Abstract
The N7-methylguanosine (m7G) modification, an epigenetic transcriptional regulatory mechanism, plays a crucial role in the development of gastrointestinal malignant tumors. This modification, mediated by enzyme complexes such as methyltransferase-like 1 (METTL1)/WD repeat domain 4 (WDR4) and williams-beuren syndrome chromosome region 22 (WBSCR22)/tRNA methyl transferase activator subunit 11-2 (TRMT12), is widely distributed in messenger RNA (mRNA), transfer RNA (tRNA), ribosomal RNA (rRNA), and non-coding RNA. Its abnormal expression is closely associated with the pathogenesis of various gastrointestinal tumors, including hepatocellular carcinoma, colorectal cancer, pancreatic cancer, and esophageal cancer. The METTL1/WDR4 complex enhances the translation efficiency of oncogenes by promoting tRNA m7G modification, thereby facilitating tumor cell proliferation, metastasis, and chemotherapy resistance. More importantly, the m7G modification significantly influences tumor cell resistance to chemotherapy, targeted therapy, and radiation therapy by regulating the epidermal growth factor receptor (EGFR) signaling pathway, autophagy processes, and DNA repair mechanisms. Therefore, m7G modification has dual potential as both a prognostic biomarker and a therapeutic target, and may provide a molecular basis for precision medicine in the treatment of gastrointestinal tumors.
keywords
Introduction
Advances in RNA and DNA sequencing technologies have established a cornerstone of contemporary genomics. Investigation of the epigenome has become feasible with the emergence of high-throughput sequencing technology. Subsequently, the concept of the epitranscriptome was introduced, thus representing an additional layer of biological regulation situated between DNA and proteins, characterized by diverse RNA modifications1–3. These modifications occur on nucleobases or ribose moieties across various RNA species, including ribosomal RNA (rRNA), transfer RNA (tRNA), messenger RNA (mRNA), small nuclear RNA (snRNA), microRNA (miRNAs), and long non-coding RNA (lncRNA)4. Among these RNA modifications, methylation is the predominant form, encompassing N1-methyladenosine (m1A), N5-Methylcytosine (m5C), N6-methyladenosine (m6A), and N7-methylguanosine (m7G) modifications (Figure 1). Among them, m6A is the most extensively studied modification. However, in recent years, m7G has garnered increasing scientific interest. This modification, which was initially identified as critical in the 5′ cap structure of mRNA (m7GpppN), plays an essential role in protecting mRNA against degradation by nucleic acid endonucleases, thereby facilitating efficient gene expression5. Advances in sequencing technologies have enabled the detection of low-abundance RNA modifications and led to the identification of internal m7G in diverse RNA species, including mRNA, rRNA, tRNA, miRNA, and pre-miRNA6,7. Similarly to other epitranscriptomic marks, internal m7G can be viewed within a writer, reader, and eraser framework: writers deposit the modification, readers interpret it to shape RNA fate, and erasers remove or functionally reset the signal. Whereas the writer enzymes have been relatively well defined, the reader and eraser mechanisms continue to be mapped, particularly for internal m7G. These technological advancements have further accelerated investigation of the functional roles of m7G modifications in various pathological conditions, particularly in cancer.
Molecular structures of representative RNA methylation modifications, including m1A, m5C, m6A, and m7G (adapted from the MODOMICS database).
The incidence of gastrointestinal (GI) tumors continues to rise globally and currently accounts for more than 25% of all cancer cases8. More than 1.9 million new cases of colorectal cancer occur annually9, and the annual incidence rate for stomach cancer is approximately 1 million cases10. Increased incidence of early-onset cancers, particularly among younger populations, has prompted urgent public health concerns11. Therefore, the clinical importance of studying GI tumors is self-evident. A deeper understanding of their epidemiology, pathogenesis, and risk factors would inform effective prevention and screening strategies, and facilitate the development of new therapies to improve survival and quality of life. This review provides a systematic analysis of recent progress in elucidating the roles of internal m7G modifications in GI malignancies.
Major enzymes catalyzing internal m7G RNA modification in Homo sapiens
Methyltransferase-like 1 (METTL1)/WD repeat domain 4 protein (WDR4)
METTL1, located on chromosome 12q13, is a highly conserved protein characterized by a conserved amino acid motif [S-adenosyl-methionine (SAM) binding motif], which suggests its potential role as a methyltransferase12. In vitro studies have demonstrated that METTL1 alone cannot catalyze m7G modification; however, it exhibits robust m7G-methyltransferase activity in complex with WDR413,14. WDR4, a WD repeat protein family member encoded on chromosome 21q22.3, comprises 412 amino acids and adopts a structural conformation consisting of 4 α-helices and 28 β-sheets. It is functionally homologous to the yeast Trm8p/Trm82p complex15.
The N-terminal domain of METTL1 is critically involved in mediating m7G46 tRNA modification through regulation of SAM binding, RNA interaction, and conformational dynamics within the METTL1-WDR4 complex14. Notably, phosphorylation of Ser27 within this domain, catalyzed by AKT (protein kinase B) and ribosomal S6 kinase (RSK), results in functional inactivation of METTL116. On the basis of findings from Li et al.17, functional inactivation of METTL1 or site-specific mutations within its peptide sequence might impair the proliferative and migratory ability of cancer cells, thereby inhibiting the malignant progression of tumors. Structural and functional analyses have identified that residues R170 and E167 of WDR4, along with K143 of METTL1, are critical for their mutual interaction and methyltransferase activity, given that substitutions at these sites markedly decrease enzymatic function. Structural studies, including the crystalline form of the METTL1-WDR4 complex and the cryo-electron microscopy configuration of the METTL1-WDR4-tRNA complex, have elucidated that tRNA recognition is achieved through structural complementarity, particularly at the tRNA elbow region14. Additionally, Zhang et al.7 have provided evidence that the METTL1-WDR4 complex catalyzes internal m7G modifications in mRNAs under in vitro conditions. Furthermore, emerging evidence suggests that miRNAs and lncRNAs also serve as substrates for m7G modification mediated by the METTL1-WDR4 complex18–20.
Emerging evidence indicates the critical involvement of METTL1 and WDR4 in the pathogenesis of various diseases. Notably, R170 mutations in WDR4 have been genetically associated with primordial dwarfism, manifesting as facial dysmorphism, cerebral malformations, and seizure disorders; therefore, the WDR4 R170 mutation impairs WDR4-mediated m7G46 modification and might contribute to aberrant neurological function in these patients21. In lung cancer, elevated expression of METTL1 and WDR4 has been observed and might facilitate tumor progression by enhancing the translation of cell-cycle regulatory genes through m7G-mediated stabilization of tRNA-Val-AAC and tRNA-Pro-AGG22. Furthermore, Zhao et al.23 have revealed that METTL1 crucially influences the regulation of m7G levels in ischemic tissue mRNA: increased METTL1 expression promotes VEGFA mRNA translation in an m7G-dependent manner, thereby enhancing HUVEC proliferation, migration, and angiogenesis, and offering novel therapeutic perspectives for ischemic diseases. METTL1 has been implicated in the pathogenesis of multiple malignancies, including prostate carcinoma24, bladder carcinoma25, and head and neck squamous cell carcinoma26. Subsequent sections comprehensively discuss the specific mechanisms underlying the METTL1-WDR4 complex in GI malignancies.
Williams-Beuren Syndrome Critical Region (WBSCR)22/TRMT112
The WBSCR22 gene, situated within the WBSCR on chromosome 7, encodes a protein characterized by a conserved SAM binding motif and a nuclear localization signal, both of which are hallmarks of methyltransferase family proteins27,28. WBSCR22 forms a heterodimeric methyltransferase complex with TRMT112, a diminutive, evolutionarily preserved protein serving as a cofactor and activator for multiple SAM-dependent methyltransferases acting on rRNA, tRNA, and proteins. This complex specifically catalyzes N7-methylation of G1639 in 18S rRNA, a critical step in 18S rRNA maturation29–31. TRMT112 plays an essential role in stabilizing WBSCR22 and facilitating its enzymatic activity. Beyond its role in rRNA m7G modification, WBSCR22 has been associated with inflammation and cancer, most clearly in lung tissues, where it modulates glucocorticoid receptor (GR) function. Mechanistically, WBSCR22 interacts with the GR coactivator GRIP1 and tunes GR-dependent transcriptional activation and repression. Inflammatory cues such as TNFα and IFNγ promote WBSCR22 ubiquitination and degradation, thereby decreasing cellular glucocorticoid responsiveness. Notably, this GR-related axis has been established primarily in lung models, and whether an analogous WBSCR22-GRIP1-GR program might operate in GI tumors remains to be defined32 (Figure 2).
Functions of m7G methylation in RNA metabolism. tRNA: The METTL1/WDR4 enzyme complex plays critical roles in tRNA biology by augmenting tRNA abundance and stability, thus enhancing the translation efficiency of associated mRNAs and contributing to oncogenic potential. mRNA: The quaking protein QKI-7 specifically interacts with m7G-modified mRNAs and directs them to stress granules, thereby suppressing their translation under cellular stress conditions. Pri-miRNA: The m7G modification facilitates the processing of let-7e-5p from its primary transcript to precursor miRNA, thereby promoting miRNA maturation. 18S rRNA: The WBSCR22/TRMT112 complex catalyzes methylation at the G1639 site in 18S rRNA, although the functions of this modification remain to be fully elucidated. tRNA, transfer RNA; mRNA, messenger RNA; pri-miRNA, primary microRNA; 18S rRNA, 18S ribosomal RNA; METTL1, methyltransferase-like 1; WDR4, WD repeat domain 4; WBSCR22, Williams-Beuren syndrome chromosome region 22; TRMT112, tRNA methyltransferase activator subunit 11-2; QKI-7, quaking protein isoform 7. Figure created with Adobe Illustrator 2025.
The m7G readers and erasers
The regulatory system of m7G modification is frequently described as a triad of writers, readers, and erasers. Writers such as METTL1/WDR4 and WBSCR22/TRMT112 catalyze m7G deposition on tRNA and rRNA, respectively (Sections METTL1/WDR4 and WBSCR22/TRMT112), whereas the mRNA 5′ cap is generated through sequential capping reactions including N7 methylation of guanosine. For the 5′ m7G cap, eukaryotic initiation factor 4E (eIF4E) is a canonical reader that binds the methylated guanine through an aromatic “sandwich” configuration, thereby promoting cap-dependent translation initiation33. In the nucleus, the cap-binding complex, composed of CBP20 and CBP80, recognizes the m7G cap co-transcriptionally and couples capping to downstream steps such as pre-mRNA processing and nuclear export34. Beyond cap recognition, the reader landscape for internal m7G has begun to emerge. Notably, Quaking protein (QKI) 7 has been reported as a reader that preferentially recognizes internal m7G-marked transcripts and links this mark to stress granule dynamics and translation control35. Unlike m6A, m7G currently has no evidence indicating the existence of a specific endogenous demethylase in mammalian cells. Functional removal of the m7G cap is achieved primarily via decapping rather than demethylation: DCP2 catalyzes hydrolysis of the 5′ cap, thus releasing m7GDP and exposing a 5′-monophosphate RNA for exonucleolytic decay36. NUDT16 has also been reported to cleave m7G-cap structures in specific RNA contexts37. However, how the internal m7G signal is removed or reset is largely unclear and remains an important gap to be addressed in future work.
The m7G modification in RNA
Internal m7G modification in mRNA
Beyond its well-characterized localization at the 5′ cap site of mRNA, m7G has been identified within internal regions of mRNA. Zhang and colleagues have demonstrated that mammalian mRNA retains a low level of m7G/G (0.02%–0.05%) even after removal of the m7G at the 5′ cap, thus indicating the presence of m7G within the internal regions of mRNA. Furthermore, they have developed a novel sequencing technology, m7G-seq, to identify these internal m7G modifications. In this approach, m7G sites are initially transformed into abasic sites (AP sites) through NaBH4 reduction and depurination. Subsequently, reverse transcription introduces mutations at the AP sites. By comparing the cDNA sequencing results before and after the treatment (NaBH4 reduction and depurination), the researchers successfully identified the m7G sites38. Using this technology, they observed a evident enrichment of m7G in the coding sequence (CDS) and 3′ untranslated region (UTR) near stop codons. Notably, this enrichment decreased after METTL1 knockout7. Moreover, Malbec et al.39 have established a more specific method for detecting internal mRNA-m7G modifications, known as m7G-miCLIP-seq technology. In this approach, m7G-modified mRNA is first enriched through immunoprecipitation with m7G-specific antibodies. After UV cross-linking to stabilize RNA-protein interactions, the RNA is released through Proteinase K digestion. After digestion, the peptides remaining bound to the RNA can cause truncation or mutations during subsequent reverse transcription. The resulting cDNA libraries, both before and after treatment, are amplified and sequenced to identify m7G-modified sites. This method has demonstrated significant enrichment of m7G in the 5′ UTR and in AG-rich regions39 (Figure 3). Moreover, the positioning of internal m7G modifications within mRNAs exhibits dynamic changes in response to various stimuli. For example, heat shock significantly elevates m7G levels in the CDS and 3′ UTR, while decreasing m7G levels in the 5′ UTR in the mRNAs of genes in the DNA repair pathway and the cell stress response pathway. Furthermore, the authors investigated the effects of the positioning of internal m7G modifications on mRNA translation efficiency. In comparison to unmethylated mRNA, internal m7G modifications in all mRNA regions (5′ UTR, CDS, and 3′ UTR) appreciably enhanced translation efficiency39. Overall, these findings have demonstrated the presence of internal m7G methylation of mRNA in higher eukaryotes and indicated that internal m7G methylation of mRNA might be a novel epigenetic transcriptomic marker potentially regulating translation efficiency.
Schematic diagram of the m7G-seq and m7G-miCLIP-seq technologies. (A) Schematic diagram of m7G-seq. Because of the positive charge on the five-membered ring, NaBH4-mediated reduction selectively converts m7G to its reduced form without affecting unmodified G. Subsequently, under acidic conditions (pH 4.5), the reduced m7G undergoes depurination, thus generating an abasic site that can be captured by biotin-ligated hydrazide to produce a biotinylated AP site. After reverse transcription with HIV RT, these gap sites (or biotinylated AP sites) are recognized primarily by RT as T and C (T and C are predominantly incorporated at corresponding positions in the reverse transcription product). Internal m7G sites can be identified according to these error incorporation signatures, at single-base resolution. (B) Schematic diagram of m7G miCLIP-seq. RNA is fragmented and incubated with anti-m7G. After UV cross-linking to stabilize RNA-protein interactions, immunoprecipitation selectively isolates the antibody-bound RNA. The RNA is then released by Proteinase K digestion and subjected to reverse transcription. During this process, residual peptides on the RNA induce truncations or mutations in the cDNA. The cDNA library is subsequently amplified by PCR and analyzed through high-throughput sequencing. AP site, abasic site; RT, reverse transcriptase. Figure created with BioRender (www.biorender.com).
Notably, Zhao et al.35 have recently reported Quaking protein 7 (QKI7) as the first identified reader of internal m7G modifications in mRNA. This protein specifically recognizes internal m7G modifications on mRNA, thereby explaining the mechanism through which internal m7G modifications in mRNA regulate translational efficiency. QKI7, an isoform of QKI belonging to the STAR family of RNA-binding proteins with a KH domain35, is essential for the formation of stress granules (SGs) under stress conditions35,40,41. Using comprehensive high-throughput sequencing techniques and experimental corroboration, Zhao et al.35, in 2023, revealed that QKI7 facilitates sequestration of internal m7G-modified mRNA into SGs by interacting with the SG core protein G3BP1, thereby modulating mRNA stability and translation in stress environments. Notably, through this mechanism, QKI7 decreases the translation efficiency of essential genes in the Hippo signaling pathway and consequently enhances cancer cell susceptibility to chemotherapeutic agents35. In light of these findings, QKI7, the first identified reader of internal m7G modifications on mRNAs, has immense potential for future studies in mRNA metabolism and cellular drug resistance mechanisms.
The m7G modification in non-coding RNAs
The m7G modification is not confined to internal regions of mRNA but also exists in various types of non-coding RNAs (ncRNAs). Frequently used assays for mapping and validating m7G across different RNA classes are summarized in Table 1.
Frequently used assays for detecting m7G modification
The m7G modification in tRNA
Currently, the most extensively studied m7G modification is m7G46 in tRNA, which is catalyzed by the METTL1/WDR4 enzyme complex. The crystal structure of tRNA-Phe reveals a base-triple interaction formed by C13, G22, and m7G4647. Under physiological conditions, G46 remains uncharged; however, after its m7G modification, it acquires a positive charge and forms hydrogen bonds with G22 and C13. This tertiary base-triple interaction is essential for maintaining the stable three-dimensional architecture of tRNA48. Beyond m7G46, the METTL1/WDR4 enzyme complex modifies a specific subgroup of tRNAs that decode m7G-tRNA-dependent codons. This modification ensures efficient ribosomal elongation and enhances the translation efficiency of associated mRNAs, including those involved in oncogenesis and cell-cycle regulation, thereby enhancing carcinogenic potential49. In contrast, diminished activity of METTL1/WDR4 results in ribosomal stalling and particularly affects the m7G-modified ARG-TCT-4-1 tRNA, thus decreasing translation of these mRNAs and consequently inhibiting tumor transformation and cancer progression49,50. Through m7G-modified tRNA immunoprecipitation sequencing (MeRIP-Seq) and m7G tRNA reduction and cleavage sequencing (TRAC-Seq), Lin et al.50 have successfully identified the m7G tRNA methylome in mESCs. Their study identified 22 tRNAs bearing m7G modifications at a conserved RAGGU motif in the variable loop, which is critical for accurate translation of genes involved in the cell-cycle as well as genes associated with brain function abnormalities in mESCs. Furthermore, knockout of METTL1 or WDR4 decreased mESC self-renewal and differentiation into neural lineages; therefore, m7G tRNA appears to be crucial for normal differentiation of the neural lineage50.
The m7G modification in rRNA
The m7G modification also occurs in rRNA. In budding yeast, the G1575 site, located at the ridge between the P site and the E site tRNA at the back of the small subunit head of the 18S rRNA, is methylated by the Bud23-Trm112 complex30. Within this complex, Bud23 is the catalytic subunit, and Trm112 is a co-activator enhancing the stability of Bud2330. In humans, m7G modification has been identified at the G1639 site of the small subunit rRNA through m7G-MaP-seq analysis, in a process catalyzed by the WBSCR22-TRMT112 complex45. WBSCR22 has been identified as the human homolog of yeast Bud23, according to sequence similarity51. Haag et al.31 have demonstrated that WBSCR22 has dual independent functions in human small nucleolar ribosomal subunit biogenesis. Beyond its role in catalyzing m7G methylation at G1639 in 18S rRNA, WBSCR22 is involved in the 3′-5′ trimming of 18SE pre-rRNA31, a function independent of its methyltransferase activity. In the absence of WBSCR22, impeded 3′ end trimming of the 18SE pre-rRNA results in buildup of 3′ end-extended 18SE intermediates within the nucleus. Furthermore, bacterial studies have demonstrated that m7G1405 deposition in 16S rRNA, mediated by sisomicin-gentamicin methyltransferase (Sgm), confers ribosome resistance to 4,6-disubstituted deoxystreptamine aminoglycosides, such as gentamicin and kanamycin52.
The m7G modification in miRNA
Emerging evidence indicates the presence of m7G modifications in miRNAs, thereby expanding understanding of RNA epigenetics. Pandolfini et al.6 have used borohydride reduction sequencing (BoRed-seq) coupled with RNA immunoprecipitation techniques to identify m7G in a subset of miRNAs that inhibit cell migration. METTL1 methylates a specific subset of tumor suppressor miRNAs (TS-miRNAs), such as the let-7 family, thereby facilitating their conversion from primary transcripts to precursor miRNAs6. Specifically, m7G modification enhances the propensity of miRNAs to form G-quadruplex structures, and G11 in let-7e-5p is a critical methylation site. Methylation at this site influences the formation of G-quadruplexes, thereby facilitating processing of precursor miRNAs. Furthermore, METTL1 depletion induces notable alterations in gene expression and cellular phenotypes through its effects on miRNA structural dynamics6.
Comparative analysis and limitations of RNA modifications sequencing techniques
Several m7G modification detection technologies, each with distinct advantages and limitations, have been developed for various RNA types, including mRNA, tRNA, rRNA, and miRNA. Most techniques, such as m7G-seq and m7G miCLIP-seq, offer single-base resolution and high sensitivity. Among these, chemically dependent methods (e.g., TRAC-Seq and AlkAniline-Seq) require no antibodies and exhibit high specificity, although they involve complex workflows and sometimes hazardous reagents. Antibody-dependent approaches (e.g., m7G miCLIP-seq and MeRIP-Seq) enable specific enrichment or streamlined procedures, yet present challenges including complex workflows, high costs, and risks of non-specific binding. MeRIP-Seq and RIP-seq achieve only RNA fragment-level resolution with relatively low sensitivity; however, the former can enrich m7G sites with varying modification ratios, whereas the latter offers greater versatility. Furthermore, novel techniques such as m7G-MaP-seq and BoRed-seq demonstrate advantages including high sensitivity, yet lack extensive validation and require further refinement.
Future directions of m7G detection techniques
To address the limitations discussed above, future m7G profiling is expected to move toward higher specificity, better quantification, and stronger cross-platform comparability. In practice, orthogonal validation workflows that pair antibody-independent chemistry with quantitative mass spectrometry can help separate true m7G signals from background and, importantly, begin to estimate site occupancy rather than reporting presence vs. absence alone. In parallel, long-read strategies together with more reliable reverse-transcription signatures have potential to improve isoform-resolved mapping, which is essential when alternative splicing or transcript heterogeneity changes the interpretation of “site-level” signals. For cancer samples, low-input and single-cell compatible protocols would be particularly valuable, because bulk measurements often average out rare but therapy-relevant cell states within tumors and their microenvironments. Finally, community-wide standardization, including spike-in controls, benchmarking datasets, and transparent reporting of sensitivity and specificity, will be necessary to facilitate the comparability of results across studies and support reproducible biological conclusions.
The m7G modification and GI cancers
Regulation of m7G depends on the coordinated activity of key enzymes, and dysregulation of this system is closely associated with GI cancer burden. Elucidating aberrant m7G patterns can clarify their prognostic value and help define the roles of m7G writers in tumor biology. Collectively, these insights have potential to guide the development of targeted therapeutic strategies.
Potential application of m7G writers as useful prognosis markers for GI cancers
The m7G RNA modification and its associated regulatory proteins, particularly METTL1/WDR4, exert critical regulatory functions in hepatocellular carcinoma (HCC), colorectal cancer (CRC), pancreatic ductal adenocarcinoma (PDAC), and esophageal squamous cell carcinoma (ESCC). These effects are mediated through the modulation of key oncogenic processes, such as tumor cell proliferation, metastatic progression, metabolic reprogramming, and tumor microenvironment interactions. Figure 4 provides an integrated summary of the HCC-related mechanisms discussed in this section, including epithelial-mesenchymal transition (EMT) regulation, myeloid-derived suppressor cell (MDSC)-associated immunosuppression, and DNA damage repair-related pathways.
Regulatory mechanisms of m7G modification in liver cancer development. (A) Under heat treatment conditions, upregulated expression of METTL1 leads to increased m7G modification levels and enhanced translation of tRNALys-CTT, thereby supporting the translation of SLUG and SNAIL mRNA, and ultimately enhancing the migration and invasion capabilities of cancer cells. (B) Cancer cells activate TGF-β2 at the translational level through METTL1-mediated tRNA m7G modification, thereby inducing the accumulation of MDSCs and forming an immunosuppressive microenvironment. (C) METTL1-mediated tRNA m7G modification selectively regulates the translation of DNA-PKcs mRNA, thereby promoting DNA repair via non-homologous end-joining (NHEJ). IR/IRFA, radiofrequency ablation/incomplete radiofrequency ablation; METTL1, methyltransferase-like 1; MDSCs, myeloid-derived suppressor cells; SLUG, snail family transcriptional repressor 2; SNAIL, snail family transcriptional repressor 1; TGF-β, transforming growth factor beta; PMN-MDSCs, polymorphonuclear myeloid-derived suppressor cells; DNA-PKcs, DNA-dependent protein kinase catalytic subunit; DNA-PK, DNA-dependent protein kinase; XRCC4, X-ray repair cross-complementing protein 4. Figure created with BioGDP (www.biogdp.com).
In various cancers, METTL1 and its associated molecules have important oncogenic roles, and their regulatory mechanisms exhibit notable specificity across cancer types. Research in HCC has revealed elevated METTL1 and WDR4 expression in HCC tissues, which is closely associated with negative clinical results such as diminished overall survival, disease-specific survival, and progression-free survival. More importantly, METTL1/WDR4 knockdown effectively inhibits the proliferative, migratory, and invasive phenotypes of HCC cells, thereby further validating the critical role of this complex in the malignant progression of HCC53. Similarly, in CRC, METTL1 promotes CRC cell growth and cell-cycle progression by suppressing CHEK2 expression. The CHEK2 pharmacological inhibitor BML-277 promoted cell proliferation and the G1/S phase transition, whereas this effect was attenuated by METTL1 inhibition in HCT-116 cells54. Notably, the m7G-related lncRNA SNHG8 is a key regulator in PDAC, in which its down-regulated expression directly correlates with enhanced tumor proliferative ability. Functional validation experiments have demonstrated that SNHG8 silencing enhances the proliferative, migratory, and invasive capabilities of PANC-1 and MIA PaCa-2 cell lines55. Collectively, these findings indicate the central regulatory role of the m7G modification network in tumorigenesis and malignant progression across cancer types.
During invasion and metastasis, m7G modification regulates the EMT program and cell motility, as demonstrated in HCC models. After incomplete radiofrequency ablation, both METTL1 expression and tRNA m7G levels have been found to increase in both in vivo and in vitro models. In-depth analysis has revealed that METTL1 upregulation closely correlates with enhanced translation of crucial EMT genes, particularly SLUG and SNAIL. This effect is particularly pronounced in heat-treated HCC cells, in which METTL1 might drive tumor metastasis by promoting the translation of the mRNAs of these genes. Notably, in SLUG and SNAIL, the lysine codon AAG decoded by m7G tRNA is the most frequently used, and the m7G modification level in the corresponding tRNA-Lys-CTT increases after heat treatment. Importantly, rescue experiments have indicated that ectopic overexpression of SLUG and SNAIL in METTL1-deficient HCC cells restores migration and invasion, thereby supporting the METTL1-m7G-SLUG/SNAIL axis in HCC metastasis56.
The involvement of m7G-associated enzymes in tumorigenesis and their associations with clinical outcomes highlight their potential as prognostic biomarkers and therapeutic targets. Through integrative analysis of The Cancer Genome Atlas database and single-cell RNA sequencing datasets, Jiang et al.54, in 2023, identified that METTL1 is markedly upregulated in CRC, and its expression strongly correlates with unfavorable clinical prognosis. This pattern aligns with the upregulation of METTL1/WDR4 expression in HCC and its association with poor clinical outcomes. Notably, in pancreatic cancer (PC), the m7G-associated molecule WBSCR22 exhibits a contradictory phenomenon entirely opposite from that observed in HCC/CRC: although its protein levels are substantially downregulated, by 50%, in tumor tissue relative to adjacent non-tumor tissue, WBSCR22 expression has significant prognostic relevance, and high WBSCR22 expression correlates with a 40% increase in 5-year survival rates57. Furthermore, investigation of an m7G-related lncRNA prognostic model (m7G-LPS) has revealed that SNHG8 expression levels are closely associated with clinicopathological characteristics, including TNM stage and tumor grade, as well as poor prognosis in PDAC55. The distinct expression patterns of SNHG8 across risk groups provide novel insights for clinical stratification and personalized treatment approaches. These seminal findings have not only enhanced understanding of the m7G modification regulatory network but also established a critical foundation for the development of precise prognostic evaluation systems and targeted therapeutic interventions.
Mechanism of m7G and its writers in regulating GI cancer development
After the regulatory roles of m7G RNA modification and its core complexes (such as METTL1/WDR4) in HCC, CRC, PDAC, and ESCC were established, subsequent studies suggested that its effects extend beyond tumor initiation and progression. Such modifications not only participate in classical carcinogenic processes but also dynamically respond to tumor microenvironmental changes (such as hypoxia and nutrient deprivation) by precisely regulating adaptive survival strategies in tumor cells. These strategies include flexible metabolic mode switching, activation of metastatic potential, and specific responses to microenvironmental signals. These extended mechanisms have revealed a multi-layered regulatory network of m7G modification within complex tumor pathologies, and offered novel entry points for elucidating the underlying molecular regulatory logic and translational application potential.
Recent studies have elucidated a critical molecular pathway in HCC, wherein MYC facilitates transcriptional activation of WDR4 by directly binding its promoter region. This interaction leads to upregulation of WDR4, which subsequently enhances the translation of CCNB1 by promoting binding of EIF2A to CCNB1 mRNA. This cascade inhibits phosphorylation of the DNA damage response proteins CHK1 and ATR, thereby attenuating DNA damage, activating PI3K/AKT signaling, inducing P53 ubiquitination, and ultimately driving HCC cell proliferation and metastasis58. Importantly, similar molecular mechanisms have been observed in other malignancies. For example, in CRC, the hypoxic environment has important effects on tumor progression. Mi et al.59, in 2024, demonstrated that, in CRC, hypoxia significantly decreases the abundance of tRNA m7G methylation and mRNA levels of the METTL1/WDR4 complex. Moreover, hypoxia-inducible factor (HIF)-1α binds the HRE sequence within the METTL1 promoter region and transcriptionally represses the expression of METTL1 during intratumoral hypoxia. Hypoxia and HIF-1α have been reported to drive CRC development, encompassing the reprogramming of metabolism, migratory ability, invasive potential, and chemoresistance60–63. Therefore, hypoxia-driven suppression of METTL1-WDR4 and the accompanying decrease in tRNA m7G might serve as a translational reprogramming mechanism in CRC. By decreasing the pool of m7G-modified tRNAs, cells might curb energetically expensive global translation while reshaping codon-dependent translation of stress-adaptive transcripts, thereby improving fitness in oxygen-limited environments59.
In PC, researchers have also identified important molecular mechanisms. Experimental data have confirmed lower expression of WBSCR22 in human PC cell lines (PANC-1, BXPC-3, and AsPC-1) than the normal pancreatic ductal epithelial cell line HPDE6-C7. Functional assays have demonstrated that overexpression of WBSCR22 effectively suppresses the proliferative, migratory, and invasive capabilities, and the tumor-forming ability, of PANC-1, BXPC3, and ASPC1 pancreatic cancer cells, both in vitro and in vivo. RNA sequencing analysis has indicated that WBSCR22 negatively regulates mRNA synthesis of the downstream interferon-stimulated gene 15 (ISG15). Similarly, experiments have confirmed that ISG15 overexpression promotes PC proliferation, migration, invasion, and tumorigenesis, and the effects of WBSCR22 overexpression are rescued by ISG15 overexpression. Notably, WBSCR22 and its cofactor tRNA methyltransferase activator subunit 11-2 (TRMT112) have synergistic roles in pancreatic cancer, and simultaneous overexpression of WBSCR22 and TRMT112 further promotes pancreatic cancer progression and enhances the ability of WBSCR22 to suppress tumors in PC. The WBSCR22/ISG15 signaling axis is therefore a potential therapeutic target offering novel intervention strategies for pancreatic cancer management57.
Recent research has begun to reveal the molecular mechanisms driving ESCC progression. Emerging evidence suggests that RNA modifications, such as m7G-modified tRNAs, are essential to this process. Downregulation of METTL1 decreases the m7G modification of tRNAs, thereby inhibiting ESCC progression. This inhibition is attributed to the decreased translational efficiency (TE-down) of certain mRNAs after a decrease in m7G-modified tRNA levels in ESCC cells under METTL1 depletion. These TE-down mRNAs are significantly enriched in mRNAs associated with autophagy and the mTOR signaling pathway. Notably, these mRNAs, including regulatory related protein 1 of the mTOR complex (RPTOR), have abundant m7G tRNA-dependent codons. The observation that METTL1 depletion was followed by decreased RPTOR expression and increased cell death and autophagy suggested that METTL1 enhances the protein synthesis of genes associated with mTOR pathway activity and regulates autophagy in a codon-dependent manner with m7G-related codons. The growth of ESCC cells was restored after overexpression of tRNA-ValAAC and/or tRNA-LysCTT, which were downregulated because of METTL1 depletion, and whose corresponding codons were highly enriched in TE-down mRNAs. Furthermore, upregulation of RPTOR and suppression of ULK1 restored the proliferative and colony-forming ability of METTL1-depleted ESCC cells and eliminated autophagic flux in these cells. Finally, conditional knockout and knock-in of METTL1 in mice robustly and directly substantiated the significant oncogenic function of METTL1 and tRNA modifications in the onset and development of ESCC64. These findings collectively demonstrated that m7G tRNA modification mediated by METTL1 accelerates ESCC development by regulating the RPTOR/ULK1 axis64.
Mechanisms of m7G and its writer in GI cancer treatment resistance
The management of GI tumors encompasses a multimodal approach including surgical resection, chemotherapy, radiotherapy, bio-immunotherapy, and targeted therapy. Despite continued advancements in cancer research and the emergence of novel therapeutic strategies, major challenges and limitations persist across all current treatment modalities65–67. A critical barrier to the efficacy of chemotherapy is the development of drug resistance68. Chemotherapeutic agents based on platinum, including cisplatin, carboplatin, and oxaliplatin, are broadly approved and constitute the first-line treatment for numerous malignant tumors. However, the frequent occurrence of resistance to platinum-based therapies remains a substantial clinical challenge69,70.
In cisplatin-resistant colorectal cancer cells (CR-CC), the expression of METTL1 is lower than that in cisplatin-sensitive colon cancer cells (CS-CC)71. Overexpression of METTL1 in CR-CC cells upregulates p53 levels through regulating the miR-149-3p/S100A4 axis, thereby affecting the cells’ resistance to cisplatin. Specifically, METTL1 promotes expression of the downstream target miR-149-3p, which subsequently binds the 3′ UTR region of S100A4 and inhibits the expression of S100A4. Notably, S100A4 modulates cisplatin resistance across various cancer types through p53-dependent pathways72,73. METTL1 or miR-149-3p overexpression has been found to increase p53 expression in CR-CC cells, thereby supporting cell death in CR-CC, whereas up-regulation of S100A4 reverses this effect. These findings collectively indicate that METTL1-mediated regulation of the miR-149-3p/S100A4/p53 axis enhances cisplatin sensitivity in CR-CC71.
When cisplatin induces DNA double-strand breaks (DSBs), activation of p53, which is essential for DNA repair and cell fate decisions, triggers apoptosis. This molecular mechanism potentiates cisplatin chemosensitivity in CR-CC cells71,74. Furthermore, knockdown of WBSCR22, a regulator involved in rRNA m7G methylation, promotes oxaliplatin-induced apoptosis in CRC cells. WBSCR22 knockdown in oxaliplatin-treated CRC cells increases intracellular reactive oxygen species production, which is critical for oxaliplatin-induced cell death75. Given the critical involvement of regulatory factors such as METTL1 and WBSCR22 in RNA m7G modification processes, therapeutic targeting of m7G modification pathways might offer a potential therapeutic approach to combat platinum-based chemoresistance in CRC.
Beyond conventional chemotherapeutic agents, targeted therapies are critical components in cancer treatment strategies. Lenvatinib, a multi-targeted tyrosine kinase inhibitor, inhibits several receptor tyrosine kinases including vascular endothelial growth factor receptors (VEGFR1–3), fibroblast growth factor receptors (FGFR1–4), platelet-derived growth factor receptor α (PDGFRα), KIT, and RET. Lenvatinib received approval for HCC treatment in 201876. Compared with sorafenib, another tyrosine kinase inhibitor, lenvatinib demonstrates superior clinical efficacy, as evidenced by its higher objective response rate and disease control rate. Additionally, lenvatinib’s enhanced safety and tolerability profiles in patients with HCC further support its therapeutic advantage77,78.
Recent investigations by Huang et al.79 have elucidated the molecular basis of lenvatinib resistance in HCC. Their findings have revealed significant upregulation of METTL1 and WDR4, the core elements of the tRNA m7G methyltransferase complex, in HCC cells resistant to lenvatinib, accompanied by elevated tRNA m7G modification. Functional studies have demonstrated that upregulation of wild-type METTL1, but not the catalytically inactive variant, in lenvatinib-sensitive Huh7 cells increases tRNA m7G modification and confers resistance to lenvatinib while promoting HCC cell proliferation. Therefore, METTL1 promotes lenvatinib resistance in a manner dependent on its enzyme activity79. Furthermore, METTL1-depleted lenvatinib-resistant HCC cells exhibit enhanced EGFR mRNA translational activity, mirroring observations in intrahepatic cholangiocarcinoma cells80. Cell culture and animal model evidence has established the critical role of the EGFR pathway in METTL1-mediated lenvatinib resistance in HCC79: EGFR activation has been identified as a limiting factor in the therapeutic response to lenvatinib81. Together, these results establish that METTL1 substantially contributes to lenvatinib resistance in HCC by orchestrating tRNA m7G modification and functionally regulating EGFR pathway activity.
Beyond lenvatinib, several EGFR-targeted therapeutic agents, including gefitinib, erlotinib, cetuximab, and panitumumab, have been developed. EGFR tyrosine kinase activity is reversibly inhibited by erlotinib and gefitinib via competitive binding at the ATP-binding domain. In contrast, monoclonal antibodies such as cetuximab and panitumumab prevent ligand engagement with EGFR’s ectodomain, enhance receptor internalization, and trigger immune-mediated cytotoxicity via antibody and complement pathways82. However, despite the demonstrated anticancer efficacy of EGFR inhibitors alone or in combination with chemotherapy and radiotherapy, the intricate mechanisms underlying resistance to these agents remain a major obstacle in cancer treatment82–88. In this context, METTL1 has emerged as a potential therapeutic target to overcome resistance to EGFR inhibitors.
Additionally, METTL1/WDR4 might be involved in lenvatinib resistance by regulating the generation of cancer stem cells (CSCs). CSCs possess a fundamental ability to maintain tumor propagation and may inherently resist medical therapy and contribute to tumor relapse89. TRIM28, a critical regulator of stem cell pluripotency, is crucial for maintaining the breast cancer stem cell population90,91. Numerous studies have reported that it might function as a protumorigenic factor in cancer, because of its ability to promote oncogenic progression and metastatic invasion via EMT activation92–94, mediate adaptation to stress conditions95–97, and suppress p53 activity98–100. Notably, lenvatinib resistance has been associated with the enrichment of CSCs101. Han et al.102 have discovered that in HCC, WDR4 directly regulates, and is positively correlated with, TRIM28 expression. Subsequent investigations have revealed that TRIM28 mediates WDR4-induced lenvatinib resistance and the upregulation of CSC markers. These findings collectively indicate that the WDR4/TRIM28 axis promotes lenvatinib resistance by enhancing the stemness properties of HCC cells102. Consequently, WDR4 has emerged as a potential biomarker for predicting lenvatinib efficacy and guiding personalized therapy.
CSCs are deemed a substantial cause of chemoresistance and radioresistance in cancer therapy103,104. Moreover, multiple molecular mechanisms underlie radiotherapy resistance, including enhanced DNA damage repair pathways, cell-cycle checkpoint activation, autophagy induction, tumor microenvironment remodeling, and hypoxia-mediated adaptive responses105. These factors that are detrimental to cancer treatment often appear after drug treatment or radiotherapy, but also have complex mechanisms of occurrence and a wide range of influencing factors. Elucidating the molecular pathways governing these resistance mechanisms and developing strategies to modulate them will be essential directions for optimizing cancer treatment outcomes.
Currently, radiotherapy has emerged as a growingly essential treatment strategy for HCC. The primary mechanism of cell death induced by radiation therapy is DNA damage, and DSBs are the most lethal form. DSBs are repaired through 2 main pathways: non-homologous end joining (NHEJ) and homologous recombination106,107. NHEJ is the predominant repair mechanism for DSBs in both dividing and quiescent somatic cells108. Mutation or dysregulation of NHEJ-associated proteins increases cellular sensitivity to ionizing radiation and DSB-inducing chemotherapeutics, thus potentially triggering premature B-cell and T-cell apoptosis, and consequently immune deficiency109,110. In NHEJ, DSBs are initially detected by the Ku70/Ku80 heterodimer (Ku), which acts as a molecular scaffold for recruitment of additional NHEJ proteins, including XRCC4, XLF, and DNA ligase IV (LIG4), to the DNA ends111. Encoded by PRKDC, the DNA-dependent protein kinase catalytic subunit (DNA-PKcs) displays high binding affinity for Ku-bound DNA termini and forms a complex with Ku that facilitates DNA repair112. Recent studies by Liao et al.113 have demonstrated that METTL1-regulated augmentation of LIG4 protein synthesis is indispensable for sustaining DNA ligase IV (LIG4) protein levels and ligation function after ionizing radiation (IR) treatment. Immediately after IR exposure, tRNA m7G modification mediated by METTL1 selectively regulates the translational efficiency of either DNA ligase IV or DNA-PKcs mRNA, thereby supporting NHEJ-based DNA repair and conferring resistance to radiotherapy in HCC cells113. The kinase activity of DNA-PKcs is essential for DNA end processing114. Further research has revealed that enhanced METTL1-mediated PRKDC mRNA translation results in significant upregulation of DNA-PKcs levels immediately after IR treatment. This upregulation markedly improves DNA end processing and repair efficiency, thereby underscoring the necessity of METTL1-mediated m7G tRNA modification for effective NHEJ repair113. METTL1, a crucial anti-radiation factor that protects HCC cells, directly participates in NHEJ-mediated DNA repair through a mechanism dependent on the frequency of m7G tRNA codons, thereby enhancing DNA DSB repair in HCC cells. Deficiency in METTL1 decreases IR-induced m7G modification and NHEJ-specific gene expression, owing to impaired translation of codon frequency-dependent mechanisms113. Therefore, METTL1 might serve as a predictive biomarker for sensitivity to radiotherapy and as a molecular target for the radio-sensitization of HCC, and has potential for clinical application.
In tumor treatment, radiofrequency ablation (RFA), because of its distinctive treatment mechanism, has emerged as a critical therapeutic modality alongside radiation therapy. RFA induces extensive cellular necrosis, which facilitates the release of tumor antigens and creates conditions conducive to anti-tumor immunity and elimination of residual tumor cells. Under heat shock conditions, dynamically upregulated expression of METTL1 and the m7G modification of internal RNA increases mRNA translation efficiency39. Moreover, overexpression of both METTL1 and TGF-β2; increased numbers of CD11b, CD15, and PMN-MDSC; and decreased numbers of CD8 T cells have been observed in recurrent HCC after RFA. These findings substantiate the existence of the METTL1-TGF-β2-MDSC axis in recurrent HCC post-RFA. Specifically, TGF-β2 is translationally activated in HCC cells through METTL1-mediated tRNA m7G modification under heat stress, which induces the accumulation of MDSCs, thereby establishing an immunosuppressive milieu115. This phenomenon might be crucial for immune monitoring after RFA treatment and for preventing the recurrence of HCC (Figure 4).
Furthermore, under hypoxic conditions, the alterations in m7G modifications in tRNA within CRC cells, as well as the ability of HIF-1α to bind the HRE in the METTL1 promoter, have been elucidated59. HIF-1α, a principal regulator of intratumoral hypoxia, plays critical roles in hypoxia-triggered cellular mechanisms and has also been associated with drug resistance processes in CRC60,62.
In the context of cell-cycle regulation, METTL1 downregulates CHEK2 (human CHEK2; murine Chk2)54. CHEK2 plays a crucial role in maintaining genomic stability and preventing carcinogenesis by mediating cell-cycle arrest and DNA damage repair through signaling cascades involving downstream substrates such as p53 and CDC25C, or by promoting apoptosis in response to inadequate or aberrant DNA repair. Truncating variants in the CHEK2 gene are associated with an intermediate risk of breast cancer116. METTL1 contributes to poor prognosis in CRC by downregulating CHEK2, thereby promoting the G1/S transition and aberrant cellular proliferation in CRC cells54 (Table 2).
Mechanisms mediated by m7G regulators associated with oncotherapy
Contradictory effects of METTL1/WDR4 function
The seemingly contradictory roles of METTL1/WDR4 can be reconciled by codon usage bias-driven translational control. METTL1/WDR4 installs m7G on a defined subset of tRNAs, thus enhancing decoding efficiency for specific codons and favoring translation of transcripts enriched in those codons. Under this framework, the context dependence arises because proliferation-associated genes and stress-response genes often differ in codon composition and consequently in their reliance on m7G-modified tRNAs. Under normoxic conditions, METTL1/WDR4 activity might preferentially support translation of growth and cell-cycle programs, in agreement with oncogenic outputs in CRC. In contrast, hypoxia represses METTL1 via HIF-1α and decreases tRNA m7G, thus shifting translational priorities away from proliferation toward hypoxia-adaptive programs, in alignment with an “adaptive” phenotype. This principle is exemplified in HCC, in which elevated tRNA m7G has been associated with codon-dependent translation of EMT drivers such as SLUG and SNAIL, whereas cancer-type-specific pathway wiring (e.g., PI3K/AKT-EGFR in HCC vs. CHEK2/p53-linked states in CRC) further shapes how these codon-biased programs translate into drug-response phenotypes.
Discussion
In recent years, substantial progress in the understanding of m7G methylation in GI cancers has opened new avenues for clarifying tumorigenesis mechanisms and developing precision treatment strategies. Evidence that m7G-modifying enzymes, particularly METTL1, exert multi-level regulatory effects in the tumor microenvironment has not only revealed the complexity of epigenetic regulation but also provided a theoretical basis for targeted interventions.
At the fundamental research level, m7G methylation has been discovered to regulate tumor progression by influencing mRNA stability and translation fidelity. This finding has spurred exploration of targeted drug development. For METTL1, early inhibitor compounds has now been reported. A docking-guided and enzymatic assay strategy identified the first small-molecule METTL1 inhibitors, and multiple hit chemotypes showed micromolar activity (approximately 40–300 μM)117. A probe-enabled high-throughput screen yielded additional micromolar hits, thus further supporting METTL1 as a tractable target for early discovery118. In contrast, to our knowledge, no specific, biologically active small-molecule inhibitors targeting WBSCR22 have been reported in the relevant literature to date. However, major challenges persist in this field: selectivity remains difficult, cellular engagement is not ensured, and m7G modification also has physiological roles in normal cells. Therefore, an effective therapeutic strategy must achieve sufficient inhibition of oncogenic m7G-related pathways in tumour cells while minimizing disruption of physiological m7G functions in normal tissues. In practice, because tumor heterogeneity is also likely to produce variable responses, biomarker-guided stratification is necessary. Concurrently, the regulatory role of m7G methylation within the tumor immune microenvironment is increasingly evident, particularly in its influence on the recruitment and function of immune cells such as MDSCs and regulatory T cells. These findings offer novel avenues for combination immunotherapies. The integration of such epigenetic regulation with immunotherapy might provide a key future strategy for overcoming tumor immune suppression. At the clinical translation level, m7G methylation analysis increasingly shows diagnostic and prognostic value. High-throughput sequencing has identified multiple GI cancer-specific m7G methylation sites, and these biomarkers exhibit high sensitivity and specificity in early diagnosis, thereby complementing existing detection methods. Future efforts should integrate these markers into clinical diagnostic workflows to establish standardized testing methods for screening high-risk populations and early diagnosis. Concurrently, constructing prognostic models based on m7G methylation profiles, by predicting patient relapse risk and treatment response, could provide a basis for personalized therapeutic planning. Furthermore, establishing dynamic monitoring platforms for m7G modifications would enable real-time assessment of treatment efficacy, by guiding protocol adjustments to enhance therapeutic precision. This research paradigm, which closely integrates fundamental discoveries with clinical needs, is advancing m7G methylation from the laboratory toward clinical application.
Exploring the therapeutic modulation of m7G modification requires advances on multiple fronts. Optimizing the structural design of small-molecule inhibitors to enhance specificity and bioavailability is essential, and novel targeting technologies such as PROTACs must simultaneously be investigated to achieve degradation-based regulation of modifying enzymes. Concurrently, combined therapeutic strategies leveraging the interaction between m7G and the immune microenvironment should be developed. For example, after robust METTL1 inhibitors or degraders become available, rational combinations with immune checkpoint blockade could be tested to evaluate whether they relieve tumor-associated immunosuppression in defined contexts. Furthermore, to target downstream genes affected by abnormal m7G modification, nucleic acid therapeutics could be designed for precise intervention, thereby establishing a comprehensive enzyme-target-drug regulatory chain.
In summary, m7G methylation research is bringing new hope for cancer treatment, given its unique perspectives and broad application prospects. From elucidating the underlying mechanisms to advancing clinical translation, research in this field has not only enriched understanding of epigenetic regulation but also provided a feasible path to improve patient outcomes and quality of life. As research continues to advance, m7G methylation is expected to become an indispensable component of precision oncology.
Conflict of interest statement
No potential conflicts of interest are disclosed.
Author contributions
Topic and drawing: Xianli Shi, Rui Li, Xiaoqing Lu.
Writing: Rui Li, Xiaoqing Lu.
Revision and proofreading: Rui Li, Rongxin Zhang.
Design and revision: Rui Li, Rongxin Zhang, Zhiyu Guan.
Acknowledgments
We thank all Zhang laboratory members for advice and support on the preparation of this work.
- Received November 17, 2025.
- Accepted March 12, 2026.
- Copyright: © 2026, The Authors
This work is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License.
References
- 1.↵
- 2.
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.
- 43.
- 44.
- 45.↵
- 46.
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.
- 84.
- 85.
- 86.
- 87.
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.
- 94.↵
- 95.↵
- 96.
- 97.↵
- 98.↵
- 99.
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.↵
- 106.↵
- 107.↵
- 108.↵
- 109.↵
- 110.↵
- 111.↵
- 112.↵
- 113.↵
- 114.↵
- 115.↵
- 116.↵
- 117.↵
- 118.↵











