METTL3-mediated m6A modification of TACSTD2 mRNA inhibits papillary thyroid cancer progression by activating the TNF signaling and blocking epithelial–mesenchymal transition Authors: Jianting Man, Zongyuan Zhang, Tianxing Cui, Tiefeng Shi Published: 2025 ABSTRACT AbstractPapillary thyroid cancer (PTC) represents a prevalent endocrine neoplasm. Dysregulated tumor-associated calcium signal transducer 2 (TACSTD2) expression is functionally involved in the progression of multiple cancers. Nonetheless, the biological molecular mechanisms underlying TACSTD2 in PTC remain unclear. To verify the biological function of TACSTD2 in PTC, we conducted loss- and gain-of-function experiments, namely, cell proliferation (Cell Counting Kit-8), EdU incorporation, colony formation, and transwell migration/invasion assays, along with establishing a nude mouse xenograft tumor model. Our findings revealed that TACSTD2 knockout impeded PTC cell proliferation, migration, and invasion in vitro and reduced tumorigenicity in vivo. In contrast, TACSTD2 overexpression reversed these effects in vitro. Through integrated bioinformatics analysis and experimental validation, TACSTD2 was recognized as a downstream target of methyltransferase-like 3 (METTL3)-modulated N6-methyladenosine (m6A) modification. In an m6A-dependent manner, METTL3 reduced TACSTD2 mRNA stability. m6A-mediated degradation of TACSTD2 mRNA was reliant on the m6A reader protein YTHDC2. As evidenced by rescue experiments, TACSTD2 overexpression resulted in a partial restoration of the METTL3-induced tumor-suppressive phenotype. Finally, the METTL3/TACSTD2/YTHDC2 axis was confirmed to be an inhibitor in PTC and activates the TNF signaling to promote cell apoptosis and inhibit the epithelial–mesenchymal transition. Collectively, METTL3-modulated m6A modification of TACSTD2 exerts an indispensable tumor-suppressive function in PTC progression. The METTL3/TACSTD2/YTHDC2 axis may serve as a molecular target for PTC therapy.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-30448-w.Keywords: Papillary thyroid cancer, TACSTD2, m6A, METTL3, YTHDC2Subject terms: Cancer, Cell biology, Molecular biology, Oncology Introduction Arising from thyroid follicles or the surrounding thyroid cells, thyroid cancer (THCA) constitutes a prevalent endocrine neoplasm worldwide1. Over the past few decades, its incidence has steadily increased globally. Papillary thyroid cancer (PTC), the most frequent THCA type, accounts for approximately 85%–90% of cases and generally responds well to treatment, resulting in a favorable prognosis2,3. However, approximately 1 in 10 patients with invasive PTC experience recurrence or distant metastasis within 10 years. Additionally, because of local recurrence and distant metastasis, patients with advanced PTC have a 5-year survival rate does not exceed 59%4,5. Research into molecular mechanisms provides valuable insights into the diagnosis and treatment of PTC. Accordingly, further elucidation of the molecular mechanisms behind PTC progression and effective therapeutic target identification is indispensable. Initially identified in human placental trophoblasts, tumor-associated calcium signal transducer 2 (TACSTD2), also termed trophoblast cell surface antigen 2, represents a transmembrane glycoprotein I6. As a cell surface receptor, TACSTD2 is essential in calcium signaling and activates the MAPK signaling pathway, facilitating cell cycle progression7. Studies show that TACSTD2 mRNA and/or protein levels are considerably elevated in colorectal, ovarian, gastric, breast, prostate, endometrial, and bladder cancers, relative to normal tissues8–11. Furthermore, overexpressed TACSTD2 is linked with poor overall survival rates in patients with gastric, colorectal, ovarian, and cervical cancers12–15, and is related to aggressive pathological features, namely, lymph node metastasis and enhanced invasiveness11,12,14,15. These findings suggest that TACSTD2 overexpression functions as an oncogenic driver in multiple solid tumors. Although the upregulation of TACSTD2 in THCA has been documented16, the role of TACSTD2 mRNA methylation and its association with clinical metastasis in THCA remain largely unexplored. THCA development is associated with various epigenetic and epitranscriptomic alterations. N6-methyladenosine (m6A) is the most prevalent internal RNA alteration in higher eukaryotes, occurring post-transcriptionally and being critical in regulating mRNA export, splicing, translation, localization, and stability17,18. The m6A alternation is dynamically governed by methyltransferases (“writers”), demethylases (“erasers”), and binding proteins (“readers”). Key methyltransferases include METTL3/14, and RBM15; demethylases include ALKBH5 and FTO; and readers comprise IGF2BP1/2/3, YTHDF1/2/3, and the heterogeneous nuclear ribonucleoprotein (hnRNP) family19. METTL3, an m6A methyltransferase complex core component, was the first enzyme to be identified as being essential for m6A modification. METTL3 regulates the progression of multiple cancers, including bladder20, gastric21, cervical22, colorectal23, and intrahepatic cholangiocarcinoma24; however, it exhibits tumor-suppressive functions in endometrial25 and triple-negative breast cancers26. These findings suggest that METTL3 may play context-dependent functions in diverse cancers. Recently, METTL3 has been found to repress PTC progression by modulating neutrophil infiltration, inhibiting the Hedgehog signaling pathway, and blocking epithelial–mesenchymal transition (EMT)27,28. Nevertheless, it is unclear whether METTL3 regulates the abundance of TACSTD2 via regulation of m6A-related mRNA processing. The current study focused on the implications of TACSTD2 in PTC tumorigenesis and dissected the cause of TACSTD2 dysregulation. Moreover, we elucidated the functional role of the METTL3/TACSTD2/YTHDC2 regulatory axis in PTC pathogenesis, uncovering a novel molecular mechanism. Therefore, TACSTD2 may be an innovative PTC prognostic predictor and therapeutic target. Results Pan-cancer analysis and differential expression of TACSTD2 in THCA The TIMER2 web tool was deployed to analyze the TACSTD2 expression profile in 32 human malignancies. As illustrated in Fig. 1A, TACSTD2 displayed significant differential expression across most cancer types compared to their corresponding normal tissues. TACSTD2 was markedly overexpressed in thyroid carcinoma tissues. To further characterize TACSTD2 expression patterns and clinical relevance in THCA, we integrated data from TCGA, GTEx, and GEO public datasets. Our combined analysis revealed significant upregulation of TACSTD2 mRNA in THCA samples (Figs. 1B–E). Additionally, HPA-retrieved IHC data confirmed significantly overexpressed TACSTD2 protein in THCA tissues than in normal tissues (Figs. 1F, G). Transcriptome sequencing was conducted on 4 pairs of PTC tissues and their corresponding neighboring non-tumor tissues, which manifested a clear TACSTD2 transcripts upregulation in tumor tissues (Fig. 1H). To expand the sample size for validation, we systematically analyzed tumors and paired adjacent tissues from 55 patients with PTC. RT-qPCR analysis of 25 of these samples further confirmed the overexpression of TACSTD2 in PTC (Fig. 1I), while 30 WB samples revealed elevated TACSTD2 protein levels in tumor tissues (Fig. 1J). Pan-cancer analysis and TACSTD2 differential expression in THCA. A: TACSTD2 differential expression in thyroid carcinoma and many other cancers. B–E: TCGA and GEO database (GSE3467, GSE60542): Differentially expressed TACSTD2 between PTC and normal controls. F–G: TACSTD2 protein levels in normal thyroid and THCA tissues. H: Volcano plot: Differential gene expression between THCA and adjacent normal tissues, n = 4, |log2(FC)|> 1 and P < 0.05. I: RT-qPCR: TACSTD2 expression levels in 25 paired PTC and normal tissues. J: WB assess TACSTD2 protein levels in 30 paired PTC and normal tissues. Silencing TACSTD2 reduces PTC cell migration, proliferation, and invasion in vitro RT-qPCR and WB demonstrated downregulated TACSTD2 in normal thyroid epithelium, whereas PTC cell lines showed a significant elevation in expression (Fig. 2A). Consequently, we carried out loss-of-function experiments in the B-CPAP and KTC-1 cell lines, which exhibited high TACSTD2 expression. To minimize off-target effects, two shRNAs specifically targeting TACSTD2 were designed and delivered into B-CPAP and KTC-1 cells via lentiviral vectors, ensuring effective gene silencing. The knockdown efficiency was confirmed via RT-qPCR and WB (Fig. 2B). Following TACSTD2 knockdown, cell proliferation was ascertained with CCK-8, colony formation, and EdU incorporation assays. Importantly, knocking down TACSTD2 significantly impeded PTC cells’ proliferative capacity (Figs. 2C–E). In addition, transwell and wound-healing assays demonstrated that knocking down TACSTD2 markedly reduced both the speed of cell migration and the invasive capacity (Figs. 2F–H). Conditioned medium from TACSTD2-silenced PTC cells significantly impaired tube formation in human lymphatic endothelial cells relative to the control (Fig. 2I). Additionally, TACSTD2 silencing significantly downregulated metastasis-related proteins, including MMP2, MMP3, MMP7, and MMP9 (Figs. 2J, K). Collectively, TACSTD2 knockdown significantly inhibits PTC cells proliferation, migration, and invasion in vitro. TACSTD2 loss-of-function impedes PTC cell proliferation and migration. A: RT-qPCR and WB verified TACSTD2 mRNA and protein expression in normal human thyroid epithelial cells and PTC cell lines. B: RT-qPCR and WB: The knockdown effect on TACSTD2 following B-CPAP and KTC-1 cell transfection with structured RNA targeting TACSTD2 (LV-sh-TACSTD2#1 and LV-sh-TACSTD2#2) or NC (Lv-sh-NC). C–E: Detection of B-CPAP and KTC-1 cell viability after TACSTD2 silencing assessed via CCK-8, colony formation, and EdU assays. F–H: Evaluating the migration ability of B-CPAP and KTC-1 cells after TACSTD2 silencing by transwell migration/invasion alongside wound healing assays. I: Influence of altered TACSTD2 levels in PTC cells on HLEC tube formation. J–K: WB analysis of metastasis-mediating proteins (MMP-2, MMP-3, MMP-7 and MMP-9) expression in PTC cells following pre-determined transfection. Gain-of-function TACSTD2 enhances migration, proliferation, and invasiveness in vitro Besides loss-of-function experiments, TACSTD2 was ectopically overexpressed in FTC-133 and TPC-1 cells, which exhibited relatively diminished endogenous TACSTD2 levels. The overexpression efficiency was confirmed using RT-qPCR and WB (Figs. 3A, B). Notably, TACSTD2 overexpression significantly enhanced both cell proliferation and colony formation (Figs. 3C, D). The EdU incorporation assay demonstrated that TACSTD2 overexpression significantly increased DNA synthesis (Fig. 3E) and improved FTC-133 and TPC-1 cell migration and invasion (Figs. 3F–H). Altogether, TACSTD2 promotes an aggressive tumor phenotype in PTC cells. TACSTD2 gain-of-function promotes PTC cell proliferation and migration. A-B: RT-qPCR and WB assess the activation effect on TACSTD2 following FTC-133 and TPC-1 cell transfection with pcDNA3.1-TACSTD2 or pcDNA3.1-NC. C–E: Determination of the viability of FTC-133 and TPC-1 cells after TACSTD2 overexpression through CCK-8, colony formation, and EdU assays. F–H: Evaluation of the migration ability of FTC-133 and TPC-1 cells after TACSTD2 overexpression by wound healing and transwell migration/invasion assays. TACSTD2 loss-of-function inhibits in vivo tumorigenicity In a subcutaneous xenograft mouse model (Fig. 4A), TACSTD2 knockdown reduced the tumor volume and final tumor weight, and slowed tumor growth (Figs. 4B, C, D). IHC staining confirmed the successful depletion of TACSTD2 in vivo (Fig. 4E). Additionally, Ki-67 and PCNA staining revealed markedly impaired proliferative activity in the TACSTD2-deficient tumor group (Fig. 4E). In summary, these findings demonstrate that TACSTD2 depletion suppresses the tumorigenic potential of PTC cells in vivo. Knockdown of TACSTD2 inhibits PTC xenograft growth in vivo. A: In vivo tumor metastasis assay: Subcutaneous injection of KTC-1 cells into the mice flanks, followed by multipoint intratumor injection with Lv-sh-TACSTD2 or Lv-sh-NC. B: Representative images of tumors in nude mice. Xenograft tumor weight (C) and growth curve (D) at 28 days after injection. E: IHC analysis of TACSTD2, Ki-67, and PCNA expression in xenograft sections. TACSTD2 loss-of-function suppresses tumors by activating TNF signaling pathway, promoting apoptosis and inhibiting EMT To identify downstream signaling pathways linked to TACSTD2 silencing, we performed mRNA sequencing, which revealed 737 upregulated and 247 downregulated genes following TACSTD2 knockdown (Fig. 5A). KEGG analysis29,30 of the DGEs showcased a strong positive relationship between reduced TACSTD2 and aberrant TNF signaling activation (Fig. 5B). The TNF signaling pathway is critical in apoptosis induction31. In vitro, WB analysis elucidated that knocking down TACSTD2 significantly upregulated key TNF pathway markers, including TNF-α, TNF-R1, TNF-R2 and TRADD (Figs. 5C, D). Furthermore, silencing TACSTD2 reduced the anti-apoptotic protein BCL-2, while significantly overexpressing pro-apoptotic markers, such as Bax, cleaved caspase-3/caspase-3, and cleaved caspase-9/caspase-9 (Fig. 5E). Having established that TACSTD2 suppression diminishes the invasive capacity of PTC cells and considering the pivotal role of EMT in metastasis32, we further investigated whether silencing TACSTD2 is involved in EMT. In vitro, WB analysis revealed that TACSTD2 knockdown reduced mesenchymal markers (N-cadherin, vimentin, Snail, and β-catenin) while overexpressing the epithelial marker E-cadherin (Fig. 5F). Collectively, TACSTD2 inhibition exerts tumor-suppressive impacts by activating TNF signaling pathway, thereby inducing apoptosis and suppressing EMT. Knockdown of TACSTD2 promotes apoptosis by activating the TNF signaling pathway and blocking EMT to inhibit PTC progression. A: Difference volcano plot of KTC-1 cells transfected with Lv-sh-NC and Lv-sh-TACSTD2#2, n = 3, |log2(FC)|> 1 and P < 0.05. B: KEGG analysis: TNF signaling pathway in PTC with TACSTD2-low expression was the top pathway based on mRNA-seq analysis. C–D: WB was used to measure TNF-α, TNF-R1, TNF-R2 and TRADD expression in PTC cells transfected with Lv-sh-NC and Lv-sh-TACSTD2#1/2. E: WB was used to measure Bax, Bcl-2, Caspase-3, cl-Caspase-3, Caspase-9 and cl-Caspase-9 expression in PTC cells transfected with Lv-sh-NC and Lv-sh-TACSTD2#1/2. F: WB was used to measure E-cadherin, N-cadherin, β-catenin, Snail, and Vimentin expression in PTC cells transfected with Lv-sh-NC and Lv-sh-TACSTD2#1/2. Furthermore, we further evaluated the effect of the knockdown of TNF-R1 and TRADD on the PTC cells with TACSTD2 knocked down. The knockdown efficiency was confirmed via WB (Fig S1, A). Functional rescue experiments showed that knockdown of TNF-R1 and TRADD partially reversed the inhibitory effect of TACSTD2 knockdown on cell proliferation in PTC cells (Fig S1, B), and can rescue the inhibitory effect of TACSTD2 knockdown on PTC cell migration and invasion (Figs S1, C-E). These results indicate that TACSTD2 knockdown can specifically activate the TNF signaling pathway. METTL3-regulated m6A modifications reduce TACSTD2 mRNA stability in a YTHDC2-dependent manner m6A modification represents a novel regulatory mechanism of gene expression that is crucial in tumor progression. Hence, we hypothesized that m6A modification participated in the post-transcriptional TACSTD2 regulation. To investigate this, we used the SRAMP prediction tool (http://www.cuilab.cn/sramp), which recognized putative m6A modification sites within the TACSTD2 transcript (Figs. 6 A, B). The m6A modification is primarily catalyzed by methyltransferase complexes—RBM15 and METTL3/14—and is reversed by demethylases, ALKBH5 and FTO. Then, the GEPIA2 database (http://gepia.cancer-pku.cn/) was accessed to analyze these methylation-related enzyme differential expression between PTC and normal thyroid tissues, showing significant METTL3 downregulation in PTC. In contrast, other enzymes exhibited no notable changes (Fig. 6C, Fig. S2A), consistent with our qRT-PCR validation (Fig. 6D). These findings suggest that METTL3 is a key m6A writer responsible for regulating TACSTD2, further reinforced by the inverse relation between TACSTD2 and METTL3 levels in the TCGA dataset (Fig. S2B). TACSTD2 expression was significantly decreased after METTL3 reintroduction, whereas it was enhanced with METTL3 knockdown in PTC cells at both mRNA and protein levels (Figs. 6E, F). Moreover, to validate whether TACSTD2 mRNA undergoes METTL3 modulated m6A modification, we performed MeRIP-qPCR, showing that the m6A abundance of TACSTD2 mRNA was significantly increased following METTL3 overexpression, while it was notably decreased upon METTL3 silencing (Fig. 6G). Subsequently, we used Actinomycin D (Act D) assay to deal with METTL3 overexpressed cells. The results found that overexpressed METTL3 shortened the half-life of TACSTD2 transcripts in B-CPAP and KTC-1 cells (Figs. 6H, I). It suggested that METTL3 reduced the stability of TACSTD2 mRNA. Taken together, all these results suggested that METTL3 reduced the stability of TACSTD2 mRNA in a m6A-dependent manner. METTL3 reduces TACSTD2 mRNA stability via an m6A-YTHDC2-dependent signaling. A: Prediction of m6A sites of TACSTD2 by SRAMP database. B: Analyzing m6A modification sites in the TACSTD2 coding sequence region through database. C: METTL3 expression level of THCA in GEPIA2 databases. D: RT-qPCR was used to evaluate the mRNA level adjusted to GAPDH of methyltransferases and demethylases in THCA and paired tissues. The expression levels of TACSTD2 in METTL3-overexpressing and METTL3 knockdown PTC cells were detected by WB (E) and qRT-PCR (F). G: MeRIP-qPCR analysis was employed to demonstrate METTL3-mediated TACSTD2 m6A modifications. The m6A modification of TACSTD2 was increased on upregulation of METTL3, while it was depleted on METTL3 silencing. H-I: TACSTD2 mRNA level in METTL3-overexpressing PTC cells treated with actinomycin D at the indicated time points were detected by qRT-PCR. METTL3 overexpression reduced the half-life of TACSTD2 transcript. J: YTHDC2 expression level of THCA in GEPIA2 databases. K: RT-qPCR was used to evaluate the mRNA level adjusted to GAPDH of N6-methyladenosine readers in THCA and paired tissues. L: The expression levels of TACSTD2 in YTHDC2-overexpressing PTC cells were detected by qRT-PCR. M: The expression levels of TACSTD2 in YTHDC2-overexpressing and YTHDC2 knockdown PTC cells were detected by WB. N: RIP-qPCR evaluated enrichment level of TACSTD2 in PTC cells after YTHDF2 knockdown. O: TACSTD2 mRNA level in YTHDC2 overexpression PTC cells treated with actinomycin D at the indicated time points were detected by qRT-PCR. YTHDC2 overexpression reduced the half-life of TACSTD2 transcript. P: WB was used to assess TACSTD2 level in PTC cells transfected with NC or OE-METTL3 with sh-NC or sh-YTHDC2. It is widely accepted that m6A modifications exert their biological effects through the recruitment of specific m6A “reader” proteins. Among the determined m6A readers, YTHDF2/3 and YTHDC2 have been shown to enhance mRNA degradation33–35. Integrating GEPIA2 database with our qRT-PCR results revealed that only YTHDC2 displayed differential expression in THCA and was negatively correlated with TACSTD2 expression (Figs. 6J, K, Fig S3). Consistent with our hypothesis, the overexpression of YTHDC2 significantly decreased TACSTD2 expression, whereas it was enhanced with YTHDC2 knockdown in PTC cells (Figs. 6L, M). A quantitative RIP assay was performed to assess the levels of TACSTD2 mRNA bound to YTHDC2 in PTC cells. TACSTD2 enrichment was drastically reduced in PTC cells upon the elimination of YTHDC2 expression (Fig. 6N). The Act D assay results found that overexpressed YTHDC2 shortened the half-life of TACSTD2 transcripts in PTC cells (Fig. 6O), suggesting that YTHDC2 can reduce the stability of TACSTD2 mRNA. Conversely, rescue experiments with knock downing YTHDC2 and overexpressing TACSTD2 revealed that silencing YTHDC2 significantly counteracted METTL3 induced TACSTD2 downregulation (Fig. 6P). To conclude, METTL3 modulated m6A modification facilitates TACSTD2 mRNA degradation in a YTHDC2-dependent manner. TACSTD2 overexpression reverses METTL3 tumor suppressor effect We further investigated the potential of TACSTD2 overexpression to modulate the tumor-suppressive effects of METTL3 in PTC. In B-CPAP and KTC-1 cells, METTL3 overexpression resulted in decreased TACSTD2 expression (Fig. 7A). The rescue experiments demonstrated that transient transfection with a TACSTD2-overexpression plasmid partially reversed METTL3’s inhibitory influences on cell proliferation (Figs. 7B–D). Additionally, TACSTD2 overexpression partially mitigated METTL3’s suppressive effects on PTC cell migration and invasion (Figs. 7E–G). Furthermore, we evaluated the effect of the METTL3–TACSTD2 axis on the TNF signaling pathway and apoptosis. The results suggested that METTL3 upregulation promoted TNF signaling pathway and apoptosis activation, while TACSTD2 overexpression partially inhibited the promotional effects on the TNF signaling pathway and apoptosis in METTL3 overexpression PTC cells (Figs. 7H–I). METTL3 inhibits PTC malignant progression by downregulating TACSTD2. A: B-CPAP and KTC-1 cell line transfection with OE-METTL3 plasmids and OE-TACSTD2 plasmids and METTL3 and TACSTD2 expression levels were quantified by WB. CCK-8 (B), colony formation (C) and EdU (D) assays to analyze METTL3 and TACSTD2 influence PTC cells proliferation. Transwell migration (E), invasion (F) and wound healing (G) assays to analyze METTL3 and TACSTD2 impacts on PTC invasion. H-I: Effect of METTL3/TACSTD2 axis on the expression of TNF signaling pathway-related proteins and apoptosis markers in PTC cells were quantified by WB assay. Discussion Molecular targeting of cell surface proteins is an emerging therapeutic strategy for cancer, inhibiting tumor proliferation and metastasis while promoting apoptosis by selectively targeting aberrantly overexpressed proteins during tumorigenesis36–38. In THCA, early detection, accurate diagnosis, appropriate management, and the identification of tumor biomarkers are essential for effective prognostic assessment. TACSTD2, a single-pass transmembrane protein, is highly expressed in various epithelial tumors39,40. Previous studies on TACSTD2 in cancer have primarily explored its roles in cell adhesion, signal transduction, and tumor cell proliferation41. Nonetheless, the molecular mechanisms behind its function in PTC remain poorly understood. Clarifying its role in THCA could be implicated in the development of improved diagnostic and treatment strategies. The current study comprehensively analyzed TACSTD2 expression and discovered its significant upregulation in papillary PTC tissues. Loss-of-function experiments demonstrated that TACSTD2 silencing hampered PTC cell invasion and proliferation in vitro. Consistently, in vivo assays confirmed that TACSTD2 knockdown inhibited tumor growth, whereas TACSTD2 overexpression produced the opposite effects. This implies that TACSTD2 functions as an oncogene in PTC progression. Research has reported TACSTD2 overexpression in breast, lung, gastric, cervical, ovarian, and nasopharyngeal cancers, where it is associated with tumor cell proliferation, invasion, and migration42. Similarly, Guan et al.16 and Sun et al.43 reported significantly increased TACSTD2 expression in THCA tissues, corroborating our findings. Interestingly, animal studies have shown that TACSTD2 silencing can enhance the malignant phenotype of skin cancer by promoting EMT44. In contrast, Wang et al.45 found that TACSTD2 deficiency enhanced squamous cell carcinoma by attenuating TAp63-dependent apoptosis. These observations suggest that TACSTD2 may play context-dependent roles due to cancer-specific heterogeneity. Previous research has shown that TACSTD2 drives tumor initiation and progression through multiple signaling pathways, correlating with poor prognosis and reduced survival rates. TACSTD2 plays a crucial role in transmembrane signal transduction, with its tumor-associated functions closely linked to subcellular localization. TACSTD2 interacts with phosphatidylinositol (4,5)-bisphosphate or undergoes phosphorylation by protein kinase C, suggesting its implication in calcium signaling46. Tight junction proteins 1 and 7 bind to the TACSTD2 extracellular domain, reinforcing epithelial cell adhesion and preventing the degradation of tight junction components47. Furthermore, TACSTD2 activates the ERK1/2-MAPK signaling, thereby promoting cell cycle progression48. The TNF signaling is critical in apoptosis induction31, whereas EMT facilitates the acquisition of mesenchymal traits by epithelial cells, driving cancer metastasis32. Our study underscores a new mechanism by which TACSTD2 knockout activates the TNF signaling, induces apoptosis, and inhibits EMT. However, further research should delve into the underlying crosstalk mechanisms. Currently, no research has reported the m6A-mediated regulation of aberrant TACSTD2 expression in THCA. We investigated whether TACSTD2 downregulation in PTC is associated with m6A modification. Epitranscriptomic analysis revealed methylation sites on TACSTD2, demonstrating a strong positive correlation between m6A levels and METTL3 expression. We confirmed that METTL3-induced m6A methylation decreased TACSTD2 mRNA stability and promoted its degradation via a YTHDC2-dependent pathway. Previous research indicated that METTL3 exhibits both oncogenic and tumor-suppressive functions49. According to Zhu et al.28, METTL3 is impeded in PTC and inhibits THCA progression by suppressing EMT and Hedgehog signaling, which aligns with our findings. Herein, rescue experiments demonstrated that TACSTD2 overexpression partially reversed the METTL3 overexpression-triggered tumor-suppressive phenotype. Our findings further indicate that YTHDC2 overexpression reduces TACSTD2 mRNA stability, whereas silencing YTHDC2 significantly reverses METTL3-mediated TACSTD2 downregulation. A previous study showed that YTHDC2 facilitates mRNA degradation35. Collectively, these findings demonstrate that METTL3 promotes TACSTD2 degradation through an m6A-dependent mechanism, leading to TNF signaling activation, apoptosis induction, EMT suppression, and inhibition of aggressive tumor phenotypes. Our identification of the METTL3/TACSTD2/YTHDC2 axis and its role in PTC progression may provide innovative perspectives to develop more effective therapeutic approaches. Conclusion To our knowledge, this study is the first comprehensive analysis of m6A modification of TACSTD2 within PTC, demonstrating it as a potential oncogene that promotes tumor proliferation and metastasis in PTC. Additionally, METTL3 epigenetically downregulates TACSTD2 via an m6A-YTHDC2-dependent pathway. Identifying the METTL3–TACSTD2 regulatory axis and its role in PTC progression provides a foundation for developing more effective therapeutic strategies for aggressive PTC. Methods Clinical tissue samples This study enrolled 55 patients who had undergone surgery for PTC at the Second Affiliated Hospital of Harbin Medical University between May 2024 and May 2025. The study adhered to the Declaration of Helsinki and was approved by the Ethics Committee of the Second Affiliated Hospital of Harbin Medical University (No. YJSKY2024-249, May 29, 2024). Before gathering tissue samples, all participants provided written informed consent. We confirm that all methods were performed in accordance with the relevant guidelines and regulations. Public database analysis We obtained THCA mRNA expression data and related clinical information by accessing the Cancer Genome Atlas (TCGA) database and analyzed GSE3467/GSE60542 datasets from the Gene Expression Omnibus (GEO). Furthermore, the Human Protein Atlas (HPA) was used to compare TACSTD2 expression between normal and THCA tissues. The GEPIA2 web-based analysis tool (http://gepia2.cancer-pku.cn/) enabled visualization and network analysis. Cell culture and transfection Human PTC cell lines (KTC-1, FTC-133, B-CPAP, and TPC-1) and the normal line (Nthy-ori3-1) were sourced from iCell Bioscience Inc. (Shanghai, China). All cells were subjected to culture in RPMI 1640 complete medium (Gibco, NY, USA) that contained 10% fetal bovine serum (FBS) under humidified conditions with 5% CO2 at 37 °C. The TACSTD2 was knocked down using lentiviral vectors (Lv-sh-NC and Lv-sh-TACSTD2#1/2; GeneChem, Shanghai, China). The expression plasmids pcDNA3.1-TACSTD2/METTL3/YTHDC2/empty vector, short hairpin RNA targeting METTL3/YTHDC2/TNF-R1/TRADD, and its negative control were synthesized by GenePharma (Shanghai, China). Once PTC cells reached approximately 70% confluence, they were cultured into six-well plates (1.5 × 105 cells/well), transfected with 2 µg of overexpression plasmids (OE-NC/TACSTD2/METTL3/YTHDC2) or sh-YTHDC2 via Lipofectamine™ 2000 (Thermo Fisher Scientific, MA, USA), and incubated at 37 °C for 48 h post transfection, following the instructions. Subsequent experiments were conducted after incubation. The sh-RNA sequences used are presented in Table S1. Cell counting Kit-8 (CCK-8) assay, 5-ethynyl-2′-deoxyuridine (EdU) analysis, and clonogenic assay The CCK-8, EdU incorporation, and colony formation assays were used to assess cell viability. After seeding PTC cells into 96-well plates (1 × 104 cells/mL), CCK-8 reagent (Solarbio, China) was added into each well at 24, 48, 72, and 96 h post-incubation, and the absorbance was determined at 450 nm with a microplate reader. Following established protocols, an EdU assay was conducted (Beyotime Biotechnology, China). Stained cells were visualized with a fluorescence microscope, followed by quantifying EdU-positive cells in at least three randomly selected fields. The colony formation assay was carried out by culturing cells in six-well plates and incubating them in complete medium for 14 days, with medium refreshed every 3 days. Colonies were subsequently fixed with paraformaldehyde, stained with 0.1% crystal violet for half an hour, and counted for quantitative analysis. RT-qPCR Total cellular RNA extraction was conducted with TRIzol reagent (Invitrogen, Waltham, MA, USA), followed by conducting reverse transcription at 42 °C with a Takara reverse transcription kit (Dalian, China). Quantitative real-time PCR (qRT-PCR) was performed using the TB Green Premix Ex Taq II kit (Takara, Dalian, China) following a standard two-step protocol, with GAPDH serving as an internal reference to normalize target gene expressions. The 2-ΔΔCt methodologies were used to analyze the data. The primer sequences used are presented in Table S2. Western blot (WB) Utilizing RIPA lysis buffer enriched with protease inhibitors, total protein was extracted from cells and then assessed by a BCA Protein Assay Kit (Beyotime Biotechnology, China). Equal protein amounts were separated by SDS-PAGE and transferred onto PVDF membranes (Millipore, Germany), blocked with 5% non-fat milk at room temperature for 2 h prior to incubation overnight at 4 °C with primary antibodies against the following proteins: TACSTD2 (CST, #76730), METTL3 (CST, #86132), YTHDC2 (CST, #46324), MMP2 (CST, #40994), MMP3 (CST, #14351), MMP7 (CST, #3801), MMP9 (CST, #24317), TNF-α (CST, #11948), TNF-R1 (CST, #3736), TNF-R2 (CST, #72337), TRADD (CST, #3684), Bax (CST, #5023), Bcl-2 (CST, #4223), caspase-3 (CST, #9662), cleaved caspase-3 (CST, #9661), caspase-9, and cleaved caspase-9 (Proteintech, 10380-1-AP), E-cadherin (Proteintech, 20874-1-AP), N-cadherin (Proteintech, 22018-1-AP), β-catenin (Proteintech, 51067-2-AP), Snail (CST, #3879), vimentin (CST, #5741), GAPDH (CST, #2118), and β-actin (CST, #4970). Membranes were subsequently incubated with suitable secondary antibodies, and immunoreactive bands were visualized using an ECL detection system (Beyotime, China) and analyzed quantitatively with ImageJ software. Transwell migration/invasion assays The transwell assay was employed to assess cell migration and invasion by seeding 1 × 104 cells into the upper chamber of a 24-well transwell insert (BD Biosciences, USA) containing serum-free medium. For the invasion assays, the upper chamber was pre-coated with 2% Matrigel, while the lower chamber was filled with medium containing 15% FBS as a chemoattractant. Following 48 h of incubation, non-migrated cells on the upper surface of the membrane were gently eliminated. Migrated or invaded cells on the lower surface were subjected to fixation, staining with crystal violet, imaged, and quantified with a light microscope. In vivo tumor xenograft model The Second Affiliated Hospital of Harbin Medical University Animal Ethics Committee (Approval No. SYDW2024-029) authorized the animal experiments and were performed in accordance with the ARRIVE guidelines. Male BALB/c nude mice (6–8 weeks; Jiangsu Huachuang Xinnuo Pharmaceutical Technology Co., Ltd.) were habituated under specific pathogen-free conditions. KTC-1 cells transduced with Lv-sh-TACSTD2 or Lv-sh-NC (5 × 106 cells/mouse) were subjected to subcutaneous injection into each mouse flank region (n = 7/group). Tumor volume (V) was measured every week and calculated as follows: V = (L × W2)/2, where L: length and W: width. On day 28, mice were euthanized by cervical dislocation after intraperitoneal injection of pentobarbital (150 mg/kg) and tumors were surgically removed and weighed. In our study, the maximum tumor length in all mice did not exceed 20 mm (Table S3). Immunohistochemistry (IHC) analysis Slides were exposed to 0.3% hydrogen peroxide treatment for half an hour to impede endogenous peroxidase activity. Tissue sections were permeabilized with 0.5% Triton X-100 for 30 min, followed by antigen retrieval using citrate buffer. After washing three times with phosphate-buffered saline, 5% normal goat serum was utilized for non-specific binding site blocking. Samples were incubated with primary antibodies against TACSTD2 (Proteintech, 1:100, 27360-1-AP), Ki-67 (Proteintech, 1:100, 27309-1-AP), and PCNA (Proteintech, 1:100, 10205-2-AP), incubated for 2 h at room temperature, and incubated with HRP-conjugated anti-rabbit IgG (H + L) secondary antibody for 30 min, before counterstaining the nuclei with DAPI. Methylated RNA Immunoprecipitation (MeRIP)-qPCR m6A RNA methylation levels were assessed using a MeRIP assay with RIP Kit (Bersinbio, Guangzhou, China) following the instructions. PTC cells were lysed utilizing TRIzol reagent for total RNA extraction and immunoprecipitated using an anti-m6A antibody. Protein A/G magnetic beads were introduced to bind the antibody–RNA complex. The co-precipitated RNA was purified and dissolved in RNase-free water. The m6A-modified enriched RNA was quantified via qRT–PCR. Data were analyzed using the ΔΔCt or 2^–ΔΔCt method. MeRIP-qPCR primers for TACSTD2 were synthesized (Bersinbio, Guangzhou, China) using the sequences: F: 5′-CTTCGAGAGGGACATCAAGG-3′ and R: 5′-GAGAACTTCGGGGGAATCTC-3′. RNA immunoprecipitation‑PCR (RIP‑qPCR) assay An RNA immunoprecipitation kit (Bersinbio RIP Kit, Guangzhou, China) was used to perform the RIP assays. The cells were lysed with lysis buffer according to the manufacturer’s instructions. Immunoprecipitation of endogenous factors was performed using a primary antibody overnight at 4℃. Protein A/G beads were then added to capture the primary antibody. RNAs were extracted using TRIzol reagent and quantified by qRT-PCR using primers for TACSTD2. The data were normalized to input or %IgG of input. RNA stability assay Forty-eight hours after transfection, PTC cells overexpressing METTL3, YTHDC2, or their respective controls were treated with actinomycin D (M4881, Abmole, USA) at a final concentration of 10 µg/mL. Cells were harvested at 0, 2, 4, 6, 8 and 12 h after treatment. Total RNA extraction was carried out using TRIzol reagent, and TACSTD2 expression was quantified by qRT–PCR using specific primers. Statistical analysis To ensure reproducibility, experiments were conducted in triplicate. Statistical analyses were performed through GraphPad Prism 7.0 (GraphPad Software, USA), with data expressed as the mean ± standard deviation (SD). Comparisons between two groups were carried out with the unpaired Student’s t-test or the nonparametric Mann–Whitney U test. For multiple-group comparisons, we applied a one-way or two-way ANOVA followed by a Bonferroni post hoc test. Pearson’s correlation analysis was deployed to assess the linear relationship between METTL3, YTHDC2, and TACSTD2 expression levels. Statistical tests were two-tailed, with P < 0.05 deemed statistically significant: *P < 0.05, **P < 0.01, and ***P < 0.001. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 Acknowledgements We thank LetPub (www.letpub.com.cn) for its linguistic assistance during manuscript preparation. Author contributions TFS designed the study; JTM drafted the manuscript; ZYZ and TXC acquired and analyzed the data; and TFS and JTM confirmed the authenticity of the raw data. All authors have read and approved the final manuscript. Funding Postgraduate Research & Practice Innovation Program of Harbin Medical University (grant no. YJSCX2024-49HYD). Data availability All data supporting the findings of this study are available within the paper. The RNA-Seq sequencing data has been submitted to the Genome Sequence Archive (Genomics, Proteomics & Bioinformatics 2021) in National Genomics Data Center (Nucleic Acids Res 2022), China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences that are publicly accessible at [https://ngdc.cncb.ac.cn/gsa-human_under_accession_number_HRA012215]. Declarations Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Supplementary Materials Supplementary Material 1 Data Availability Statement All data supporting the findings of this study are available within the paper. The RNA-Seq sequencing data has been submitted to the Genome Sequence Archive (Genomics, Proteomics & Bioinformatics 2021) in National Genomics Data Center (Nucleic Acids Res 2022), China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences that are publicly accessible at [https://ngdc.cncb.ac.cn/gsa-human_under_accession_number_HRA012215].