Post-translational modifications in the oral microenvironment: Stem cell regulation from periodontal regeneration to oral cancer therapy

Abstract

The oral microenvironment plays a pivotal role in determining stem cell fate, driving both regeneration and pathological transformation. Emerging evidence suggests that post-translational modifications (PTMs) play a role as dynamic molecular signatures that regulate key signaling networks in dental-derived mesenchymal stem cells. These PTMs not only influence stem cell self-renewal and differentiation in periodontal tissue regeneration but also contribute to cancer stem cell plasticity and therapeutic resistance in oral squamous cell carcinoma (OSCC). At the pathway level, PTM programs interface with Wnt/β-catenin and bone morphogenetic protein/SMAD axis and integrate mitogen-activated protein kinase (p38/c-Jun N-terminal kinase) → runt-related transcription factor 2 in regeneration, whereas in OSCC/cancer stem cell they converge on Janus kinase/signal transducer and activator of transcription 3, phosphatidylinositol 3-kinase/protein kinase B/mammalian target of the rapamycin, and transforming growth factor-beta/SMAD-driven epithelial-mesenchymal transition. This review expounds on recent advances in PTM-mediated regulatory mechanisms in dental-derived mesenchymal stem cells, outlines their functional implications in inflammatory and tumor microenvironments, and discusses translational strategies-including localized, time-staged PTM modulation for regeneration and pathway-anchored combinations for OSCC-for regenerative medicine and targeted cancer therapies. Future research directions emphasize the integration of single-cell and spatial multi-omics with PTM profiling as a new approach to precision-based dental and oncological therapies.

Key Words: Dental-derived mesenchymal stem cells; Post-translational modifications; Periodontal regeneration; Oral squamous cell carcinoma; Cancer stem-cell plasticity

Core Tip: Post-translational modifications act as dynamic molecular codes that steer dental-derived mesenchymal stem cells through a continuum from periodontal regeneration to oral squamous cell carcinoma stem-cell plasticity. This review integrates single-cell and spatial multi-omics evidence to map a sequential cascade-acetylation, crotonylation, phosphorylation, trimethylation, ubiquitination and N6-methyladenosine RNA methylation-that reprograms lineage commitment, immune evasion and therapy resistance. Identifying tractable post-translational modification enzymes (e.g., general control non-depressible 5, enhancer of zeste homolog 2, methyltransferase-like 3) clarifies actionable levers for regeneration and oncology.

INTRODUCTION

The oral microenvironment encompasses multifaceted and dynamic interactions among immunologic, mechanical, hypoxic, and metabolic cues that collectively regulate the behavior of dental-derived mesenchymal stem cells (DMSCs), including periodontal ligament stem cells (PDLSCs), dental pulp stem cells (DPSCs), and stem cells from the apical papilla (SCAPs). DMSCs exhibit intrinsic multipotency and robust regenerative potential within their native niches, thereby rendering them essential for the preservation and restoration of periodontal architecture[1,2]. Nevertheless, the cellular plasticity that underpins their reparative functions may be co-leveraged under pathological microenvironmental influences, facilitating malignant transformation, especially in the development and progression of oral squamous cell carcinoma (OSCC)[3].

A growing body of evidence suggests that post-translational modifications (PTMs) play a crucial role in modulating stem cell fate. Interestingly, PTMs, including lysine acetylation, serine/threonine/tyrosine phosphorylation, ubiquitination and related ubiquitin-like modifications (e.g., ISGylation), DNA and histone methylation, and N6-methyladenosine (m6A) RNA methylation, serve as molecular codes that modulate gene expression, protein stability, signal transduction, and chromatin architecture[4-6]. These modifications are not static but operate in a combinatorial and temporally dynamic manner, integrating microenvironmental cues into functional cellular outcomes.

In periodontal regeneration, for example, general control non-depressible 5 (GCN5)-dependent histone acetylation has been shown to modulate the Wnt antagonist Dickkopf-1 (DKK1), thereby regulating osteogenic differentiation via the Wnt/β-catenin axis[7]. Ubiquitin E3 ligases such as Smurf1 have been reported to negatively regulate runt-related transcription factor 2 (RUNX2), a master transcription factor in osteogenesis, through proteasomal degradation in response to mechanical or inflammatory signals[8]. Similarly, recent discoveries have revealed that RNA m6A methylation, catalyzed by enzymes such as methyltransferase-like 3 (METTL3) and KIAA1429 (VIRMA), modulates odontogenic differentiation by stabilizing mRNA transcripts involved in lineage commitment, such as distal-less homeobox 3 (DLX3) and ATP citrate lyase (ACLY), through readers like insulin-like growth factor 2 mRNA binding protein 2 (IGF2BP2)[9,10].

Interestingly, many of these PTM-regulated pathways are reactivated or subverted in oral cancer. OSCC cells, particularly cancer stem cell (CSC)-like subpopulations, harness PTM-mediated networks to sustain epithelial-mesenchymal transition (EMT), immune evasion, metabolic plasticity, and resistance to chemotherapy or radiotherapy[11]. E3 ligases, including neural precursor cell expressed developmentally downregulated protein 4 (NEDD4) L, S-phase kinase-associated protein 2 (Skp2), STIP1 homology and U-box-containing protein 1, HECT and RCC-like domain-containing protein 5, and RNF139, have been implicated in regulating tumor cell stemness and therapeutic tolerance via proteostasis and the modulation of key oncogenic pathways[12-14].

This review synthesizes the current understanding of PTM-mediated regulation in DMSCs, positioning these modifications along a molecular continuum from tissue regeneration to oncogenic transformation. It first outlines the core features of DMSCs and their surrounding microenvironment, then examines how PTMs are modulated under inflammatory, mechanical, and metabolic stress. It further explores how PTM patterns shape regenerative responses and how dysregulation of these patterns initiates a shift from repair to malignancy. Emphasis is placed on the role of PTMs in maintaining CSC plasticity and therapy resistance in OSCC. The review also highlights advances in single-cell and spatial multi-omics for dissecting PTM dynamics within heterogeneous tissues and concludes by evaluating translational strategies that leverage PTM-targeted approaches for regenerative and oncologic applications.

OVERVIEW OF DMSCS & ORAL MICROENVIRONMENT

DMSCs represent a unique subpopulation of somatic stem cells isolated from craniofacial tissues, characterized by their multilineage differentiation capacity, immunomodulatory potential, and neural crest origin. Among these, PDLSCs, DPSCs, and SCAPs have been extensively studied for their regenerative properties and responsiveness to microenvironmental cues. PDLSCs play a pivotal role in alveolar bone remodeling and periodontal ligament regeneration. Besides, DPSCs play a crucial role in maintaining the dentin-pulp complex and promoting neurovascular regeneration, whereas SCAPs, situated at the root apex of immature permanent teeth, exhibit enhanced proliferative potential and differentiation plasticity, particularly under hypoxic and developmental conditions[15,16]. Table 1 summarizes the tissue origins, predominant PTM patterns, key regulatory enzymes and relative regenerative capacity of the major dental-derived stem-cell populations introduced above.

Table 1 Dental-derived stem-cell types and post-translational modification signatures.

Stem-cell type	Tissue source	Dominant PTMs reported	Key regulatory enzymes	Regenerative potency	Ref.
PDLSCs	Periodontal ligament	Histone H3K9/H3K14 acetylation; m6A RNA methylation; K48-linked ubiquitination	GCN5 (HAT); METTL3 (m6A writer); Smurf1 (E3 ligase)	High-drives alveolar bone, cementum and ligament regeneration	[7,24,36,41]
DPSCs	Dental pulp (dentin-pulp complex)	m6A RNA methylation; histone H3K27/H3K9 acetylation; p38/JNK phosphorylation	METTL3 (m6A writer); p300/GCN5 (HATs); p38 MAPK (kinase)	High-supports dentin-pulp repair and neurovascular regeneration	[10,25,26,29,43]
SCAPs	Apical papilla of immature permanent teeth	m6A RNA methylation; stage-specific H3K9 acetylation	METTL3 (m6A writer); HAT (e.g., GCN5)	Very high-promotes root development, angiogenic osteogenesis	[40,50,51]

PTM: Post-translational modification; PDLSCs: Periodontal ligament stem cells; DPSCs: Dental pulp stem cells; SCAPs: Stem cells from the apical papilla; m6A: N6-methyladenosine; JNK: C-Jun N-terminal kinase; GCN5: General control non-depressible 5; METTL3: Methyltransferase-like 3; MAPK: Mitogen-activated protein kinases; HAT: Histone acetyltransferase.

The oral microenvironment exerts dynamic control over the behavior of these stem cell types through a combination of biomechanical forces, inflammatory signals, hypoxic gradients, microbial metabolites, and nutrient availability. These factors form a highly specialized niche that can either promote tissue regeneration or trigger aberrant cellular responses leading to fibrosis, immunopathology, or even malignant transformation. Inflammation is now understood to be a central modifier of stem cell behavior in the periodontium. Exposure to pro-inflammatory cytokines, such as tumor necrosis factor (TNF)-α, interleukin (IL)-6, and IL-1β, has been shown to inhibit osteogenesis in PDLSCs while promoting senescence and apoptotic cascades, often via nuclear factor-kappa B (NF-κB) and mitogen-activated protein kinase (MAPK) signaling[17]. Interestingly, short-term or low-grade inflammation may transiently enhance stemness or differentiation in PDLSCs via PTM-sensitive pathways such as signal transducer and activator of transcription 3 (STAT3) phosphorylation or H3K27 acetylation[18].

Oxygen tension plays a critical role in the functional heterogeneity of DMSCs. Mild hypoxia (1%-5% O₂) has been reported to enhance the proliferation and stemness of DPSCs and SCAPs via stabilization of hypoxia-inducible factors (HIFs) and downstream modulation of glycolytic enzymes and angiogenic mediators. This hypoxia-responsive signaling is tightly associated with PTMs, particularly hydroxylation, ubiquitination, and acetylation of HIF-1α, which determines its transcriptional activity and proteasomal degradation[19,20]. Moreover, hypoxic preconditioning can epigenetically reprogram DPSCs through chromatin remodeling and increased expression of ten-eleven translocation enzymes, resulting in global DNA demethylation and activation of pro-regenerative gene networks[21].

These contextual signals not only affect lineage commitment and regenerative potential but also determine susceptibility to oncogenic reprogramming. Microenvironmental stressors, especially chronic inflammation and metabolic stress, can induce transcriptional noise and phenotypic drift in DMSCs, thereby facilitating dedifferentiation or transformation into a CSC-like phenotype. Importantly, recent advances have identified that PTMs serve as both mediators and sensors of these microenvironmental perturbations, integrating extracellular stimuli into intracellular signaling and chromatin states that ultimately shape cell fate trajectories[22,23].

OVERVIEW OF PTM CLASSES AND CORE ENZYMES (WRITERS/READERS/ERASERS)

To connect microenvironmental inputs with the downstream signaling programs, we first summarize the principal PTM classes and their core writer/reader/eraser enzymes. Post-translational and epitranscriptomic modifications function as rapid, reversible regulators that translate microenvironmental cues into lineage-defining programs in DMSCs. In this review we consider five widely studied classes- acetylation/crotonylation, phosphorylation, histone/DNA methylation, ubiquitination, and m6A RNA methylation-framed by their writer, reader, and eraser enzymes. Collectively, these modules interface with Wnt/β-catenin and bone morphogenetic protein (BMP)/SMAD axis and integrate MAPK [p38/c-Jun N-terminal kinase (JNK)] → RUNX2 in regeneration, whereas in OSCC/CSC they converge on Janus kinase (JAK)/STAT3, phosphatidylinositol 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of the rapamycin, and transforming growth factor-beta (TGF-β)/SMAD-driven EMT, shaping plasticity and therapy response (Figure 1). This overview sets up the subsequent sections on microenvironmental modulators and context-specific outcomes.

Figure 1 Post-translational modification classes and their associated pathways shaping dental-derived mesenchymal stem cell outcomes in periodontal regeneration vs oral squamous cell carcinoma/cancer stem cell. Regeneration: Acetylation/crotonylation, phosphorylation, trimethylation, ubiquitination, and N6-methyladenosine act on representative pathways including Wnt/β-catenin, bone morphogenetic protein/SMAD1/5/8, mitogen-activated protein kinase (p38/c-Jun N-terminal kinase) → runt-related transcription factor 2, Notch/Hedgehog, signal transducer and activator of transcription 3 (pro-repair), and nuclear factor-kappa B (resolution) to bias lineage commitment and tissue integration. Oral squamous cell carcinoma/cancer stem cell: Post-translational modification (PTM)-driven rewiring converges on Janus kinase/signal transducer and activator of transcription 3, phosphatidylinositol 3-kinase/protein kinase B/mammalian target of the rapamycin, transforming growth factor-β/SMAD2/3 → epithelial-mesenchymal transition, Hippo/YAP-TAZ, nuclear factor-kappa B, and β-catenin activation, promoting plasticity, immune evasion, and therapy resistance. Straight connectors indicate regulatory influence (not arrows); colored groupings label PTM classes; listed pathways are representative rather than exhaustive. This schematic complements the main text by summarizing how PTM categories interface with core signaling networks under regenerative vs malignant contexts. GCN5: General control non-depressible 5; H3K9ac: Histone H3 lysine-9 acetylation; Kcr: Lysine crotonylation; STAT3: Signal transducer and activator of transcription 3; OSCC: Oral squamous cell carcinoma; CSC: Cancer stem cell; EZH2: Enhancer of zeste homolog 2; H3K27me3: Histone H3 lysine-27 trimethylation; Skp2: S-phase kinase-associated protein 2; K48: Lysine-48-linked polyubiquitin; Ub: Ubiquitin; METTL3: Methyltransferase-like 3; m6A: N6-methyladenosine; UBA1: Ubiquitin-activating enzyme E1; UBE2T/UBE2O: E2 ubiquitin-conjugating enzymes; EMT: Epithelial-mesenchymal transition; E1/E2/E3: Ubiquitin-activating/-conjugating/-ligase enzymes.

INFLAMMATORY & MICROENVIRONMENTAL MODULATION OF PTMS

Inflammation

The oral microenvironment is frequently subject to inflammatory insults, ranging from acute periodontal inflammation to chronic systemic conditions such as diabetes mellitus. These inflammatory cues are potent regulators of PTMs, thereby influencing stem cell behavior at both transcriptional and epigenetic levels. Among DMSCs, PDLSCs are particularly sensitive to the inflammatory milieu, which alters histone acetylation, protein phosphorylation, ubiquitin tagging, and m6A RNA methylation in a context-dependent manner.

Proinflammatory cytokines, such as TNF-α, IL-1β, and interferon-γ, suppress the osteogenic differentiation of PDLSCs by modulating histone acetylation levels through the inhibition of histone acetyltransferases (HATs), including GCN5 and p300. In inflamed periodontal tissues, downregulation of GCN5 leads to decreased acetylation of histone H3K9 and H3K14 at the DKK1 promoter, resulting in overexpression of this Wnt antagonist and subsequent repression of osteogenic Wnt/β-catenin signaling[7,24]. Similarly, p300 deficiency has been shown to impair chromatin accessibility of RUNX2 and Osterix (OSX) promoters, further hindering osteogenesis in DPSCs under inflammatory stress[25].

Phosphorylation cascades activated by MAPKs, especially p38 and JNK, also contribute to inflammation-driven stem cell fate shifts. For example, p38-mediated phosphorylation of Smad1/5 suppresses BMP2-induced osteogenesis, while JNK activation impairs TGF-β-mediated odontoblastic differentiation via inhibitory phosphorylation of Smad2/3[26]. These phosphorylation-dependent effects are reversible, and inhibition of upstream kinases can restore osteogenic potential, highlighting the plastic nature of these PTMs.

Ubiquitination and deubiquitination represent another crucial axis in regulating the inflammatory response in DMSCs. Current evidence suggests that E3 ligases such as Smurf1, WWP1, and ITCH are upregulated under inflammatory conditions and target key osteogenic transcription factors, including RUNX2 and DLX5, for proteasomal degradation[27]. Smurf1 is specifically activated in response to proinflammatory signaling and catalyzes K48-linked ubiquitination of RUNX2, thereby antagonizing osteoblast differentiation in PDLSCs[28]. Conversely, deubiquitinating enzymes such as ubiquitin-specific protease 9X and ubiquitin-specific protease 7 are widely thought to counteract this degradation, although their exact role in oral stem cell biology warrants further research.

RNA m6A modification has emerged as a critical epitranscriptomic mechanism linking microenvironmental stress with gene expression dynamics. Inflammatory stimuli have been reported to elevate METTL3 expression in PDLSCs and DPSCs, increasing m6A deposition on transcripts such as ETS1, DLX3, and TGF-β receptor type-1[2,29]. These modifications enhance transcript stability and translation efficiency through m6A readers such as IGF2BP1/2, promoting pro-regenerative phenotypes under transient stress. However, chronic inflammatory conditions may induce pathological m6A methylation patterns that drive maladaptive cellular responses, such as senescence or dedifferentiation, suggesting a biphasic role that depends on context and duration.

Mechanical stimuli

Mechanical stimuli represent another key modulator of PTMs in the periodontal microenvironment. Physiological tensile strain promotes osteogenic differentiation in PDLSCs by activating histone deacetylase (HDAC) inhibition, H3 acetylation, and Wnt signaling, while excessive or pathological loading triggers stress kinases and shifts PTM profiles toward a catabolic state[30]. Integrin-mediated mechanotransduction leads to the phosphorylation of FAK and YAP, resulting in subsequent chromatin remodeling through the acetylation and methylation of key gene loci involved in cytoskeletal organization and lineage specification[31].

Hypoxia and metabolism

Hypoxia and metabolic signals further modulate PTMs through oxygen-sensitive enzymes and metabolite-dependent cofactors. For instance, hypoxia reduces the activity of prolyl hydroxylases, stabilizing HIF-1α, which undergoes PTM crosstalk via acetylation (p300), ubiquitination (VHL), and sumoylation (PIASy), thereby fine-tuning angiogenic and glycolytic gene expression[20]. In the meantime, metabolic intermediates such as acetyl-CoA, SAM, α-ketoglutarate, and NAD+ serve as cofactors or inhibitors for acetyltransferases, methyltransferases, and deacetylases, suggesting that metabolic rewiring in inflamed or ischemic niches directly influences the PTM machinery regulating stem cell plasticity[32].

Microbial factors

Finally, microbial components in the oral cavity, including lipopolysaccharide (LPS), butyrate, and short-chain fatty acids, have been shown to directly alter PTM landscapes. LPS exposure increases HDAC1 expression in DPSCs and reduces global histone acetylation, leading to decreased expression of odontogenic markers[33]. Butyrate, paradoxically, functions as a HDAC inhibitor and may transiently promote stemness-associated genes; however, chronic exposure impairs differentiation and increases DNA damage via altered chromatin organization[34]. Collectively, these findings underscore the tightly interwoven relationship between the oral microenvironment and PTMs in shaping stem cell fate. Inflammation, hypoxia, mechanical strain, metabolism, and microbial factors all converge on PTM-mediated regulatory networks, dictating whether stem cells contribute to regeneration or pathological transformation.

PTM-MEDIATED REGULATION IN PERIODONTAL/DENTAL REGENERATION

The regenerative capacity of DMSCs is tightly orchestrated by temporally and spatially coordinated PTMs. These dynamic chemical modifications modulate protein function, stability, localization, and interaction potential, thereby integrating microenvironmental signals into stem cell fate decisions critical for osteogenesis, odontogenesis, and vascularization. Accumulating evidence suggests that PTMs form multilayered regulatory networks that are not only lineage-specific but also responsive to inflammatory, mechanical, hypoxic, metabolic, and microbial niche perturbations[35].

Role of acetylation in dental regeneration

Among the acetylation-related regulatory axes, the HAT GCN5 plays a crucial role in periodontal tissue regeneration by enhancing the transcriptional accessibility of osteogenic loci. In inflamed microenvironments, GCN5 expression and activity are suppressed, resulting in reduced acetylation of H3K9 and H3K14 at promoters such as DKK1, thereby derepressing this Wnt antagonist and impairing Wnt/β-catenin signaling in PDLSCs[7]. Functional rescue using aspirin, which enhances GCN5 activity, restores histone acetylation and osteogenic gene expression (RUNX2, osteocalcin, alkaline phosphatase), improving alveolar bone regeneration in vivo. Notably, chromatin immunoprecipitation (ChIP) quantitative polymerase chain reaction confirmed histone modification changes at osteogenic gene loci, while liquid chromatography-tandem mass spectrometry (LC-MS/MS) proteomic analysis demonstrated global shifts in H3K9ac and H3K14ac levels following aspirin treatment[24].

Role of ubiquitination in dental regeneration

Ubiquitination represents another fundamental PTM significantly influencing stem cell commitment. Current evidence suggests that Smurf1, an E3 ubiquitin ligase, targets RUNX2 and DLX5 for K48-linked polyubiquitination and subsequent degradation, particularly under conditions of inflammatory or mechanical stress. Mechanical loading experiments in PDLSCs have demonstrated that Smurf1 expression increases under 10%-15% cyclic tensile strain, leading to a reduction in RUNX2 protein half-life, as confirmed by cycloheximide chase assays and immunoprecipitation-based ubiquitin profiling[36]. Conversely, CRISPR-mediated knockout of Smurf1 resulted in sustained osteogenic differentiation and matrix mineralization under mechanical load, thereby supporting a negative regulatory role for Smurf1[37]. Furthermore, Smurf1 regulates the degradation of HDAC1, indirectly enhancing global histone H3 acetylation and establishing a PTM interplay loop between ubiquitination and acetylation[38]. In addition, recent evidence indicates that palmitoylation, another PTM catalyzed by DHHC-type acyltransferases, participates in the regulation of DPSC fate. Specifically, inhibition of zinc finger DHHC-type palmitoyltransferase 16 (ZDHHC16), a palmitoyltransferase, promotes the osteogenic differentiation of DPSCs and reduces ferroptosis by activating the cyclic adenosine monophosphate response element-binding protein (CREB) pathway. ZDHHC16 suppresses p-CREB expression, thereby limiting the expression of osteogenic genes and compromising cell survival, while CREB activation can counteract these inhibitory effects, highlighting a critical ZDHHC16-CREB regulatory axis in osteogenic commitment and resistance to ferroptosis[39].

Role of m6A RNA methylation in dental regeneration

Epitranscriptomic regulation via m6A introduces an additional post-transcriptional layer of control over stem cell fate. METTL3, the core methyltransferase responsible for m6A deposition, modulates odontogenic differentiation by m6A methylation of transcripts encoding proteins, including DLX3, ACLY, and vascular endothelial growth factor A (VEGFA). In SCAPs, METTL3 overexpression increases pre-miR-665 methylation, enhancing DLX3 expression through a METTL3/miR-665 axis. Dual-luciferase reporter assays and methylated RNA immunoprecipitation (MeRIP)-quantitative polymerase chain reaction confirmed direct methylation sites[40]. Similarly, in DPSCs, the METTL3-IGF2BP2-mediated stabilization of ACLY mRNA supports histone acetylation by increasing acetyl-CoA availability, thus coupling metabolism with chromatin regulation[41]. These findings highlight the intricate interplay between RNA methylation, metabolic signaling, and transcriptional control in the context of dental regeneration.

Role of DNA/histone methylation in dental regeneration

Both DNA and histone methylation are now understood to contribute to epigenetic lineage priming. Inflammation-induced upregulation of DNA methyltransferases (DNMT)1 and DNMT3B leads to hypermethylation of RUNX2, DSPP, and ALPL promoters, reducing the accessibility and transcription of these genes. Whole-genome bisulfite sequencing performed on inflamed DPSCs has identified hypermethylated CpG islands in key osteogenic gene promoters, while 5-azacytidine treatment reversed these methylation patterns and restored mineral deposition[42]. Complementarily, enhancer of zeste homolog 2 (EZH2)-mediated H3K27me3 accumulation under chronic inflammation was found to repress OSX and OCN, an effect reversed by KDM6A overexpression or EZH2 knockdown[43]. Besides, CUT&RUN profiling further revealed bivalent chromatin states at various lineage-specific loci, characterized by the co-enrichment of H3K27me3 and H3K4me3, indicative of poised differentiation states[44].

Role of phosphorylation in dental regeneration

Phosphorylation cascades transmit microenvironmental inputs to lineage regulators. For instance, BMP2 signaling activates p38 MAPK, which subsequently phosphorylates Smad1/5, enhancing their transcriptional activity in DPSCs. Co-immunoprecipitation (Co-IP) and phospho-specific western blotting validated formation of a p38-Smad1 complex during odontogenic induction[44]. JNK phosphorylation of RUNX2 at Ser319 enhances DNA binding and promotes collagen type 1 alpha 1 transcription; inhibition of JNK blocks mineralization even in the presence of osteoinductive stimuli[45]. These studies overlap in their assertion that phosphorylation acts not only as a rapid-response modulator but also as a gatekeeper for chromatin engagement by transcription factors. The intricate crosstalk among different PTM types enables a combinatorial regulatory encoding of regenerative decisions. For example, the m6A-mediated upregulation of ACLY increases intracellular acetyl-CoA levels, which enhances histone acetylation by increasing p300/GCN5 activity, thereby forming a positive feedback loop between RNA methylation and chromatin remodeling[46]. Besides, the Smurf1-mediated degradation of HDAC1 results in increased histone acetylation at RUNX2 loci, which in turn activates Wnt target genes, exemplifying a complex ubiquitin-acetylation-transcription cascade[38]. These synergistic PTM networks have been rigorously validated by integrating multi-omics datasets that combine RNA-sequencing, ChIP-sequencing (ChIP-seq), and MeRIP-sequencing (MeRIP-seq), across time points during cellular differentiation[21].

Role of histone lysine lactylation in dental regeneration

Histone lysine lactylation (Kla) emerges as a glycolysis-linked acylation that directly activates repair and developmental gene programs. Elevated lactate in hypoxic or inflamed oral niches can drive Kla at enhancers, coordinating chromatin accessibility with lineage trajectories relevant to craniofacial development and wound resolution[47]. Developmental systems demonstrate that Kla marks tissue-specific enhancers and couples metabolic state with deployment of gene regulatory networks; analogous conditions are expected in periodontal microenvironments where DMSC fate decisions intersect with immune remodeling and matrix dynamics[48]. Recognizing lactylation alongside acetylation and phosphorylation refines the metabolic-epigenetic model governing regenerative success vs malignant drift.

Subtype-specific PTM responses in DMSCs

Notably, PTM responses vary significantly among different stem cell subtypes. PDLSCs, when subjected to inflammatory cytokines or LPS, primarily exhibit dominant Smurf1-RUNX2-BMP2 suppression and HDAC1 recruitment, whereas DPSCs tend to activate STAT3-p300 circuits, enhancing chromatin accessibility at odontoblastic loci[49]. In contrast, SCAPs demonstrate pronounced m6A sensitivity, with METTL3-dependent stabilization of angiogenic transcripts (VEGFA, angiopoietin-1) and odontogenic effectors (DLX3, SP7), which can be precisely traced via co-profiling with single-cell RNA sequencing and ATAC-sequencing (ATAC-seq)[50]. In a recent single-cell time-series study, SCAPs exhibited a biphasic PTM landscape where early m6A gains were followed by delayed H3K9ac increases, suggesting temporal layering of PTMs during lineage progression[51].

ChIP-seq has enabled high-resolution localization of histone PTMs such as H3K9ac and H3K27me3 at osteogenic gene loci under various niche conditions[7]. Proteomic analyses via tandem mass tag (TMT)-based LC-MS/MS have quantified dynamic PTM shifts (acetylation, ubiquitination) in response to osteoinductive vs inflammatory stimuli[52]. Regarding m6A mapping, techniques including MeRIP-seq and SCARLET have enabled the identification of context-specific methylation sites relevant to cellular differentiation. Functional perturbation studies using CRISPR-Cas9-mediated knockout (e.g., METTL3, Smurf1, EZH2) and site-directed mutagenesis (e.g., RUNX2-S319A) have confirmed the causal roles of PTMs[53]. Moreover, protein-protein interaction studies (e.g., Co-IP, bimolecular fluorescence complementation) have substantiated that PTM writers/readers form dynamic complexes with transcription factors and chromatin remodelers during lineage commitment[54].

Biomaterial-guided PTM modulation and delivery systems

From a translational standpoint, biofunctionalized scaffolds incorporating PTM-targeting molecules (e.g., HDAC inhibitors, EZH2 inhibitors, or m6A modulators) have been developed to enhance in situ differentiation. For instance, polydopamine-modified scaffolds loaded with trichostatin A (a known HDAC inhibitor) successfully promoted H3K9 acetylation and bone formation in PDLSC-seeded defect models[55]. Similarly, exosome-based delivery systems carrying METTL3-overexpressing constructs have demonstrated enhanced pulp-like tissue regeneration and angiogenesis in vivo when introduced into DPSCs. These biomaterial-guided delivery systems offer spatial and temporal control over PTM-modulated differentiation in clinical contexts.

Within oral grafts, spatial-temporal PTM cues have been engineered using dental-relevant matrices. A chitosan/biphasic calcium phosphate scaffold locally delivering the HDAC inhibitor trichostatin A increased histone acetylation, upregulated osteogenic programs in human periodontal ligament cells, and accelerated calvarial repair, exemplifying matrix-bound PTM presentation[56]. A complementary temporal strategy uses MI192-preconditioned mesenchymal stromal cells encapsulated in GelMA and reinforced by a 3D-printed PEGT/PBT frame, which preserves viability and mechanics while enabling staged PTM control from priming to engraftment[57]. To approximate endogenous vesicle-borne modifiers, nano-hydroxyapatite/chitosan/PLGA constructs act as sustained exosome depots in mandibular defects, yielding durable release and robust bone fill in vivo[58]. Together, these systems operationalize graded or sequential PTM modulation-surface-bound HDAC inhibitor and pre-programmed cells, plus depot-mediated vesicular signaling-within clinically tractable materials for periodontal and craniofacial regeneration.

PTMs serve as multi-functional regulators in both dental and periodontal regeneration by encoding niche signals into gene regulatory programs through modular and dynamic interactions. Acetylation, ubiquitination, phosphorylation, methylation, and m6A methylation do not function in isolation; rather, they interact through feedback, convergence, and hierarchical coordination. These processes are typically context-dependent, cell-type specific, and temporally layered, accurately reflecting the biological complexity of stem cell plasticity. A comprehensive understanding of these PTM networks, and their integration into single-cell epigenomic and proteomic frameworks, will indeed open new avenues for developing targeted regenerative strategies and advanced bioengineered therapies tailored to specific tissue contexts.

FROM REGENERATION TO PATHOLOGICAL TRANSFORMATION

The precisely orchestrated sequence of PTMs that underlies DMSC-mediated regeneration can transition into a maladaptive code under chronic stress, driving phenotypes akin to CSCs. In healthy regenerative responses, PDLSCs, DPSCs, and SCAPs execute a PTM “program” arranged hierarchically and temporally. This program begins with early acetylation by HATs (GCN5/p300), which establishes chromatin permissiveness. Next, ubiquitination via Smurf1 ensures proper transcription factor turnover. Phosphorylation by MAPKs subsequently configures signaling kinetics, and m6A-mediated transcript stabilization guides differentiation commitment. However, dysregulation along this axis, often triggered by sustained inflammation, mechanical strain, metabolic derangement, or microbial persistence, can tip the balance toward oncogenic transformation[59].

A pivotal mechanism driving this shift involves chronic inflammation, wherein cytokines such as TNF-α, IL-6, and interferon-γ sustain the activity of E3 ubiquitin ligases, including Smurf1 and NEDD4 L, as well as DNMT1 and DNMT3B. This combination undermines proper differentiation and promotes dedifferentiation, ultimately leading to survival. Single-cell transcriptomic analysis of inflamed periodontal tissues has revealed distinct cell clusters characterized by co-upregulation of Smurf1 and phosphorylation-related signaling genes, accompanied by suppression of osteogenic markers[60]. Functionally, prolonged Smurf1 activity not only depletes key transcription factors, such as RUNX2, but also stabilizes HDAC1, which represses tumor suppressor loci via deacetylation, thereby reinforcing EMT and proliferation during early neoplastic events[61].

Ubiquitin signaling further contributes to pathological cellular transformation. While NEDD4 L initially confers a protective effect by degrading pro-apoptotic proteins, its target preference gradually shifts to key regulatory proteins, such as phosphatase and tensin homolog (PTEN) and TGF-β receptors. This transition activates PI3K/AKT signaling, facilitates EMT, and promotes dissemination phenotypes, as demonstrated in long-term LPS-stimulated SCAP cultures[44]. In parallel, Skp2 promotes degradation of cyclin-dependent kinase inhibitors p21 and p27, thereby lifting proliferative restraints and enabling CSC-like behavior. This effect is further reinforced by UBD, which stabilizes oncogenic proteins marked for degradation and enhances radioresistance[62].

m6A RNA methylation also plays a dual role in this context. While beneficial during early regeneration, chronic METTL3 upregulation in PDLSCs and SCAPs leads to the aberrant stabilization of transcripts involved in proliferation and self-renewal, including SALL4, SOX2, and TWIST1. Experimental data in OSCC models have further indicated that METTL3-dependent m6A enrichment on SLC7A11 and CD44 mRNAs contributes to enhanced stemness and therapy resistance. Pharmacological inhibition of METTL3 in precancerous epithelial cultures has been shown to reverse these changes and restore expression of differentiation-related genes, highlighting a potential therapeutic window[62].

FTO and ALKBH5, the two canonical m6A “erasers”, exert context-dependent control over cancer stem-like states. In head and neck squamous cell carcinoma, FTO upregulation increases CTNNB1 in an m6A-dependent manner, thereby reinforcing β-catenin signaling that is tightly coupled to self-renewal and migratory behavior[58]. Across hematologic and solid tumor models, genetic or pharmacologic FTO blockade reduces sphere-forming capacity, ALDH^high fractions, and immune evasion programs, indicating a causal role in CSC maintenance[63]. ALKBH5 likewise shapes stemness: In glioblastoma stem-like cells, ALKBH5 demethylates nascent FOXM1 transcripts to sustain tumorigenicity and self-renewal[64], while in non-small cell lung cancer it supports EMT-and stemness-linked programs through a p53-dependent axis, with ALKBH5 silencing downregulating NANOG/OCT4 and increasing E-cadherin[65]. Emerging head-and-neck evidence also connects ALKBH5 to therapy resistance and RNA-helicase-dependent regulation in OSCC cells, underscoring that eraser activity can reprogram plasticity within the oral microenvironment and should be carefully monitored when considering m6A-targeted interventions in regenerative settings (see also REF3/4 for mechanistic parallels). Collectively, these data support a working model in which m6A erasers dynamically tune DMSC/CSC states and may tip the balance between periodontal repair and malignant transformation.

Pathological shifts in phosphorylation cascades further potentiate malignancy. Under sustained inflammatory stimuli, MAPKs, such as p38 and JNK, shift from differentiation-supportive roles to pro-oncogenic functions through interactions with NF-κB, STAT3, and AP-1. For instance, JNK-mediated phosphorylation of STAT3 enhances its interaction with p300, increasing histone acetylation and the transcription of oncogenic genes[60]. In vitro models of chronic inflammatory exposure in DPSCs have shown persistent occupancy of STAT3-p300 complexes at promoters such as matrix metalloproteinase 9 (MMP9) and IL-8, thereby reinforcing invasive phenotypes.

Epigenetic reprogramming is now understood to be significantly driven by chromatin modifiers. In this respect, EZH2-mediated H3K27me3 accumulation suppresses osteogenic genes including OSX and RUNX2, then progressively targets loci such as CDKN2A and CDH1, promoting proliferation and EMT. CUT&RUN profiling has confirmed the redistribution of H3K27me3 from differentiation-related to proliferation-related genes as malignant transformation advances[66]. This epigenetic repression is further compounded by the DNA hypermethylation of tumor suppressor promoters, as mapped via whole-genome bisulfite sequencing, committing cells to dedifferentiation and malignant behavior[67].

The metabolic state of the microenvironment also influences PTM patterns. Increased acetyl-CoA levels, often driven by ACLY overexpression, facilitate hyperacetylation of oncogenic loci such as MYC and CCND1. Besides, fumarate and other metabolites inhibit ten-eleven translocation enzymes, thereby impeding DNA demethylation and solidifying pro-tumorigenic epigenotypes. Metabolomic profiling of dysplastic tissues has revealed a shift from anti-inflammatory intermediates such as itaconate to pro-oncogenic metabolites, including lactate and acetyl-CoA[68].

Mechanotransduction under pathological tension further exacerbates malignant transitions. Within inflamed niches, increased mechanical stress activates FAK-YAP signaling, leading to YAP/TAZ nuclear translocation and chromatin activation. Enhanced H3K9ac at oncogenic enhancers has been detected in mechanically stressed PDLSCs, linking mechanical cues with PTM-driven plasticity[37].

Integrative single-cell and spatial omics approaches have been instrumental in characterizing the dynamic remodeling of PTM landscapes. Regenerative modules, such as GCN5-acetylation and METTL3-mediated m6A modification, are now understood to be progressively replaced by ubiquitination and DNA hypermethylation signatures. These are accompanied by the emergence of inflammatory and EMT-associated phosphorylation clusters. Such findings validate the concept of a PTM continuum, where regenerative signals eventually reprogram into malignant codes. Tipping points are defined by cumulative PTM alterations, including persistent E3 ligase activity, irreversible DNA methylation, and the m6A stabilization of oncogenic transcripts[69].

From a translational perspective, the early detection of these inflection points in PTM profiles could inform timely therapeutic intervention. Restoring acetylation balance via HDAC inhibitors or HAT activators, modulating E3 ligases such as Smurf1 and NEDD4 L, inhibiting METTL3-mediated m6A methylation, or targeting JNK/STAT3 pathways offer viable strategies to prevent or reverse malignant transformation. Preclinical studies utilizing CRISPR-modified PDLSCs and nanoparticle-mediated delivery systems further support the feasibility of reversing early oncogenic traits while restoring regenerative potential[70]. Ultimately, the same PTMs that orchestrate regeneration may, when chronically dysregulated, rewire cellular programs toward malignancy. Decoding the temporal dynamics of PTM progression and defining actionable molecular thresholds may lay the foundation for regenerative-oncology paradigms that both preserve tissue integrity and forestall oncogenic drift.

During early periodontal repair, permissive acetylation events, driven chiefly by GCN5-mediated H3K9ac, prime osteogenic loci and cooperate with emerging crotonylation to enhance PI3K-AKT-dependent mineralization. However, as differentiation proceeds under chronic inflammatory and metabolic stress, the chromatin landscape gradually shifts: EZH2-catalyzed H3K27me3 accrues, repressing osteogenic programs while re-engaging Wnt/β-catenin and EMT cues. Parallel hyperactivation of the Skp2 ubiquitin ligase stabilizes pro-invasive effectors, and rising METTL3-dependent m6A deposition consolidates stemness and immune evasion. Collectively, these stepwise epigenetic changes trace a continuum from reversible regeneration to malignant potential, providing a conceptual bridge to the forthcoming discussion on OSCC CSC plasticity. Figure 2 illustrates PTMdriven progression, mapping each modification along a temporal axis while contrasting the corresponding regenerative or CSC phenotypes, thereby establishing the framework for the forthcoming discussion of OSCC CSC plasticity.

Figure 2 Sequential post-translational modification cascade driving the shift from regeneration to malignancy in dental-derived mesenchymal stem cells. A vertical timeline (centre) depicts six key post-translational modifications (PTMs) that accumulate in a fixed order: General control non-depressible 5-mediated histone-H3 lysine-9 acetylation, lysine crotonylation, signal transducer and activator of transcription 3 tyrosine-705 phosphorylation, enhancer of zeste homolog 2-catalysed histone-H3 lysine-27 trimethylation, S-phase kinase-associated protein 2-directed K48-linked ubiquitination, and methyltransferase-like 3-dependent N6-methyladenosine RNA methylation. Left-hand annotations summarize regenerative outcomes of each modification (e.g., osteogenic switch-on, extracellular-matrix mineral build-up), whereas right-hand annotations indicate the corresponding cancer stem cell or malignant traits that emerge as the same PTMs intensify (e.g., invasion programme silent → epithelial-mesenchymal transition onset, polycomb repression of bone genes, immune-escape programme locked). The downward arrow signifies temporal progression: Early, reversible PTMs support tissue repair; late, irreversible PTMs consolidate oral squamous cell carcinoma cancer stem cell phenotypes. CSC: Cancer stem cell; GCN5: General control non-depressible 5; H3K9ac: Histone H3 lysine-9 acetylation; ECM: Extracellular matrix; STAT3: Signal transducer and activator of transcription 3; EMT: Epithelial-mesenchymal transition; EZH2: Enhancer of zeste homolog 2; H3K27me3: Histone H3 lysine-27 trimethylation; Skp2: S-phase kinase-associated protein 2; METTL3: Methyltransferase-like 3; m6A: N6-methyladenosine.

PTMS IN OSCC CSC PLASTICITY & THERAPY RESISTANCE

CSCs in OSCC are characterized by their abilities for self-renewal, metastasis, and resistance to therapy. Mounting evidence implicates PTMs as critical regulators of CSC plasticity and therapeutic failure. These PTMs modulate CSC phenotypes by reprogramming protein stability, epigenetic landscapes, signal transduction, and metabolic pathways[59,61,71]. Table 2 compiles the principal PTMs, regulatory enzymes and functional consequences that define malignancy-associated DMSC populations discussed in this section.

Table 2 Post-translational modification signatures in malignancy-associated dental stem-cell populations.

Stem-cell type	Tissue source	Dominant PTMs reported	Key regulatory enzymes	Regenerative potency/tumorigenic capacity	Ref.
OCSCs (CD44+/ALDH1+)	Primary OSCC tumour tissues & spheres	m6A RNA methylation; H3K27 trimethylation; K63/K48-linked ubiquitination; STAT3 Tyr705 phosphorylation	METTL3; EZH2; Skp2; STAT3	High tumour-initiation, chemoresistance, EMT plasticity	[59,76,82-84]
Inflammation-transformed PDLSCs	Chronic periodontitis-affected periodontal ligament	H3K27 trimethylation; H3K9 acetylation imbalance; Smurf1-mediated ubiquitination	EZH2; GCN5; Smurf1	Diminished osteogenic potential; pro-malignant phenotypes	[27,36,41,105]

PTM: Post-translational modification; OCSCs: Oral cancer stem cells; ALDH: Aldehyde dehydrogenase; PDLSCs: Periodontal ligament stem cells; OSCC: Oral squamous cell carcinoma; K63/K48: Lysine-63/-48-linked polyubiquitin; STAT3: Signal transducer and activator of transcription 3; METTL3: Methyltransferase-like 3; EZH2: Enhancer of zeste homolog 2; Skp2: S-phase kinase-associated protein 2; GCN5: General control non-depressible 5; EMT: Epithelial-mesenchymal transition.

Ubiquitination-driven CSC maintenance and resistance

Ubiquitin E3 ligases, such as NEDD4 L and Skp2, are significantly overexpressed in OSCC CSC populations and contribute to tumor aggressiveness by degrading tumor suppressors and maintaining proliferative signals. NEDD4 L was found to specifically target PTEN and TGF-β receptors for ubiquitination, thereby enhancing PI3K/AKT pathway activity and triggering EMT and cisplatin resistance. Loss of NEDD4 L markedly decreased CD44+/ALDH+ CSC fractions and restored chemosensitivity in vitro and in vivo[61]. Skp2 similarly promoted cell cycle progression by ubiquitinating p27 and p21. Skp2 inhibition suppressed CSC features, attenuated metabolic reprogramming, and diminished radioresistance in cancer models, attributing effects to altered AKT/Wee1/CDK1 signaling and increased survivin ubiquitination by FBXL7. Moreover, the ubiquitin-binding protein UBD (FAT10) potentiates the stability of Skp2 and survivin, further enhancing chemoresistance and self-renewal capabilities[72].

Ubiquitin-activating enzyme E1 (UBA1) constrains ubiquitin flux at the apex of the cascade; in immuno-oncology models, a UBA1-STIP1 homology and U-box-containing protein 1 signaling axis depresses intratumoral cytotoxic T-cell activity and associates with resistance to immune checkpoint blockade, whereas UBA1 inhibition stabilizes JAK1 and augments antitumor responses[12]. These observations argue that UBA1 copy-number gain or overexpression may be informative for stratification in lesions where stemness and immune evasion converge. Pharmacologic tractability at the E1 node is further supported by the first-in-class small molecule TAK-243, which selectively engages UBA1, reduces global ubiquitylation, impairs leukemic stem cell fitness, and shows in vivo activity-establishing a framework for dose scheduling and resistance surveillance[73]. That also implicates E2 enzymes as functional drivers. In oral epithelial malignancy models, UBE2T enhances motility and EMT through induction of IL-6 and activation of the IL-6/JAK/STAT3 axis, mechanistically linking inflammatory cues to plasticity and invasion[74]. In head and neck squamous cell carcinoma, UBE2O potentiates EMT, invasion and tumor growth in vitro and in vivo, indicating that E2 dysregulation cooperates with E3 circuitry to stabilize pro-oncogenic states[75]. Incorporating E1/E2 alongside E3 therefore widens the actionable continuum: Profiling E1/E2 expression or copy-number to inform risk, and selectively modulating these upstream steps-alone or with E3-targeting or PTM-rebalancing strategies-to blunt EMT/CSC programs while preserving regenerative capacity.

Notably, ubiquitin-mediated regulation exhibits extensive interaction with phosphorylation and acetylation networks. The ubiquitination of signal transduction proteins amplifies signaling loops that reinforce CSC traits. Quantitative ubiquitin proteomic profiling (TMT-label LC-MS/MS) in OSCC CSC subpopulations revealed global increases in ubiquitination of PTEN and SMAD4. Co-IP and proximity ligation assays confirmed direct interactions between NEDD4 L and AKT pathway components. CRISPR-Cas9-mediated knockout of NEDD4 L and Skp2 in CSC-enriched OSCC spheres reduced sphere formation, downregulated MSC markers (CD44, SOX2), and enhanced therapeutic susceptibility[61].

RNA m6A methylation reprogramming stemness

An m6A-dependent mechanism has been shown to upregulate SLC7A11 and CD44 in OSCC, mediated by METTL3, which enhances radio-resistance and stemness. In xenograft models, the knockdown of METTL3, combined with irradiation, significantly reduced tumor volume and CSC marker expression[61]. Further research revealed a METTL3-miR-146a-5p-SMAD4 axis, where METTL3-mediated methylation promoted pre-miR-146a processing. The mature miR-146a suppressed SMAD4, thereby disrupting TGF-β signaling, and triggering EMT and increased invasion in OSCC cells[76].

Mechanistically, m6A modifications are interpreted by reader proteins YTHDF1/2, which recruit translation initiation factors, thereby enhancing the expression of stemness and EMT regulators, such as TWIST1 and SOX2[77]. MeRIP-seq analyses of OSCC CSCs revealed thousands of differentially methylated transcripts enriched in cell-cycle control, antioxidant defense, and stem-cell maintenance; among the antioxidant genes, SLC7A11 emerged as a prominent METTL3-m6A target. Function-blocking of YTHDF1/2 via shRNA attenuated CSC marker expression and restored chemosensitivity in 3D collagen sphere assays[59].

5-methylcytosine (m5C) on mRNA stabilizes pro-oncogenic transcripts through the RNA methyltransferase NSUN2 and cognate readers. In head-and-neck squamous cell carcinoma, NSUN2-dependent m5C of LAMC2 enhances focal adhesion/FAK-SRC signaling and epithelial-mesenchymal dynamics, aligning with cancer stem-like programs observed in OSCC contexts[78]. m5C readers such as YBX1 further reinforce transcript stability and translation, providing a mechanistic route by which metabolic and inflammatory cues in the oral niche can tune stemness, invasion, and therapy response[79]. Incorporating m5C into the PTM-epitranscriptomic axis therefore links DMSC plasticity under regenerative conditions with malignant reprogramming when checkpoints fail.

Phosphorylation-driven cross-talks

Phosphorylation pathways, mediated by kinases such as JNK, p38, and STAT3, play pivotal roles in transferring extracellular stress signals into CSC transcriptional programs. A STAT3-p300 acetylation axis has been identified, in which JNK-phosphorylated STAT3 recruits p300 to tumor sphere cells under inflammatory conditions, resulting in increased H3K27 acetylation at the promoters of MMP9 and IL-8, and consequently promoting invasion and stemness. Epidermal growth factor receptor-triggered phosphorylation of AKT synergizes with Skp2 to degrade cell cycle inhibitors and boost survival signaling[80].

Proteomic phospho-profiling studies have shown elevated phosphorylation of STAT3 (Tyr705), AKT (Ser473), and epidermal growth factor receptor (Tyr1068) in OSCC CSCs, corresponding with upregulated ubiquitination signals and histone acetylation at promoters of proliferation and survival genes[81]. JAK/STAT3 inhibitors, in combination with Skp2 blockade, have shown synergistic inhibition of CSC traits in vitro and enhanced therapeutic responses in preclinical models[71].

Acetylation and histone modifications in tumor evolution

Histone acetylation orchestrates chromatin accessibility and transcription factor recruitment. In OSCC CSCs, increased acetylation of histones H3K27 and H3K9, mediated by p300 and restrained by HDAC1, interacts with transcriptional circuits promoting malignancy. For example, STAT3-p300 activity increases H3K27ac at the promoters of stem cell genes. Inhibitor studies have shown that combining HDAC inhibitors (e.g., vorinostat) with METTL3 knockdown attenuates CSC markers (CD44, SOX2) and tumorigenicity in OSCC xenografts[82]. Interestingly, m6A methylation and histone acetylation are widely thought to exhibit a synergetic relationship: METTL3 knockdown reduces expression of p300 and GCN5 HATs, leading to decreased global acetylation levels[83].

DNA and histone methylation reprogram CSC epigenome

Overexpression of EZH2, the H3K27 methyltransferase, is a defining feature in aggressive OSCC. EZH2 enhances H3K27me3 to silence differentiation genes, including E-cadherin (CDH1), CDKN2A (p16), and other tumor suppressors, sustaining dedifferentiation[84]. DNMT1-driven promoter hypermethylation in CDH1 and p16 reinforces this phenotype[85]. ChIP-seq revealed EZH2/H3K27me3 enrichment at these loci in CD44+ CSC fractions, while DNMT inhibition (5-azacytidine) or EZH2 knockdown restored expression and reduced CSC abundance and tumor growth[84,85].

PTM synergy and resistance loops

Current evidence suggests that PTMs interact in co-dependent regulatory loops. For instance, Skp2-mediated ubiquitination activates the AKT pathway, which phosphorylates and stabilizes EZH2, enhancing methylation of differentiation genes. Concurrently, METTL3-dependent m6A stabilization is required for expression of Skp2 and EZH2, creating self-reinforcing loops. Acetylation and phosphorylation act additively to drive transcriptional circuits favoring CSC survival. Systems biology modeling based on multi-omic datasets predicts that simultaneous targeting of METTL3, Skp2, and EZH2 is necessary to disrupt the CSC network integrity[86].

Translational implications and targeted therapies

PTM enzymes are emerging as therapeutic targets in cancer. METTL3 inhibitors, such as STM2457, have entered preclinical trials, demonstrating tumor regression and a reduction in CSC burden[87]. Skp2 inhibitors (e.g., C1) similarly sensitize OSCC to radiotherapy and reduce metastasis in xenograft models[72]. The EZH2 inhibitor tazemetostat has demonstrated efficacy in reversing lymphoma traits and enhancing immune checkpoint blockade efficacy[88]. STAT3 blockade combined with HDAC inhibition was found to deplete CSC markers and restore chemosensitivity[71]. Notably, dual inhibition of METTL3 and Skp2 synergistically reduced sphere formation, promoted differentiation, and enhanced apoptosis in CSC-enriched OSCC models[89]. PTM enzyme perturbations have been validated using CRISPR-Cas9, siRNA/shRNA, small-molecule inhibitors, and proteolysis-targeting chimeras (PROTACs). PTM interaction assays, including Co-IP, ChIP-seq, MeRIP-seq, proteomics, metabolomics, and single-cell multi-omic approaches, can be leveraged to profile CSC and non-CSC subpopulations pre- and post-treatment.

Transition to single-cell and spatial profiling: Toward resolution of PTM heterogeneity

Despite accumulating evidence that PTMs orchestrate CSC plasticity and therapeutic resistance in OSCC, current understanding remains hindered by the limitations of bulk-tissue analyses, which obscure intratumoral heterogeneity and dynamic epigenetic reprogramming. CSC-associated PTM signatures, including METTL3-mediated m6A patterns, Skp2-dependent ubiquitin substrates, and EZH2/H3K27me3-modulated chromatin landscapes, are likely to vary significantly across spatial compartments and cell lineages within the tumor microenvironment[86-89]. Moreover, treatment-induced reprogramming, such as the emergence of therapy-resistant CSC subclones, may involve transient and spatially restricted PTM cascades that remain undetectable by conventional bulk omics.

To fully resolve these complexities, there is a growing need to integrate single-cell, spatially resolved, and multi-modal omics technologies. Such approaches can map PTM landscapes at cellular resolution, reveal lineage trajectories, and identify regulatory hubs that control transitions between regenerative, pre-malignant, and CSC states. In the following section, we will examine recent innovations in spatial epigenomics, single-cell phosphor-/acetyl-/ubiquitin-proteomics, and emerging integrative frameworks capable of decoding the combinatorial PTM codes governing oral tissue homeostasis and its transformation into malignancy.

INTEGRATED SINGLE-CELL & SPATIAL MULTI-OMICS OF PTMS

The promise of decoding PTM landscapes at both cellular and spatial resolutions has catalyzed significant advances in regenerative dentistry and oral oncology. Traditional bulk profiling obscures the critical heterogeneity inherent in the oral microenvironment, especially at the intersection between regenerative niches and malignant transformation foci. A robust solution to this limitation has emerged through the development of integrated singlecell and spatial multiomic technologies designed to measure PTMs in situ and dynamically[90].

Single-cell advancements have been spearheaded by high-throughput proteomic workflows like SCoPE2 and timsTOF PASEF-enhanced mass spectrometry, which quantify PTM fractions in individual cells. The application of these platforms to human tumors revealed STAT3-pY705 and H3K27ac high subpopulations within OSCC, consistent with CD44+ CSC clusters[91]. Single-cell MeRIP-seq analysis of OSCC organoids has revealed progenitor-like clusters enriched for METTL3-mediated m6A on SLC7A11 and CD44, as well as transitional cell populations displaying increased EZH2-mediated H3K27me3 levels, highlighting key PTM shifts during cellular transformation[59]. Importantly, temporal profiling in PDLSCs undergoing periodontal regeneration detected early STAT3-phosphorylation and H3K9ac increases at osteogenic gene loci, followed by a second wave of EZH2/H3K27me3 during resolution, reinforcing the concept of staged PTM encoding[92].

Spatial technologies have been established as crucial complements to single-cell data by preserving tissue architecture. One study effectively utilized multiplex imaging to spatially resolve STAT3-pY705, p300-mediated H3K27ac, and YAP acetylation in OSCC sections, revealing enriching gradients of activated PTM signals at invasive fronts near programmed death-ligand 1 positive immune cells[93]. Another investigation adapted CUT&Tag for spatial histone PTM mapping, uncovering interweaving domains of H3K9ac and H3K27me3 in regenerating mouse periodontal tissue that coincided with zones of active PDLSCs. In contrast, early OSCC lesions displayed discrete clusters of H3K27me3-rich cells adjacent to invasive margins[94]. A recent study further showcased the application of spatial phosphoproteomics by imaging phosphorylated AKT, STAT3, and YAP in situ, thereby defining stem-cell niche-like regions in prostate cancer[95].

While one study merged scRNA-seq and single-cell ATAC-seq in OSCC to reconstruct differentiation trajectories, even deeper insights emerge when PTM layers are incorporated. Using MOFA+ for integrative modeling, researchers traced bifurcating trajectories in PDLSC regeneration, where early nodes displayed METTL3-m6A and STAT3 phosphorylation, diverging into either a mineralizing regenerative pathway or a malignant-prone lineage with PTM signatures including Skp2-ubiquitination and EZH2/H3K27me3 accrual[96]. This mapping has been further validated in patient-derived xenografts and organoids with lineage barcoding, showing that cells exhibiting a combined METTL3-high/SKP2-high/EZH2-high signature more frequently seed metastases[97].

The technical landscape for analyzing post-transcriptional modifications continues to evolve rapidly. Enhanced sensitivity techniques now achieve sub-femtomole levels, enabling quantitative analysis from single-cell lysates. An ultra-sensitive MS workflow has been introduced to capture mismatched PTM forms, such as acetylated vs phosphorylated peptides, thereby boosting detection in rare CSC populations[98]. Antibody panels for multiplex imaging are continually refined to increase PTM specificity, dissecting modifications like H3K9ac from H3K23ac via dual-validation strategies combining epitope competition and mass tag confirmation[99]. Standardization efforts, including fixation protocols designed to preserve post-fixation PTM signals, have been codified in platforms like SPaRTAN[100] to ensure robust spatial assessments.

Over the years, analytical frameworks have substantially matured to support multi-layer data integration. Tools, including MOFA+, symphony, and linked matrix factorization approaches, enable the effective integration of scRNA-seq, ATAC-seq, proteomics, and spatial-PTM imaging modalities, facilitating the extraction of latent factors corresponding to shared biological axes, such as stemness or EMT. Besides, sophisticated data harmonization pipelines address batch effects, while lineage-plasticity inference methods trace PTM-defined bone and dental-related cell states that would elude capture by transcriptomics alone[101].

The synergistic application of these methodologies provides actionable insights for clinical translation. Composite PTM signatures, such as METTL3-high/H3K27ac-high/pSTAT3-high, demonstrate significant predictive value for poor prognosis and CRT resistance in OSCC cohorts, exceeding performance of genomic or transcriptomic markers alone[102]. In periodontal regenerative research, spatially resolved PTM profiling has been harnessed to guide scaffold placement, targeting areas of active regeneration marked by H3K9ac and low EZH2/H3K27me3, with biomaterials engineered for temporal release of HDAC inhibitors or METTL3 modulators[103]. Longitudinal spatial PTM monitoring after therapy can enable the early detection of reemergent therapy-resistant CSC clones, thereby facilitating adaptive treatment strategies.

Emerging innovations in PTM analysis include live-cell PTM biosensors designed to track PTM dynamics in real time. For instance, FRET-based reporters that detect acetylation of H3 tail peptides illuminate temporal activation during regeneration and revert before malignant reactivation[104]. Besides, multi-epitope mass cytometry integrates isotopic labels for ubiquitin, phosphorylation, and acetylation measurements in single cells to define combinatorial PTM states during CSC development. Artificial intelligence-driven frameworks such as PTMscape leverage deep learning to predict functional outcomes from PTM profiles in diverse cellular contexts, enabling in silico PTM-target prioritization and drug target identification[103].

Indeed, achieving a comprehensive “oral PTM atlas”, profiling healthy dental regeneration, inflammatory perturbation, pulp repair, dysplasia, early OSCC, CSC expansion, and metastasis, will require the high-throughput application of integrated single-cell and spatial PTM techniques across longitudinal patient cohorts. Such atlases could fundamentally power precision interventions in dentistry and oncology by enabling spatiotemporally resolved therapeutic targeting of PTM-dependent states, rather than relying on cell lineage or mutational profiles alone.

In summary, the revolution in PTM analysis, driven by single-cell, spatial, and multi-omic approaches, has huge potential for transforming our mechanistic understanding of oral biology, connecting static signaling snapshots to dynamic trajectories. By providing unprecedented resolution of cell-state heterogeneity, spatiotemporal PTM modulation, and downstream regulatory consequences, these approaches establish a foundation for innovative diagnostic, regenerative, and therapeutic strategies. The next section will synthesize these insights into actionable translational modalities, focusing on PTM-targeted biomaterials, combinatorial therapeutics, and their clinical application.

TRANSLATIONAL & THERAPEUTIC STRATEGIES

The elucidation of PTM regulatory networks in oral tissue regeneration and OSCC oncogenesis presents a compelling opportunity to translate molecular insights into dual-purpose therapies that can simultaneously promote tissue repair and prevent malignant progression. This translational strategy encompasses regenerative dentistry, through epigenetically functionalized biomaterials, metabolic supplementation, and targeted epigenetic enzyme modulation, and precision oncology, by inhibiting CSC-associated PTMs via small-molecule inhibitors, PROTACs, combination regimens, and personalized stratification guided by PTM signatures.

In regenerative dentistry, the spatial and temporal orchestration of PTMs is critical for guided tissue repair, particularly within the periodontal microenvironment. Significant efforts have been made to engineer biomaterials that mimic endogenous PTM gradients. In this respect, polydopamine-functionalized collagen scaffolds loaded with the HDAC inhibitor trichostatin A enable prolonged local H3K9 acetylation at osteogenic gene promoters, leading to enhanced osteogenic differentiation of PDLSCs and improved alveolar bone regeneration in preclinical models[105]. Concurrently, injectable thermosensitive hydrogels releasing METTL3 activators support epitranscriptomic programming via m6A methylation of angiogenic and osteogenic transcripts, including DLX3 and VEGFA, thus improving scaffold integration and vascularized bone regeneration in mandibular defects[106]. Such materials exemplify the potential for spatiotemporally controlled modulation of PTMs during regeneration, reflecting the staged hierarchical encoding (acetylation → m6A → methylation) inherent to healthy repair processes.

Beyond physical scaffolds, small-molecule modulators of epigenetic enzymes offer another promising route toward PTM-directed regeneration. That activation of p300/CBP HATs enhances chromatin accessibility and supports odontoblast differentiation. Besides, bromodomain inhibitors targeting BET proteins, shown to suppress acetylation-dependent inflammatory cascades, reduce fibrotic outcomes while preserving regeneration, thereby demonstrating dual control over inflammation and epigenetic state[107].

The oncology arm of this strategy specifically targets PTM-dependent CSC mechanisms. METTL3 inhibitors such as STM2457 effectively dismantle the m6A-dependent stabilization of transcripts like SLC7A11, CD44, and SALL4, significantly reducing CSC abundance and radioresistance in OSCC xenograft models[70]. Similarly, Skp2 inhibitors (e.g., C1) restore cell cycle inhibition (p21/p27) and attenuate ubiquitin-mediated degradation modules, thereby synergizing with radiotherapy to suppress OSCC recurrence[108]. Inhibitors targeting the methyltransferase EZH2, such as tazemetostat, not only awaken silenced tumor suppressor pathways but also reprogram CSC plasticity, increasing immunogenicity and chemosensitivity in OSCC[109]. Crucially, combined METTL3-Skp2 blockade demonstrates additive reductions in CSC viability and tumor progression, underscoring cooperative therapeutic targeting of these key PTM axis[110].

Patient stratification using PTM biomarkers is emerging as a cornerstone of precision medicine. Spatially resolved single-cell PTM profiling enables the identification of patients whose tumors exhibit high METTL3-m6A/STAT3-p300 acetylation signatures and low Skp2/EZH2 expression. Such tumors respond optimally to a METTL3 inhibitor combined with BET blockade. Conversely, tumors with elevated Skp2 and EZH2, marked by enhanced ubiquitin-phospho-methylation loops, are more likely to benefit from co-targeted Skp2/EZH2 inhibition regimens[111]. Clinical pilot studies using PTM stratification have already demonstrated improved response rates and reduced toxicities compared to mutation-based cohorts.

For instance, an EZH2-PROTAC achieves sustained CSC depletion and radiosensitization in OSCC xenografts beyond catalytic inhibition[112]. PROTACs specifically targeting Skp2-E3 ligase and METTL3 are currently under preclinical evaluation. It has been reported that their ability to induce targeted enzyme degradation may overcome adaptive resistance typically observed with traditional inhibitors.

PTM profiling offers significant diagnostic and monitoring applications, including the use of circulating biomarkers and advanced imaging techniques. Acetyl-histone signatures, particularly H3K27ac, found in circulating extracellular vesicles, along with mRNA transcripts of METTL3 and Skp2, correlate with OSCC recurrence risk and therapeutic response[111]. Non-invasive imaging modalities, including PET probes specific for acetylated histones or phosphorylated STAT3, enable real-time visualization of residual CSC niches, guiding precise surgical resection or adjuvant therapy.

Despite significant translational advances, challenges remain, particularly in ensuring systemic specificity for epigenetic and PTM-modulator molecules used in regenerative or oncological settings, which require localized delivery systems or depot formulations. Biomaterials offer spatial restriction, while nanoparticle coating and targeted PROTACs may limit systemic exposure. Combination therapy regimens must be carefully designed to prevent antagonism and minimize off-target effects; combination indices derived from PTM signal synergy profiling in organoid models can guide therapeutic design. Indeed, PROTACs must meet safety, bioavailability, and tissue-penetration benchmarks before their clinical application[112].

Long-term safety considerations include avoiding inadvertent oncogenic transformation when modulating the same PTM pathways in regenerative contexts. Biomaterials that deliver HAT activators or METTL3 modulators should be transiently active and developed with stringent release kinetics to prevent malignant conversion. Comprehensive preclinical assessment, including carcinogenesis monitoring, must accompany all translational pipelines[113].

Modulating core PTM regulators (e.g., HDACs, EZH2, METTL3/FTO) can shift lineage programs and, in permissive niches, favor EMT, immune evasion, or CSC expansion. Clinical experience with pan-HDAC inhibitors shows systemic liabilities-hematologic cytopenias, gastrointestinal intolerance, and QTc effects-requiring dose-limiting strategies and cardiac monitoring[114]. RNA-m6A-targeting agents remain highly pleiotropic; both “writers” and “erasers” remodel broad gene sets with context-dependent tumor-suppressive or pro-oncogenic consequences, arguing for transient, localized exposure in regenerative indications[115]. Risk-mitigation should prioritize: (1) Target selection supported by human-genetic evidence and absence of safety-concordant phenotypes; (2) Confinement of exposure using degradable scaffolds or nanoparticles with controlled release; and (3) A translational safety plan with orthogonal on-target biomarkers plus early EMT/CSC readouts and predefined stopping rules[116,117]. These guardrails reduce the chance that PTM modulation intended for repair inadvertently selects for malignant traits.

In conclusion, PTM-directed approaches in oral health encompass therapeutic control over DNA packaging, signal transduction, and transcript stability within regenerative and carcinogenic processes. Regenerative interventions modulate the PTM trajectory toward mineralization and inflammation resolution, while anti-cancer strategies disrupt malignant PTM circuitry. Indeed, while PROTACs and combination regimens hold promise, they will require rigorous PK/PD evaluation and PTM-informed patient selection. The fusion of PTM stratification, localized delivery, and combinatory regimens represents a forward-looking strategy in precision oral medicine, uniting therapy and prevention in a single PTM-aware paradigm. As clinical trials leverage these approaches, the next section will explore the limitations, knowledge gaps, and prospective research directions shaping the future of PTM-informed oral therapeutics.

LIMITATIONS, KNOWLEDGE GAPS & FUTURE PERSPECTIVES

Despite substantial progress in characterizing the regulatory role of PTMs in oral tissue biology and carcinogenesis, several limitations continue to constrain the depth, resolution, and translational readiness of current approaches. These limitations span technical and methodological hurdles in PTM detection, as well as conceptual and clinical challenges in integrating PTM data into therapeutic frameworks.

A critical shortcoming lies in the lack of temporal granularity in PTM studies. Most current investigations capture PTM landscapes as static endpoints, thereby missing transient but biologically significant changes during processes such as regeneration or oncogenic transformation. In this respect, early waves of GCN5-dependent acetylation or METTL3-mediated m6A modification may peak and dissipate within narrow windows that are not sampled[118]. Capturing such events will require time-resolved experimental designs that couple inducible lineage tracing with high-frequency sampling and advanced PTM quantification methods. Incorporating PTM-responsive fluorescent sensors or optogenetic reporters into in vivo models may further enhance the ability to capture dynamic transitions relevant to stem cell fate and malignant progression[119].

Another major technical limitation stems from insufficient sensitivity and breadth in single-cell PTM profiling. Although methods such as SCoPE2 and timsTOF-PASEF have improved proteomic resolution, they remain constrained in quantifying low-abundance PTM isoforms, such as site-specific lysine acetylations or non-canonical ubiquitin linkages[120]. Inadequate peptide enrichment, the loss of labile modifications, and suboptimal chromatographic separation further reduce detection accuracy. To overcome these issues, emerging strategies involve the use of microfluidic mass spectrometry, enhanced TMT multiplexing, and magnetic nanoparticle-based PTM peptide isolation. These approaches, particularly when combined with complementary antibody-based platforms such as mass cytometry or CODEX imaging, may allow for more comprehensive profiling of PTM heterogeneity at the single-cell level[121].

The integration of multi-modal PTM data is complicated by batch effects and technical inconsistencies across different platforms. While computational tools such as MOFA+, Seurat v4, and Harmony can mitigate these biases to some extent, PTM-specific challenges, including variability in ionization efficiency, antibody cross-reactivity, and epitope masking, introduce additional noise that is not fully accounted for by current correction algorithms[98]. Standardization initiatives, including the use of synthetic spike-in standards and cross-laboratory benchmarking datasets, will be essential to achieve robust reproducibility and interoperability among PTM datasets generated from different research centers or patient cohorts.

Spatial resolution remains an equally pressing concern, particularly in preserving PTM integrity during sample processing. Techniques such as CUT&Tag or multiplexed immunofluorescence often require aggressive fixation protocols that may degrade phospho-sites or deacetylate histone tails[122]. Balancing tissue preservation with molecular integrity is a persistent challenge. Recent advances in cryosectioning, combined with vacuum-assisted drying or rapid UV crosslinking, may offer improved molecular fidelity; however, further validation is necessary. Cross-comparisons between mass spectrometry and spatial antibody-based imaging from the same tissues could help to establish reference baselines for signal retention and spatial concordance.

Equally underdeveloped is the toolkit for live-cell visualization of PTMs during regeneration or cancer progression. While transcriptional reporters and real-time calcium or metabolic sensors are well established, equivalent biosensors for PTMs, particularly methylation, acetylation, or ubiquitination, are still in early development. Nanobody-based fluorescence sensors, as well as split-luciferase or FRET-tagged binding domain systems, are being tested in vitro, but they remain rarely applied in stem cell or tumor models within the oral cavity[123]. Generating stable cell lines or transgenic models expressing PTM-sensitive reporters in DMSCs, or OSCC-derived CSCs could yield real-time insights into how PTM fluctuations orchestrate lineage decisions, therapeutic resistance, and cellular plasticity.

From a systems biology perspective, the combinatorial logic underlying PTM crosstalk remains under-characterized. Key mechanistic pathways, such as ubiquitination by Skp2, enhancing AKT phosphorylation, followed by EZH2-driven histone methylation and compensatory STAT3 acetylation, highlight the layered nature of PTM signaling in OSCC[124]. However, few studies model these interactions quantitatively. Multiscale computational frameworks, such as those employing dynamic Bayesian networks or differential equation modeling, could simulate such PTM hierarchies. These models must be iteratively refined through experimental perturbations, e.g., sequential inhibition of METTL3 and EZH2, to identify rate-limiting steps and predict synergistic therapeutic vulnerabilities[125].

On the clinical translation front, only a limited number of PTM-targeting agents have progressed beyond preclinical validation. The METTL3 inhibitor STM2457, currently entering early-phase trials for leukemia, has yet to be systematically evaluated in OSCC patients[87]. PROTACs directed at EZH2 have shown efficacy in depleting CSCs in xenograft models; however, they face concerns regarding off-target degradation in healthy proliferating tissues, such as gingival basal epithelia or mucosal stem cell compartments. Achieving tissue selectivity will likely depend on the development of conditional PROTAC constructs that are activated only under tumor-specific cues, such as MMP activity, redox state, or acidic pH[126].

A key translational bottleneck also lies in the deployment of PTM-based diagnostics and biomarkers. Although circulating histone acetylation marks and extracellular vesicle-associated transcripts have demonstrated prognostic value, these assays typically lack standardized quantification methods, threshold cutoffs, and cross-platform validation. Future clinical trials incorporating PTM readouts as stratification variables, alongside genomic and immune signatures, will be necessary to test the predictive power of PTM signatures in therapeutic decision-making[112].

Looking ahead, several promising directions emerge. Longitudinal multi-modal animal studies that combine single-cell transcriptomics, chromatin accessibility, proteomics, and spatial PTM imaging across regenerative and malignant timepoints are essential to resolve causal PTM shifts. Live-cell biosensor development for PTM dynamics in the oral microenvironment will offer a transformative tool for tracking reprogramming and resistance in situ. On the analytical side, development of oral-specific deep learning frameworks trained on PTM-integrated datasets can generate predictive models for regenerative success or tumor recurrence risk. In therapeutics, conditional and reversible PROTACs, as well as PTM-sensitive scaffolds designed to degrade after delivering an epigenetic payload, will minimize toxicity while preserving efficacy. Crucially, future phase II trials for OSCC and periodontal disease must include PTM-based endpoints to validate their clinical relevance, integrating plasma biomarker analysis, imaging, and spatial biopsy sequencing. Together, these developments will push the field toward a new frontier in PTM-informed oral medicine, one that aligns molecular precision with regenerative and oncologic goals.

CONCLUSION

PTMs function as dynamic molecular codes that drive DMSCs along a continuum from regenerative commitment to malignant transformation, integrating signals from acetylation, phosphorylation, ubiquitination, methylation, and m6A RNA marking. Their inherent reversibility and combinatorial nature make PTMs both sensitive indicators of cellular state and tractable therapeutic nodes. By modulating key enzymes, such as METTL3, Skp2, and EZH2, emerging pharmacologic inhibitors, PROTAC degraders, and epigenetically functionalized biomaterials can simultaneously enhance periodontal and pulp regeneration while curbing CSC plasticity and therapy resistance. The precise deployment of these interventions will hinge on singlecell, spatial, and longitudinal PTM profiling platforms that resolve transient epigenetic inflection points and guide patient stratification. Looking ahead, PTM-aware clinical trials incorporating plasma biomarkers, image-guided biopsies, and artificial intelligence-driven multi-omic integration promise to align regenerative dentistry with oncologic safety, positioning PTMs as central regulators and actionable sentinels of oral tissue health and disease.

ACKNOWLEDGEMENTS

We would like to thank all the professionals who contributed to the discussion and elaboration of this review.