The m⁵C demethylase TET1 is identified as a key regulator of scleral remodeling in myopia

Abstract

Background

The global prevalence of myopia has risen sharply, particularly among children and adolescents in urban and highly educated populations. RNA modifications have been recognized as important regulators in myopia progression. However, the role of m⁵C, a crucial RNA modification, remains largely unexplored in myopia progression.

Methods

Key RNA modification-related genes were first identified through weighted gene co-expression network analysis (WGCNA), based on expression profiles of m⁶A‑, m¹A‑, m⁵C‑, and m⁷G‑associated genes from the mouse myopic sclera dataset PRJCA000717. Pathway enrichment analysis was performed via Gene Set Enrichment Analysis (GSEA), and immune infiltration characteristics were evaluated using CIBERSORT. Further experimental validation included RT-qPCR and Western blotting to confirm TET1 up-regulation in a hypoxia-induced model of human scleral fibroblasts (HSFs), and functional assessments via knockdown experiments.

Results

Utilizing WGCNA, we identified Tet1, an m⁵C demethylase, as a key gene associated with myopia. Its expression was significantly elevated in myopic scleral samples. We further confirmed the increased TET1 expression in hypoxia-induced HSFs. TET1 was found to be involved in pivotal biological processes and the scleral microenvironment. Immune cell infiltration analysis revealed a significant negative correlation between Tet1 and eosinophils. GSEA indicated that Tet1‑related pathways include the proteasome, HIF‑1 signaling, and TGF‑β signaling. Knockdown of TET1 reversed hypoxia‑induced increases in apoptosis, restored m⁵C modification levels, and promoted HSF proliferation. Additionally, TET1 down‑regulation elevated COL1A1 protein expression while reducing α‑SMA, a key marker of scleral remodeling. Data from the GeneCards database showed that Tet1 expression significantly correlated with several myopia‑related progression genes, including Col2a1 (p = 0.002) and Fbn1 (p < 0.001).

Conclusion

Our study reveals substantial alterations in TET1 expression and m⁵C modification in myopic sclera, providing new insights that may inform the development of novel diagnostic and therapeutic approaches for myopia management.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13148-026-02099-9.

Introduction

Myopia, a leading cause of visual impairment worldwide, is characterized by blurred distance vision while near vision remains clear. In recent decades, its global prevalence has surged dramatically, particularly among children and adolescents in urban and educated populations. [1] This trend is concerning, as high myopia elevates the lifelong risk of irreversible visual impairment or blindness from conditions such as retinal detachment, glaucoma, and myopic maculopathy [2, 3].

The pathogenesis of myopia is complex, involving interactions between genetic predispositions and environmental factors that influence ocular growth. A strong heritable component is well-established, with research confirming that variations in genes regulating scleral remodeling and retinal signaling can predispose individuals to myopia [2]. Environmental factors, particularly visual demand and lighting, modulate this risk. Sustained near-work activities may promote axial elongation, whereas increased time outdoors has been consistently shown to protect against myopia onset and progression [3–5].

Epigenetics serves as a bridge between life experiences and phenotypic outcomes, referring to heritable modifications that regulate gene expression without altering the DNA sequence [6]. RNA methylation, a key post-transcriptional modification, is catalyzed by RNA methyltransferases and occurs at various nucleotide positions. Prominent modifications include N7-methylguanosine (m⁷G), N6-methyladenosine (m⁶A), 5-methylcytidine (m⁵C), and N1-methyladenosine (m¹A), all of which play pivotal roles in modulating gene expression, cellular differentiation, and disease pathogenesis. Wen et al. compared the expression profiles of m⁶A-related regulatory factors in the anterior capsules between patients with simple nuclear cataract and those with high myopia-associated nuclear cataract. They found that ALKBH5 expression was downregulated in the latter group, and differentially methylated genes were enriched in the extracellular matrix (ECM) pathway [7]. This highlights the close association between RNA modifications and myopia development.

Alterations in the sclera constitute a crucial pathophysiological basis for myopia progression. ECM remodeling leads to characteristic scleral thinning, reduced biomechanical strength, and axial elongation. In this study, we comprehensively profiled the expression of regulators associated with m⁶A, m¹A, m⁵C, and m⁷G in myopia scleral samples. Using weighted gene co-expression network analysis (WGCNA), we identified the m⁵C demethylase Tet1 as a key gene and further investigated its biological functions and immune correlations. By establishing a hypoxia-induced model of human scleral fibroblasts (HSFs), we demonstrated that elevated TET1 expression reduces m⁵C modification levels in scleral cells and accelerates apoptosis. To our knowledge, this work is the first to report altered TET1 expression and m⁵C modification in myopic scleral samples, potentially offering new perspectives for the diagnosis and treatment of myopia.

Materials and methods

Data download

We obtained the myopic sclera-related single-cell RNA sequencing (scRNA-seq) dataset (PRJCA000717, derived from a mouse model of experimental myopia) [8] from the National Genomics Data Center (https://ngdc.cncb.ac.cn/gsa/browse/CRA000775). The gene expression data analyzed in this study originated from a published scRNA-seq dataset: Wu et al. induced experimental myopia in male C57BL/6 mice through monocular form deprivation (FD). Two days after FD, the entire scleral tissue from each eye was separately collected for scRNA-seq analysis. The sclera from the eye subjected to FD-induced myopia was considered the myopic sample, while the sclera from the contralateral untreated eye served as the control sample [8]. Although the original data capture transcriptional heterogeneity at single-cell resolution, the data were analyzed here in an aggregate, pseudo-bulk manner. Specifically, gene expression values across all cells within each sample were averaged/summed to generate a sample-level expression profile, which was then used for subsequent transcriptomic analyses, including differential expression and pathway enrichment.

WGCNA analysis

By constructing a WGCNA, we can identify co-expressed gene modules, explore associations within gene networks, and pinpoint key genes within these networks. The WGCNA-R package was employed to build the co-expression network for all genes in the dataset. Using this algorithm, genes were screened for further analysis with the soft threshold set to three. The weighted adjacency matrix was transformed into a topological overlap matrix (TOM) to estimate network connectivity, and a hierarchical clustering method was applied to construct the clustering tree structure of the TOM matrix. Different branches of the clustering tree represent distinct gene modules, with different colors indicating different modules. Based on the weighted correlation coefficients of genes, genes were categorized according to their expression patterns; genes with similar patterns were grouped into a single module, and all genes were ultimately partitioned into multiple modules based on their expression profiles.

Gene function enrichment analysis

We utilized the Metascape database to perform functional annotations on the module genes, enabling a comprehensive exploration of their functional relevance. Gene ontology (GO) pathway analysis was conducted on the specified gene sets, applying screening criteria of a minimum overlap of ≥ 3 and a p-value of ≤ 0.01.

Immune cell infiltration analysis

The CIBERSORT method is a widely utilized tool for assessing immune cell composition within tissue microenvironments. Based on the principle of support vector regression, it performs deconvolution analysis on gene expression matrices to estimate the relative abundances of immune cell subtypes. In this study, leveraging the mouse scRNA-seq dataset obtained from the National Genomics Data Center (PRJCA000717) [8], we applied the CIBERSORT algorithm to perform deconvolution analysis on the immune cell composition of each individual sample. This was achieved by first annotating cell types based on the scRNA-seq data and then calculating the proportion of each cell type per sample, thereby revealing the overall differences in immune infiltration patterns between the two groups. Subsequently, Pearson correlation analysis was conducted to examine the relationship between the expression of specific genes and the inferred immune cell abundances.

Gene set enrichment analysis (GSEA)

The differences in signal pathways between the high and low expression groups were further analyzed by Gene set enrichment analysis (GSEA). The background gene set was downloaded from the MsigDB database as the annotation gene set of the subtype pathway, and the differential expression analysis of the pathways between subtypes was performed. The significantly enriched gene sets (adjusted p < 0.05) were ranked according to the consistency score.

Regulatory network analysis and drug interaction prediction analysis

The Cistrome DB database, which integrates ChIP-seq and DNase-seq data, was used to investigate the regulatory relationships between transcription factors and key genes. The genome assembly was set to hg38, and the region analyzed extended 10 kb from the transcription start site. The resulting network was visualized using Cytoscape. Additionally, miRNAs targeting the key genes were identified using the TargetScan database, and their regulatory networks were similarly constructed and visualized with Cytoscape. The Comparative Toxicogenomics Database (CTD) was utilized to predict potential drugs based on interactions between key genes and chemicals, as it provides manually curated data on chemical-gene, chemical-disease, and gene-disease relationships. The gene-drug interaction network derived from this analysis was also visualized using Cytoscape.

Cell culture and hypoxic treatment

HSFs (Aiyou Biotechnology Center, Shanghai, China) were cultured in DMEM (Gibco) supplemented with 10% fetal bovine serum (FBS; Gibco), 100 U/mL penicillin, and 100 mg/mL streptomycin, and maintained at 37℃ in a humidified atmosphere of 5% CO₂. After vimentin and keratin identification, HSFs at the third to fourth passages were adopted for the experiments.

The hypoxic treatment in this study was conducted following the procedure described previously by Wu [8] HSFs were maintained under normoxic conditions (21% O₂) or exposed to hypoxia (5% O₂) for 1, 3, 6, or 9 h. Upon reaching the designated time points, cells were lysed for subsequent experimental analysis.

For transfection, HSFs in the logarithmic growth phase were trypsinized (0.25% trypsin), counted, and seeded into 6-well plates. When cells reached approximately 60% confluency, they were transfected with siRNA targeting TET1 (si-TET1) or a non-targeting control siRNA (si-NC) using Lipofectamine 3000 (Invitrogen, Thermo Fisher Scientific) according to the manufacturer’s protocol. Following transfection, cells were exposed to either normoxic or hypoxic conditions for subsequent experiments. siRNA sequences are listed in Supplementary Fig. 1.

RNA isolation and quantitative real-time PCR

Total RNA was extracted from cultured cells using the EZpress RNA Purification Kit (B0004) following the manufacturer’s instructions. Complementary DNA (cDNA) was synthesized with the PrimeScript RT Reagent Kit (TaKaRa Bio, Otsu, Japan). Quantitative real-time PCR was performed using PowerUp SYBR Green Master Mix (Life Technologies) on a real-time PCR system (Applied Biosystems, Irvine, CA, USA). ACTB was used as the internal control for normalization. Primer sequences are listed in Supplementary Fig. 1.

Western blotting assay

HSFs were lysed in RIPA lysis buffer (Biosharp, Hefei, China, BL504A) and centrifuged at 13,000×g for 30 min at 4℃. Then, protein samples were separated by 7.5% (wt/vol) sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to polyvinylidene fluoride membranes (Millipore Corporation, Billerica, MA, USA). After blocking with 5% milk for 1 h at room temperature (RT), the membranes were incubated with primary antibody against TET1 (A21914, Abclonal), COL1A1 (67288-1-Ig, Proteintech), α-SMA (14395-1-AP, Proteintech), Paxillin (10029-1-Ig, Proteintech), Vinculin (ab129002, Abcam), and ACTB (30101ES60, Yeasen) at 4℃ overnight. The membranes were then incubated with secondary antibodies. The membranes were scanned to acquire western blotting images, and the bands were quantified in grayscale using Image J (National Institutes of Health, Bethesda, MD, USA).

Dot blot assay

Total RNA was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), adhering strictly to the manufacturer’s standard protocol. Predetermined amounts of RNA samples were then loaded onto Hybond-N+ membranes (FFN10, Beyotime, China). The nylon membranes were UV-crosslinked, followed by blocked with 5% milk for 1.5 h at RT. Next, the membrane was incubated overnight at 4 °C with an anti-m⁵C antibody (ab214727, Abcam). After thorough rinsing three times with TBST (Thermo Fisher Scientific, USA), the membrane was incubated with HRP-conjugated anti-rabbit IgG (SA00001-2, Proteintech) for an additional 1.5 h at RT. Specific antibody-antigen complexes were detected using an enhanced chemiluminescence kit (WBKLS0100, Thermo Fisher Scientific).

Cell counting Kit-8 assay

HSFs were seeded into 96-well plates (Corning) at a density of 5,000 ~ 6,000 cells per well in 100 µL of complete medium. Cell proliferation was assessed at 0, 24, 48, and 72 h using the Cell Counting Kit-8 (CCK-8; Dojindo) according to the manufacturer’s instructions. Briefly, 10 µL of CCK-8 reagent was added to each well, followed by incubation for 3 ~ 4 h at 37 °C. Absorbance was measured at 450 nm, and growth curves were plotted.

Apoptosis assay

Apoptosis was evaluated using the FITC Annexin V Apoptosis Detection Kit (BD Biosciences) according to the manufacturer’s instructions. Briefly, cells were washed twice with cold PBS and stained with FITC-Annexin V and propidium iodide (PI) for 5 min on ice. Samples were analyzed immediately using a BD LSRFortessa flow cytometer.

Statistical analysis

Statistical analyses were performed using GraphPad Prism (version 10.0) and R software (version 4.3.0). All statistical tests were two-sided, and a p-value < 0.05 was considered statistically significant.

Results

WGCNA analysis and key genes screening

The workflow of this study is illustrated in Fig. 1. We first constructed a WGCNA network using the scRNA‑seq dataset derived from the sclera of a FD‑induced myopia mouse model (PRJCA000717), setting the soft threshold β to 3 (Fig. 2A). Gene modules were then identified based on the TOM, yielding five distinct modules (Fig. 2B, Supplementary Table S1): blue (362 genes), brown (356 genes), grey (428 genes), turquoise (2,808 genes), and yellow (345 genes). We further analyzed the relationship between modules and traits and found that the yellow module had the highest correlation (p = 0.01) (Fig. 2C). Next, we used the yellow module genes to perform pathway analysis using the Metascape database. The results showed that these genes were mainly enriched in pathways such as response to peptide, protein-DNA complex disassembly, response to steroid hormone, endomembrane system organization, and intracellular protein transport (Fig. 2D), suggesting that the yellow module likely represents a critical set of co-expressed genes involved in the pathological mechanisms of myopia.

Fig. 1.

The design and flow chart of this study

Fig. 2.

The identification of Tet1 as a key gene in myopia. A Scale-free index and average connectivity of each soft threshold. B Dendrogram of gene clustering, different colors represent different modules. C Heat map of the correlation between module characteristic genes and diseases, blue indicates negative correlation and red indicates positive correlation. D GO-KEGG enrichment analysis using Metascape database. E The expression status of regulators associated with m⁶A, m¹A, m⁵C, and m⁷G in control and disease samples. F Venn diagram showing the overlap between the characteristic genes of the yellow module identified by WGCNA and the regulatory genes associated with m⁶A/m⁵C/m⁷G/m¹A modifications. G Correlation circle diagram of important genes. H Differential expression of important genes in control and disease samples, blue represents control samples, pink represents disease samples (statistical tests were labeled: ns, P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001)

In order to identify the RNA modification related key genes that affect myopia, we then obtained the genes related to methylated m⁶A/m⁵C/m⁷G/m¹A from the references [9–12]. Their expression profiles in myopia and control samples were analyzed, and differentially expressed genes (DEGs) were identified (Fig. 2E). Intersection of these key genes related to RNA modification with the yellow module genes yielded two overlapping genes: Mbd2 and Tet1 (Fig. 2F). A co-expression network was constructed based on the expression levels of Mbd2 and Tet1 through correlation analysis (Fig. 2G). Of these two genes, only Tet1 showed significantly elevated expression in the myopia group (Fig. 2H). Therefore, Tet1 was selected as the key gene for further investigation.

Increased TET1 expression in hypoxia-treated HSFs.

We treated HSFs with hypoxia. Immunofluorescence staining confirmed the identity of the cultured cells as HSFs, showing positivity for vimentin and negativity for keratin (Fig. 3A). Following hypoxia exposure, myofibroblast transdifferentiation and collagen production were assessed (Fig. 3B). After 9 h of hypoxia, protein levels of α‑SMA and focal adhesion proteins (vinculin and paxillin) were significantly increased, while COL1A1 expression was markedly reduced, confirming successful establishment of the myopic cell model (Fig. 3C). Subsequent Western blot and RT‑qPCR analyses demonstrated that TET1 expression was upregulated in hypoxia‑treated HSFs (Fig. 3D and E), consistent with the bioinformatics findings.

Fig. 3.

Validation of up-regulated TET1 in hypoxia-treated HSFs. A Immunofluorescence evaluation of vimentin and keratin in HSFs. B, C Western blotting examination of α-SMA, focal adhesion proteins (vinculin and paxillin) and COL1A1 in hypoxia-treated HSFs. D, E Western blotting examination and RT-qPCR examination of TET1 expression in hypoxia- and normoxia-treated HSFs (statistical tests were labeled: ns, P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, N = 3)

Immune infiltration analysis

The microenvironment is mainly composed of immune cells, ECM, various growth factors, inflammatory factors and special physical and chemical characteristics, which significantly affects disease pathogenesis and treatment response. Firstly, in order to investigate the immune microenvironment of the myopic sclera, we performed deconvolution analysis on the immune cell composition of each sample using the CIBERSORT algorithm based on the mouse scRNA-seq dataset. Figure 4A illustrates the relative percentages of 25 immune cell subsets in the myopic group and the control group. Figure 4B presents the correlations among different immune cell types. Compared with controls, myopic scleral samples showed significant enrichment of activated dendritic cells (DCs) (Fig. 4C). To further investigate the role of Tet1 in myopia progression, we examined its correlation with immune cell infiltration. Tet1 expression was significantly negatively correlated with eosinophil infiltration (p = 0.04, Supplementary Fig. 1B). Using the TISIDB database, we also analyzed correlations between Tet1 and immunomodulatory factors, including immunostimulators, chemokines, major histocompatibility complex (MHC) molecules, and receptors (Fig. 5A-D). These results suggest that Tet1 is closely associated with immune cell infiltration and may influence the immune microenvironment in myopia.

Fig. 4.

Immune infiltration analysis. A Relative percentage of immune cell subsets. B Correlation between immune cells, blue indicates negative correlation, red indicates positive correlation. C Difference in immune cell content between control samples and disease samples. (statistical tests were labeled: ns, P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001)

Fig. 5.

Relationship between Tet1 and immune factors. A–D Correlation between Tet1 and chemokine, immunostimulator, MHC and receptor

Tet1-specific signaling pathways and metabolic pathways

We next investigated Tet1‑related signaling pathways to explore its potential molecular mechanisms. GSEA indicated that Tet1 expression was associated with proteasome activity, HIF‑1 signaling, and TGF‑β signaling pathways (Fig. 6A and B).

Fig. 6.

GSEA analysis of Tet1. A, B KEGG signaling pathways involved in Tet1, as well as pathway regulation and genes involved. C Heat map of the correlation between Tet1 and metabolic pathways, blue indicates low expression and red indicates high expression

Additionally, correlation analysis between Tet1 and metabolic pathways was performed. A heatmap of metabolic signatures revealed that samples with high Tet1 expression exhibited enriched activity in amino acid metabolism, lipid metabolism, drug metabolism, and C3-specific metabolic pathways. Metabolic scores were significantly higher in myopic samples than in controls (Fig. 6C).

TET1 suppresses cell proliferation and decreases m⁵C modification level

To determine whether TET1 influences HSF proliferation under hypoxia, we transfected HSFs with si-TET1 and confirmed knockdown efficiency (Fig. 7A and B). Given that TET1 encodes an m⁵C demethylase [13], we evaluated global m⁵C RNA modification levels via dot-blot assay. Hypoxia treatment reduced overall m⁵C modification, an effect that was reversed upon TET1 knockdown (Fig. 7C). We then assessed the impact of TET1 knockdown on fibrotic remodeling markers. Western blot analysis showed that si-TET1 transfection decreased α‑SMA and increased COL1A1 protein levels in hypoxia‑exposed HSFs (Fig. 7D). CCK‑8 and flow cytometry assays indicated that hypoxia inhibited proliferation and promoted early apoptosis in HSFs, whereas TET1 knockdown mitigated these effects (Fig. 7E and G). Together, these findings suggest that TET1 inhibits proliferation, promotes early apoptosis, and reduces global m⁵C RNA modification in hypoxia‑treated HSFs.

Fig. 7.

TET1 silencing increase the m⁵C modification levels and abrogate the influences of hypoxia treatment on proliferation and apoptosis of HSFs. A, B Assessment of TET1 knockdown in HSFs by RT-qPCR and Western blotting. C The m⁵C dot intensity of HSFs treated with normoxia, hypoxia, or hypoxia + si-TET1. D Western blotting examination of myofibroblast transdifferentiation and collagen production in HSFs treated with normoxia, hypoxia, or hypoxia + si-TET1. E CCK-8 assay to determine the proliferation of HSFs treated with normoxia, hypoxia, or hypoxia + si-TET1. F, G Flow cytometry to determine the apoptosis of HSFs treated with normoxia, hypoxia, or hypoxia + si-TET1 (statistical tests were labeled: ns, P > 0.05; *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001, N = 3)

Analysis of transcriptional regulation, correlation between Tet1 and myopia progression genes, and drug interaction network.

To elucidate the transcriptional regulation of Tet1, we used the Cistrome DB toolkit to predict potential transcription factors, identifying 65 candidates. A regulatory network was constructed and visualized using Cytoscape (Fig. 8A). In addition, 38 miRNAs targeting Tet1 were predicted via the TargetScan database, and the mRNA-miRNA interaction network was similarly visualized (Fig. 8B).

Fig. 8.

Regulation analysis of Tet1. A, B Transcriptional regulation related to Tet1 and miRNA network. A Transcriptional regulation related to Tet1, red represents key genes, green represents transcription factors. B miRNA network of Tet1, yellow represents mRNA, blue represents miRNA. C, D Correlation between Tet1 and disease progression genes. C Expression differences of disease regulatory genes, blue represents control samples, pink represents disease samples. D Correlation analysis between Tet1 and disease genes, blue represents negative correlation, red represents positive correlation

Myopia progression-associated genes were retrieved from the GeneCards database. Expression analysis of the top 20 genes ranked by relevance score revealed significant differential expression for Fbn1 and Plod1 (Fig. 8C). Moreover, Tet1 expression correlated significantly with several progression genes—negatively with Col2a1 (p = 0.002) and positively with Fbn1 (p < 0.001) (Fig. 8D). Moreover, we used the CTD database to analyze drugs that may interact with the Tet1 gene. Six drugs were identified as interacting with Tet1, suggesting possible avenues for therapeutic targeting; these interactions were visualized as a network (Supplementary Fig. 1C).

Discussion

Our study comprehensively profiled regulators of m⁶A, m¹A, m⁵C, and m⁷G modifications in the myopic sclera and identified the m5C demethylase Tet1 as a key regulatory gene. We further investigated its biological functions and immune characteristics. Elevated TET1 expression was found to reduce global m⁵C modification levels in scleral cells and promote apoptosis. To our knowledge, this is the first study to focus on altered m⁵C modification in myopia, providing novel insights into its pathogenesis and potential treatment strategies.

RNA modifications, such as m⁶A, have been reported to participate in myopia progression. For instance, comparison between patients with high myopic nuclear cataract and those with simple nuclear cataract revealed reduced expression of ALKBH5 and FTO, along with differential methylation of genes enriched in ECM-related pathways [7] Another recent study demonstrated that inhibiting METTL3 transcription and reducing m⁶A-modified transcripts recognized by YTHDF2 can enhance APOA1 mRNA stability and transcription, thereby contributing to scleral remodeling [14] While m⁵C modification levels are regulated by methyltransferases, demethylases, and reader proteins, their role in myopia remains poorly understood. In this study, we observed increased expression of the m⁵C demethylase TET1 in a myopic cell model, which in turn reduced global m⁵C modification levels and promoted scleral cell apoptosis. Our findings suggest that m⁵C modification plays a crucial role in maintaining scleral homeostasis.

The immune microenvironment is also known to contribute to myopia progression. Compared with non-myopic eyes, myopic eyes exhibit elevated levels of inflammatory cytokines in the aqueous humor and vitreous. For example, increased expression of interferon‑γ (IFN‑γ), IL‑6, eotaxin, monocyte chemoattractant protein‑1 (MCP‑1), macrophage inflammatory protein‑1α (MIP‑1α), and MIP‑1β has been reported in the vitreous of patients with high myopia and in highly myopic eyes with macular holes [15, 16]. Elevated MIP levels—originating from macrophages, DCs, and lymphocytes—indicate active involvement of immune cells in myopia development. Moreover, MCP-1 expression is significantly higher in the aqueous humor of patients with high myopic cataract compared to those with age-related cataract [17]. In line with these observations, our study revealed enrichment of activated dendritic cells in myopic scleral tissues. Consistent with previous reports that scleral immune cells help maintain ocular function and vascular homeostasis, [18] we speculate that activated DCs may disrupt local microenvironmental balance and thereby promote myopia progression.

RNA modifications are known to influence the biology of DCs, monocytes, and macrophages, as well as modulate inflammatory cytokine secretion [19, 20]. In particular, m⁵C RNA modification has been shown to shape immune cell function and subsequent inflammatory responses within the microenvironment [21] For example, For example, DCs exposed to m⁵C-modified RNAs exhibit reduced cytokine production and activation marker expression, suggesting that nucleoside modifications can suppress the immunostimulatory potential of RNA [19] In microglia mediated neuroinflammation, increased TET1 expression reduces m⁵C RNA modification levels, promoting M1 polarization while inhibiting M2 polarization [22]. Based on these findings, we propose that TET1 may regulate immune responses in myopia progression through an m⁵C-dependent mechanism.

The hypoxic microenvironment has emerged as a compelling pathophysiological mechanism in myopia progression. Clinical observations consistently indicate significant choroidal thinning, reduced choroidal blood flow and partial pressure of oxygen in highly myopic eyes, strongly suggesting the presence of chronic scleral hypoxia [23, 24]. This hypoxic state is postulated to activate key signaling pathways, most notably those mediated by HIF-1α—a critical transcription factor upregulated in the myopic sclera that orchestrates cellular adaptation to low oxygen conditions [25]. Its upregulation has been linked to the expression of factors such as vascular endothelial growth factor (VEGF), which contributes to vision-threatening complications including choroidal neovascularization [8] More fundamentally, hypoxia-driven signaling appears to directly influence scleral fibroblasts, disrupting the equilibrium of ECM synthesis and degradation. This imbalance promotes scleral remodeling, thinning, and a loss of biomechanical integrity—structural alterations central to axial elongation in myopia. Supporting this mechanistic view, animal studies demonstrate that inhibiting HIF-1α in the sclera induces hyperopia, whereas its upregulation promotes myopic changes [25].

The TGF-β signaling pathway plays a central role in regulating scleral remodeling during myopia progression. Specifically, TGF-β2 influences myopic development by modulating collagen production [26]. In scleral fibroblasts, collagen metabolism is regulated by TGF-β1, and both TGF-β1 and type I collagen expression gradually decline during myopia-associated scleral remodeling [27, 28]. Evidence also suggests that the TGF-β1 pathway upregulates type I collagen synthesis via transcription factor-specific protein 1, thereby contributing to the pathological progression of myopia [29]. In this study, we demonstrate that Tet1 is involved in both the HIF-1α and TGF-β signaling pathways, offering novel insights into the molecular mechanisms driving myopia and identifying potential therapeutic targets.

Emerging research on TET proteins reveals an increasingly complex picture, demonstrating that their functions extend far beyond their canonical role, forming a sophisticated regulatory network across multiple layers including epigenetics and cellular signaling. Traditionally TET proteins are recognized as key executors of active DNA demethylation. The study by Arroyo et al. provides a detailed mechanistic insight into this classical function, showing that different isoforms of TET1 (TET1 and TET1s) achieve spatiotemporal-specific nuclear localization and catalytic targeting through distinct structural domains and post-translational modifications [30] Recent studies have further expanded the functional repertoire of TET proteins to the realm of RNA modifications. Hastert et al. demonstrated that TET1/2/3 localize to splicing speckles, interact with splicing factors, and enhance RNA splicing efficiency in a manner independent of their catalytic activity, suggesting their role as structural components within RNA processing complexes. Concurrently, their catalytic activity (oxidizing RNA-m⁵C) can also restore splicing efficiency in vitro, highlighting the potential for direct chemical modification-mediated regulation of RNA processing [31]. In the context of DNA damage repair, TET1 acts as an “eraser” of RNA-m⁵C, coordinating with the “writer” TRDMT1 and the “reader” FMRP to collaboratively facilitate the completion of DNA repair [13]. The functional scope of TET proteins also extends to cellular responses to external physical signals. In vascular endothelial cells, TET1s mediates hemodynamic shear stress-induced endothelial planar cell polarity by modulating the Wnt signaling pathway and regulating F-actin polymerization [32]. This discovery links the function of TET1 to cellular mechanosensing and cytoskeletal reorganization mechanisms, providing critical clues for elucidating the biomechanical etiology of myopia. Future studies are warranted to employ in vivo models to further investigate the specific mechanisms by which m⁵C-related regulatory factors contribute to scleral remodeling in myopia, thereby facilitating the translation of these findings into clinical applications.

In conclusion, to our knowledge, this study presents the first evidence of differential TET1 expression and reduced m⁵C modification levels in hypoxia-induced HSFs. However, several limitations should be acknowledged. First, the analysis was based on publicly available single-cell RNA-seq data with a limited sample size. Furthermore, the transcriptomic data used in this study were optimized for quantifying gene expression levels and lack the junction-spanning read coverage necessary for reliably detecting alternative splicing events. Additionally, in vivo intervention studies are needed to assess the therapeutic potential of targeting TET1. Second, the specific genomic sites of m⁵C modification and the precise molecular mechanisms by which TET1 regulates m⁵C dynamics remain to be elucidated. Finally, the broader functional consequences of m⁵C alterations—such as effects on RNA stability, translation efficiency, and other cellular processes—are still unclear and merit further investigation. Elucidating these aspects will be crucial for understanding the full biological relevance of m⁵C modifications and their role in disease pathogenesis. Collectively, our findings suggest that RNA methylation regulation represents a novel and promising therapeutic target for myopia.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

SJ and YW were responsible for the conception and design of the study. SJ, LY and PW analyzed the data and wrote the manuscript. SJ, LY and PW conducted the experiment, collected samples. JY, GH and YW reviewed and revised the paper. All authors read and approved the final manuscript for submission.

Funding

This work was supported by the National Natural Science Foundation of China (82301240, 82271118, 82303106, and 12505410), the National Program on Key Research Project of China (2022YFC2404502), the Eye Institute of Nankai University (NKYKK202305), the Tianjin Health Research Project (TJWJ2023QN078), the Science and Technology Commission of Shanghai Municipality (24YF2723700), the China Postdoctoral Science Foundation (2024M752045), the Nankai University Optometry & Visual science Institute foundation (NKSGY202413) and the Tianjin Key Medical Discipline (Specialty) Construction Project (TJYXZDXK-016 A).

Data availability

The data used to support this study are included within the article.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors critically reviewed and approved the final manuscript.

Competing interests

The authors declare no competing interests.