A previous study employing genome-wide methylation analysis identified hypermethylation in advanced degenerative discs compared with early-stage discs (10). Further evidence has demonstrated that numerous genes related to stress responses, matrix catabolism and cell death are regulated by aberrant DNA methylation in degenerative discs (11).

In particular, the upregulation of the DNA methyltransferase (DNMT)3B has received considerable attention. DNMT3B expression is regarded as a prominent feature of degenerative nucleus pulposus (NP) cells (12). In both in vitro and in vivo models, blocking DNMT activity [e.g., using the inhibitor 5-azacytidine (5-AZA)] can reduce ferroptosis and oxidative stress, ultimately slowing NP cell degeneration and ECM breakdown (12), as schematically illustrated in Fig. 1, which depicts the regulatory roles of DNMTs and TET enzymes in methylation-mediated ferroptosis and ECM metabolism.

Research has indicated that secreted protein acidic and cysteine rich (SPARC), an important regulator of ECM homeostasis, is hypermethylated in aging and degenerative discs, leading to suppressed SPARC expression, which disrupts matrix integrity and is closely associated with chronic pain (13). Growth differentiation factor 5 (GDF5) has a critical CpG site in its 5′ untranslated region (UTR), where increased methylation alters transcription factor (SP1/SP3) binding, reducing GDF5 expression and increasing the risk of multiple musculoskeletal disorders, including lumbar disc degeneration (14).

Regarding ferroptosis and oxidative stress, silencing of the solute carrier family 40 member 1 (SLC40A1) gene by DNMT3B-mediated hypermethylation makes NP cells more vulnerable to ferroptosis and oxidative stress (12). Conversely, restoring SLC40A1 expression by inhibiting DNMT3B can ease disc damage (15). In addition, epigenetic regulation of apoptotic pathways is evident in the methylation-dependent upregulation of certain microRNAs (miRNAs/miRs; for example, miR-143), which suppress BCL2 and increase NP cell apoptosis (16).

Autophagy and senescence pathways also display altered methylation patterns. For instance, miR-129-5p is downregulated in degenerative discs due to hypermethylation in its promoter region; consequently, its reduced suppression of Beclin-1 affects normal autophagy homeostasis (17). E4F transcription factor 1, another key factor in cell senescence, is also epigenetically inhibited when DNMT3B is upregulated by ALKBH5-mediated N⁶-methyladenosine (m⁶A) demethylation, compounding the degenerative process (18).

DNA methylation not only affects gene transcription but also interacts with other epigenetic processes, such as histone modifications, ncRNA-mediated regulation and RNA methylation, to amplify injury from senescence, inflammation and oxidative stress (6). Inflammatory signals can stimulate DNMT and ten-eleven translocation (TET) enzyme expression, while oxidative stress can modulate the activity of methyltransferases via the p38/MAPK pathway, increasing catabolic enzyme expression (19).

Histone acetylation, mediated by HATs, reduces the positive charge of lysine residues, leading to chromatin relaxation and promoting gene transcription (23). Conversely, HDACs remove acetyl groups, resulting in chromatin condensation and transcriptional repression. Emerging evidence suggests that dysregulation of HDACs serves a crucial role in IDD (22).

Studies have demonstrated that HDAC9 expression is reduced in degenerative NP cells (22,24). Under normal conditions, HDAC9 deacetylates and stabilizes RUNX family transcription factor 3 (RUNX3) to maintain cell viability. Loss of HDAC9 expression in IDD diminishes RUNX3 stability, thereby promoting NP cell apoptosis (22). By contrast, HDAC4 has been shown to promote IDD progression by upregulating the Krüppel-like factor 5 (KLF5) and apoptosis signal-regulating kinase 1 signaling axis, which accelerates ECM degradation (24). These findings highlight the complex role of histone acetylation/deacetylation in IDD pathogenesis and suggest that selective HDAC inhibition may offer novel therapeutic potential for mitigating disc degeneration.

In the pathogenesis and progression of IDD, histone methylation has been shown to serve a crucial regulatory role in key processes such as ECM homeostasis, cellular senescence, autophagy and inflammation (21). This modification is primarily mediated by methyltransferases [such as enhancer of zeste homolog 2 (EZH2), lysine methyltransferase 2A (KMT2A) and SUV39H2] and demethylases [such as lysine demethylase (KDM)2/7, KDM4B and KDM3A], which add or remove methyl groups on specific lysine or arginine residues of histones, thereby influencing chromatin conformation and gene transcription (25,26).

Multiple studies have demonstrated that EZH2, a methyltransferase responsible for trimethylation of histone H3 lysine 27 (H3K27), is elevated in degenerative discs (27,28). By enriching trimethylation of histone H3 lysine 27 (H3K27me3) in specific gene promoter regions, EZH2 suppresses gene expression, accelerating the senescence of NP cells and promoting ECM degradation (28). For example, inhibition of EZH2 can remove H3K27me3 marks at the SOX9 promoter, upregulate cartilage phenotype-related genes such as SOX9, and curb ECM breakdown, thus counteracting endplate and NP degeneration (27). Additionally, EZH2-mediated H3K27me3 can silence key regulatory factors such as dickkopf Wnt signaling pathway inhibitor 1 or miR-129-5p, thereby activating the NLRP3, NLR family apoptosis inhibitory protein/NLR family CARD domain containing 4 or MAPK1 pathways, and further exacerbating pyroptosis and inflammation in NP cells (21,28). These findings suggest that inhibiting EZH2 expression or activity may be a potential strategy to delay disc degeneration.

Besides EZH2, other methyltransferases also serve essential roles in the pathological progression of IDD. SUV39H2 can specifically monomethylate protein phosphatase 1 catalytic subunit α at K141, disrupting its interaction with transcription factor EB (TFEB), and hampering TFEB nuclear translocation and autophagy gene activation, ultimately inducing cell senescence and disc degeneration (29). KMT2A (MLL1) enhances METTL3 expression via H3K4me3 enrichment at the METTL3 promoter, which increases m⁶A modification and downregulates autophagy related (ATG)4a transcription. This leads to impaired autophagy, GATA binding protein 4 activation and a senescent phenotype in NP cells (30). Such findings highlight the interplay among histone methylation, autophagy and cellular senescence, offering novel insights into IDD mechanisms.

In contrast to methyltransferases, histone demethylases (for example, the KDM2, KDM4 and KDM3 families) remove methyl groups from specific residues, thus either activating or repressing target genes (31). Research has shown that targeted inhibition of KDM2/7, using either gene editing or pharmacological approaches, may promote the differentiation of human induced pluripotent stem cells into notochord-like cells, potentially opening novel avenues for regenerative therapy against IDD (32). Furthermore, in a 400 mOsm environment, KDM4B upregulates NP marker genes and enhances stem cell differentiation into NP-like cells (33). This finding highlights the crucial role of histone demethylation in maintaining disc cell identity under osmotic stress, a key pathological factor in IDD. By promoting notochordal and NP-like differentiation, KDM4B-mediated demethylation may help restore the regenerative potential of degenerated discs. Therefore, targeting KDM4B activity or mimicking its epigenetic effects could represent a promising therapeutic approach for disc repair and regeneration. However, excessive demethylase activity could trigger abnormal regulation of downstream genes. For instance, increased KDM3A may elevate hypoxia-inducible factor 1α levels, provoking disturbances in autophagy and apoptosis among NP cells; thus, precisely regulating demethylase activities remains a crucial therapeutic goal (34).

miRNAs are endogenous small RNAs, 20–22 nucleotides in length, and regulate gene expression by binding to the 3′UTR of target mRNAs, thereby inhibiting translation or promoting degradation (36,37). During IDD, miRNAs serve key regulatory roles in NP cells, annulus fibrosus cells and inflammation-related processes (38). Studies have demonstrated that miRNAs, through their specific targeting mechanisms, influence ECM synthesis, autophagy and inflammatory pathways (39,40). For example, a regulatory network involving multiple miRNAs has been identified as crucial in IDD progression (41).

During IDD pathogenesis, some miRNAs affect matrix metabolism and apoptosis. For instance, miR-141-5p promotes IDD progression by regulating circRNA_0000253 (42). Additionally, miR-141 has been shown to accelerate IDD by inhibiting the sirtuin (SIRT)1/NF-κB pathway (43).

In IDD-related inflammation regulation, studies have indicated that miRNAs influence the expression of inflammatory factors by modulating specific pathways (44,45). For example, miRNA-222 serves a role in IDD progression by regulating MMP1 expression (46). Furthermore, the interaction between miRNAs and autophagy is also considered a key factor in IDD progression (47).

Notably, the function of certain miRNAs may depend on different stages of IDD or specific microenvironments (38). This regulatory complexity suggests that miRNAs could serve as potential therapeutic targets for IDD (48). As illustrated in Fig. 3, miRNAs repress target messenger RNAs (mRNAs), thereby regulating key biological processes such as apoptosis, inflammation, and ECM metabolism. In addition, long non-coding RNAs (lncRNAs) and circular RNAs (circRNAs) act as competing endogenous RNAs (ceRNAs) that modulate miRNA activity, further influencing IDD pathogenesis (49).

lncRNAs, which are transcripts longer than 200 nucleotides without significant protein-coding potential, serve pivotal roles in IDD. By acting as competing endogenous RNAs (ceRNAs), interacting with proteins or regulating RNA modifications (such as m⁶A), lncRNAs influence key processes such as NP cell proliferation, apoptosis, ECM catabolism and inflammatory responses (50,51). Current studies have identified a variety of lncRNAs that either promote or ameliorate IDD (52,53).

Several lncRNAs have been shown to aggravate IDD by promoting NP cell apoptosis, inflammation or ECM degradation. For instance, OIP5-AS1 is highly expressed in degenerative discs and exacerbates these pathological changes by sponging miR-25-3p (54). LINC01121 is also induced by IL-1 and TNF-α, and accelerates disc degeneration through the miR-150-5p/MMP16 axis, elevating inflammation-related enzymes such as MMP-3 and ADAM metallopeptidase with thrombospondin type 1 motif 5 (ADAMTS5) (55). Similarly, LINC00324 is upregulated in IDD and enhances Fas ligand expression, thereby exacerbating NP cell apoptosis (56). HOTAIR promotes pro-degenerative events by targeting miR-130b and inhibiting the PTEN/AKT pathway, leading to reduced CyclinD1 expression and impaired NP-cell proliferation (57). In addition, HCG18 augments apoptosis and inflammation via the miR-495-3p/follistatin like 1 axis (58), while FAF1 (referring to a lncRNA transcript in that region) is associated with advanced degeneration grades and activates the Erk signaling pathway (59). Beyond ceRNA interactions, aberrant m⁶A modification also contributes to IDD. WTAP-mediated hypermethylation of NORAD accelerates its degradation via YTH N⁶-methyladenosine RNA binding protein F2, reducing the ability of NORAD to restrain PUMILIO and thereby promoting cellular senescence (51).

By contrast, some lncRNAs exhibit protective functions against IDD. KLF3-AS1 is downregulated in degenerative discs but can improve NP cell viability and suppress apoptosis and ECM breakdown by sequestering miR-10a-3p, thereby upregulating zinc finger and BTB domain containing 20 (60). RP11-81H3.2 also serves a protective role by binding miR-1539, relieving its inhibitory effect on COL2A1 and mitigating NP cell apoptosis (61). Additionally, LINC00689 counters disc degeneration by sponging miR-3127-5p and activating the autophagy-related gene ATG7, which promotes autophagy and diminishes apoptosis in NP cells (62).

ncRNAs, particularly circRNAs, have attracted attention in studies of IDD. With their high stability and resistance to degradation, circRNAs can bind to proteins or miRNAs in ways that influence NP-cell proliferation, apoptosis, ECM metabolism and inflammatory responses, thereby contributing to the development and progression of IDD (63,64).

Multiple studies have demonstrated that circRNAs frequently show aberrant expression in degenerative NP tissues and can aggravate IDD by regulating crucial pathways. For example, Chen et al (65) found that circGPATCH2L was highly expressed in degenerated NP tissues when in a hypomethylated (m⁶A) state, enabling it to bind to and block the phosphorylation of tripartite motif containing 28, thereby inhibiting P53 degradation and promoting DNA damage and apoptosis. Du et al (66) reported that hsa_circ_0083756 was upregulated in degenerative NP tissues and cells and, by sponging miR-558, it upregulated triggering receptor expressed on myeloid cells 1 expression, ultimately suppressing NP cell proliferation and ECM formation. Meng and Xu (67) demonstrated that hsa_circ_0001658 exerted a protective effect in IDD by sponging miR-181c-5p and enhancing FAS expression, thereby inhibiting degeneration. Wang et al (68) identified circEYA3 as a key regulator of intervertebral disc degeneration. circEYA3 acts as a sponge for miR-196a-5p to upregulate EBF1 (early B-cell factor 1), thereby activating NF-κB signaling and promoting NP cell apoptosis and ECM degradation. Yan et al (69) reported that hsa_circ_0134111 was upregulated in IDD tissues and could bind to miR-578, thereby increasing the levels of ADAMTS5 and MMP-9 to promote ECM degradation and inflammation.

These findings collectively indicate that circRNAs serve pivotal regulatory roles in IDD. They often serve as miRNA ‘sponges’ or interact with specific proteins to disrupt the dynamic balance of ECM metabolism, cell proliferation and apoptosis, and inflammatory responses (70). At the same time, they offer novel avenues for potential clinical interventions. Strategies, such as inhibiting the upregulation of particular circRNAs or enhancing their physiological degradation, could represent promising approaches for slowing or preventing disc degeneration in the future.

Excessive reactive oxygen species (ROS) can activate p38/MAPK, ERK and other signaling pathways, thereby affecting the stability and transcriptional activity of histone methyltransferases (such as KMT2D) and other epigenetic factors (for example, DNMTs and HDACs). This leads to the upregulation of matrix-degrading genes (MMPs and ADAMTS) and exacerbates ECM degradation (19). Notably, oxidative stress is not only a product of metabolic imbalance but also amplifies degeneration via DNA and histone modifications (19).

NAD⁺ is a cofactor for the SIRT family, directly influencing the catalytic efficiency of deacetylases such as SIRT1 and SIRT6. When NAD⁺ levels decline, SIRT activity is diminished, weakening the resilience of intervertebral disc cells against external stress. Conversely, maintaining higher NAD⁺ levels helps regulate autophagy and ECM homeostasis (72). Additionally, α-ketoglutarate (α-KG) can serve as a cofactor for TET demethylases, influencing DNA and histone demethylation (73). Recent findings suggest that α-KG inhibits excessive expression of inflammatory or fibrotic genes in disc cells and possibly slows degenerative processes by upregulating demethylase activity (74).

Lysine lactylation (Kla), a histone modification dependent on the intracellular lactate concentration, is increasingly recognized (75). In the hypoxic environment of the disc, anaerobic glycolysis leads to lactate accumulation, and the resulting Kla modifications may regulate key biological processes such as cellular aging, ECM metabolic imbalance and inflammatory responses. Recent evidence from NP cells further confirms that Kla is dynamically regulated under hypoxic and high-lactate conditions, highlighting its critical role in intervertebral disc degeneration (76). Additionally, m⁶A RNA modification intersects with various metabolic pathways (including ferroptosis and oxidative stress): ALKBH5 and FTO can alter the stability or translation of key gene mRNAs, thereby affecting the cellular stress response and degenerative progression (18).

Taken together, IDD is no longer viewed merely as a structural degradation resulting from mechanical loading and inflammation, but rather as a disease driven by the bidirectional regulatory network involving metabolic intermediates and pathways, including those associated with NAD⁺ metabolism, α-ketoglutarate-dependent reactions, and lactate-mediated signaling, as well as epigenetic modifications (including DNA/histone methylation, m⁶A, and Kla) (76). A deeper understanding of these ‘metabolism-epigenetics’ interactions will aid in identifying novel molecular targets and strategies for early intervention or even potential reversal of disc degeneration (77).

Targeting epigenetic modifications offers novel therapeutic approaches for IDD. DNA methylation inhibitors, such as 5-AZA, have been shown to restore the expression of key genes involved in disc homeostasis by reversing hypermethylation-induced silencing (12). Similarly, histone modification regulators, such as EZH2 inhibitors, have demonstrated the ability to attenuate inflammation and ECM degradation, offering another promising avenue for treatment (6,27).

RNA modifications, particularly m⁶A methylation, serve a crucial role in IDD progression by influencing mRNA stability and translation (78). The modulation of key m⁶A enzymes, such as METTL3, FTO and ALKBH5, may provide an effective means to regulate NP cell apoptosis and ECM homeostasis (18). RNA-based therapies, including small interfering RNA and antisense oligonucleotides, could offer precise interventions to correct abnormal RNA modifications and restore gene function (79).

Epigenetic regulation can be combined with biological therapies, such as stem cell transplantation and gene editing, to enhance IDD treatment outcomes. For instance, mesenchymal stem cell (MSC) transplantation has been explored as a regenerative therapy for IDD, but its efficacy is often limited by the harsh degenerative microenvironment (80,81). Preconditioning stem cells with epigenetic modulators, such as DNMT3A or EZH2 inhibitors, could improve their survival and therapeutic potential in IDD treatment (6,27).

Another promising approach involves integrating epigenetic regulation with nanotechnology-based drug delivery systems. Nanocarriers, such as liposomes and biodegradable polymers, could facilitate the targeted delivery of epigenetic drugs, reducing systemic side effects while enhancing local efficacy (82). Additionally, gene-editing technologies, such as clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9), offer the potential to directly modify epigenetic regulatory genes, providing a long-term therapeutic solution for IDD (83).

Beyond these approaches, bioengineered nanomaterials have been applied to co-deliver MSCs and epigenetic modulators for disc regeneration. Novel nanofibrous spongy microspheres (NF-SMS) enhanced MSC seeding, proliferation and differentiation compared with conventional microcarriers. A hyperbranched polymer (HP) with strong miRNA-binding affinity was used to complex with anti-miR-199a, forming ‘double shell’ polyplexes with high transfection efficiency. Encapsulation of these polyplexes within biodegradable nanospheres (NS) enabled sustained release of anti-miR-199a. The composite system (MSC/HP-anti-miR-199a/NS/NF-SMS) promoted NP-like phenotypes, resisted calcification in vitro and in vivo, and, in a rabbit lumbar degeneration model, preserved disc height, maintained ECM function and prevented IDD calcification (84).

Advances in epigenomic profiling have paved the way for personalized IDD treatment strategies. By analyzing epigenetic signatures, such as DNA methylation patterns and histone modification profiles, clinicians can stratify patients based on their molecular characteristics and tailor treatment plans accordingly (13,27,85). For example, patients with elevated EZH2 expression and suppressed SOX9 levels may benefit from EZH2 inhibitors in combination with MSC therapy, whereas those with increased DNMT3b activity might respond better to DNA methylation inhibitors (18).

Dynamic monitoring of epigenetic modifications during treatment could enable real-time adjustments to therapeutic strategies. This ‘precise-dynamic-reversible’ approach refers to the concept of precisely targeting disease-related epigenetic modifications, dynamically monitoring their changes during treatment and reversibly modulating these marks in response to therapeutic needs. In the context of IDD, this strategy emphasizes individualized, adaptable interventions that can fine-tune gene expression and cellular responses in real time, potentially improving both safety and efficacy in clinical management (15,49).

To synthesize the aforementioned translational insights, Table I (6,81,86–88) summarizes the major candidate epigenetic interventions for IDD. Table I compiles current strategies, their targets, experimental systems, therapeutic outcomes, adverse effects and delivery options, providing a comprehensive overview of the preclinical landscape and guiding future clinical development.

Despite encouraging preclinical findings, several translational bottlenecks hinder the clinical application of epigenetic therapies for IDD. First, off-target effects remain a major concern. Epigenetic modulators such as DNA methylation inhibitors or HDAC inhibitors often act broadly rather than in a locus-specific manner, potentially leading to dysregulation of unrelated genes and adverse biological outcomes in non-disc tissues (89). Second, immune responses induced by RNA-based therapies or CRISPR/deactivated Cas9 epigenetic editors can activate innate immunity or exacerbate inflammation, which is particularly problematic in the already inflamed disc microenvironment (90,91). Third, long-term safety is insufficiently understood. While epigenetic modifications are theoretically reversible, persistent or cumulative alterations may increase the risk of tumorigenesis, ectopic tissue remodeling or interference with normal aging pathways (92). Finally, the lack of standardized outcome measures complicates translational progress. Preclinical studies employ heterogeneous endpoints, from disc height index and MRI signals to histological and molecular markers, making cross-study comparisons difficult and limiting their predictive value for clinical outcomes such as pain relief and functional improvement (93).

Overcoming these challenges will require the development of precision-targeted delivery systems, immune-evasive biomaterials, long-term safety monitoring protocols and consensus-driven standardized evaluation criteria to guide both preclinical and clinical research (94).