Sun et al. (2020) proposed that the AGE/RAGE axis can inhibit the osteogenic differentiation of BMSCs. Their study employed inhibition of the AGE/RAGE axis to mitigate dysfunction in SCs differentiation. Okazaki et al. found that AGEs constrained the osteogenic differentiation of mouse stromal ST2 cells by inhibiting Osterix (OSX) expression and partially increasing RAGE expression. Furthermore, AGEs interfere with the process of SCs differentiation into bone cells by potentially reducing osteocalcin production while increasing RAGE expression (Okazaki et al., 2012). AGEs exert a more pronounced impact during the immature stage of osteoblasts compared to the differentiation stage, inhibiting differentiation and reducing the number of mature osteoblasts (Ogawa et al., 2007). Stolzing et al. (2010) added different doses of AGEs to cultured MSCs and found that the self-renewal and osteogenic differentiation of MSCs were significantly reduced. Under osteogenic differentiation conditions, the extent of this effect depended on the concentration of AGEs in the culture medium. Furthermore, the proliferation of MSCs significantly increased in the low-concentration group, while normal proliferation and osteogenic differentiation of MSCs were impaired in the high-concentration group. They also observed that AGEs suppressed osteocalcin mRNA expression in rat MSCs, thereby hindering their differentiation. Lin et al. (2016) reported that HMGB1 facilitate the osteogenic differentiation of BMSCs while also increasing the expression of RAGE and Toll-like receptors 2 and 4 (TLR2/4) bound to HMGB1. RAGE, a high-affinity receptor for HMGB1, can activate the p38/MAPK and NF-κB pathways upon binding to HMGB1, thereby promoting the osteogenic differentiation of BMSCs (Park et al., 2004). The p38/MAPK pathway plays a crucial role in cell cycle regulation (Barnum and O’Connell, 2014); Kim et al. reported (Kim and Kwon, 2013) that COMP-Ang1 induces the upregulation of the PI3K/AKT and p38/MAPK pathways, thereby facilitating the attenuation of osteogenic differentiation of MSCs by AGEs via the Ang1/Tie2 pathway.

TGF-β is a crucial factor in regulating the differentiation of MSCs and plays a vital role in stem cell differentiation (Li et al., 2024). Notsu et al. considered that AGEs increase TGF-β by binding to RAGE, and the AGE-TGF-β pathway has a negative effect on the differentiation of MSCs into osteoblasts, impairing their differentiation. This indicates that TGF-β is one of the factors influencing the differentiation potential of SCs (Notsu et al., 2014). In recent years, joint cartilage regeneration technology has advanced significantly, driven by continuous improvements in biological scaffold materials. TGF-β3, as an important isoform of the TGF-β family, plays a pivotal role in mesenchymal stem cell differentiation through both Smad-dependent and non-Smad pathways. Its active involvement and tightly regulated role in the bone healing process have been widely recognized. In recent years, there has been increasing interest in the potential of TGF-β3 to promote and induce the proliferation, osteogenesis, and chondrogenic differentiation of adult SCs in biological scaffold materials. In particular, the induction of TGF-β3 is particularly significant in the early stages of the osteogenic process, providing new therapeutic strategies and research ideas for bone tissue regeneration. These studies not only help us to understand the mechanism of cartilage repair and regeneration deeply, but also provide a solid theoretical basis and experimental foundation for future clinical applications (Li et al., 2018; Roth et al., 2019; Martin et al., 2021).

In general, the role of TGF-β3 in cartilage formation is cell-type specific. Jin et al.'s reported that the inhibitory effect of TGF-β3 on chondrocytes is achieved through the activation of Notch signaling, which inhibits the proliferation of mesenchymal cells and pre-cartilage condensation (Jin et al., 2007). In another study, they also reported a similar finding regarding the inhibitory effect of TGF-β3 on the differentiation of MSCs, which is that TGF-β3 downregulates Protein Kinase C-α (PKC-α) mediated activation of connexin 43, integrin β4, and ERK, inhibiting chondrogenic differentiation of mesenchymal cells (Jin et al., 2008). In contrast to the inhibitory effect of TGF-β3 on MSC differentiation mentioned above, Zheng et al. found that knocking out the TβRIII gene can promote TGF-β3-induced MSCs cartilage differentiation, demonstrating the positive induction effect of TGF-β3 on mesenchymal stem cell differentiation (Zheng et al., 2018). Similarly, Jin et al. found that TGF-β3 stimulates the differentiation of MSCs into chondrocytes and inhibits the differentiation of chondrocytes. This is because TGF-β3 promotes chondrogenic differentiation of mesenchymal cells by activating the PKC-α and p38 MAPK pathways (Jin et al., 2006). Based on the multiple studies on TGF-β3 specified above, it can be concluded that the differences in TGF-β3 are due to its various functions, manifested as a mixed effect of induction and inhibition on the differentiation process of MSCs.

Overall, it is crucial to investigate the complex signaling pathways and mechanisms by which AGEs affect the differentiation process of SCs. This will not only aid in revealing the mechanisms by which AGEs influence SC differentiation, but also provide insight into potential strategies to reverse the toxic effects of AGEs.

The Wnt/β-catenin pathway, a central signaling pathway, precisely regulates cell polarity, determines the differentiation fate of cells, guides the migration process of cells, and has a profound impact on spindle formation, organ development, and stem cell renewal (Nayak et al., 2016). Currently, 19 Wnt ligands have been identified, and all of these ligands specifically bind to a seven-transmembrane Wnt receptor named Frizzled (FZD) (Houschyar et al., 2019). The Wnt pathway, a well-established osteogenic differentiation pathway, is a complex system comprising three distinct pathways, which are believed to be activated upon Wnt receptor activation: the canonical Wnt/β-catenin cascade, the noncanonical planar cell polarity (PCP) pathway, and the Wnt/Ca2+ pathway.

In the canonical Wnt/β-catenin cascade, the central event is the nuclear translocation of the β-catenin protein and its regulation of target genes. In the absence of Wnt ligands, β-catenin is degraded by intracellular complexes-primarily composed of glycogen synthase kinase 3 (GSK-3). However, once the canonical Wnt/β-catenin cascade is activated, Dishevelled proteins (Dvl) are triggered, which in turn inhibit GSK-3, thereby stabilizing β-catenin and promoting its nuclear translocation and target gene expression. Atypical Wnt signaling also plays a crucial role in the differentiation of bone tissue. Unlike the canonical Wnt/β-catenin cascade, the atypical Ca2⁺ dependent Wnt pathway uniquely promotes osteogenesis in MSCs. When the Wnt ligand binds to the FZD receptor, it activates G proteins, triggering the release of Ca2⁺ ions from the endoplasmic reticulum. This process initiates the Protein Kinase C (PKC) pathway and continue signaling to promote osteogenesis (Ahmadi et al., 2022). The Wnt pathway is initiated when Wnt ligands bind to FZD receptors, activating G proteins that subsequently trigger the release of Ca2⁺ ions from the endoplasmic reticulum, which then initiates the PKC pathway. Low-density lipoprotein receptor associated protein 5/6 (LRP5/6) or receptor tyrosine kinase-like orphan receptors (RORs) function as common receptors alongside FZD, facilitating the binding of Wnt proteins to their receptors. The involvement of these co-receptors dictates the downstream effects following successful ligand binding, initiating either the canonical Wnt/β-catenin cascade or the noncanonical planar cell polarity (PCP) pathway (Houschyar et al., 2019; Ahmadi et al., 2022) (Figure 4). Zhou et al. (2023) found that AGEs can impair the osteogenic differentiation process of BMSCs by upregulating the expression of fat and obesity-related gene FTO. This process is regulated by FTO to modify the SOST transcript with m6A, increase the mRNA stability recognized by YTHDF2, inhibit the Wnt signaling pathway, and ultimately disrupt the differentiation of BMSCs into bone. Minear et al. (2010) amplified the cell Wnt response by removing the Axin 2 gene in a mouse model and found that delivering liposome vesicles containing purified Wnt-3a protein can promote the Wnt pathway, leading to increased proliferation and early differentiation of BMSCs, thereby accelerating fracture healing.

Growth factors activate aspects of the Wnt pathway. The TGF-β pathway is a membrane-to-nucleus signaling cascade activated by receptor-mediated transcription factors. Due to structural and functional considerations, the 32 family members are classified into TGF-β and bone morphogenetic proteins (BMPs) subfamilies, along with other variations (David and Massagué, 2018). BMPs have been extensively studied, with BMPs 2, 6, and 9 being the primary isoforms. As potent growth factor, BMPs stimulate MSCs to differentiate into osteoblasts (Carreira et al., 2014). The functional Wnt signaling pathway constitutes the core mechanism of BMP-induced osteogenic differentiation of MSCs. There is a significant interaction between the Wnt and the TGF-β pathways, as they share some key regulatory targets, thus forming a complex signaling network. Among them, β-catenin, a key node in this network, plays a crucial role in regulation (Case and Rubin, 2010). β-catenin plays various roles during different phases of bone repair. In the initial stages following injury, it modulates the osteoblast-to-chondrocyte ratio within the callus tissue induced by MSCs, ensuring a balanced and coordinated repair process (Bao et al., 2017). In the later stages of bone healing, β-catenin induces osteoblasts to differentiate and produce an osteogenic matrix, promoting bone reconstruction and regeneration (Wang et al., 2017). Zhang et al. (2009) suggest that BMP2 regulates β-catenin by stimulating the expression of Lrp5 in osteoblasts and inhibiting the expression of β-Trcp. Chen et al. (2019) also found that the key growth factor BMP2 stimulates the Wnt/β-catenin pathway to promote the osteogenic differentiation of BMSCs. The addition of Wnt-3a enhances the osteogenic effect of BMP9. However, it is counteracted by the downregulation of β-catenin or the increased expression of FrzB, which acts as an antagonist of the FZD receptor (Boland et al., 2004).

The glyoxalase system is integral, serving as a key enzyme system present in all mammalian cells. This system consists of two enzymes that act in concern: glyoxalase 1 (Glo1) and glyoxalase 2 (Glo2). The AGE/RAGE pathway alleviate differentiation dysfunction of BMSCs by enhancing the activity of Glo1. AGEs activate complex signaling pathways by binding to RAGE, thereby triggering various toxic effects in the organism. These enzymes catalyze sequential reactions, with reduced glutathione (GSH) serving as a catalytically active and essential component. The Glo1 enzyme plays a crucial role in the metabolic process by catalyzing the nonenzymatic isomerization of the active dicarbonyl metabolite methylglyoxal (MG) with glutathione to produce dithiol acetaldehyde. Additionally, Glo2 catalyzes the hydrolysis of S-D-lactoylglutathione to generate D-lactic acid, thereby efficiently supplementing the glutathione consumed in the Glo1-catalyzed process. Due to this synergistic action, the glyoxalase system is able to efficiently process dicarbonyl compounds in vivo and maintain normal metabolic functions of cells (Rabbani and Thornalley, 2019). Wang et al. (2019) used glycine to inhibit the formation of AGEs, and the study described that Glo1 also mediates this effect. In Jandial et al. mouse model (Jandial et al., 2018), blocking Glo1 resulted in increased AGE production and upregulation of RAGE expression. Consequently, Glo1 inhibition caused cellular accumulation of MG, triggering rapid modifications of proteins, lipids, and DNA, ultimately inducing apoptosis. To counteract the adverse effects of AGEs on primary SCs, an effective strategy involves blocking the interaction between AGE and RAGE. The discovery of Zhang et al. can be utilized with the RAGE inhibitor FPS-ZM1, which can attenuate the adverse effects of AGEs on the osteogenic potential of SCs (Zhang et al., 2018). Rasheed et al. (2011) found that knocking down RAGE by pre-treating soluble RAGE (sRAGE) or using siRNAs effectively reduced the cytotoxicity of AGEs. Based on the above research, activating Glo1 or directly inhibiting the binding of AGEs to RAGE is a highly feasible strategy to mitigate the toxic effects of AGEs.

Alterations in the Wnt/β-catenin pathway significantly influence bone metabolism. Notably, elevated concentrations of Wnt3a can inhibit the osteogenic differentiation of BMSCs, indicating that regulating the Wnt/β-catenin pathway needs to be fully balanced to avoid adverse effects on stem cell differentiation. Therefore, comprehensive research on the Wnt/β-catenin pathway and its interaction with DNA methylation processes is highly important for optimizing the physiological functions of SCs in AGEs -induced environments (Boland et al., 2004). In Liang et al.'s study, DNA methylation was found to have a substantial impact on the expression of Wnt/β-catenin signaling pathway genes, which also proves that the impact of DNA methylation on the physiological function of MSCs is achieved through the Wnt/β-catenin pathway (Liang et al., 2015). DNA methylation is a molecular modification that determines cell identity and lineages by regulating gene expression and maintaining genomic stability. Under the action of DNA methyltransferase, the covalent bond at the cytosine 5 carbon position of the CpG dinucleotide in the genome binds to a methyl group. DNA methylation induces changes in chromatin structure, DNA conformation, stability, and the dynamics of DNA-protein interactions, thus exerting control over gene expression (Nishiyama and Nakanishi, 2021). DNMT1, DNMT3a and DNMT3b play indispensable roles in DNA methylation. Recent research by Zhang et al. (2018) demonstrated that the expression of DNMT1 and DNMT3a was upregulated, indicating that AGEs increased the level of DNA methylation in ADSCs. To reverse this effect, the investigators used FPS-ZM1, which successfully rescued the loss of osteogenic differentiation in ADSCs by inhibiting AGEs induced DNA methylation. In a study by Li et al. (2020), when ADSCs were cultured in a medium containing AGEs, they exhibited high levels of 5-mC and DNMTs, accompanied by a significant reduction in osteogenic differentiation capacity in vitro. However, by applying DNMT inhibitors (5-aza-dC), investigators found that the osteogenic differentiation potential of ADSCs was improved. The promotion of DNA demethylation enhanced the osteogenic differentiation of ADSCs, highlighting the critical role of DNA methylation levels in regulating this process.