The crosstalk network of non-coding RNAs: Emerging opportunities for the clinical application of osteoporosis.

Introduction

Osteoporosis is a systemic skeletal disease that is mainly characterised by a decrease in bone mass and damage to the bone microstructure, which leads to a reduction in bone strength and an increased risk of fractures [ 1 ]. As individuals age, bone mass is gradually lost, and the risk of developing osteoporosis increases. Notably, the disease does not present with clear precursors or symptoms, and patients often seek medical treatment only after fractures have occurred. As a result, osteoporosis is frequently referred to as the ‘silent disease’ [ 2 ]. Fractures are one of the most severe consequences of osteoporosis, causing significant pain and economic burden, and profoundly impacting the quality of life for patients. Therefore, the effective prevention and early diagnosis of osteoporosis are crucial. However, commonly used diagnostic methods, such as imaging examinations, are limited in that they provide only partial information about bone strength and lack the sensitivity needed to detect early bone mass loss. Additionally, drugs such as bisphosphonates, calcitonin preparations, and parathyroid hormone analogues, which are recognised by the World Health Organization for osteoporosis treatment, all have associated adverse effects. For example, bisphosphonates can cause gastrointestinal reactions such as nausea, vomiting, abdominal pain, and diarrhoea [ 3 ]. Therefore, the development of more effective and safer anti-bone loss drugs is crucial for preventing and treating osteoporosis. Additionally, biomarkers with both diagnostic specificity and sensitivity can be used to complement the limitations of existing detection methods and thus warrant urgent exploration.

Non-coding ribonucleic acids (ncRNAs) are a class of RNA molecules that do not encode proteins. They include microRNAs (miRNAs), long non-coding RNAs (lncRNAs), circular RNAs (circRNAs), small interfering RNAs (siRNAs), PIWI-interacting RNAs (piRNAs), and small nucleolar RNAs (snoRNAs). Far from being isolated entities, these diverse types of ncRNAs are intricately interconnected, resembling a highly intertwined network. Amongst them, miRNAs, lncRNAs, and circRNAs, are the principal components, forming a highly complex regulatory network. These molecules collaborate with and regulate one another, influencing gene transcription processes and intricately participating in various molecular signalling pathways [ [4] , [5] , [6] ]. Importantly, the construction of ncRNA networks often enables the simultaneous regulation of multiple targets and signalling pathways in osteoporosis [ 7 , 8 ]. This prospect brings new hope for potential breakthroughs in gaining a deeper understanding of the disease, making ncRNAs a highly promising research focus in the field of osteoporosis. Therefore, a thorough analysis of the mechanisms by which ncRNAs function in osteoporosis is expected to open up new avenues for the diagnosis, treatment, and prevention of the disease.

Extracellular vesicles (EVs) are a collection of nanoscale membrane structures, including exosomes, microvesicles, and apoptotic bodies, which are widely present in biological systems. Almost all cell types in the body are capable of secreting EVs. These structures carry various biomolecules (e.g., lipids, proteins, RNA, etc.) originating from the parent cell, are transmitted through bodily fluids such as blood and lymph, and participate in intercellular communication and regulation [ 9 ]. The effects of EVs vary depending on their source and type and can even differ among those derived from the same cell type at different life stages. An increasing body of evidence has indicated that EVs can regulate bone metabolism through the biomolecules they carry, especially ncRNAs.

Therefore, through their mediation of cross-talk mechanisms regulating various ncRNAs, EVs hold promise as a novel therapeutic strategy for osteoporosis. In this article, knowledge gaps regarding these networks are addressed by reviewing ncRNAs used in the prevention, diagnosis, and treatment of osteoporosis over the past five years and outlining the potential applications of EV-secreted ncRNAs in treating the disease. Additionally, the challenges faced in the research of EV-secreted ncRNAs in recent years are critically discussed, and the future development direction of the clinical application of the cross-talk network of EV-secreted ncRNAs in osteoporosis is highlighted.

ncRNAs

Despite the fact that ncRNAs do not encode proteins, they nevertheless play diverse roles in organisms, including regulating gene expression, participating in cell differentiation and development, and influencing disease pathogenesis. Moreover, they are involved in the protein translation process and are key in helping RNA fulfil its functions. The most common types of ncRNAs are miRNAs, lncRNAs, and circRNAs, but siRNAs, piRNAs, and snoRNAs have also become increasingly recognised. Initially, these ncRNAs were considered by-products of RNA metabolism and degradation. However, a growing body of evidence has shown that they play regulatory roles in various cells and signalling pathways and are essential for the development of multiple diseases [ [10] , [11] , [12] ]. The isolation [ [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] ], function [ [23] , [24] , [25] , [26] , [27] , [28] , [29] , [30] , [31] , [32] ], and species specificity [ [33] , [34] , [35] , [36] , [37] , [38] , [39] , [40] , [41] ] of ncRNAs are briefly summarised in the subsections below ( Table 1 ).

Biogenesis of ncRNAs

Biogenesis of miRNAs

The biogenesis of miRNAs involves multiple steps [ 42 ] ( Fig. 1 ): (1) Generation of primary miRNAs (pri-miRNAs): In the nucleus, miRNA genes are transcribed by RNA polymerase II (RNA Pol II) into pri-miRNAs, which have complex secondary structures ranging in length from several hundred to several thousand nucleotides. This process is similar to the transcription mechanism of messenger RNA (mRNA). The initiation of transcription requires the participation of regulatory elements in the promoter region, which can be regulated by various transcription factors. (2) Generation of precursor miRNAs (pre-miRNAs): The RNase III endonuclease Drosha recognises specific structural features at the junction between the double-stranded region and the single-stranded loop of pri-miRNA and cleaves it, releasing pre-miRNA which has a characteristic hairpin-like secondary structure [ 43 ]. This precise cutting is crucial for generating a pre-miRNA with the correct structure and functional potential. (3) Generation of mature miRNAs: The pre-miRNA is transported from the nucleus to the cytoplasm, where the RNase III endonuclease Dicer recognises its double-stranded structure and cleaves it at the loop region, generating a double-stranded miRNA product of approximately 20–25 nucleotides in length [ 43 ]. Subsequently, the double-stranded miRNA is integrated into the RNA-induced silencing complex (RISC). During RISC assembly, one strand is degraded or released, whereas the other strand becomes the mature miRNA and binds tightly to the Argonaute (AGO) protein in RISC. In summary, miRNA biogenesis is a multi-step process involving the coordinated action of various molecules and mechanisms in both the nucleus and cytoplasm and is affected by multiple factors.

Biogenesis of lncRNAs

The biogenesis of lncRNAs is relatively complex and mainly includes the following key steps [ 44 ] ( Fig. 1 ): (1) Transcription initiation: lncRNAs can be transcribed using an independent promoter or the promoter of other genes. RNA Pol II may mediate transcription initiation, and the transcription start site may be located at different positions in the genome, including the intergenic and intronic regions. The transcription initiation of some lncRNAs may be regulated by specific transcription factors that bind to the promoter region of the lncRNA to initiate transcription. (2) Transcription extension: RNA Pol II moves along the DNA template to synthesise the lncRNA transcript. During this process, the transcript may undergo various processing steps and modifications. (3) Transcription termination: Specific termination factors bind to the lncRNA transcripts and promote the termination of transcription. (4) Post-transcriptional processing: The post-transcriptional processing of lncRNAs includes splicing, modification, and transport. The lncRNA transcript may undergo different modifications, such as methylation and acetylation [ 45 ]. Additionally, the transcript needs to be transported to specific cellular regions, such as the nucleus or cytoplasm, to exert its function. This transport process may be regulated by specific transport factors. In summary, the biogenesis of lncRNAs is a complex and multi-step process involving various stages, including transcription initiation, transcription elongation, transcription termination, and post-transcriptional processing. These stages are regulated by different mechanisms that jointly determine the expression level and function of lncRNAs. In-depth research on the biogenesis of lncRNAs enhances our understanding of their biological functions and provides new insights and methods for the diagnosis and treatment of diseases.

Biogenesis of circRNAs

The biogenesis of circRNAs generally includes four steps ( Fig. 1 ): transcription initiation, transcription elongation, reverse splicing, and post-transcriptional processing and accumulation. In the normal linear RNA splicing process, exons are connected sequentially. However, in the biogenesis of circRNAs, the 5′ end of the downstream exon is connected to the 3′ end of the upstream exon, forming a circular structure [ 46 ]. This reverse-splicing process competes with traditional linear RNA splicing, and its occurrence efficiency is typically low [ 46 ]. Complementary sequences in the flanking introns surrounding circular exons act as important cis-regulatory elements. These complementary sequences can bring the ends of the exons closer together through base pairing, thereby promoting reverse splicing. RNA-binding proteins with binding sites on flanking introns can also bring the splicing sites on both sides of the exon closer, further promoting circRNA formation [ 47 ]. CircRNAs are formed from pre-mRNAs through the reverse splicing of introns or exons during transcription, which produces either intron-retaining or exon-only circRNAs, unlike linear RNA [ 48 ]. In conclusion, elucidating the biogenesis of circRNAs is essential for deciphering their biological roles and disease mechanisms and gaining strategies for therapy development.

Biogenesis of other ncRNAs

As an effective gene silencing tool, the demand for siRNAs has increased following the development of siRNA drugs. One research group proposed the use of Almac's 3-2-3-2 hybrid RNA ligase enzyme strategy to synthesise target siRNAs by sequentially ligating short oligonucleotide fragments (blockmers), providing an effective approach for their large-scale production ( Fig. 2 ) [ 49 ]. This method can be carried out under relatively mild conditions, avoiding the side reactions and product damage that may occur in traditional chemical synthesis methods.

The biogenesis of piRNAs is a complex and important biological process. Their production in germ cells involves the transcription of specific genomic regions (e.g. piRNA clusters), followed by a series of complex processing steps, including primary and secondary processing, which ultimately lead to the formation of mature piRNAs ( Fig. 2 ). Although the mechanism of piRNA production in somatic cells shares some similarities with that in germ cells, there are also distinct dissimilarities, such as differences in transcription start sites and the utilisation of different processing enzymes [ 50 ]. Studies have indicated the existence of a unique protein complex, named the Caenorhabditis elegans piRNA biogenesis and chromosome segregation (PICS) complex, which is composed of multiple subunits, including two mutually exclusive factors. The PICS complex has been proven effective in mediating piRNA biogenesis and cell division [ 51 ].

In eukaryotes, the transcription of snoRNA genes is typically initiated by RNA Pol II, which recognises and binds to the promoter region of the gene ( Fig. 2 ). During transcription elongation, snoRNA undergoes co-transcriptional assembly with related proteins ( Fig. 2 ). One study found that the biogenesis of snoRNAs is tightly coupled with the transcription process in Saccharomyces cerevisiae . The assembly factor Naf1p and the core proteins Cbf5p and Nhp2p are recruited to chromatin during the early transcription stage in a Ctk1p-dependent manner. Their correct assembly is essential for the proper 3′-end formation of snoRNAs and effectively prevents transcriptional read-through [ 52 ]. During post-transcriptional processing, some snoRNA transcripts may need to be spliced to remove non-coding sequences (e.g., introns) to form the mature snoRNA structure. Additionally, the endonuclease Rnt1 plays a crucial role in determining the nuclear localisation fate of yeast snoRNAs. During the co-transcriptional process, Rnt1 performs specific enzymatic cleavage operations on the snoRNA precursor, resulting in the subsequent processing and localisation of snoRNA [ 53 ].

In summary, ncRNA biosynthesis is a complex multi-step process involving diverse RNA polymerases and post-transcriptional modifications. These ncRNAs are essential regulators of cellular functions and disease mechanisms ( Fig. 1 , Fig. 2 ).

Clinical applications of ncRNAs and their networks in osteoporosis

Osteoporosis is a multi-factorial disease marked by low bone density and poor microstructure, which increases the susceptibility to fragility fractures [ 54 ]. The human body reaches peak bone mass at approximately 28–30 years of age [ 55 ], after which the bone density begins to decline. In particular, women experience an accelerated loss of bone mass after menopause [ 56 ]. Recent studies report a 12.6 % prevalence of osteoporosis among US adults aged 50 years or older [ 57 ]. Additionally, osteoporotic fractures significantly increase disability and mortality rates amongst patients [ 58 ]. Therefore, active research on effective preventive and treatment methods for the disease and the exploration of more accurate diagnostic indicators are important areas of focus. Despite the emerging importance of ncRNAs in osteoporosis management, their exact mechanisms of action remain incompletely understood.

Clinical applications of miRNAs in osteoporosis

miRNAs and osteoporosis prevention

As key regulators of bone homeostasis, miRNAs influence the progression of osteoporosis via targeted gene networks, offering diagnostic and therapeutic avenues. Chen et al. [ 59 ] treated human skeletal muscle myoblasts (HSMMs) with the histone deacetylases (HDACs) inhibitor Trichostatin A (TSA) and evaluated its impact on myotube formation and myogenic gene expression. They found that after TSA treatment, myotube fusion was enhanced, the cell volume increased, and the expression of myogenic genes was upregulated, indicating that TSA promotes muscle differentiation [ 59 ]. Importantly, the upregulation of these myogenic genes and the increase in EV secretion occurred simultaneously [ 59 ]. Furthermore, in vivo experiments showed that HDAC inhibition triggered the release of miR-873-3p-enriched EVs from HSMMs, which stimulated bone formation and counteracted ovariectomisation (OVX)-induced bone loss [ 59 ]. Additionally, TSA-carrying EVs (TSA-EVs) concurrently inhibited osteoclastogenesis [ 59 ]. The above-mentioned results demonstrate the dual mechanism of TSA-EVs in regulating bone remodelling. Therefore, the upregulation of myogenic genes such as myogenic differentiation 1 (Myod) and myogenic regulatory factor 4 (Mrf4) reflects an improved functional status of muscle cells, enhancing their ability to act as “secretory organs” to release EVs that promote bone function. The upregulation of myogenic genes is closely related to the osteogenic effect of TSA-EVs, which is mediated by paracrine pathways in the muscle–skeleton axis. Moreover, as key drug carriers, EVs can transfer HSMM-derived miR-873-3p to human bone marrow mesenchymal stem cells (hBMSCs) [ 59 ]. Specifically, the high expression of miR-873-3p in HSMM-derived EVs further promotes its expression in hBMSCs, and its upregulated expression is closely linked to bone density [ 59 ]. The researchers found that muscle-derived EVs overexpressing miR-873-3p effectively increased the bone density and promoted bone formation in a mouse model of osteoporosis. By contrast, knockdown of miR-873-3p eliminated the osteogenic benefits of TSA-EVs [ 59 ]. Moreover, miR-873-3p exerts its effect by directly targeting and inhibiting the expression of H2 calponin (CNN2), a negative regulator of osteogenesis [ 59 ]. The downregulation of CNN2 significantly promotes the proliferation, differentiation, and mineralisation of hBMSCs, thus driving bone formation. Therefore, targeting miR-873-3p may be a promising approach for treating bone loss diseases, and has significant potential as a therapeutic strategy to enhance bone formation and prevent osteoporosis. In the future, miRNAs may serve as key target molecules for preventing bone diseases ( Fig. 3 ).

miRNAs and osteoporosis diagnosis

The quantitative polymerase chain reaction (qPCR)-based detection of changes in the miRNA expression of osteoporosis-related genes offers a non-invasive diagnostic approach. For example, the overexpression of miR-122-5p and miR-4516 in patients with osteoporosis positions both miRNAs as reliable diagnostic biomarkers [ 60 ]. You et al. [ 61 ] detected decreased miR-181a/c expression in bone tissues and circulating plasma from patients with osteoporosis as well as in mouse models of OVX-induced bone loss. Notably, BMSCs undergoing osteogenic differentiation exhibited elevated miR-181a/c levels [ 61 ]. Inhibition of miR-181a/c suppressed osteogenic gene expression, decreased alkaline phosphatase (ALP) activity, and impaired extracellular matrix mineralisation in BMSCs [ 61 ]. These findings demonstrate the critical role of miR-181a/c in BMSC differentiation and its potential as a diagnostic marker for osteoporosis. Gao et al. discovered 79 exosomal miRNAs (exo-miRNAs) with diagnostic potential, including postmenopausal osteoporosis (PMOP)/osteopenia (OPNA)-discriminative (9), PMOP-selective (17), and OPNA-selective (53) subsets [ 62 ]. These findings indicate that exo-miRNAs have the potential to serve as biomarkers for the clinical diagnosis of PMOP and OPNA. Overlapping of the differentially expressed miRNAs that showed the same change trend in the OPNA and PMOP groups revealed that the expression levels of six miRNAs and their 92 target genes were upregulated, whereas those of 10 miRNAs and their 145 target genes were downregulated [ 62 ]. The results of Gene Ontology and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses showed that the different concentrations of the overlapping upregulated circulating exo-miRNAs may reflect the progression of PMOP and OPNA mediated by the negative regulation of osteoclast differentiation as well as through the hypoxia-inducible factor-1 (HIF-1), mitogen-activated protein kinase (MAPK), and transforming growth factor-beta (TGF-β) signalling pathways. By contrast, different concentrations of the overlapping downregulated circulating exo-miRNAs may contribute to the progression of PMOP and OPNA by affecting osteoblast differentiation, the Forkhead box O (FoxO) signalling pathway, and cell apoptosis, among other mechanisms. Therefore, the exo-miRNAs in the blood circulation may participate in the development of PMOP and OPNA through multiple biological processes and show a concentration-dependent correlation [ 62 ]. PPlasma exo-miRNAs have emerged as game-changing diagnostic biomarkers for postmenopausal bone disorders. In conclusion, miRNAs as biomarkers have the advantages of high specificity, sensitivity, and suitability for the non-invasive detection and early diagnosis of osteoporosis ( Fig. 3 ). However, the lack of standardised detection methods and clinical verification remains a challenge to overcome.

miRNAs and osteoporosis treatment

miRNAs participate in both skeletal and osteoporosis development by regulating multiple targets and signalling pathways [ 63 , 64 ]. For instance, miR-181c-5p activates the Wnt3a/β-catenin signalling pathway to promote osteogenesis and inhibit osteoclastogenesis [ 65 ]. Taipaleenmäki et al. demonstrated that antagonising miR-19a/b in vivo can increase bone formation parameters. To investigate the molecular mechanisms, osteoblasts from the MC3T3-E1 cell line were transfected with miR-19a/b inhibitors or mimics. The results confirmed that Tgif1 protein abundance was diametrically modulated by miR-19a/b inhibitors (upregulation) and mimics (downregulation), verifying its direct targeting by the miRNA [ 66 ]. Therefore, targeting miR-19a/b could expand the current therapeutic arsenal for patients with osteoporosis with unmet clinical needs. In this study, combined treatment with parathyroid hormone and anti-miR-19a/b was found to have a strong synergistic effect in increasing all parameters of tibial osteoblast function and bone formation [ 66 ]. This indicates that miR-19a/b is an important downstream component of the parathyroid hormone signalling cascade, which opens up several avenues for improving the therapeutic effect of this hormone. Using in vivo and in vitro experiments, You et al. demonstrated that Wilms tumor 1-associated protein (WTAP) upregulates the N6-methyladenosine (m6A) ‘reader’ YTH N6-methyladenosine RNA-binding protein 1 (YTHDF1) to enhance the m6A modification of miR-181a/c, consequently suppressing secreted frizzled-related protein 1 (SFRP1) to drive BMSC commitment towards osteogenesis while suppressing adipogenesis ( Fig. 4 ). These findings position WTAP as a promising molecular target for regulating bone-fat balance in metabolic bone disorders [ 61 ]. Hong et al. reported that nuclear factor erythroid 2-related factor 2 (Nrf2; an epigenetic regulatory factor) can inhibit miR-214-3p levels when activated in bone marrow macrophages. This inhibition also suppressed osteoclast differentiation in vitro and can thus exert protective effects on skeletal health. Therefore, the negative regulation of miR-214-3p by Nrf2 may represent a promising therapeutic target for osteoporosis [ 67 ]. In recent years, several studies have confirmed that HDACs mainly regulate osteoclast differentiation and activity, making them potential treatment targets for osteoporosis. In vivo and in vitro studies by Zhang et al. have demonstrated that miR-26a-5p, delivered by EVs from urine-derived stem cells, targets HDAC4 to trigger HIF-1α–vascular endothelial growth factor A (VEGFA) signalling, coordinately enhancing bone formation and suppressing resorption, hence offering a potential diabetic osteoporosis treatment strategy [ 68 ]. Yan et al. demonstrated that the miRNA let-7g simultaneously suppresses foam cell generation and BMSC adipogenesis, thus positioning it as a dual target for osteoporosis intervention [ 69 ]. Therefore, advancing miRNA-based drugs that exploit these regulatory networks represents a promising frontier for precision osteoporosis treatment ( Fig. 3 ).

In summary, in terms of prevention, detecting the expression levels of specific miRNAs helps in identifying high-risk patient populations ( Fig. 3 ); in terms of diagnosis, the abnormal expression of miRNAs such as miR-122-5p can serve as potential biomarkers, the detected levels of which can assist in diagnosis since they are abnormally expressed in patients ( Fig. 3 ); in terms of treatment, the targeting of specific miRNAs is expected to provide new strategies for the treatment of osteoporosis and improve the prognosis of patients ( Fig. 3 ).

Clinical applications of lncRNAs in osteoporosis

lncRNAs and osteoporosis prevention

lncRNAs demonstrate significant potential for osteoporosis prevention. Zhao et al. found that the lncRNA GM15416 can regulate Fas cell surface death receptor (Fas) by binding to the gene transcription factor FBJ murine osteosarcoma viral oncogene homolog (c-Fos). In vivo experiments showed that GM15416 knockdown could increase the levels of osteogenic differentiation markers and correct osteogenic differentiation dysfunction [ 70 ]. This is expected to become an important factor in preventing bone loss. The osteoclast-suppressing activity of metastasis-associated lung adenocarcinoma transcript 1 (Malat1), evidenced by its downregulation during osteoclast differentiation, positions this conserved lncRNA as a potential bone-loss inhibitor [ 71 ]. To explore its specific mechanism of action, Zhao et al. knocked out Malat1 in bone marrow macrophages and RAW264.7 murine macrophages and stimulated the cells with receptor activator of nuclear factor kappa-B ligand (RANKL). They found that nuclear factor of activated T cells 1 (Nfatc1) showed an upward trend. The results of chromatin immunoprecipitation and qPCR assays further confirmed that the occupancy rate of Nfatc1-binding sites in the cathepsin K (Ctsk) (an Nfatc1 target gene) promoter region in Malat1-knockout RAW264.7 cells was increased after RANKL treatment [ 71 ]. Notably, in osteoclast precursor cells without Malat1, knocking out Nfatc1 reversed both the RANKL-induced osteoclastogenesis and increase in Ctsk expression [ 71 ]. Therefore, Malat1 knockout can increase the expression level of Nfatc1 and promote osteoclast differentiation. Additionally, co-immunoprecipitation detection revealed that knocking out Malat1 significantly promoted the interaction between TEA domain transcription factor 3 (Tead3) and Nfatc1. Knocking out Tead3 in RAW264.7 cells weakened the osteoclast differentiation induced by RANKL and downregulated the expression of Nfatc1 and Ctsk, whereas overexpressing Tead3 in RAW264.7 cells reversed these effects [ 71 ]. Therefore, the deletion of Malat1 enhanced the Tead3–Nfatc1 interaction to promote osteoclast differentiation. However, Malat1 cannot directly regulate the expression level of Tead3. In conclusion, this study confirmed the important role of the Malat1–Tead3–Nfatc1 axis in regulating osteoclastogenesis. Malat1 inhibited osteoclastogenesis in both mouse and human models, demonstrating therapeutic potential for preventing osteoporosis. Conversely, because Nfatc1 activation is required for RANKL-induced osteoclastogenesis, drugs that target the Tead3–Nfatc1 interaction also have great potential in preventing osteoporosis.

lncRNAs and osteoporosis diagnosis

lncRNAs are versatile regulators of gene expression at multiple levels. They participate in the occurrence and development of osteoporosis through various mechanisms. lncRNAs can serve as candidate causal genes for genome-wide association studies of bone mineral density. Abood et al. [ 72 ] identified 58 known and new lncRNAs that can serve as potential effector transcripts through methods such as transcriptome-wide association studies and expression quantitative trait loci co-localisation analysis. Although several lncRNAs are known to modulate bone remodelling, their precise actions and utility as osteoporosis biomarkers remain to be elucidated. A clinical analysis of BMSCs derived from patients with osteoporosis revealed elevated levels of the lncRNA DANCR, the overexpression of which impaired osteogenic differentiation and showed positive correlation with the disease severity, suggesting its pathogenic relevance [ 73 ]. This indicates that lncRNAs can be used as potential biomarkers to identify osteoporosis or as indicators for monitoring disease progression and treatment effects. Additionally, the authors of this study reported that the expression levels of both DANCR and miR-320a increased during the development of osteoporosis, whereas that of catenin beta 1 (CTNNB1) decreased; however, the opposite was true during osteogenic differentiation [ 73 ]. miR-320a directly targets CTNNB1 (as validated by luciferase assay) and downregulates its expression at both the mRNA and protein levels [ 73 ]. Despite lacking expression correlation, DANCR and miR-320a co-operatively suppress CTNNB1/β-catenin to inhibit Wnt signalling and impair osteogenesis. However, studies on their upstream regulatory network still need to be conducted. A high-throughput RNA sequencing study revealed 51 differentially expressed lncRNAs in blood samples from patients with PMOP versus normal controls [ 74 ]. This not only provides new clues for understanding the functions of lncRNAs but also helps towards the development of new diagnostic biomarkers for PMOP [ 74 ]. Using the reverse transcription-quantitative polymerase chain reaction (RT-qPCR), Huang et al. [ 75 ] detected that PAX-binding protein 1 antisense RNA 1 (PAXB-AS1) was significantly upregulated in the mouse model of OVX-induced osteoporosis. PAXB-AS1 impairs osteoblast function in osteoporosis through the miR-1252-5p–G protein subunit beta 1 (GNB1)-mediated activation of autophagy and suppression of viability [ 75 ], revealing the inseparable connection between PAXB-AS1 and osteoporosis and hence the potential of the protein to become a new diagnostic index for the disease. Furthermore, some lncRNAs that are differentially expressed in patients with osteoporosis may be similarly differentially expressed in other bone-related or systemic diseases. Therefore, clarifying the specificity of lncRNAs for the diagnosis of osteoporosis is also a key point that cannot be ignored. Moreover, some lncRNAs can form complexes with miRNAs to jointly affect gene expression. Whether such complexes can be used as new detection markers in clinical applications is an important research direction that needs to be explored.

lncRNAs and osteoporosis treatment

lncRNAs regulate bone metabolism through multiple mechanisms, including affecting stem cell differentiation and the functions of osteoblasts and osteoclasts, and regulating related signalling pathways. Zhao et al. [ 70 ] confirmed a significant correlation between c-Fos and the lncRNA GM15416 using RNA immunoprecipitation and RNA pull-down assays. Furthermore, immunofluorescence staining analysis showed that GM15416 overexpression significantly increased the relative expression of c-Fos, and this upregulated expression could be reversed by si-Fos, indicating that the expression levels of GM15416 and c-Fos are positively correlated [ 70 ]. The results of Western blot analysis indicated that the overexpression of GM15416 increased the expression levels of Fas, Bcl-2-associated X protein (Bax), caspase-3, and cleaved-caspase-3, whereas si-Fos promoted a decrease in the expression of these apoptotic proteins [ 70 ]. Correspondingly, knockdown of GM15416 partially alleviated the reduction in osteogenic differentiation ability caused by the overexpression of c-Fos [ 70 ]. Additionally, knockout of GM15416 inhibited Fas protein expression and partially rescued the cell apoptosis and impairment of osteogenic differentiation induced by c-Fos. This indicates that GM15416 regulates downstream Fas through c-Fos to transmit apoptotic signals to mediate osteoblast apoptosis. The results of in vivo experiments demonstrated that compared with the OVX + siRNA negative control group, the OVX + si-GM15416 group showed a significant increase in bone mass, a decrease in Fos and Fas protein levels, and inhibition of cell apoptosis [ 70 ]. Additionally, knockdown of lncRNA GM15416 increased the expression of runt-related transcription factor 2 (Runx2), osterix (Osx), and osteopontin (Opn), indicating that the dysfunction of osteogenic differentiation was corrected [ 70 ]. Overall, GM15416 emerges as a dual-function lncRNA in osteoporosis, modulating osteogenic differentiation while serving as a promising clinical biomarker. In another study, researchers found through RT-qPCR analysis of peripheral blood mononuclear cells (PBMCs) that lncMIAT gain- or loss-of-function significantly modulated miR-216a levels [ 76 ]. Moreover, the expression levels of lncMIAT and miR-216a were negatively correlated. Furthermore, the overexpression of miR-216a in PBMCs downregulated the expression of p38MAPK, and this effect could be weakened by an miR-216a inhibitor. RT-qPCR detection revealed that the co-transfection of lncMIAT and miR-216a mimics into PBMCs downregulated the overexpression of p38MAPK induced by lncMIAT alone, and Western blot analysis supported this result [ 76 ]. Additional cell experiments confirmed that lncMIAT promotes the secretion of the important inflammatory cytokines interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-α), and interleukin-1 beta (IL-1β) in the progression of osteoporosis [ 76 ]. These aforementioned results indicate that lncMIAT regulates the expression level of p38MAPK through cross-talk with miR-216a, leading to an increase in the secretion of inflammatory cytokines in PBMCs of patients with PMOP. Therefore, silencing lncMIAT to inhibit the inflammatory cytokines may be a promising strategy for treating osteoporosis [ 76 ]. As potential biomarkers, lncRNAs have important application prospects in the diagnosis and treatment of osteoporosis.

Currently, research studies on the use of lncRNAs for the prevention of osteoporosis are relatively limited. However, a deeper understanding of their normal expression and regulatory mechanisms will help towards identifying high-risk patient populations and implementing early interventions in the future. Additionally, lncRNAs can be used as potential biomarkers ( Fig. 3 ). For example, the level of the lncRNA DANCR is high in BMSCs of patients with osteoporosis, 51 differentially expressed lncRNAs have been detected in the blood of patients with PMOP, and PAXB-AS1 is upregulated in the mouse model of OVX-induced osteoporosis, which can assist in diagnosis ( Fig. 3 ). Moreover, exploring targeted therapies related to osteoporosis-associated lncRNAs is an important research direction ( Fig. 3 ).

Clinical applications of circRNAs in osteoporosis

circRNAs and osteoporosis prevention

The role of circRNAs as multi-functional molecules in bone metastasis has been confirmed by many studies. For example, circIKBKB can promote breast cancer bone metastasis by activating nuclear factor kappa B (NF-κB)–bone remodelling factor signal transduction [ 77 ]. Additionally, circRNAs have great research potential in the field of osteoporosis, particularly in exploring their regulatory effects on bone mass from the perspective of prevention. Using an RNA pull-down assay, Wu et al. [ 78 ] confirmed the interaction between miR-186-5p and circEIF4B. In this study, a series of in vitro experiments revealed that silencing circEIF4B could upregulate the expression of miR-186-5p [ 78 ] ( Fig. 4 ). Conversely, circEIF4B overexpression led to the downregulation of miR-186-5p, indicating a competitive relationship between miR-186-5p and circEIF4B [ 78 ]. Additionally, this study showed that knocking down circEIF4B in BMSCs reduced the expression levels of osteogenesis-related markers, such as ALP and collagen type I (COL1), whereas circEIF4B overexpression increased the expression levels of these osteogenesis-related markers, indicating that circEIF4B has a positive regulatory effect on cell osteogenesis, which may be achieved through its interaction with miR-186-5p [ 78 ]. In another in vivo study, tartrate-resistant acid phosphatase (TRAP) staining results revealed that the number of Nfatc1-positive cells was increased in the OVX group, and this increase was weakened after circZNF367 knockdown. Western blot detection confirmed that the expression of osteoclast transcription factors in the OVX group was enhanced and partially reversed by circZNF367 knockdown [ 79 ]. In vitro experiments showed that the knockdown of circZNF367 inhibited the osteoclastic differentiation of BMDMs induced by RANKL and macrophage colony-stimulating factor (M-CSF), an effect manifested as a decrease in the number of osteoclasts and the downregulation of osteoclast-related proteins (c-Fos and TRAP) [ 79 ]. These findings revealed that regulating the expression or activity of circZNF367 may help reduce bone loss, highlighting the prospect of new preventive interventions beneficial to skeletal health [ 79 ]. In conclusion, circRNAs play crucial roles in regulating the activities of osteoblasts and osteoclasts and may become new targets for the prevention of osteoporosis. Additionally, with research on the correlation between osteoporosis and the gut microbiota gaining increasing attention, exploring how the gut microbiota regulates bone metabolism by affecting the expression of circRNAs will be a new research field in the future.

circRNAs and osteoporosis diagnosis

Numerous studies on the involvement of circRNAs in the pathogenesis of osteoporosis have been reported, and research on their applications is also becoming more extensive. circRNAs are closed loops of RNA molecules generated through special splicing. On the one hand, they exhibit high stability and tissue specificity and can exist stably in various body fluid samples, such as serum, plasma, PBMCs, and plasma exosomes. These characteristics render circRNAs as potential new biomarkers for osteoporosis. On the other hand, researchers using Spearman correlation analysis found significant differences in the levels of circRNAs in patients with OPNA and osteoporosis [ 80 ]; therefore, circRNAs are expected to become biomarkers for the early detection and diagnosis of osteoporosis. The expression of human circRNA 0076690 (hsa_circ_0076690) in patients with osteoporosis is reportedly significantly reduced. This particular circRNA promotes the osteogenic differentiation of hBMSCs by sponging miR-152 and upregulating the expression of Runx2, indicating that it has diagnostic value for osteoporosis [ 81 ]. Additionally, the expression of hsa_circ_0001275 is upregulated in patients with PMOP, and its knockdown has a certain therapeutic effect on osteoporosis, and it can thus also be used as a potential diagnostic biomarker for this disease [ 82 , 83 ]. Notably, some circRNAs have been found to regulate osteoblasts and/or osteoclasts in osteoporosis, thereby affecting bone metabolism. For example, in vitro experiments performed by Chen et al. [ 84 ] revealed that circFam190a is positively correlated with osteoclastogenesis and is a key regulator of osteoclast differentiation and function, but does not affect bone formation. Through transcriptome analysis, fused in sarcoma (FUS) was found to be significantly increased during osteoclastogenesis, and its knockdown could reduce the level of circFam190a; therefore, FUS can drive the expression of circFam190a and was confirmed to be its upstream target [ 84 ]. Additionally, RNA immunoprecipitation analysis confirmed robust interactions amongst circFam190a, AKT serine/threonine kinase 1 (AKT1), and heat shock protein 90 beta family member 1 (HSP90β). Notably, circFam190a knockdown impaired, whereas its overexpression strengthened, HSP90β–AKT1 binding, thereby modulating AKT1 protein stability [ 84 ]. qPCR and Western blot analyses of the TRAP RNA and protein levels revealed that the reduction in osteoclast formation induced by circFam190a knockdown recovered after AKT1 overexpression [ 84 ]. These results confirm that circRNAs play a role in the regulation of osteoclastogenesis through their interaction with related proteins. Moreover, circ-Plod2 can specifically bind to insulin-like growth factor 2 mRNA-binding protein 2 (IGF2BP2) to form an RNA–protein complex that destabilises myeloperoxidase (Mpo) mRNA, thereby promoting the osteogenic differentiation of BMSCs. Therefore, evaluating circ-Plod2 in BMSCs or bone marrow tissue is helpful for the early diagnosis of osteoporosis [ 85 ]. The advantages of circRNAs as biomarkers are that they are more stable than traditional mRNA markers and are not easily affected by RNA exonucleases; therefore, they are easier to detect and quantify in clinical practice. Moreover, the expression levels of circRNAs are specific to different disease stages and tissues, thereby also providing the possibility for the early diagnosis and treatment of osteoporosis. Although research on circRNAs as biomarkers for osteoporosis is still in the early stages, existing lines of evidence show that they can play important roles in the development and diagnosis of the disease and are expected to become important targets in future diagnostic strategies.

circRNAs and osteoporosis treatment

circRNAs show great potential in the treatment of osteoporosis, and multiple studies have confirmed their promising utility as treatment targets for the disease. For example, hsa_circ_0006859 is a competing endogenous RNA (ceRNA) in the cytoplasm of BMSCs that can directly bind to miR-642b-5p and miR-483-3p and downregulate their expression [ 86 ] ( Fig. 4 ). The knockdown of hsa_circ_0006859 upregulated the expression of osteogenesis-related genes (Runx2, osteocalcin (OCN), Opn, etc.), whereas its overexpression reduced the levels of these genes, indicating that this circRNA inhibits osteogenesis, which was also confirmed by the results of in vivo experiments [ 86 ]. Additionally, the knockdown of miR-642b-5p and miR-483-3p downregulated the expression of catenin beta 1 (β-catenin), MYC proto-oncogene (c-Myc), and G1/S-specific cyclin-D1 (cyclinD1), whereas co-transfection with sh-circ_0006859 increased the expression of these Wnt signal-related proteins [ 86 ]. Moreover, miR-642b-5p and miR-483-3p target ephrin A2 (EFNA2) and dedicator of cytokinesis 3 (DOCK3), respectively, and are negatively correlated with their expression levels [ 86 ]. Therefore, hsa_circ_0006859 activates the Wnt signalling pathway by targeting the miR-642b-5p–EFNA2 and miR-483-3p–DOCK3 axes; hence, it is expected to become a new target and biomarker for the treatment of PMOP. Furthermore, other human circRNAs such as hsa_circ_0114581 [ 87 ], hsa_circ_0076690 [ 81 ], and hsa_circ_0006215 [ 88 ] can bind to specific miRNAs to affect the differentiation and function of BMSCs, thereby playing key roles in the occurrence and development of osteoporosis. In vitro studies have demonstrated that circ_0024097 is positively correlated with the expression level of Yes-associated protein 1 (YAP1), and the downregulation of YAP1 can reverse the increase in Runx2, OCN, and Opn levels and enhancement of ALP protein activity and staining intensity caused by the upregulation of circ_0024097; that is, the depletion of YAP1 cancels the effect of circ_0024097 upregulation in promoting osteoblast differentiation [ 89 ]. Therefore, YAP1 is essential for circ_0024097-mediated osteoblast differentiation. Moreover, RT-qPCR assay results confirmed that silencing circ_0024097 in BMSCs and a murine pre-osteoblastic cell line (MC3T3-E1) significantly reduces the β-catenin protein level [ 89 ]. Luciferase assay detection revealed that the relative luciferase activity of the T-cell factor/lymphoid enhancer factor (TCF/LEF) optimal promoter in circ_0024097-silenced cells was reduced [ 89 ]. Therefore, the increase in circ_0024097 and YAP1 expression can activate the Wnt–β-catenin pathway to promote the differentiation of BMSC and MC3T3-E1 cells, a fact also confirmed by rescue experiment results. However, the molecular mechanisms and signalling pathways through which numerous other circRNAs modulate osteoporosis still need to be explored. Yao et al. [ 85 ] reported that exon-type circ-Plod2 generated by the Plod2 gene can transfer from the nucleus to the cytoplasm and form a complex with the RNA-binding protein IGF2BP2 [ 85 ]. This complex destabilises the mRNA of Mpo to downregulate its expression, thereby promoting the osteogenic differentiation of BMSCs to relieve osteoporosis. Therefore, circ-Plod2 can promote the osteogenic differentiation of BMSCs and is crucial for skeletal health and osteoporosis treatment [ 85 ]. Furthermore, whether circRNAs can combine with other proteins to form complexes and participate in the development of osteoporosis remains to be explored.

In terms of prevention, the above series of experimental results fully prove that some circRNAs can promote the functions of cells related to bone formation ( Fig. 3 ). In terms of diagnosis, circRNAs not only have a high degree of sequence conservation and stability, but some (e.g. hsa_circ_0076690, circ_0001275, and circ_0001445) are expressed in significantly different levels in patients with osteoporosis compared with non-patients. Therefore, these RNA molecules have great potential as new clinical biomarkers for the early diagnosis of osteoporosis ( Fig. 3 ). In terms of therapeutics, circRNAs can provide a new direction for the treatment of osteoporosis by regulating cells related to bone metabolism ( Fig. 3 ).

Clinical applications of circRNA–miRNA networks in osteoporosis

In hepatocellular carcinoma research, circ-0091579 is a newly found circRNA. Bioinformatic analysis identified complementary binding sites between circ-0091579 and miR-1225-5p [ 80 ]. Correlation analysis of molecular and clinical biochemical markers in patients with OPNA and osteoporosis revealed that circ-0091579 exhibited a significant positive association with bone resorption markers (RANKL, RANKL-to-osteoprotegerin (OPG) ratio), inflammatory mediators (NF-κB p65, IL-1β, IL-6), and oxidative stress-related malondialdehyde, while showing a negative correlation with bone formation and antioxidant markers (bone mineral density, OPG, glutathione). Conversely, miR-1225-5p displayed an inverse relationship with these factors. Notably, elevated circ-0091579 and reduced miR-1225-5p levels were linked to increased oxidative stress, as indicated by their respective associations with malondialdehyde accumulation and glutathione depletion [ 80 ]. The study also found that circ-HIPK3 was positively correlated with bone formation and negatively correlated with bone destruction, whereas its corresponding miRNA (miR-338-3p) had the opposite regulatory effects on bone formation and bone destruction [ 80 ]. These results confirm the relationship between the circRNA–miRNA network and osteoporosis severity and its potential as a diagnostic marker. Yin et al. [ 86 ] discovered the interaction between miR-642b-5p and miR-483-3p and circ_0006859 using RNA immunoprecipitation experiments. Subsequent experiments confirmed that hsa_circ_0006859 acts as a sponge for miR-642b-5p and miR-483-3p in BMSCs and competitively binds to them [ 86 ]. Additionally, hsa_circ_0006859 suppressed the osteogenic differentiation of BMSCs and downregulated osteogenesis-related genes, as validated through in vivo and in vitro studies. To explore whether hsa_circ_0006859 affects bone loss, the researchers established a mouse model of OVX-induced osteoporosis [ 86 ]. Through micro-computed tomography imaging and hematoxylin and eosin staining, they found that knocking down circ_0006859 could restore the bone mass of the ovariectomised mice [ 86 ]. Therefore, silencing circ_0006859 shows great potential as a strategy for alleviating bone loss. The underlying mechanism was proven to involve hsa_circ_0006859 competitively binding to miR-642b-5p and miR-483-3p, positively regulating the expression of EFNA2 and DOCK3 in BMSCs, and inhibiting Wnt–β-catenin signal transduction, thereby inhibiting BMSC differentiation [ 86 ]. Li et al. [ 87 ] explored the pathogenesis of osteoporosis from the perspective of the circRNA–miRNA–mRNA network and confirmed the important role of the hsa_circ_0114581–miR-155-5p–heterogeneous nuclear ribonucleoprotein A3 (HNRNPA3) axis in promoting osteogenic differentiation. It can be seen that the circRNA–miRNA network plays an important role in the regulation of bone metabolism in osteoporosis. Hence, studying the specific regulatory mechanism of this network on bone metabolism helps gain an in-depth understanding of the pathogenesis of osteoporosis and provides new ideas for the prevention and clinical treatment of the disease ( Fig. 5 ).

Clinical applications of lncRNA–miRNA networks in osteoporosis

lncRNAs and miRNAs interact with each other in the cell through a ceRNA network [ 90 ]. For example, the lncRNA SNHG14 can act as a molecular sponge for miR-493-5p to regulate gene expression and plays an important role in the osteogenic differentiation of BMSCs [ 91 ] ( Fig. 4 ). Using an ovariectomised mouse model, Yu et al. [ 65 ] demonstrated that short hairpin RNA targeting SNHG1 (sh-SNHG1) enhanced bone regeneration by silencing the gene, an effect reversible by miR-181c-5p inhibition [ 65 ]. Mechanistically, SNHG1 suppresses miR-181c-5p, leading to the upregulation of SFRP1 and subsequent activation of the Wnt3a–β-catenin signalling pathway, which promotes the osteoclastic differentiation of BMSCs [ 65 ] ( Fig. 4 ). Therefore, silencing SNHG1 by targeting miR-181c-5p can eliminate the negative impact on bone balance in ovariectomised mice. Liu et al. [ 92 ] found that the lncRNA AW011738 in osteoclast-derived exosomes could inhibit osteogenic differentiation and exacerbate osteoporosis in ovariectomised mice. RT-qPCR verification revealed that the expression of miR-24-2-5p was significantly reduced in MC3T3-E1 cells transfected with the AW011738 overexpression vector, indicating that AW011738 and miR-24-2-5p expression levels were negatively correlated. Additionally, luciferase reporter gene assay results confirmed that miR-24-2-5p could bind directly to AW011738 [ 92 ] ( Fig. 4 ). Moreover, miR-24-2-5p overexpression increased the expression of osteogenic genes (collagen type I alpha 1 chain (COL1A1) and bone gamma-carboxyglutamate protein (BGLAP)) and significantly reduced the expression of the triggering receptor expressed on myeloid cells 1 (TREM1) inflammasome, indicating that miR-24-2-5p could enhance osteogenesis and inhibit the inflammatory response of TREM1. Western blot analysis revealed that TREM1 knockout led to a significant increase in ALP staining, indicating that TREM1 plays an inhibitory role in osteogenic differentiation [ 92 ]. Therefore, AW011738 can hinder osteoblast differentiation by inhibiting miR-24-2-5p to increase the expression of TREM1 and has the potential to become a new target for the treatment of PMOP. AK039312 and AK079370 are two osteogenesis-related lncRNAs that inhibit the osteogenic transcription factors TCF7/LEF1 by sequestering miR-199b-5p, thereby hindering osteoblast differentiation and bone formation [ 93 ]. Both in vitro and in vivo experimental results have demonstrated that the combined inhibition of bone formation by AK039312 and AK079370 is more potent than either lncRNA acting alone. Notably, the combined application of si- AK039312 and si- AK079370 had a more obvious improvement effect on osteoporosis than the single sequence application, which provides a new strategy for treatment of the disease [ 93 ] ( Fig. 4 ). Collectively, the findings of recent studies show that lncRNAs can sponge miRNAs, act as potential pri-miRNAs to generate mature miRNAs, and indirectly regulate the expression level of miRNAs through other proteins. Thus, with in-depth research on the relationship between the lncRNA–miRNA network and osteoporosis, more treatment targets and diagnostic markers may be discovered, providing new strategies and methods for disease prevention and treatment ( Fig. 5 ).

Clinical applications of other ncRNAs in osteoporosis

piRNAs not only play crucial roles in germ cell development, transposable element silencing, and epigenetic memory but also regulate key signalling pathways at the transcriptional and post-transcriptional levels. One study showed that piRNA-36741 and methyltransferase-like 3 (METTL3) were upregulated during the osteogenic differentiation of BMSCs, and their interference significantly hindered osteogenic differentiation [ 94 ]. In RNA pull-down assays, piR-36741 bound directly to PIWI-like RNA-mediated gene silencing 4 (PIWIL-4) rather than to METTL3, whereas the piRNA-36741–PIWIL4 complex could directly interact with METTL3 [ 94 ]. At high expression levels, piR-36741 could competitively bind to METTL3 to inhibit its binding to bone morphogenetic protein 2 (BMP2) mRNA and reduce m6A modification, thus promoting the expression of BMP2, a positive regulator of osteoblast differentiation [ 94 ] ( Fig. 4 ). Additionally, intervention with a piRNA-36741 mimic alleviated the osteoporosis induced by OVX in mice. Therefore, piR-36741 regulates the expression of BMP2 mRNA through METTL3-mediated m6A modification, thereby promoting the osteogenic differentiation of BMSCs, and can thus be a new molecular target for the treatment of osteoporosis. A recent study proposed a new method using alendronic acid-modified mesoporous silica nanoparticles to deliver sodium bicarbonate and piR7472 to cells [ 95 ]. On the one hand, it can stably target hydroxyapatite and accumulate in the osteoporotic area. On the other hand, sodium bicarbonate can inhibit osteoclast formation, and piR7472 can promote the osteogenic differentiation of BMSCs by enhancing m6A methylation, thereby ameliorating osteoporosis [ 95 ] ( Fig. 4 ). These findings suggest that piRNAs are critically involved in the pathogenesis of osteoporosis and thus represent a potential therapeutic target for treatment of the disease.

siRNAs have also become a powerful treatment tool because of their ability to specifically regulate the expression of target genes. However, most currently approved siRNA drugs target only liver hepatocytes. Therefore, the development of siRNA drugs that target other organs remains an ongoing research goal [ 96 ]. Although small extracellular vesicles (sEVs) are promising delivery systems for siRNAs, they have limited cell- or site-targeting ability. Kang et al. [ 97 ] developed a new generation of heart-targeted sEVs, loaded them with NADPH oxidase 4 (NOX4) siRNA, and confirmed the anti-myocardial hypertrophy effect of the sEV-delivered siRNA drug in vitro and its abilities to improve cardiac function and reduce both fibrosis and the myocardial cell cross-sectional area in vivo, indicating that sEV-delivered NOX4-siRNA can be used for the targeted treatment of myocardial hypertrophy [ 97 ]. Luo et al. [ 98 ] successfully developed a bioactive microdroplet–Nfatc1–polylactic acid–exosome scaffold by (i) encapsulating Nfatc1-siRNA in microdroplets to regulate the excessive activity of osteoclasts and (ii) fixing mesenchymal stem cell-derived exosomes inside the polylactic acid surface to regulate the bone-forming environment and osteogenesis. This scaffold had a significant function in promoting the osteogenic differentiation of hBMSCs and is expected to become an innovative structure for bone tissue regeneration [ 98 ]. Therefore, developing EVs that target bone-related cells and exploring siRNAs that are beneficial to bone balance are important research directions in the future, and EV cross-talk with siRNAs has great potential. However, the regulatory roles that piRNAs, siRNAs, and snoRNAs play in the prevention, diagnosis, and treatment of osteoporosis have not been fully explored both domestically and internationally and warrant investigation.

Role of extracellular vesicles in regulating the proliferation and differentiation of bone-related cells

EVs, which are nanoscale-sized membrane particles released into the extracellular environment, are classified into three subtypes: exosomes, microvesicles, and apoptotic bodies. As key mediators of intercellular communication, EVs hold significant promise as therapeutic agents, drug carriers, and disease biomarkers [ 99 ]. Recent research has emphasised the key functions of EVs in the bone–organ axis (bidirectional cross-talk between the skeleton and other organs) [ 100 ] and provided new strategies and methods for their use in the clinical treatment of osteoporosis. With further research, their application prospects in the treatment of bone diseases will be even broader.

Extracellular vesicles and osteoblast proliferation and differentiation

EVs play important roles in the proliferation and differentiation of osteoblasts and may become potential tools for the treatment of osteoporosis and promotion of bone regeneration. However, EVs from different sources and types have different effects on osteoblasts, which may be related to their species specificity and complex network-regulatory mechanisms. A team of researchers found that colonisation of the gut of osteoporotic mice with children-derived gut microbiota (CGM) promoted bone formation by osteoblasts and inhibited bone resorption by osteoclasts, thereby improving bone mass and microstructure in the animals [ 101 ]. Micro-computed tomography analysis further revealed that the CGM-induced increases in bone mineral density, trabecular bone volume per total volume, trabecular number, cortical area per total area, and cortical thickness and decrease in trabecular separation could be eliminated by pre-treatment with a neutral sphingomyelinase inhibitor (GW4869; inhibits EV release), indicating that EVs are necessary for the bone-protective effect of CGM [ 101 ]. Furesi et al. [ 102 ] used EVs derived from the RM1-BM mouse bone metastatic prostate cancer cell line to treat murine osteoprogenitors and observed through Alizarin Red staining that the concentration of the EVs was inversely proportional to the mineral deposition of osteoblasts. Furthermore, qPCR analysis confirmed that the levels of the osteoblast markers Runx2 and Osx were significantly decreased after RM1-BM-derived EV treatment, indicating that these vesicles inhibited osteoblast differentiation and mineralisation, which was also confirmed using in vivo experiments [ 102 ]. Other studies have shown that EVs can affect osteoblast activity through different mechanisms. For example, oyster mantle-derived exosomes can promote osteogenesis by activating the phosphatidylinositol 3-kinase (PI3K)–AKT serine/threonine kinase (Akt)–β-catenin pathway in osteoblasts [ 103 ]. Another research group showed that young bone-derived EVs (YB-EVs) can enhance bone formation, whereas aged bone-derived EVs (AB-EVs) not only lose this ability but also promote the adipogenic differentiation of BMSCs, indicating that the regulatory effect of bone-derived EVs on the skeleton is age dependent [ 104 ]. This result indicates that EVs from the same cell type at different age stages can have different effects on osteogenic differentiation. Although EVs have been well studied, research to explore their impact on the proliferation and differentiation of osteoblasts at the molecular level remains lacking.

Extracellular vesicles and osteoclast proliferation and differentiation

Osteoclasts, which are derived from precursor cells, are specialised bone-resorbing cells that degrade the bone matrix. Emerging evidence has indicated that EVs from various sources differentially regulate osteoclast proliferation and differentiation. For example, EVs derived from Propionibacterium freudenreichii MJ2 inhibit osteoclastogenesis [ 105 ], whereas those derived from BMSCs can activate osteoclast function [ 106 ]. Moreover, EVs are considered important messengers for long-distance intercellular communication and play an important role in the cross-talk between osteoblasts and osteoclasts. Osteoblast-derived EVs can transmit signals to osteoclast precursors and promote the formation of osteoclasts through the cell-surface receptor RANK [ 107 ]. Therefore, it is obvious that EVs not only play an important role in bone homeostasis but also show great potential in bone disease treatment. One research group found that oyster mantle-derived exosomes had a dual effect of promoting osteogenesis and inhibiting osteoclastogenesis [ 103 ]. In their study, oral treatment with oyster mantle-derived exosomes effectively lessened bone loss and enhanced the bone microstructure in osteoporotic model mice, thereby preventing and treating osteoporosis [ 103 ]. Therefore, the potential of edible exosomes for disease treatment cannot be ignored, and oral administration of beneficial flora containing EVs may also be an effective means to prevent osteoporosis in the future. It is worth noting that RT-qPCR revealed that the expression level of circ_0008542 in exosomes increased after tensile stimulation, and the expression of Ctsk, matrix metallopeptidase 9 (MMP9), and TRAP mRNAs induced by RANKL also increased [ 108 ]. Western blot results further showed that exosomes with increased circ_0008542 levels after tensile stimulation promoted the upregulation of key RANKL-induced regulators. Tensile-stimulated exosomes also amplified TRAP-positive osteoclast formation and multi-nucleation, suggesting a pro-osteoclastogenic role [ 108 ]. In summary, EVs have a complex association with the proliferation and differentiation of osteoclasts, which play an important role in the maintenance of bone homeostasis and the occurrence and development of bone-related diseases.

Extracellular vesicles and BMSC proliferation and differentiation

EVs derived from BMSCs can be transported to bone-related and other tissue cells to produce physiological or pathological effects. For example, bone-derived EVs promote fracture recovery and osteoblast differentiation by releasing miR-335 to activate vesicle-associated membrane protein B (VapB) and the Wnt–β-catenin signalling pathway, and BMSC-derived exosomes can reduce doxorubicin-induced heart damage and cardiomyocyte pyroptosis [ 109 ]. Additionally, these vesicles have the potential to treat osteoporosis because of their role in promoting angiogenesis and regulating immune responses and the proliferation and differentiation of osteoblasts and osteoclasts. Correspondingly, EVs from other sources can also affect the proliferation and differentiation of BMSCs and regulate their functions. Cell membrane vesicles (CMVs) derived from hBMSCs and human umbilical vein endothelial cells can enhance bone regeneration [ 110 ]. In one study, hBMSCs co-cultured with endothelial cell-derived CMVs (EC-CMVs) showed increased ALP activity and calcium nodule formation, indicating that EC-CMVs can promote the osteogenic differentiation of hBMSCs [ 110 ]. Moreover, EC-CMVs contain the key growth factors BMP2 and VEGF that promote osteogenesis [ 110 ]. These factors are similar to those in their parent cells, indicating that EC-CMVs contain beneficial proteins for bone formation and enhance osteogenic activity after being endocytosed by hBMSCs. Another research group demonstrated that exosomes boost cell traction and cytoskeleton organisation by modulating the Ras-related C3 botulinum toxin substrate 1 (Rac1)–Ras homolog family member A (RhoA) signalling pathway, leading to enhanced cell proliferation [ 111 ], which indirectly confirms the potential of EVs to promote the proliferation of BMSCs. The possibility of using the analysis of the mechanical signals of cells to quickly assess the early treatment effects of exosomes was also confirmed in this study, adding an advantage to the clinical application of these particles for disease treatment. These studies provide an in-depth understanding of the relationship between EVs and the proliferation and differentiation of BMSCs and reveal the potential applications of such vesicles in bone regeneration.

Extracellular vesicles and chondrocyte proliferation and differentiation

A team of researchers proposed that exosomes derived from subcutaneous fat-derived mesenchymal stem cells (SC-MSCs) could promote chondrocyte proliferation by delivering miR-199a-3p [ 112 ]. Their KEGG enrichment data analysis and immunofluorescence assay results confirmed that the cellular phagosome and endocytic processes of osteoarthritic rats treated with SC-MSC-derived exosomes were significantly regulated, and the protein expression level of microtubule-associated protein 1 light chain 3 beta (LC3B), a key component of autophagosomes in cartilage, was significantly increased, indicating that the effect of these exosomes on cartilage was closely related to an improvement in the autophagy level, which was also confirmed by in vitro experimental results [ 112 ]. Further studies by those researchers revealed that the exosomes enhanced autophagy in chondrocytes by downregulating mechanistic target of rapamycin (mTOR) kinase expression and the mTOR signalling pathway, and this enhancement was highly dependent on miR-199a-3p. Antagonising miR-199a-3p eliminated the therapeutic effect of the SC-MSC-derived exosomes on damaged cartilage [ 112 ]. Additionally, the researchers developed engineered SC-MSC-derived exosomes that could specifically deliver miR-199a-3p to articular chondrocytes [ 112 ], revealing the potential applications of EVs in cartilage tissue engineering and regenerative medicine. In another study, plasma exosomes from children with idiopathic short stature (ISS) suppressed the proliferation and bone-forming activity of normal human chondrocytes [ 113 ]. This inhibition was mainly attributed to the specific lncRNA ISSRL, which inhibits the expression of miR-877-3p through Argonaute RISC catalytic component 2 (AGO2) and upregulates the expression of granzyme B (GZMB). Its effects manifested as a decrease in chondrocyte proliferation; a reduction in the mRNA and protein expression levels of collagen type X alpha 1 (COL10A1), Runx2, Opn, and OCN; a decline in ALP activity; and a decrease in mineralisation in the overexpression-idiopathic short stature-related lncRNA (OE-ISSRL) group. However, this trend was reversed in the OE-ISSRL + miR-877-3p mimic and OE-ISSRL + si-GZMB groups. Therefore, the authors of this study innovatively used growth plate cartilage-targeted exosomes — loaded with siRNAs against ISSRL and overexpressing growth hormone — to deliver the siRNA drug to alleviate the inhibitory effect of abnormally expressed ncRNAs on the development of growth plate cartilage [ 113 ]. Combined with the precise delivery of growth hormones, this approach has become a new treatment strategy for ISS. Additionally, in another study, overexpression of the exosome circ_0001236 was found to increase the levels of the chondrogenesis markers collagen type II alpha 1 chain (Col2a1) and SRY-box transcription factor 9 (Sox9), indicating its potential to promote mesenchymal stem cell chondrogenesis [ 114 ] while also proving that EVs play an indispensable role in the process of chondrocyte proliferation and repair.

Extracellular vesicles and skeletal muscle cell proliferation and differentiation

Skeletal muscle-derived EVs have been shown to transport glycolytic enzymes, reach the bones through the bloodstream, and be phagocytosed by BMSCs, thereby exerting a protective effect on disuse osteoporosis in mice [ 115 ]. The question arises as to what effects EVs from other sources have on skeletal muscle. One research team found that young serum-circulating EVs had a beneficial effect on aged muscle regeneration and mitochondrial function, and this effect was inseparable from that of circulating EVs [ 116 ]. The EV delivery of mRNA of the longevity protein Klotho to muscle cells promoted muscle cell regeneration and repair after injury. Moreover, the level of Klotho mRNA-containing EVs decreased significantly in aged blood [ 116 ], which explains why muscle regeneration ability weakens with age, suggesting that EVs could be used as therapeutic drugs to offset the functional defects caused by ageing. Additionally, the findings from this study verified the hypothesis that the information stored in circulating EVs in young serum contains young signals to alleviate tissue ageing, whereas harmful signals are transmitted by circulating EVs under ageing conditions [ 116 ]. This indicates that during the shedding of EVs from their originating cells, the vesicles carry away information about the cellular age and health status (i.e., proteins, lipids, and genetic materials), transmit this information when received by the corresponding cells, and trigger responses. Furthermore, exosomes derived from healthy tendon stem cells have a similar anti-ageing effect. They can break the positive feedback cross-talk between ageing senescent tendon stem cells and macrophages by polarising the latter from the classically activated macrophage phenotype (M1 phenotype) to the alternatively activated phenotype (M2 phenotype), thereby alleviating senescent tendon stem cell ageing and benefiting tendon-bone healing [ 117 ]. In conclusion, EVs not only affect the regeneration and repair of skeletal muscle but may also participate in regulating the functions of skeletal muscle cells through interaction with other tissues, such as bone, playing an important role in the process of skeletal muscle cell proliferation and differentiation.

Potential applications of ncRNAs secreted by extracellular vesicles in osteoporosis

As a unique class of lipid bilayer-bound particles released by various cells, EVs are related to processes such as immune regulation [ 118 ], tissue repair [ 119 ], and tumour metastasis [ 120 ]. EVs secreted by different cells have been shown to play important roles in the diagnosis and treatment of diseases of the lungs [ 121 ], kidneys [ 122 ], eyes [ 123 ], and cardiovascular [ 124 ] and nervous systems [ 125 , 126 ]. Previous studies have revealed that EVs from different sources have varying effects on osteoporosis [ 127 ]. Additionally, EVs transport various cellular components such as proteins, mRNA, miRNA, DNA, and lipids across distances, playing a crucial role in cell communication and affecting various physiological and pathological events [ 99 ]. Amongst such transported molecules, ncRNAs (including miRNAs, lncRNAs, and circRNAs) occupy an important position in EVs [ 128 ]. Although studies have already demonstrated the role of EVs from different sources in bone homeostasis [ 127 ], how ncRNAs, as key regulatory factors in this process, modulate the progression of osteoporosis needs to be further explored.

Osteoporosis treatment by miRNAs secreted by extracellular vesicles

Reports on the cross-talk between EVs and miRNAs in various skeletal diseases are common. For example, EVs participate in tumour bone invasion by regulating the pre-metastatic bone microenvironment, and EV-derived miRNAs (miR-26a-5p, miR-27a-3p, and miR-30e-5p) play a crucial role in osteogenic inhibition mediated by prostate cancer [ 102 ]. Clinically, vascular calcification usually occurs together with osteoporosis, known as the ‘calcification paradox’. This may be because in aged individuals, AB-EVs promote the adipogenesis of BMSCs and enhance the osteogenic trans-differentiation of vascular smooth muscle cells [ 104 ]. Amongst the miRNAs, miR-483-5p and miR-2861 are abundant in AB-EVs and act as mediators for the promotion of adipogenic differentiation in BMSCs and calcification in vascular smooth muscle cells [ 104 ]. On the one hand, AB-EVs are transported via the blood circulation and deposited in the vascular system. By transferring miR-2861, they stimulate the expression of Runx2, resulting in vascular calcification. On the other hand, AB-EVs contain calcium and phosphorus, which are also transported to the vascular system during AB-EV transfer, thereby contributing to vascular calcification. YB-EVs from rats significantly enhance the formation of calcium nodules in BMSCs and tend to induce a decrease in the number of bone marrow adipocytes. Notably, YB-EVs are favourable for the osteogenesis of BMSCs rather than adipogenesis [ 104 ]. Therefore, reducing the release of AB-EVs or increasing that of YB-EVs are potential strategies for protecting bones and are expected to become a new target for the treatment of osteoporosis. Additionally, in both in vivo and in vitro experiments, EVs derived from BMSCs were shown to transfer the miR-143/145 cluster to osteoclasts, which activated their proliferation and differentiation, and knocking down miR-143/145 could offset the adverse effect of oestrogen deficiency on the osteogenesis of BMSCs [ 106 ]. The miR-143/145 cluster is an effective target for simultaneously regulating bone catabolism and anabolism. Developing drugs with the dual function of promoting osteogenesis and inhibiting osteoclastogenesis will be the focus of future clinical translation research. Notably, exosomes from osteogenically differentiated BMSCs promoted H-type vessel formation via miR-150-5p-mediated metabolic reprogramming, with 7-day differentiated BMSC-derived exosomes demonstrating enhanced potential to accelerate bone regeneration and reduce bone loss through angiogenesis–osteogenesis coupling [ 129 ]. These findings provide new insights into the role of EV-mediated miRNAs in bone regeneration and injury diseases. In summary, research on miRNAs secreted by EVs is an important entry point for exploring the stability of the bone microenvironment, and these RNA molecules have an inescapable impact on ameliorating osteoporosis and will become the key to the prevention and treatment of bone loss diseases ( Fig. 6 ).

Osteoporosis treatment by lncRNAs secreted by extracellular vesicles

In recent years, increasing amounts of evidence have confirmed that EV-derived lncRNAs play a crucial role in the development of various diseases, particularly tumours. For example, the exosome-derived lncRNA MEG3 has an inhibitory effect on osteosarcoma [ 130 ]; however, its influence on osteoporosis cannot be ignored. The lncRNAs derived from exosomes of inflammatory osteoclasts (iOCLs) (referred to as lncRNAs from inflammatory osteoclast-derived exosomes (LIOCE)) have been proven to mediate iOCL–osteoblast communication in inflammatory bone diseases. Using qPCR and Western blot assays, Ren et al. [ 131 ] determined that when iOCL-derived exosomes were co-cultured with MC3T3-E1 cells, the relative mRNA and protein expression levels of osteoblast-related genes such as Runx2, Osx, and COL1A1 were significantly increased, indicating that these exosomes help enhance osteoblast activity [ 131 ]. To explore the specific mechanism, the researchers knocked down lnc-LIOCE and found through qPCR analysis that the expression levels of ALP, Runx2, BGLAP, and COL1A1 were downregulated, and the corresponding Western blot results revealed that the protein levels of ALP, Runx2, and COL1A1 were also significantly decreased [ 131 ]. Further quantitative analysis of ALP and Alizarin Red staining showed that knockdown of lnc-LIOCE also significantly reduced ALP activity and mineralisation [ 131 ]. At the same time, lnc-LIOCE overexpression significantly upregulated the expression of osteogenic genes, including ALP, Runx2, COL1A1, and BGLAP [ 131 ]. Therefore, it is beyond doubt that the osteogenic ability of iOCL-derived exosomes depends at least partly on lnc-LIOCE. Additionally, injecting exosomes embedded with lnc-LIOCE into a mouse model of lipopolysaccharide-induced osteolysis and bone defect significantly alleviated bone resorption, revealing the potential of this exosome system to become an effective therapy for low-bone-mass diseases, particularly osteoporosis. To explore the impact of cross-talk between exosomes derived from multiple myeloma cells and osteoclasts on bone lesions, researchers used TRAP staining to quantify osteoclasts induced by RAW264.7 cells [ 132 ]. They found that exosomes significantly increased the number of osteoclasts, whereas exosome inhibitor treatment reduced the number of TRAP-positive osteoclasts, confirming that exosomes from multiple myeloma cells were positively correlated with the number of osteoclasts and promoted their differentiation. The bone-destroying ability of multiple myeloma-derived exosomes was verified by the results of in vivo experiments [ 132 ]. Furthermore, Malat1, the highest-expressing lncRNA in multiple myeloma-derived exosomes, was able to target RAW264.7 cells. Immunohistochemical analysis of femur sections from exosome-injected mice revealed that the expression of RANKL and the phosphorylation of Akt, c-Jun N-terminal kinase (JNK), extracellular signal-regulated kinase (ERK), and p38 were significantly increased, and the knockdown of Malat1 rescued their high expression [ 132 ], indicating that Malat1 promotes osteoclast differentiation through the positive regulation of RANKL expression and its downstream Akt and MAPK signalling pathways. These findings were further confirmed by results from in vitro experiments. Moreover, RNA 5-methylcytosine (m5C) modification mediated by the m5C methyltransferase NOP2/Sun RNA methyltransferase 2 (NSUN2) can enhance the methylation of Malat1 and promote its expression [ 132 ]. Therefore, reducing NSUN2 to inhibit Malat1 expression to reduce bone loss may be a feasible entry point for future clinical translation. In summary, the lncRNAs secreted by EVs play important roles in the occurrence and development of osteoporosis and may also participate in epigenetic regulation, regulate the differentiation process of osteoclasts and osteoblasts, and transmit signals in the bone microenvironment ( Fig. 6 ).

Osteoporosis treatment by circRNAs secreted by extracellular vesicles

circRNAs, as a special type of ncRNA molecule, are receiving increasing attention for their role in orthopaedic diseases. Researchers found that the expression of circ_0001846 was increased in the articular cartilage tissue of patients with osteoarthritis and in a chondrocyte cell line (CHON-001) induced by the pro-inflammatory cytokine IL-1β, and the knockdown of circ_0001846 alleviated the IL-1β-induced damage to CHON-001 cells, indicating that this circRNA promotes the progression of osteoarthritis [ 133 ]. Additionally, miR-149-5p has complementary sites for circ_0001846 and Wnt family member 5B (WNT5B). The study findings confirmed that knockdown of circ_0001846 reversed the CHON-001 cell dysfunction caused by IL-1β by binding to miR-149-5p [ 133 ]. Western blot analysis also indicated that an miR-149-5p mimic blocked the IL-1β-induced upregulation of WNT5B expression, whereas a miR-149-5p inhibitor enhanced the effect [ 133 ]. Therefore, exosome-mediated circ_0001846 promotes IL-1β-induced chondrocyte damage and dysfunction by downregulating miR-149-5p and positively regulating WNT5B, indicating that silencing circ_0001846 has significant benefits for chondrocytes. Using ALP staining, RT-qPCR, and Western blot assays, Ma et al. [ 134 ] established that BMSC-derived exosomes could improve the osteogenic differentiation of MC3T3-E1 cells, as reflected by an increase in ALP activity and the upregulation of osteogenic markers such as RUNX [ 134 ]. They also found that circHIPK3 was enriched in the BMSC-derived exosomes, and that using siRNA to downregulate the expression of circHIPK3 in BMSCs and BMSC-derived exosomes significantly reduced the osteogenic differentiation of MC3T3-E1 cells [ 134 ], indicating that circRNA cross-talk via exosomes can promote osteogenic differentiation, thus expanding new ideas for the treatment of osteoporosis. Other research evidence indicates that the overexpression of demethyltransferase AlkB homolog 5 (ALKBH5) can reverse bone loss induced by circ_0008542 via osteoblast-derived exosomes [ 108 ]. Mechanistically, circ_0008542 promotes osteoclast differentiation and bone resorption by binding to miRNA-185-5p, whereas an miRNA-185-5p mimic can counteract this effect. Furthermore, compared with the circ_0008542-only group, the combination of circ_0008542 with RANK silencing significantly suppressed osteoclast differentiation and bone resorption [ 108 ]. These results imply that circ_0008542 enhances bone absorption by sponging miRNA-185-5p through the circ_0008542–miRNA-185-5p–RANK axis. Additionally, the results of in vivo experiments indicated that exosomes overexpressing ALKBH5 significantly reversed circ_0008542-induced bone loss 8 weeks after injection [ 108 ]. Mechanistically, the molecular sponge effect of circ_0008542 relies on m6A methylation, whereas ALKBH5-mediated demethylation reduces the binding capacity of the circRNA to the miR-185-5p–RANK axis, thereby inhibiting osteoclast differentiation and bone resorption ( Fig. 4 ). This highlights that the silencing of circ_0008542 to prevent it from sponging miRNA-185-5p is a potential treatment strategy for preventing bone loss and is beneficial for osteoporosis, and using exosomes that release ALKBH5 to enhance bone protection is a potentially effective approach.

In summary, circRNAs in EVs (e.g., BMSC-derived exosomes) play key roles in the occurrence and development of osteoporosis ( Fig. 6 ). On the one hand, some circRNAs (e.g., circHIPK3) promote osteoblast differentiation and are beneficial for bone formation; on the other hand, other circRNAs (e.g., circ_0008542) promote osteoclast differentiation and bone absorption, leading to bone loss. By regulating related enzymes (e.g., ALKBH5) or directly silencing circRNAs that promote bone loss, bone metabolic balance can be regulated, thus providing a strategy for the treatment of osteoporosis ( Fig. 6 ).

Network cross-talk of circRNA–miRNA secreted by extracellular vesicles in the diagnosis and treatment of osteoporosis

In recent years, the consensus is that circRNAs, which are rich in miRNA-binding sites, participate in gene regulation as miRNA sponges. Some researchers have discovered circRNA–miRNA–mRNA interactions in exosomes derived from BMSCs of patients with PMOP [ 135 ]. Additionally, KEGG enrichment analysis has revealed that the differentially expressed circRNAs in patients with PMOP may be involved in multiple pathways, such as autophagy and PI3K–Akt, FoxO, and MAPK signalling [ 135 ], revealing their great potential as predictive biomarkers and treatment targets for PMOP. In vitro experiments [ 134 ] have also revealed that circHIPK3 (present in BMSC-derived exosomes) enhances mitochondrial autophagy via the miR-29a-5p–phosphatase and tensin homolog (PTEN)-induced kinase 1 (PINK1) pathway, aiding in the osteogenic differentiation of MC3T3-E1 cells. Furthermore, exosomal circ-LMO7 can target Rho GTPase activating protein 24 (ARHGAP24) by sponging miR-21-5p to inhibit the proliferation and migration of osteosarcoma cells [ 136 ], although whether it has a similar regulatory role in osteoporosis remains to be verified. Additionally, cross-talk between EVs and the circRNA–miRNA network can play a regulatory role in chondrocytes. For example, exosomal circRNA_00000001236 acts as an miR-3677-3p sponge and can maintain cartilage function, enhance chondrogenic differentiation, and regulate cartilage metabolic balance through the targeted circ_00000001236–miR-3677-3p–Sox9 axis [ 114 ].

In summary, solid evidence exists regarding the network cross-talk role of circRNA–miRNA secreted by EVs in bone diseases ( Fig. 6 ). However, the specific mechanisms relating to osteoporosis lack corresponding literature support and remain to be explored.

Network cross-talk of lncRNA–miRNA secreted by extracellular vesicles in the diagnosis and treatment of osteoporosis

Research evidence has shown that treating aged BMSCs with osteoinductive EVs can upregulate the expression of osteogenesis-related genes (ALP, BMP2, and OCN) and downregulate that of ageing-related genes (p53 and p21), reaching levels comparable to those of young BMSCs [ 137 ]. Therefore, osteoinductive EVs can partially relieve the ageing of aged BMSCs and enhance their proliferation and osteogenic differentiation. Moreover, the study authors found that osteoinductive EV-derived lncRNA-ENSRNOG00000056625 was differentially expressed between young and aged BMSCs [ 137 ]. Overexpression of lncRNA-ENSRNOG00000056625 in aged BMSCs significantly upregulated the expression of osteogenesis-related genes, including ALP, BMP2, COL1, and OCN, whereas the expression of p53 and p21 was significantly downregulated, indicating that the osteogenesis-promoting and anti-ageing abilities of osteoinductive EVs partly stem from this particular lncRNA, which was also confirmed at the protein level by immunofluorescence staining results. Moreover, lncRNA-ENSRNOG00000056625 could sponge miR-1843a-5p and upregulate the expression of its target gene, modulator of apoptosis 3A (Mob3a) [ 137 ], thereby promoting osteogenic proliferation and differentiation and alleviating ageing. Therefore, the construction of the osteoinductive EVs–lncRNA-ENSRNOG00000056625–miR-1843a-5p regulatory network is key for promoting bone regeneration and treating aged bone defects. The importance of the cross-talk between EVs and the lncRNA–miRNA regulatory network is self-evident. Using the luciferase assay, a research team found that miR-17 has multiple binding sites with the lncRNA Malat1 and that this lncRNA in multiple myeloma-derived exosomes can act as a sponge for miR-17 to upregulate the expression of RANKL, thus having a significant effect in promoting osteoclast differentiation [ 132 ] ( Fig. 4 ). Another study confirmed that Malat1 is highly expressed in exosomes derived from endothelial progenitor cells and binds to miR-124 to competitively regulate miR-124 [ 138 ] ( Fig. 4 ). The findings from RT-qPCR and Western blot analyses showed that exosomes derived from endothelial progenitor cells could significantly upregulate the gene expression and corresponding protein levels of genetic markers of osteoclast differentiation, such as MMP9, Ctsk, TRAP, and carbonic anhydrase 2 (CAR2), whereas overexpressing miR-124 or silencing Malat1 significantly downregulated these markers in a mouse femur fracture model [ 138 ]. Additionally, the results of a luciferase reporter gene assay confirmed the interaction between miR-124 and integrin beta 1 (ITGB1), where the levels of both molecules were negatively correlated, indicating that miR-124 may inhibit osteoclast function through ITGB1 [ 138 ]. Therefore, exosomes derived from endothelial progenitor cells can promote osteoclastogenesis through the Malat1–miR-124 pathway.

It is worth noting that cancer cell-derived EVs can also function via cross-talk of the lncRNA–miRNA network. For example, exosomes derived from prostate cancer cells carry the lncRNA NEAT1 as a ceRNA to bind to miR-205-5p and upregulate its downstream target gene Runx2 (a transcription factor for osteoblast differentiation) to promote the osteogenesis of hBMSCs [ 139 ] ( Fig. 4 ). On the one hand, this provides a potential target for promoting the osteogenesis of hBMSCs in the osteoblastic bone metastasis of prostate cancer. On the other hand, it also raises the question of whether the cross-talk of the lncRNA–miRNA network mediated by cancer cell-derived EVs can be used to promote the proliferation and regeneration of other cells, and whether this is related to the tropism of cancer cell metastasis.

In summary, the network cross-talk role of lncRNA–miRNA secreted by EVs is a complex system that can regulate both osteoblasts and osteoclasts and has great potential in the treatment of osteoporosis ( Fig. 6 ). Additionally, the same lncRNA may bind to multiple miRNAs but play a similar role. For example, Malat1 promotes osteoclastogenesis through the miR-17/miR-124 pathway. Therefore, targeting such multi-path node molecules may have more excellent effects on the clinical diagnosis and treatment of bone diseases, particularly osteoporosis. However, the construction of this network still requires more research to supplement and corroborate its feasibility for this purpose.

Conclusion and perspective

Although the prevention, diagnosis, and treatment of osteoporosis based on ncRNAs have broad potential, some challenges remain, including the optimisation of ncRNA isolation, determination of the best types for targeted therapy, and ensuring the safety and effectiveness of various clinical treatment strategies. Additionally, the interactions between circRNAs and miRNAs and between lncRNAs and miRNAs also play crucial roles in the prevention, diagnosis, and treatment of osteoporosis. As a key medium for intercellular communication, EVs can carry a variety of molecules (including ncRNAs) to facilitate communication in complex biological systems. They can also act as natural carriers to package and transfer ncRNAs that can be released and taken up by different cells, thus playing a role in the development of various diseases and the regulation of microenvironments. In recent years, numerous studies on ncRNAs secreted by EVs have been conducted; however, none of them has summarised the impact of the molecular mechanisms and regulatory networks of EV–ncRNA cross-talk on osteoporosis. This review elaborates on the application of ncRNAs in the prevention, diagnosis, and treatment of osteoporosis ( Fig. 7 ). It also emphasises the mechanism by which EV-secreted ncRNAs regulate bone loss diseases through intricate cross-talk and proposes the concept of the EV–ncRNA network for more researchers to participate in its improvement ( Fig. 7 ).

A recent study confirmed that injured brains accelerate fracture healing by releasing sEVs rich in osteogenic miRNAs that target bone progenitor cells [ 140 ]. This demonstrates the potential of EV–ncRNA interactions in promoting bone healing after fractures, which should not be overlooked. Another study used exosomal nanoparticles for the targeted delivery of siRNAs and growth hormone to the growth plate. This approach promoted skeletal growth by correcting abnormal ncRNA expression in growth plate cartilage and precisely delivering the growth hormone, thereby addressing ISS [ 113 ]. These results not only confirm the significant potential of EVs in targeted ncRNA delivery for promoting skeletal growth but also suggest the possibility of using EVs to encapsulate and precisely deliver hormones to treat bone diseases. As a metabolic bone disease, osteoporosis is primarily triggered by oestrogen deficiency. Therefore, whether EVs can target and transport oestrogen via the mediation of ncRNAs to mitigate bone loss in osteoporosis is a valid research topic ( Fig. 7 ). Although existing studies have not directly confirmed that EVs can ‘load’ oestrogen itself, the fact that EVs can achieve functional compensation for oestrogen deficiency by delivering ncRNAs has been clearly established. Researchers used red blood cell-derived EVs to deliver anti-miR-214 to osteoclasts [ 141 ]. By inhibiting miR-214 and regulating the PTEN–PI3K–Akt signalling pathway, this approach reduced osteoclast activity and promoted osteoblast function, thereby achieving functional compensation for bone metabolic balance in an oestrogen-deficient osteoporosis model [ 141 ]. This finding suggests that EV–ncRNA can compensate for bone loss caused by decreased oestrogen levels without directly transporting the hormone molecules. Furthermore, the essence of osteoporosis is bone loss and structural deterioration resulting from an imbalance in bone calcium metabolism. Therefore, whether EVs can carry specific ncRNAs to regulate the effective osteoblast utilisation of calcium and enhance bone calcium deposition is a critical area of inquiry ( Fig. 7 ). Studies have shown that sEVs derived from human umbilical cord mesenchymal cells can deliver miR-23a-3p to activate the PTEN–Akt signalling pathway, thereby promoting calcium salt deposition and inducing osteogenic differentiation [ 142 ]. Exosomes from human adipose-derived stem cells deliver miR-130a-3p, which targets and inhibits sirtuin 7 (SIRT7) expression, subsequently activating the Wnt–β-catenin signalling pathway [ 143 ]. This significantly promotes the expression of osteogenesis-related genes and enhances calcium nodule formation and matrix mineralisation [ 143 ]. Thus, EV–ncRNA holds promise for driving osteogenic differentiation and calcium matrix deposition through multi-target synergistic effects, providing a direct driving force for the restoration of bone mass in patients with osteoporosis. Moreover, the bone marrow microenvironment in osteoporosis is characterised by low-grade chronic inflammation, featuring elevated pro-inflammatory cytokines that lead to excessive osteoclast activation. Whether EVs can mediate ncRNAs to regulate immune responses within the inflammatory microenvironment of osteoporosis is another topic worthy of further investigation ( Fig. 7 ). Existing research evidence has indicated that sEVs derived from dental pulp stem cells can polarise macrophages towards the pro-healing M2 phenotype, confirming the dental immunomodulatory properties of these miRNA-enriched vesicles [ 144 ]. Multiple myeloma cells can deliver various miRNAs to monocytes and macrophages via exosomes, synergistically inhibiting the cyclic GMP-AMP synthase–stimulator of interferon genes (cGAS–STING) signalling pathway, leading to reduced type I interferon production and thereby weakening the antiviral immunity of the host [ 145 ]. Therefore, EV–ncRNA cross-talk is closely related to immune responses, but research studies on whether it can play a role in osteoporosis and its specific mechanisms of action are lacking and warranted.

In summary, the multi-dimensional dissection of the molecular circuitry through which EV–ncRNA cross-talk governs the pathogenesis of osteoporosis—spanning hormonal signalling, calcium homeostasis, and immune modulation—remains at the forefront of contemporary biomedical research. Nevertheless, current investigations are largely confined to single molecules or isolated pathways. Future efforts must therefore integrate transcriptomic, proteomic, and metabolomic datasets to systematically chart a comprehensive EV–ncRNA regulatory atlas of osteoporosis. Such a systems-level landscape is not only indispensable for elucidating disease mechanisms but also constitutes the central strategy for realising the targeted prevention, precision therapy, and early diagnosis of osteoporosis.

However, it must be emphasised that translating EV–ncRNA-based therapeutics from bench to bedside faces formidable challenges. Firstly, overcoming the technical bottlenecks of large-scale EV isolation, purification, and rigorous quality control is critical for guaranteeing batch-to-batch reproducibility and product stability. Secondly, mechanistic and safety profiles for the EV-mediated delivery of specific ncRNAs for modulating bone metabolism and remodelling the skeletal microenvironment require extensive validation in both cellular and pre-clinical animal models. Most importantly, prospective multicentre clinical trials rooted in large and well-characterised osteoporosis cohorts are essential for validating promising EV–ncRNA biomarkers. Building on robust pre-clinical evidence, phase I, II, and III clinical studies must be systematically conducted to comprehensively evaluate the safety, efficacy, and optimal dosing regimens of EV–ncRNA therapeutics—an obligatory step towards successful clinical translation. Moreover, the clinical translation of EV–ncRNA therapy faces major regulatory challenges. As a potential bioactive nanoparticle drug [ 146 ], EVs require the establishment of unique evaluation standards for their heterogeneity and complexity, focusing on rigorous quality control and long-term safety monitoring. Moreover, EV-based liquid biopsy research has made certain progress in the early detection and diagnosis of various cancers [ 147 ]. Given the vast industrialisation prospects of EV–ncRNA therapy, we speculate that EV–ncRNA liquid biopsy diagnostic kits will be the first to achieve breakthroughs in their applications for early screening and treatment monitoring. In the therapeutic field, the current trend is towards developing standardised and universally applicable EV products that can load ncRNAs, ultimately providing a new and precise treatment mode for osteoporosis and potentially many other diseases, and holding significant market potential.

CRediT authorship contribution statement

Chunlu Yan: Writing – review & editing. Qiao Wan: Writing – review & editing. Zeling Fang: Resources, Formal analysis. Chuangwei Sui: Resources, Formal analysis. Xiaolong Xiao: Resources, Formal analysis. Fan Yang: Resources, Formal analysis. Baoshan Di: Conceptualization. Yajun Li: Funding acquisition, Formal analysis. Juan Yu: Validation, Supervision. Yana Zhao: Validation, Supervision. Guoliang Liu: Validation, Funding acquisition. Yibo Zhou: Validation, Resources. Wei Li: Validation, Resources. Fangyu An: Funding acquisition, Conceptualization.

Funding

This work was financially supported by the National Natural Science Foundation of China (Grant No. 82560967), Internal fund projects of Longgang Central Hospital of Shenzhen (Grant No. YN2025B009), Gansu Joint Research Fund Project (Grant No. 24JRRA879), Gansu Natural Science Foundation (Grant No. 24JRRA1018), the Scientific Research Program of Gansu Chinese Medicine Bure (Grant No. GZKG-2024-88), Dingxi Technology Plan Project (Grant No. DX2024BY017), and Teaching Reform Research Project of Gansu Higher Education (Grant No. GJJGB228).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We were very grateful for Editage ( https://editage.cn ) with English language editing and FigDraw software ( https://www.figdraw.com ) with figure editing.

Footnotes

Peer review under the responsibility of Editorial Board of Non-coding RNA Research.

Contributor Information

Baoshan Di, Email: Lgzxyydbs@163.com.

Fangyu An, Email: anfyuwsmh@163.com.

References

1. Yang K., Wang X., Zhang C., Liu D., Tao L. Metformin improves HPRT1-targeted purine metabolism and repairs NR4A1-mediated autophagic flux by modulating FoxO1 nucleocytoplasmic shuttling to treat postmenopausal osteoporosis. Cell Death Dis. 2024;15:1–30. doi: 10.1038/s41419-024-07177-5.

2. Sietsema D.L. Fighting the epidemic: bone health and osteoporosis. Nurs. Clin. 2020;55:193–202. doi: 10.1016/j.cnur.2020.02.002.

3. Du Y., Chen H., Zhou L., Guo Q., Gong S., Feng S., Guan Q., Shi P., Lv T., Guo Y., Yang C., Sun P., Li K., Xu S., Li L. REGγ is essential to maintain bone homeostasis by degrading TRAF6, preventing osteoporosis. Proc. Natl. Acad. Sci. U. S. A. 2024;121 doi: 10.1073/pnas.2405265121.

4. Farooqi A.A., Attar R., Yulaevna I.M., Berardi R. Interaction of long non-coding RNAs and circular RNAs with microRNAs for the regulation of immunological responses in human cancers. Semin. Cell Dev. Biol. 2022;124:63–71. doi: 10.1016/j.semcdb.2021.05.029.

5. Gao Y., Li X., Shang S., Guo S., Wang P., Sun D., Gan J., Sun J., Zhang Y., Wang J., Wang X., Li X., Zhang Y., Ning S. LincSNP 3.0: an updated database for linking functional variants to human long non-coding RNAs, circular RNAs and their regulatory elements. Nucleic Acids Res. 2021;49:D1244–D1250. doi: 10.1093/nar/gkaa1037.

6. Madè A., Bibi A., Garcia-Manteiga J.M., Tascini A.S., Piella S.N., Tikhomirov R., Voellenkle C., Gaetano C., Leszek P., Castelvecchio S., Menicanti L., Martelli F., Greco S. circRNA-miRNA-mRNA deregulated network in ischemic heart failure patients. Cells. 2023;12:2578. doi: 10.3390/cells12212578.

7. Wang X., Pei Z., Hao T., Ariben J., Li S., He W., Kong X., Chang J., Zhao Z., Zhang B. Prognostic analysis and validation of diagnostic marker genes in patients with osteoporosis. Front. Immunol. 2022;13 doi: 10.3389/fimmu.2022.987937.

8. Zhao X., Zhao D., Geng B., Yaobin W., Xia Y. A novel ceRNA regulatory network involving the long noncoding NEAT1, miRNA-466f-3p and its mRNA target in osteoblast autophagy and osteoporosis. J. Mol. Med. 2022;100:1629–1646. doi: 10.1007/s00109-022-02255-7.

9. Jadamba B., Jin Y., Lee H. Harmonising cellular conversations: decoding the vital roles of extracellular vesicles in respiratory system intercellular communications. Eur. Respir. Rev. 2024;33 doi: 10.1183/16000617.0272-2023.

10. Li C., Xiang Z., Hou M., Yu H., Peng P., Lv Y., Ma C., Ding H., Jiang Y., Liu Y., Zhou H., Feng S. miR-NPs-RVG promote spinal cord injury repair: implications from spinal cord-derived microvascular endothelial cells. J. Nanobiotechnol. 2024;22:590. doi: 10.1186/s12951-024-02797-7.

11. Xue Z., Liu J., Xing W., Mu F., Wu Y., Zhao J., Liu X., Wang D., Wang J., Li X., Wang J., Huang B. Hypoxic glioma-derived exosomal miR-25-3p promotes macrophage M2 polarization by activating the PI3K-AKT-mTOR signaling pathway. J. Nanobiotechnol. 2024;22:628. doi: 10.1186/s12951-024-02888-5.

12. Zhou W., Van Sinderen M., Rainczuk K., Menkhorst E., Sorby K., Osianlis T., Pangestu M., Santos L., Rombauts L., Rosello-Diez A., Dimitriadis E. Dysregulated miR-124-3p in endometrial epithelial cells reduces endometrial receptivity by altering polarity and adhesion. Proc. Natl. Acad. Sci. U. S. A. 2024;121 doi: 10.1073/pnas.2401071121.

13. Di K., Fan B., Gu X., Huang R., Khan A., Liu C., Shen H., Li Z. Highly efficient and automated isolation technology for extracellular vesicles microRNA. Front. Bioeng. Biotechnol. 2022;10 doi: 10.3389/fbioe.2022.948757.

14. Grentzinger T., Oberlin S., Schott G., Handler D., Svozil J., Barragan-Borrero V., Humbert A., Duharcourt S., Brennecke J., Voinnet O. A universal method for the rapid isolation of all known classes of functional silencing small RNAs. Nucleic Acids Res. 2020;48:e79. doi: 10.1093/nar/gkaa472.

15. Jin X., Wei Z., Liu L., Liu H., Liu J. Comparative studies of two methods for miRNA isolation from milk whey. J. Zhejiang Univ. Sci. B. 2015;16:533–540. doi: 10.1631/jzus.B1400355.

16. Kumar K., Oli A., Hallikeri K., Shilpasree A.S., Goni M. An optimized protocol for total RNA isolation from archived formalin-fixed paraffin-embedded tissues to identify the long non-coding RNA in oral squamous cell carcinomas. MethodsX. 2022;9 doi: 10.1016/j.mex.2021.101602.

17. Lardeux H., Stavenhagen K., Paris C., Dueholm R., Kurek C., De Maria L., Gnerlich F., Leek T., Czechtizky W., Guillarme D., Jora M. Unravelling the link between oligonucleotide structure and diastereomer separation in hydrophilic interaction chromatography. Anal. Chem. 2024;96:9994–10002. doi: 10.1021/acs.analchem.4c01384.

18. Li L., Foley J.P., Helmy R. Simultaneous separation of small interfering RNA and lipids using ion-pair reversed-phase liquid chromatography. J. Chromatogr. A. 2019;1601:145–154. doi: 10.1016/j.chroma.2019.04.061.

19. Panda A.C., De S., Grammatikakis I., Munk R., Yang X., Piao Y., Dudekula D.B., Abdelmohsen K., Gorospe M. High-purity circular RNA isolation method (RPAD) reveals vast collection of intronic circRNAs. Nucleic Acids Res. 2017;45 doi: 10.1093/nar/gkx297.

20. Smela D., Zelinkova E., Rehulka P., Bilkova Z., Kupcik R. Phenol/chloroform-free TiO2-Based miRNA extraction from cell lysate. Int. J. Mol. Sci. 2022;23:8848. doi: 10.3390/ijms23168848.

21. Zhang B., Li H., Kong L., Yang N., Yang S., Qi L., Liu T., Wang X., Qin W. Tandem enrichment of serum exosomes and exosomal RNA with titanium dioxide. J. Chromatogr. A. 2023;1693 doi: 10.1016/j.chroma.2023.463882.

22. Zununi Vahed S., Barzegari A., Rahbar Saadat Y., Mohammadi S., Samadi N. A microRNA isolation method from clinical samples. Bioimpacts. 2016;6:25–31. doi: 10.15171/bi.2016.04.

23. Alshaer W., Zureigat H., Al Karaki A., Al-Kadash A., Gharaibeh L., Hatmal M.M., Aljabali A.A.A., Awidi A. siRNA: mechanism of action, challenges, and therapeutic approaches. Eur. J. Pharmacol. 2021;905 doi: 10.1016/j.ejphar.2021.174178.

24. Chen S., Ben S., Xin J., Li S., Zheng R., Wang H., Fan L., Du M., Zhang Z., Wang M. The biogenesis and biological function of PIWI-interacting RNA in cancer. J. Hematol. Oncol. 2021;14:93. doi: 10.1186/s13045-021-01104-3.

25. Kaur S., Frørup C., Mirza A.H., Fløyel T., Yarani R., Colli M.L., Johannesen J., Størling J., Eizirik D.L., Pociot F. Pro-Inflammatory cytokines promote the transcription of circular RNAs in human pancreatic β cells. Non-Coding RNA. 2022;8:69. doi: 10.3390/ncrna8050069.

26. Li J., Hong X., Jiang M., Kho A.T., Tiwari A., Wang A.L., Chase R.P., Celedón J.C., Weiss S.T., McGeachie M.J., Tantisira K.G. A novel piwi-interacting RNA associates with type 2-high asthma phenotypes. J. Allergy Clin. Immunol. 2024;153:695–704. doi: 10.1016/j.jaci.2023.10.032.

27. Mattick J.S., Amaral P.P., Carninci P., Carpenter S., Chang H.Y., Chen L.-L., Chen R., Dean C., Dinger M.E., Fitzgerald K.A., Gingeras T.R., Guttman M., Hirose T., Huarte M., Johnson R., Kanduri C., Kapranov P., Lawrence J.B., Lee J.T., Mendell J.T., Mercer T.R., Moore K.J., Nakagawa S., Rinn J.L., Spector D.L., Ulitsky I., Wan Y., Wilusz J.E., Wu M. Long non-coding RNAs: definitions, functions, challenges and recommendations. Nat. Rev. Mol. Cell Biol. 2023;24:430–447. doi: 10.1038/s41580-022-00566-8.

28. Rojas-Ríos P., Chartier A., Enjolras C., Cremaschi J., Garret C., Boughlita A., Ramat A., Simonelig M. piRNAs are regulators of metabolic reprogramming in stem cells. Nat. Commun. 2024;15:8405. doi: 10.1038/s41467-024-52709-4.

29. Shen L.-P., Zhang W.-C., Deng J.-R., Qi Z.-H., Lin Z.-W., Wang Z.-D. Advances in the mechanism of small nucleolar RNA and its role in DNA damage response. Mil. Med. Res. 2024;11:53. doi: 10.1186/s40779-024-00553-4.

30. Vaxevanis C., Bachmann M., Seliger B. Immune modulatory microRNAs in tumors, their clinical relevance in diagnosis and therapy. J. Immunother. Cancer. 2024;12 doi: 10.1136/jitc-2024-009774.

31. Zhang Q., Zhu Y., Cao X., Tan W., Yu J., Lu Y., Kang R., Wang X., Li E. The epigenetic regulatory mechanism of PIWI/piRNAs in human cancers. Mol. Cancer. 2023;22:45. doi: 10.1186/s12943-023-01749-3.

32. Zou J., Yang S., He C., Deng L., Xu B., Chen S. miR-630 as a therapeutic target in pancreatic cancer stem cells: modulation of the PRKCI-Hedgehog signaling axis. Biol. Direct. 2024;19:109. doi: 10.1186/s13062-024-00539-1.

33. Huang S.-Y., Huang C.-H., Chen C.-J., Chen T.-W., Lin C.-Y., Lin Y.-T., Kuo S.-M., Huang C.-G., Lee L.-A., Chen Y.-H., Chen M.-F., Kuo R.-L., Shih S.-R. Novel role for miR-1290 in host species specificity of influenza A virus. Mol. Ther. Nucleic Acids. 2019;17:10–23. doi: 10.1016/j.omtn.2019.04.028.

34. Das J.K., Choudhury P.P., Chaudhuri A., Hassan S.S., Basu P. Analysis of purines and pyrimidines distribution over miRNAs of human, gorilla, chimpanzee, mouse and rat. Sci. Rep. 2018;8:9974. doi: 10.1038/s41598-018-28289-x.

35. Jiang Y., Li Y., Shen Q., Yuan X., Gao F., Ma B. Comprehensive characterization of long non-coding RNAs in porcine tissues: expression patterns and functional insights during oocyte development. Cells. 2025;14:1409. doi: 10.3390/cells14181409.

36. Belavilas-Trovas A., Gregoriou M.-E., Tastsoglou S., Soukia O., Giakountis A., Mathiopoulos K. A species-specific lncRNA modulates the reproductive ability of the asian tiger mosquito. Front. Bioeng. Biotechnol. 2022;10 doi: 10.3389/fbioe.2022.885767.

37. Rosspopoff O., Cazottes E., Huret C., Loda A., Collier A.J., Casanova M., Rugg-Gunn P.J., Heard E., Ouimette J.-F., Rougeulle C. Species-specific regulation of XIST by the JPX/FTX orthologs. Nucleic Acids Res. 2023;51:2177–2194. doi: 10.1093/nar/gkad029.

38. Xia S., Feng J., Lei L., Hu J., Xia L., Wang J., Xiang Y., Liu L., Zhong S., Han L., He C. Comprehensive characterization of tissue-specific circular RNAs in the human and mouse genomes. Briefings Bioinf. 2017;18:984–992. doi: 10.1093/bib/bbw081.

39. Santos-Rodriguez G., Voineagu I., Weatheritt R.J. Evolutionary dynamics of circular RNAs in primates. eLife. 2021;10 doi: 10.7554/eLife.69148.

40. Dias Mirandela M., Zoch A., Leismann J., Webb S., Berrens R.V., Valsakumar D., Kabayama Y., Auchynnikava T., Schito M., Chowdhury T., MacLeod D., Xiang X., Zou J., Rappsilber J., Allshire R.C., Voigt P., Cook A.G., Barau J., O'Carroll D. Two-factor authentication underpins the precision of the piRNA pathway. Nature. 2024;634:979–985. doi: 10.1038/s41586-024-07963-3.

41. Deryusheva S., Talhouarne G.J.S., Gall J.G. “Lost and found”: snoRNA annotation in the Xenopus genome and implications for evolutionary studies. Mol. Biol. Evol. 2020;37:149–166. doi: 10.1093/molbev/msz209.

42. Isaac R., Reis F.C.G., Ying W., Olefsky J.M. Exosomes as mediators of intercellular crosstalk in metabolism. Cell Metab. 2021;33:1744–1762. doi: 10.1016/j.cmet.2021.08.006.

43. Rao G., Peng B., Zhang G., Fu X., Tian J., Tian Y. MicroRNAs in diabetic macroangiopathy. Cardiovasc. Diabetol. 2024;23:344. doi: 10.1186/s12933-024-02405-w.

44. Nojima T., Proudfoot N.J. Mechanisms of lncRNA biogenesis as revealed by nascent transcriptomics. Nat. Rev. Mol. Cell Biol. 2022;23:389–406. doi: 10.1038/s41580-021-00447-6.

45. Zhang X., Wang W., Zhu W., Dong J., Cheng Y., Yin Z., Shen F. Mechanisms and functions of long non-coding RNAs at multiple regulatory levels. Int. J. Mol. Sci. 2019;20:5573. doi: 10.3390/ijms20225573.

46. Zhang Y., Xue W., Li X., Zhang J., Chen S., Zhang J.-L., Yang L., Chen L.-L. The biogenesis of nascent circular RNAs. Cell Rep. 2016;15:611–624. doi: 10.1016/j.celrep.2016.03.058.

47. Ashwal-Fluss R., Meyer M., Pamudurti N.R., Ivanov A., Bartok O., Hanan M., Evantal N., Memczak S., Rajewsky N., Kadener S. circRNA biogenesis competes with Pre-mRNA splicing. Mol. Cell. 2014;56:55–66. doi: 10.1016/j.molcel.2014.08.019.

48. Qian Y., Li Y., Li R., Yang T., Jia R., Ge Y.-Z. circ-ZNF609: a potent circRNA in human cancers. J. Cell Mol. Med. 2021;25:10349–10361. doi: 10.1111/jcmm.16996.

49. Paul S., Gray D., Caswell J., Brooks J., Ye W., Moody T.S., Radinov R., Nechev L. Convergent biocatalytic mediated synthesis of siRNA. ACS Chem. Biol. 2023;18:2183–2187. doi: 10.1021/acschembio.3c00071.

50. Sato K., Siomi M.C. The piRNA pathway in Drosophila ovarian germ and somatic cells. Proc. Jpn. Acad. Ser. B Phys. Biol. Sci. 2020;96:32–42. doi: 10.2183/pjab.96.003.

51. Wang X., Zeng C., Liao S., Zhu Z., Zhang J., Tu X., Yao X., Feng X., Guang S., Xu C. Molecular basis for PICS-mediated piRNA biogenesis and cell division. Nat. Commun. 2021;12:5595. doi: 10.1038/s41467-021-25896-7.

52. Ballarino M., Morlando M., Pagano F., Fatica A., Bozzoni I. The cotranscriptional assembly of snoRNPs controls the biosynthesis of H/ACA snoRNAs in Saccharomyces cerevisiae. Mol. Cell Biol. 2005;25:5396–5403. doi: 10.1128/MCB.25.13.5396-5403.2005.

53. Grzechnik P., Szczepaniak S.A., Dhir S., Pastucha A., Parslow H., Matuszek Z., Mischo H.E., Kufel J., Proudfoot N.J. Nuclear fate of yeast snoRNA is determined by co-transcriptional Rnt1 cleavage. Nat. Commun. 2018;9:1783. doi: 10.1038/s41467-018-04094-y.

54. Zhang Y.-W., Wu Y., Liu X.-F., Chen X., Su J.-C. Targeting the gut microbiota-related metabolites for osteoporosis: the inextricable connection of gut-bone axis. Ageing Res. Rev. 2024;94 doi: 10.1016/j.arr.2024.102196.

55. Lobo R.A., Gompel A. Management of menopause: a view towards prevention. Lancet Diabetes Endocrinol. 2022;10:457–470. doi: 10.1016/S2213-8587(21)00269-2.

56. Buttgereit F., Palmowski A., Bond M., Adami G., Dejaco C. Osteoporosis and fracture risk are multifactorial in patients with inflammatory rheumatic diseases. Nat. Rev. Rheumatol. 2024;20:417–431. doi: 10.1038/s41584-024-01120-w.

57. Qaseem A., Mount C.A., Campos K., McLean R.M., Tierney S., Cross J.T., Fitterman N., Performance Measurement Committee of the American College of Physicians. Andrews R.A., Basch P., Goldzweig C.L., MacDonald S.T., MacLean C.H., Mathew S.M., Powell R.E., Saini S.D. Quality indicators for osteoporosis in adults: a review of performance measures by the American College of physicians. Ann. Intern. Med. 2023;176:1386–1391. doi: 10.7326/M23-1291.

58. Khan A.A., Slart R.H.J.A., Ali D.S., Bock O., Carey J.J., Camacho P., Engelke K., Erba P.A., Harvey N.C., Lems W.F., Morgan S., Moseley K.F., O’Brien C., Probyn L., Punda M., Richmond B., Schousboe J.T., Shuhart C., Ward K.A., Lewiecki E.M. International working group on DXA best practices, osteoporotic fractures: diagnosis, evaluation, and significance from the international working group on DXA best practices. Mayo Clin. Proc. 2024;99:1127–1141. doi: 10.1016/j.mayocp.2024.01.011.

59. Chen M., Li Y., Zhang M., Ge S., Feng T., Chen R., Shen J., Li R., Wang Z., Xie Y., Wang D., Liu J., Lin Y., Chang F., Chen J., Sun X., Cheng D., Huang X., Wu F., Zhang Q., Cai P., Yin P., Zhang L., Tang P. Histone deacetylase inhibition enhances extracellular vesicles from muscle to promote osteogenesis via miR-873-3p. Signal Transduct. Targeted Ther. 2024;9:256. doi: 10.1038/s41392-024-01976-0.

60. Mandourah A.Y., Ranganath L., Barraclough R., Vinjamuri S., Hof R.V., Hamill S., Czanner G., Dera A.A., Wang D., Barraclough D.L. Circulating microRNAs as potential diagnostic biomarkers for osteoporosis. Sci. Rep. 2018;8:8421. doi: 10.1038/s41598-018-26525-y.

61. You Y., Liu J., Zhang L., Li X., Sun Z., Dai Z., Ma J., Jiao G., Chen Y. WTAP-mediated m6A modification modulates bone marrow mesenchymal stem cells differentiation potential and osteoporosis. Cell Death Dis. 2023;14:33. doi: 10.1038/s41419-023-05565-x.

62. Gao P., Li S., Dong Z., Luo Y., Zhang X., Han L., Peng S., Shen J., Xu F., Deng Z. A plasma-derived exosomal microRNA signature by small RNA sequencing for early detection of postmenopausal osteoporosis. Clin. Transl. Med. 2024;14 doi: 10.1002/ctm2.1637.

63. Wang T., Mo L., Ou J., Fang Q., Wu H., Wu Y., Nandakumar K.S. Proteus mirabilis vesicles induce mitochondrial apoptosis by regulating miR96-5p/Abca1 to inhibit osteoclastogenesis and bone loss. Front. Immunol. 2022;13 doi: 10.3389/fimmu.2022.833040.

64. Yu L., Sui B., Fan W., Lei L., Zhou L., Yang L., Diao Y., Zhang Y., Li Z., Liu J., Hao X. Exosomes derived from osteogenic tumor activate osteoclast differentiation and concurrently inhibit osteogenesis by transferring COL1A1-targeting miRNA-92a-1-5p. J. Extracell. Vesicles. 2021;10 doi: 10.1002/jev2.12056.

65. Yu X., Rong P.-Z., Song M.-S., Shi Z.-W., Feng G., Chen X.-J., Shi L., Wang C.-H., Pang Q.-J. lncRNA SNHG1 induced by SP1 regulates bone remodeling and angiogenesis via sponging miR-181c-5p and modulating SFRP1/Wnt signaling pathway. Mol. Med. 2021;27:141. doi: 10.1186/s10020-021-00392-2.

66. Taipaleenmäki H., Saito H., Schröder S., Maeda M., Mettler R., Ring M., Rollmann E., Gasser A., Haasper C., Gehrke T., Weiss A., Grimm S.K., Hesse E. Antagonizing microRNA-19a/b augments PTH anabolic action and restores bone mass in osteoporosis in mice. EMBO Mol. Med. 2022;14 doi: 10.15252/emmm.202013617.

67. Hong J., Shi Z., Li C., Ji X., Li S., Chen Y., Jiang G., Shi M., Wang W., Zhang Y., Hu B., Yan S. Virtual screening identified natural Keap1-Nrf2 PPI inhibitor alleviates inflammatory osteoporosis through Nrf2-mir214-Traf3 axis. Free Radic. Biol. Med. 2021;171:365–378. doi: 10.1016/j.freeradbiomed.2021.05.020.

68. Zhang D., Du J., Yu M., Suo L. Urine-derived stem cells-extracellular vesicles ameliorate diabetic osteoporosis through HDAC4/HIF-1α/VEGFA axis by delivering microRNA-26a-5p. Cell Biol. Toxicol. 2023;39:2243–2257. doi: 10.1007/s10565-022-09713-5.

69. Yan J., Yang Y., Liu Y., Shi X., Wu H., Dai M. MicroRNA let-7g links foam cell formation and adipogenic differentiation: a key regulator of Paeonol treating atherosclerosis-osteoporosis. Phytomedicine. 2024;126 doi: 10.1016/j.phymed.2024.155447.

70. Zhao D., He J., Zhao X., Sheng X., Feng Z., Wang X., Zhang C., Wang S., Geng B., Xia Y. A novel lncRNA GM15416 regulates osteoblast apoptosis and differentiation through the c-Fos/Fas axis and mitigates osteoporosis. Int. J. Biol. Macromol. 2024;254 doi: 10.1016/j.ijbiomac.2023.127824.

71. Zhao Y., Ning J., Teng H., Deng Y., Sheldon M., Shi L., Martinez C., Zhang J., Tian A., Sun Y., Nakagawa S., Yao F., Wang H., Ma L. Long noncoding RNA Malat1 protects against osteoporosis and bone metastasis. Nat. Commun. 2024;15:2384. doi: 10.1038/s41467-024-46602-3.

72. Abood A., Mesner L., Rosenow W., Al-Barghouthi B.M., Horowitz N., Morgan E.F., Gerstenfeld L.C., Farber C.R. Identification of known and novel long noncoding RNAs potentially responsible for the effects of bone mineral density (BMD) genomewide association study (GWAS) loci. J. Bone Miner. Res. 2022;37:1500–1510. doi: 10.1002/jbmr.4622.

73. Wang C.-G., Hu Y.-H., Su S.-L., Zhong D. LncRNA DANCR and miR-320a suppressed osteogenic differentiation in osteoporosis by directly inhibiting the Wnt/β-catenin signaling pathway. Exp. Mol. Med. 2020;52:1310–1325. doi: 10.1038/s12276-020-0475-0.

74. Fei Q., Bai X., Lin J., Meng H., Yang Y., Guo A. Identification of aberrantly expressed long non-coding RNAs in postmenopausal osteoporosis. Int. J. Mol. Med. 2018;41:3537–3550. doi: 10.3892/ijmm.2018.3575.

75. Huang C., Li R., Yang C., Ding R., Li Q., Xie D., Zhang R., Qiu Y. PAX8-AS1 knockdown facilitates cell growth and inactivates autophagy in osteoblasts via the miR-1252-5p/GNB1 axis in osteoporosis. Exp. Mol. Med. 2021;53:894–906. doi: 10.1038/s12276-021-00621-y.

76. Li R., Shi T.-T., Wang Q., Zhang Y.-X. Elevated lncRNA MIAT in peripheral blood mononuclear cells contributes to post-menopausal osteoporosis. Aging (Albany NY) 2022;14:3143–3154. doi: 10.18632/aging.204001.

77. Xu Y., Zhang S., Liao X., Li M., Chen S., Li X., Wu X., Yang M., Tang M., Hu Y., Li Z., Yu R., Huang M., Song L., Li J. Circular RNA circIKBKB promotes breast cancer bone metastasis through sustaining NF-κB/bone remodeling factors signaling. Mol. Cancer. 2021;20:98. doi: 10.1186/s12943-021-01394-8.

78. Wu J., Li X., Nie H., Shen Y., Guo Z., Huihan Chu C., Cai K., Tang C. Phytic acid promotes high glucose-mediated bone marrow mesenchymal stem cells osteogenesis via modulating circEIF4B that sponges miR-186-5p and complexes with IGF2BP3. Biochem. Pharmacol. 2024;222 doi: 10.1016/j.bcp.2024.116118.

79. Deng M., Wang Z., Luo J., Cao H., Li Y., Chen L., Liu G. CircZNF367 promotes osteoclast differentiation and osteoporosis by interacting with FUS to maintain CRY2 mRNA stability. J. Orthop. Surg. Res. 2023;18:492. doi: 10.1186/s13018-023-03955-7.

80. Ismail S.M., Abd-Elmawla M.A., Shabayek M.I., Hamoud H.S., Darwish H.A., El-Sawalhi M.M. Exploring the role of circ-0091579/miR-1225-5p and circ-HIPK3/miR-338-3p axes in the pathogenesis of postmenopausal osteoporosis. Int. J. Biol. Macromol. 2024;278 doi: 10.1016/j.ijbiomac.2024.134688.

81. Han S., Kuang M., Sun C., Wang H., Wang D., Liu Q. Circular RNA hsa_circ_0076690 acts as a prognostic biomarker in osteoporosis and regulates osteogenic differentiation of hBMSCs via sponging miR-152. Aging (Albany NY) 2020;12:15011–15020. doi: 10.18632/aging.103560.

82. Xu Y., Sun L., Hu J., Xu S. Knockdown of hsa_circ_0001275 reverses dexamethasone-induced osteoblast growth inhibition via mediation of miR-377/CDKN1B axis. PLoS One. 2021;16 doi: 10.1371/journal.pone.0252126.

83. Zhao K., Zhao Q., Guo Z., Chen Z., Hu Y., Su J., Chen L., He Z., Cai X., Chen M., Zheng L., Wang W., Wang Q. Hsa_Circ_0001275: a potential novel diagnostic biomarker for postmenopausal osteoporosis. Cell. Physiol. Biochem. 2018;46:2508–2516. doi: 10.1159/000489657.

84. Chen K., Chen X., Lang C., Yuan X., Huang J., Li Z., Xu M., Wu K., Zhou C., Li Q., Zhu C., Liu L., Shang X. CircFam190a: a critical positive regulator of osteoclast differentiation via enhancement of the AKT1/HSP90β complex. Exp. Mol. Med. 2023;55:2051–2066. doi: 10.1038/s12276-023-01085-y.

85. Yao Y., Cai X., Zhang M., Zheng Y., Fei W., Zhao M., Zheng C. Circ-Plod2 destabilizes Mpo mRNA by binding to IGF2BP2 to promote osteogenic differentiation of bone marrow mesenchymal stem cells. Eur. J. Pharmacol. 2023;961 doi: 10.1016/j.ejphar.2023.176192.

86. Yin P., Xue Y. CircRNA hsa_circ_0006859 inhibits the osteogenic differentiation of BMSCs and aggravates osteoporosis by targeting miR-642b-5p/miR-483-3p and upregulating EFNA2/DOCK3. Int. Immunopharmacol. 2023;116 doi: 10.1016/j.intimp.2023.109844.

87. Li H., Wang C., Yao J., Jin Y., Song X., Meng Q., Wu J., Liu Q., Liu M., Sun H. Circ_0114581 promotes osteogenic differentiation of BMSCs via the MiR-155-5p/HNRNPA3 axis. Life Sci. 2023;333 doi: 10.1016/j.lfs.2023.122127.

88. Ji H., Cui X., Yang Y., Zhou X. CircRNA hsa_circ_0006215 promotes osteogenic differentiation of BMSCs and enhances osteogenesis-angiogenesis coupling by competitively binding to miR-942-5p and regulating RUNX2 and VEGF. Aging (Albany NY) 2021;13:10275–10288. doi: 10.18632/aging.202791.

89. Huang Y., Xiao D., Huang S., Zhuang J., Zheng X., Chang Y., Yin D. Circular RNA YAP1 attenuates osteoporosis through up-regulation of YAP1 and activation of Wnt/β-catenin pathway. Biomed. Pharmacother. 2020;129 doi: 10.1016/j.biopha.2020.110365.

90. Ae G., K A., Z R., T S. Identification and analysis of differentially expressed lncRNAs and their ceRNA networks in DMD/mdx primary myoblasts. Sci. Rep. 2024;14 doi: 10.1038/s41598-024-75221-7.

91. Xue J., Liu L., Liu H., Li Z. LncRNA SNHG14 activates autophagy via regulating miR-493-5p/Mef2c axis to alleviate osteoporosis progression. Commun. Biol. 2023;6:1120. doi: 10.1038/s42003-023-05493-8.

92. Liu J., Wang B., Chen H., Yu X., Cao X., Zhang H. Osteoclast-derived exosomes influence osteoblast differentiation in osteoporosis progression via the lncRNA AW011738/miR-24-2-5p/TREM1 axis. Biomed. Pharmacother. 2024;178 doi: 10.1016/j.biopha.2024.117231.

93. Yin C., Tian Y., Yu Y., Li D., Miao Z., Su P., Zhao Y., Wang X., Pei J., Zhang K., Qian A. Long noncoding RNA AK039312 and AK079370 inhibits bone formation via miR-199b-5p. Pharmacol. Res. 2021;163 doi: 10.1016/j.phrs.2020.105230.

94. Liu J., Chen M., Ma L., Dang X., Du G. piRNA-36741 regulates BMP2-mediated osteoblast differentiation via METTL3 controlled m6A modification. Aging (Albany NY) 2021;13:23361–23375. doi: 10.18632/aging.203630.

95. Long Y., Ma Y., Yang H., Wang X., Fan J., Li Y., Tao B., Sun T. A bone-targeting delivery platform based on mesoporous silica loaded with piR7472 for the treatment of osteoporosis. Arab. J. Chem. 2024

96. Ahn I., Kang C.S., Han J. Where should siRNAs go: applicable organs for siRNA drugs. Exp. Mol. Med. 2023;55:1283–1292. doi: 10.1038/s12276-023-00998-y.

97. Kang J.-Y., Mun D., Chun Y., Park D.-S., Kim H., Yun N., Joung B. Engineered small extracellular vesicle-mediated NOX4 siRNA delivery for targeted therapy of cardiac hypertrophy. J. Extracell. Vesicles. 2023;12 doi: 10.1002/jev2.12371.

98. Luo P., Zhang Y., Huang M., Luo G., Ma Y., Wang X. Microdroplets encapsulated with NFATc1-siRNA and exosomes-derived from MSCs onto 3D porous PLA scaffold for regulating osteoclastogenesis and promoting osteogenesis. Int. J. Nanomed. 2024;19:3423. doi: 10.2147/IJN.S443413.

99. Kumar M.A., Baba S.K., Sadida H.Q., Marzooqi S.A., Jerobin J., Altemani F.H., Algehainy N., Alanazi M.A., Abou-Samra A.-B., Kumar R., Al-Shabeeb Akil A.S., Macha M.A., Mir R., Bhat A.A. Extracellular vesicles as tools and targets in therapy for diseases. Signal Transduct. Targeted Ther. 2024;9:27. doi: 10.1038/s41392-024-01735-1.

100. Deng A.-F., Wang F.-X., Wang S.-C., Zhang Y.-Z., Bai L., Su J.-C. Bone-organ axes: bidirectional crosstalk. Mil. Med. Res. 2024;11:37. doi: 10.1186/s40779-024-00540-9.

101. Liu J.-H., Chen C.-Y., Liu Z.-Z., Luo Z.-W., Rao S.-S., Jin L., Wan T.-F., Yue T., Tan Y.-J., Yin H., Yang F., Huang F.-Y., Guo J., Wang Y.-Y., Xia K., Cao J., Wang Z.-X., Hong C.-G., Luo M.-J., Hu X.-K., Liu Y.-W., Du W., Luo J., Hu Y., Zhang Y., Huang J., Li H.-M., Wu B., Liu H.-M., Chen T.-H., Qian Y.-X., Li Y.-Y., Feng S.-K., Chen Y., Qi L.-Y., Xu R., Tang S.-Y., Xie H. Extracellular vesicles from child gut microbiota enter into bone to preserve bone mass and strength. Adv. Sci. 2021;8 doi: 10.1002/advs.202004831.

102. Furesi G., de Jesus Domingues A.M., Alexopoulou D., Dahl A., Hackl M., Schmidt J.R., Kalkhof S., Kurth T., Taipaleenmäki H., Conrad S., Hofbauer C., Rauner M., Hofbauer L.C. Exosomal miRNAs from prostate cancer impair osteoblast function in mice. Int. J. Mol. Sci. 2022;23:1285. doi: 10.3390/ijms23031285.

103. Hu Y., Hou Z., Liu Z., Wang X., Zhong J., Li J., Guo X., Ruan C., Sang H., Zhu B. Oyster mantle-derived exosomes alleviate osteoporosis by regulating bone homeostasis. Biomaterials. 2024;311 doi: 10.1016/j.biomaterials.2024.122648.

104. Wang Z.-X., Luo Z.-W., Li F.-X.-Z., Cao J., Rao S.-S., Liu Y.-W., Wang Y.-Y., Zhu G.-Q., Gong J.-S., Zou J.-T., Wang Q., Tan Y.-J., Zhang Y., Hu Y., Li Y.-Y., Yin H., Wang X.-K., He Z.-H., Ren L., Liu Z.-Z., Hu X.-K., Yuan L.-Q., Xu R., Chen C.-Y., Xie H. Aged bone matrix-derived extracellular vesicles as a messenger for calcification paradox. Nat. Commun. 2022;13:1453. doi: 10.1038/s41467-022-29191-x.

105. Woo H.-E., Cho J.-Y., Lim Y.-H. Propionibacterium freudenreichii MJ2-derived extracellular vesicles inhibit RANKL-induced osteoclastogenesis and improve collagen-induced rheumatoid arthritis. Sci. Rep. 2024;14 doi: 10.1038/s41598-024-76911-y.

106. Xu R., Shen X., Xie H., Zhang H., Liu D., Chen X., Fu Y., Zhang P., Yang Y., Cheng J., Jiang H. Identification of the canonical and noncanonical role of miR-143/145 in estrogen-deficient bone loss. Theranostics. 2021;11:5491–5510. doi: 10.7150/thno.55041.

107. Kobayashi-Sun J., Yamamori S., Kondo M., Kuroda J., Ikegame M., Suzuki N., Kitamura K.-I., Hattori A., Yamaguchi M., Kobayashi I. Uptake of osteoblast-derived extracellular vesicles promotes the differentiation of osteoclasts in the zebrafish scale. Commun. Biol. 2020;3:190. doi: 10.1038/s42003-020-0925-1.

108. Wang W., Qiao S.-C., Wu X.-B., Sun B., Yang J.-G., Li X., Zhang X., Qian S.-J., Gu Y.-X., Lai H.-C. Circ_0008542 in osteoblast exosomes promotes osteoclast-induced bone resorption through m6A methylation. Cell Death Dis. 2021;12:628. doi: 10.1038/s41419-021-03915-1.

109. Zhai X., Zhou J., Huang X., Weng J., Lin H., Sun S., Chi J., Meng L. LncRNA GHET1 from bone mesenchymal stem cell–derived exosomes improves doxorubicin-induced pyroptosis of cardiomyocytes by mediating NLRP3. Sci. Rep. 2024;14 doi: 10.1038/s41598-024-70151-w.

110. Wang D., Guo Y., Heng B.C., Zhang X., Wei Y., He Y., Xu M., Xia B., Deng X. Cell membrane vesicles derived from hBMSCs and hUVECs enhance bone regeneration. Bone Res. 2024;12:23. doi: 10.1038/s41413-024-00325-9.

111. Chen Y., Zhang Z., Li Z., Wu W., Lan S., Yan T., Mei K., Qiao Z., Wang C., Bai C., Li Z., Wu S., Wang J., Zhang Q. Dynamic nanomechanical characterization of cells in exosome therapy. Microsyst. Nanoeng. 2024;10:97. doi: 10.1038/s41378-024-00735-z.

112. Zhao S., Xiu G., Wang J., Wen Y., Lu J., Wu B., Wang G., Yang D., Ling B., Du D., Xu J. Engineering exosomes derived from subcutaneous fat MSCs specially promote cartilage repair as miR-199a-3p delivery vehicles in Osteoarthritis. J. Nanobiotechnol. 2023;21:341. doi: 10.1186/s12951-023-02086-9.

113. Yuan J., Wang Y., Huang Y., Li S., Zhang X., Wu Z., Zhao W., Zhu J., Zhang J., Huang G., Yu P., Cheng X., Wang X., Liu X., Jia J. Investigating novel therapeutic approaches for idiopathic short stature: targeting siRNA and growth hormone delivery to the growth plate using exosome nanoparticles. Adv. Sci. 2024;11 doi: 10.1002/advs.202309559.

114. Mao G., Xu Y., Long D., Sun H., Li H., Xin R., Zhang Z., Li Z., Yang Z., Kang Y. Exosome-transported circRNA_0001236 enhances chondrogenesis and suppress cartilage degradation via the miR-3677-3p/Sox9 axis. Stem Cell Res. Ther. 2021;12:389. doi: 10.1186/s13287-021-02431-5.

115. Ma S., Xing X., Huang H., Gao X., Xu X., Yang J., Liao C., Zhang X., Liu J., Tian W., Liao L. Skeletal muscle-derived extracellular vesicles transport glycolytic enzymes to mediate muscle-to-bone crosstalk. Cell Metab. 2023;35:2028–2043.e7. doi: 10.1016/j.cmet.2023.10.013.

116. Sahu A., Clemens Z.J., Shinde S.N., Sivakumar S., Pius A., Bhatia A., Picciolini S., Carlomagno C., Gualerzi A., Bedoni M., Houten B.V., Lovalekar M., Fitz N.F., Lefterov I., Barchowsky A., Koldamova R., Ambrosio F. Regulation of aged skeletal muscle regeneration by circulating extracellular vesicles. Nat. Aging. 2021;1:1148. doi: 10.1038/s43587-021-00143-2.

117. Zhang X., Song W., Liu Y., Han K., Wu Y., Cho E., Fang Z., Jiang L., Hu Y., Zhu X., Jiang J., Huangfu X., Zhao J. Healthy tendon stem cell-derived exosomes promote tendon-to-bone healing of aged chronic rotator cuff tears by breaking the positive-feedback cross-talk between senescent tendon stem cells and macrophages through the modulation of macrophage polarization. Small. 2024;20 doi: 10.1002/smll.202311033.

118. Kalluri R. The biology and function of extracellular vesicles in immune response and immunity. Immunity. 2024;57:1752–1768. doi: 10.1016/j.immuni.2024.07.009.

119. Zhou B., Chen Q., Zhang Q., Tian W., Chen T., Liu Z. Therapeutic potential of adipose-derived stem cell extracellular vesicles: from inflammation regulation to tissue repair. Stem Cell Res. Ther. 2024;15:249. doi: 10.1186/s13287-024-03863-5.

120. Liu Z., Chen J., Ren Y., Liu S., Ba Y., Zuo A., Luo P., Cheng Q., Xu H., Han X. Multi-stage mechanisms of tumor metastasis and therapeutic strategies. Signal Transduct. Targeted Ther. 2024;9:270. doi: 10.1038/s41392-024-01955-5.

121. Ding F., Zhou M., Ren Y., Li Y., Xiang J., Li Y., Yu J., Hong Y., Fu Z., Li H., Pan Z., Liu B. Mitochondrial extracellular vesicles: a promising avenue for diagnosing and treating lung diseases. ACS Nano. 2024;18:25372–25404. doi: 10.1021/acsnano.4c02940.

122. Li B., Qi C., Zhang Y., Shi L., Zhang J., Qian H., Ji C. Frontier role of extracellular vesicles in kidney disease. J. Nanobiotechnol. 2024;22:583. doi: 10.1186/s12951-024-02852-3.

123. Su Y., Chen M., Xu W., Gu P., Fan X. Advances in extracellular-vesicles-based diagnostic and therapeutic approaches for ocular diseases. ACS Nano. 2024;18:22793–22828. doi: 10.1021/acsnano.4c08486.

124. Li H., Zhang J., Tan M., Yin Y., Song Y., Zhao Y., Yan L., Li N., Zhang X., Bai J., Jiang T., Li H. Exosomes based strategies for cardiovascular diseases: opportunities and challenges. Biomaterials. 2024;308 doi: 10.1016/j.biomaterials.2024.122544.

125. Putthanbut N., Lee J.Y., Borlongan C.V. Extracellular vesicle therapy in neurological disorders. J. Biomed. Sci. 2024;31:85. doi: 10.1186/s12929-024-01075-w.

126. Zhao H., Zhu L., Wang C., Yang Y. Extracellular vesicles-based theranostics for neurodegenerative diseases. Wiley Interdiscip. Rev. Nanomed. Nanobiotechnol. 2024;16 doi: 10.1002/wnan.1993.

127. Fang F., Yang J., Wang J., Li T., Wang E., Zhang D., Liu X., Zhou C. The role and applications of extracellular vesicles in osteoporosis. Bone Res. 2024;12:4. doi: 10.1038/s41413-023-00313-5.

128. Kotani A., Ito M., Kudo K. Non-coding RNAs and lipids mediate the function of extracellular vesicles in cancer cross-talk. Semin. Cancer Biol. 2021;74:121–133. doi: 10.1016/j.semcancer.2021.04.017.

129. Wu F., Song C., Zhen G., Jin Q., Li W., Liang X., Xu W., Guo W., Yang Y., Dong W., Jiang A., Kong P., Yan J. Exosomes derived from BMSCs in osteogenic differentiation promote type H blood vessel angiogenesis through miR-150-5p mediated metabolic reprogramming of endothelial cells. Cell. Mol. Life Sci. 2024;81:344. doi: 10.1007/s00018-024-05371-4.

130. Huang X., Wu W., Jing D., Yang L., Guo H., Wang L., Zhang W., Pu F., Shao Z. Engineered exosome as targeted lncRNA MEG3 delivery vehicles for osteosarcoma therapy. J. Contr. Release. 2022;343:107–117. doi: 10.1016/j.jconrel.2022.01.026.

131. Ren L., Zeng F., Deng J., Bai Y., Chen K., Chen L., Sun L. Inflammatory osteoclasts-derived exosomes promote bone formation by selectively transferring lncRNA LIOCE into osteoblasts to interact with and stabilize Osterix. FASEB J. 2022;36 doi: 10.1096/fj.202101106RR.

132. Yu M., Cai Z., Zhang J., Zhang Y., Fu J., Cui X. Aberrant NSUN2-mediated m5C modification of exosomal LncRNA MALAT1 induced RANKL-mediated bone destruction in multiple myeloma. Commun. Biol. 2024;7:1249. doi: 10.1038/s42003-024-06918-8.

133. Zhu C., Shen K., Zhou W., Wu H., Lu Y. Exosome-mediated circ_0001846 participates in IL-1β-induced chondrocyte cell damage by miR-149-5p-dependent regulation of WNT5B. Clin. Immunol. 2021;232 doi: 10.1016/j.clim.2021.108856.

134. Ma S., Li S., Zhang Y., Nie J., Cao J., Li A., Li Y., Pei D. BMSC-derived exosomal CircHIPK3 promotes osteogenic differentiation of MC3T3-E1 cells via mitophagy. Int. J. Mol. Sci. 2023;24:2785. doi: 10.3390/ijms24032785.

135. Fu M., Fang L., Xiang X., Fan X., Wu J., Wang J. Microarray analysis of circRNAs sequencing profile in exosomes derived from bone marrow mesenchymal stem cells in postmenopausal osteoporosis patients. J. Clin. Lab. Anal. 2022;36 doi: 10.1002/jcla.23916.

136. Luo A., Liu H., Huang C., Wei S. Exosome-transmitted circular RNA circ-LMO7 facilitates the progression of osteosarcoma by regulating miR-21-5p/ARHGAP24 axis. Cancer Biol. Ther. 2024;25 doi: 10.1080/15384047.2024.2343450.

137. Qi L., Pan C., Yan J., Ge W., Wang J., Liu L., Zhang L., Lin D., Shen S.G.F. Mesoporous bioactive glass scaffolds for the delivery of bone marrow stem cell-derived osteoinductive extracellular vesicles lncRNA promote senescent bone defect repair by targeting the miR-1843a-5p/Mob3a/YAP axis. Acta Biomater. 2024;177:486–505. doi: 10.1016/j.actbio.2024.01.044.

138. Cui Y., Fu S., Sun D., Xing J., Hou T., Wu X. EPC-derived exosomes promote osteoclastogenesis through LncRNA-MALAT1. J. Cell Mol. Med. 2019;23:3843–3854. doi: 10.1111/jcmm.14228.

139. Mo C., Huang B., Zhuang J., Jiang S., Guo S., Mao X. LncRNA nuclear-enriched abundant transcript 1 shuttled by prostate cancer cells-secreted exosomes initiates osteoblastic phenotypes in the bone metastatic microenvironment via miR-205-5p/runt-related transcription factor 2/splicing factor proline- and glutamine-rich/polypyrimidine tract-binding protein 2 axis. Clin. Transl. Med. 2021;11 doi: 10.1002/ctm2.493.

140. Xia W., Xie J., Cai Z., Liu X., Wen J., Cui Z.-K., Zhao R., Zhou X., Chen J., Mao X., Gu Z., Zou Z., Zou Z., Zhang Y., Zhao M., Mac M., Song Q., Bai X. Damaged brain accelerates bone healing by releasing small extracellular vesicles that target osteoprogenitors. Nat. Commun. 2021;12:6043. doi: 10.1038/s41467-021-26302-y.

141. Xu L., Xu X., Liang Y., Wen C., Ouyang K., Huang J., Xiao Y., Deng X., Xia J., Duan L. Osteoclast-targeted delivery of anti-miRNA oligonucleotides by red blood cell extracellular vesicles. J. Contr. Release. 2023;358:259–272. doi: 10.1016/j.jconrel.2023.04.043.

142. Hu H., Zhang H., Bu Z., Liu Z., Lv F., Pan M., Huang X., Cheng L. Small extracellular vesicles released from bioglass/hydrogel scaffold promote vascularized bone regeneration by transferring miR-23a-3p. Int. J. Nanomed. 2022;17:6201–6220. doi: 10.2147/IJN.S389471.

143. Yang S., Guo S., Tong S., Sun X. Exosomal miR-130a-3p regulates osteogenic differentiation of human adipose-derived stem cells through mediating SIRT7/Wnt/β-catenin axis. Cell Prolif. 2020;53 doi: 10.1111/cpr.12890.

144. Zheng J., Kong Y., Hu X., Li Z., Li Y., Zhong Y., Wei X., Ling J. MicroRNA-enriched small extracellular vesicles possess odonto-immunomodulatory properties for modulating the immune response of macrophages and promoting odontogenesis. Stem Cell Res. Ther. 2020;11:517. doi: 10.1186/s13287-020-02039-1.

145. Chen X., Wang L., Cheng Q., Deng Z., Tang Y., Yan Y., Xie L., Li X. Multiple myeloma exosomal miRNAs suppress cGAS-STING antiviral immunity. Biochim. Biophys. Acta Mol. Basis Dis. 2024;1870 doi: 10.1016/j.bbadis.2024.167457.

146. Li M., Fang F., Sun M., Zhang Y., Hu M., Zhang J. Extracellular vesicles as bioactive nanotherapeutics: an emerging paradigm for regenerative medicine. Theranostics. 2022;12:4879–4903. doi: 10.7150/thno.72812.

147. Su X., Shan Z., Duan S. Harnessing extracellular vesicles using liquid biopsy for cancer diagnosis and monitoring: highlights from AACR Annual Meeting 2024. J. Hematol. Oncol. 2024;17:55. doi: 10.1186/s13045-024-01577-y.