especially pronounced in postmenopausal women, in whom a more severe osteoporotic phenotype correlates with more pronounced disc degeneration [31]. These findings underscore the urgent need to elucidate the mechanisms governing bone metabolism and regeneration, which may provide novel diagnostic and therapeutic strategies for OP. Biochemical signals, such as epidermal growth factor (EGF), which maintains mesenchymal cells in an undifferentiated state, promote chondrocyte proliferation while delaying hypertrophy and orchestrate osteoblast and osteoclast activities [32, 33]. Concurrently, biophysical stimuli, such as hypoxic culture conditions, which suppress BMP2, promote MSC colony formation, and pulsed electromagnetic fields accelerate osteoblast proliferation and alkaline phosphatase activity. These findings underscore the multifaceted control of bone development and homeostasis and suggest novel combinatorial strategies for osteoporosis therapy [34, 35]. Recent advances have highlighted circular RNAs (circRNAs) as emerging regulators and potential biomarkers in OP [17, 36]. Consistent with these observations, our study identified a novel regulatory axis in which circGCLC functions as a miR-516b-5p sponge, enhancing GNAS expression and promoting osteogenic differentiation. We also revealed that these effects are mediated, at least in part, through MAPK signalling, suggesting that a previously unrecognized circGCLC/miR-516b-5p/GNAS/MAPK axis is involved in osteogenesis.

BMSCs are multipotent stem cells with robust osteogenic potential, which makes them ideal candidates for regenerative strategies in OP. Our findings indicate that circGCLC expression is progressively upregulated during osteogenic induction of BMSCs, indicating that circGCLC is involved in the initiation and maintenance of osteogenic commitment. Functional knockdown of circGCLC significantly impaired BMSC calcification and downregulated alkaline phosphatase (ALP) expression, both of which are hallmarks of osteoblast differentiation. These results agree with recent findings that circGCLC expression is reduced in osteoporotic patients [37], further supporting its proosteogenic role.

Previous studies have reported that circRNAs can regulate osteogenic markers [36, 38, 39]. Our data reinforce this hypothesis, showing that circGCLC upregulates RUNX2, OSX, and OCN protein levels during osteogenesis. Furthermore, mechanistic studies revealed that circGCLC positively regulated the MAPK signalling pathway, as evidenced by elevated p-JNK and p-p38 levels upon circGCLC expression [40, 41]. Conversely, circGCLC knockdown attenuated MAPK pathway activation, indicating that MAPK signalling serves as a downstream effector of circGCLC. Although our findings suggest a role for MAPK activation in circGCLC-mediated osteogenesis, further studies using genetic knockout models of MAPK pathway components are warranted to validate this mechanism.

In line with the function of circRNAs as miRNA sponges, we identified miR-516b-5p as a target of circGCLC. Colocalization studies confirmed their interaction in the cytoplasm of BMSCs. miR-516b-5p is downregulated in several cancers, including osteosarcoma [42], bladder cancer [43], and ESCC [44]. Importantly, a recent study showed that miR-516b-5p suppressed osteogenic differentiation in osteoporotic models [24]. Our results are consistent with these findings, demonstrating that miR-516b-5p overexpression inhibits osteogenic differentiation of BMSCs and suppresses RUNX2, OSX, and OCN expression. Given that BMSCs are the principal progenitor cells responsible for bone regeneration, these data suggest that miR-516b-5p is key for inhibiting osteoblastogenesis and may contribute to OP pathogenesis.

Through bioinformatic prediction and dual-luciferase assays, we identified GNAS as a target of miR-516b-5p. GNAS encodes the stimulatory G protein α-subunit, which is involved in multiple signalling cascades that are critical for osteogenesis. GNAS downregulation impairs the osteogenic differentiation of BMSCs [25], whereas miRNAs such as miR-196a promote osteogenesis by upregulating GNAS expression [45]. Our findings align with this evidence and suggest that circGCLC indirectly increases GNAS expression by sequestering miR-516b-5p. However, a limitation of the current study is the lack of in vivo validation. Future research utilizing osteoporotic animal models is necessary to substantiate the functional role of the circGCLC/miR-516b-5p/GNAS axis and to evaluate its translational potential in OP therapy.

In addition to its proosteogenic effects on BMSCs, GNAS signalling has been shown to play a pivotal role in osteoclast differentiation and activity, notably by modulating RANKL-induced NFATc1 activation and resorptive enzyme expression [46, 47]. Because circGCLC sponges miR-516b-5p to increase GNAS levels, this axis may also enhance osteoclastogenesis or alter resorptive function. To test this hypothesis, future work should examine osteoclast formation and bone resorption in coculture systems of BMSCs and osteoclast precursors, as well as in osteoporotic animal models following circGCLC overexpression or miR-516b-5p inhibition. These studies will help determine whether targeting the circGCLC/miR-516b-5p/GNAS pathway can rebalance both arms of bone remodelling for more comprehensive osteoporosis therapy.

A potential limitation of our study is the relatively small number of human donors from whom primary BMSCs were isolated. However, in our study, each experiment was conducted in biological triplicates by using independent isolates from three healthy adult donors, and technical duplicates were performed for each assay, thereby capturing both interdonor variability and methodological consistency.

Conclusions

In conclusion, our study identified a novel regulatory pathway in which circGCLC acts as a ceRNA to modulate the miR-516b-5p/GNAS axis, thereby promoting osteogenic differentiation of BMSCs via MAPK signalling activation. These findings not only improve the current understanding of circRNA-mediated gene regulation in bone biology but also highlight the therapeutic potential of targeting the circGCLC/miR-516b-5p/GNAS axis for treating osteoporosis. Future research is warranted to explore the clinical applicability of this pathway in OP management.

Author contributions

AL and QF designed the whole study; AL, SZ, JZ and ZP performed the analysis and experiments; AL, SZ, JZ, ZP and DG wrote and edited the manuscript; all authors approved for submission.

Funding

This study was supported by Scientific Research Project of Guangdong Provincial Bureau of Traditional Chinese Medicine (No.20211412).

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethical approval

An approval was collected from Ethics Committee of Funing People’s Hospital. Fifteen healthy volunteers were informed of possible risks and complications and signed informed consent.

Patient consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.