Pan-cancer analyses reveal the molecular and clinical characteristics of TET family members and suggests that TET3 maybe a potential therapeutic target

The Ten-Eleven Translocation (TET) family genes are implicated in a wide array of biological functions across various human cancers. Nonetheless, there is a scarcity of studies that comprehensively analyze the correlation between TET family members and the molecular phenotypes and clinical characteristics of different cancers. Leveraging updated public databases and employing several bioinformatics analysis methods, we assessed the expression levels, somatic variations, methylation levels, and prognostic values of TET family genes. Additionally, we explored the association between the expression of TET family genes and pathway activity, tumor microenvironment (TME), stemness score, immune subtype, clinical staging, and drug sensitivity in pan-cancer. Molecular biology and cytology experiments were conducted to validate the potential role of TET3 in tumor progression. Each TET family gene displayed distinct expression patterns across at least ten detected tumors. The frequency of Single Nucleotide Variant (SNV) in TET genes was found to be 91.24%, primarily comprising missense mutation types, with the main types of copy number variant (CNV) being heterozygous amplifications and deletions. TET1 gene exhibited high methylation levels, whereas TET2 and TET3 genes displayed hypomethylation in most cancers, which correlated closely with patient prognosis. Pathway activity analysis revealed the involvement of TET family genes in multiple signaling pathways, including cell cycle, apoptosis, DNA damage response, hormone AR, PI3K/AKT, and RTK. Furthermore, the expression levels of TET family genes were shown to impact the clinical staging of tumor patients, modulate the sensitivity of chemotherapy drugs, and thereby influence patient prognosis by participating in the regulation of the tumor microenvironment, cellular stemness potential, and immune subtype. Notably, TET3 was identified to promote cancer progression across various tumors, and its silencing was found to inhibit tumor malignancy and enhance chemotherapy sensitivity. These findings shed light on the role of TET family genes in cancer progression and offer insights for further research on TET3 as a potential therapeutic target for pan-cancer.


Introduction
Epigenetics, including DNA methylation, histone modifications, and non-coding RNA regulation, plays a crucial role in human disease, particularly cancer, by orchestrating gene expression dynamics without altering DNA sequence.Aberrant epigenetic changes, such as hypermethylation of tumor suppressor gene promoters and dysregulated histone modifications, disrupt normal cellular processes, fostering tumor initiation, progression, and metastasis.Despite the continuous advancements in surgical and targeted therapies in recent years, a considerable portion of tumor patients still present with advanced and widespread metastasis at the time of diagnosis, resulting in a bleak prognosis (Golemis et al., 2018).Consequently, conducting comprehensive research into the molecular mechanisms underlying cancer development holds paramount importance for tumor prevention, early screening and diagnosis, clinical treatment, as well as enhancing patient prognosis and survival rates (Alqahtani et al., 2019;Ban et al., 2021;Rodriguez et al., 2022).
Genetic variations and epigenetic changes play pivotal roles in the initiation and progression of tumors (Talib et al., 2021).Nearly all tumors involve single or tandem mutations in one or multiple genes (Kossenas and Constantinou, 2021).To devise precise treatment strategies capable of effectively targeting tumor cells while minimizing damage to normal tissues, from the myriad of small molecule chemical drugs available, a comprehensive understanding of the underlying mechanisms of the implicated genes is imperative (Srivastava and Goodwin, 2020).The advent and advancement of modern bioinformatics methodologies offer a streamlined approach for humanity to efficiently explore and analyze tumor occurrences within established gene information databases, thereby providing guidance for tumor treatment (Nooter and Stoter, 1996;Zaridze, 2008;Gonçalves et al., 2021).
Recent studies have revealed that epigenetic regulations, such as DNA methylation and demethylation modifications, are intricately linked to the onset and progression of human cancers (Kulis and Esteller, 2010;Klutstein, Nejman, Greenfield and Cedar, 2016;Nishiyama and Nakanishi, 2021).DNA methylation often occurs on the CpG islands of the promoter regions, resulting in the activation of oncogenes and the inactivation of tumor suppressor genes (Kulis and Esteller, 2010;Papanicolau-Sengos and Aldape, 2022).Aberrant DNA methylation can disrupt the apoptosis process of cells, render tumor cells insensitive to growth inhibition signals, and induce uncontrolled replication of tumor cells (Pan et al., 2018).Consequently, epigenetic research, particularly focusing on DNA methylation, holds significant value for tumor molecular diagnosis, prevention, and treatment, as well as predicting therapeutic outcomes and prognosis (C.Chen et al., 2022;Meng et al., 2015).
Studies indicate that both hypermethylation and hypomethylation play crucial roles in tumor development.The Ten-Eleven Translocation (TET) enzymes facilitate the oxidation of 5-methylcytosines (5mCs) and promote site-specific reversal of DNA methylation (Kohli and Zhang, 2013).TET family genes, including TET1, TET2, and TET3, exhibit diverse vital biological functions in mammals (Kinney and Pradhan, 2013;Zeng and Chen, 2019;Joshi, Liu, Breslin and Zhang, 2022).At their core, TET family genes consist of α-Ketoglutaric acid (α-KG) and Fe2+-dependent dioxygenases, which regulate DNA demethylation, modulate gene transcription, and influence various life processes such as embryonic development, as well as the onset of various diseases including tumors (Rasmussen and Helin, 2016;Ma et al., 2021).Additionally, studies have demonstrated that TET family genes not only facilitate active DNA demethylation but also impede methylation propagation by maintaining a low DNA methylation state, thus closely associating them with tumorigenesis (Shekhawat et al., 2021).Notably, TET family genes exhibit a dual role in different tumors, displaying both carcinogenic and anticancer effects.
While TET family genes are known to play an indispensable role in the progression of numerous tumors, there is currently scarce literature on pan-cancer analysis of these genes.Therefore, in this study, we conducted an analysis of the relationship between TET family genes and pan-cancer, exploring various aspects including mRNA expression levels, somatic cell variations, DNA methylation patterns, pathway activities, survival rates and prognoses, immune subtypes specific to individual tumors, tumor microenvironment (TME), stem cell indices, and drug sensitivities.Our aim is to provide valuable reference data and constructive suggestions based on a comprehensive understanding of the landscape.

Results
Expression and differential analysis of TET family genes in pan-cancer Firstly, utilizing transcriptome data from 33 tumors available in the UCSC Xena database, we conducted an extraction and analysis of the expression profiles of the target genes belonging to the TET family (specifically, TET1, TET2, and TET3) across various types of cancer.The results depicted in the Boxplot revealed that the expression level of the TET3 gene ranked highest across pancancer samples, followed by TET2, with TET1 exhibiting the lowest expression (Figure 1A, Supplementary Table S1).
Next, we screened 18 types of tumors, each containing at least five corresponding normal samples, and analyzed the expression differences of TET family members in pan-cancer and its normal tissues.As shown in Figure 1B, TET1 exhibited abnormal expression in 10 types of tumors, with significantly downregulated expression observed in BRCA, KICH, KIRC, KIRP, and THCA, and upregulated expression noted in CHOL, HNSC, LIHC, LUAD, and LUSC tumors compared to their corresponding normal samples.The expression of TET2, in tumor samples such as BRCA, COAD, HNSC, KIRP, READ, and THCA, was lower, while in CHOL, GBM, KIRC, and UCEC tumors, it was significantly higher than that in corresponding normal control tissues (Figure 1C).The expression of the TET3 gene varied across 15 tumors and corresponding normal samples, including BLCA, BRCA, CHOL, ESCA, HNSC, KIRC, KIRP, LIHC, LUAD, LUSC, PRAD, STAD, THCA, UCEC, and KICH.Among these, except for KICH, TET3 was significantly overexpressed in all 14 other tumors (Figure 1D).These findings suggest that the high expression of TET3 may serve as a key factor in the occurrence, development, and poor prognosis of various clinical tumors.
To further elucidate the differential expression patterns of TET family members in various tumors, we conducted Wilcox analysis on the expression differences in each tumor and obtained the aforementioned differential expression cluster analysis graph (Figure 1E; Table 1).As previously mentioned, TET3 was significantly upregulated, while TET1 and TET2 were downregulated in most tumors.It remains unclear whether there is a correlation between the expression of the three members of the TET family in pan-cancer.Therefore, we conducted a detailed analysis of the correlation between TET family genes in pan-cancer.As illustrated in Figure 1F, the expression of TET1, TET2, and TET3 positively correlated in pan-cancer, with a stronger positive correlation observed between TET2 and TET3.
Given the high expression of TET3 in numerous tumors, we speculate that it may represent a potential therapeutic target.Consequently, we assessed the expression of TET3 protein in four tumor cell lines (BGC-823, HepG2, A549, TPC-1) and their corresponding normal cell lines (GES-1, LX2, 16HBE, Nthy-ori-3-1).The results demonstrated higher expression levels of TET3 in all four tumor cell lines compared to the corresponding control cell lines (Figures 1G-I).Subsequently, we examined the expression of TET3 protein in three different pathological subtypes of lung cancer cell lines (A549, SKMES1, NCIH460).The findings revealed significantly higher expression levels of TET3 protein in lung adenocarcinoma A549 and lung squamous cell carcinoma SKMES1 compared to normal controls and large cell lung cancer cell lines.This observation suggests differential expression of TET3 protein within the same type of tumors with varying pathological subtypes.Finally, we evaluated the expression of TET3 protein in gastric cancer cell lines (NCI-N87, BGC-823, HGC-27) with differing degrees of differentiation.The results indicated that as the degree of cell differentiation decreased, the expression of TET3 protein gradually increased.This implies that the expression intensity of TET3 protein may serve as an indicator of the malignant level of gastric cancer.

Somatic cell variation landscape of TET family genes
To comprehend the genomic alterations of TET1-3 in tumors, we scrutinized single nucleotide variant data (SNV, Table 2) and copy number variant data (CNV, Table 3) from a total of 10,289 pan-cancer samples spanning 33 types of tumors, and illustrated SNV and CNV landscapes.
Initially, we analyzed the SNP data pertaining to the TET gene to ascertain the frequency and variant types across each cancer subtype.As depicted in Figures 2A, B, among these cancers, the SNV frequency of UCEC, SKCM, and COAD ranges from 19% to 57%.The SNV frequency of the TETs gene collectively is 91.24% (1052/1153).Variant analysis unveiled missense mutation as the predominant SNP type.SNV percentage analysis revealed mutation rates of 23%, 17%, and 16% for TET1-3, respectively.
To elucidate the mutation pattern of CNV, we examined the CNV data of TETs genes in the TCGA database.The distribution of CNV pie charts illustrated heterozygous amplification and heterozygous deletion as the primary types (Figure 2C).CNV percentage analysis unveiled heterozygous amplification rates of TET1 in ACC and UCS, TET2 in ACC and KICH, and TET3 in ESCA, LUSC, OV, TGCT, and UCS, all exceeding 20% (Figure 2D).However, homozygous amplification was weak, observed solely in CHOL for TET1 (Figure 2E).Heterozygous deletions of TET1 in GBM, KICH, SARC, SKCM, and TGCT, TET2 in CHOL, ESCA, LIHC, LUSC, MESO, OV, READ, TGCT, and UCS, and TET3 in KICH and SARC, all surpassed 40% (Figure 2D).No homozygous deletions were noted across all tumors (Figure 2E).Correlation analysis revealed a positive association between the mRNA expression of TETs and CNV, notably with TET2 in CHOL and ACC.Conversely, TET2 in LAML and TET3 in THMY exhibited negative correlations with CNV (Figure 2F; Table 4).This suggests that CNV of the TETs genes mediate abnormal expression in pancancer, potentially playing a pivotal role in cancer progression.

Methylation variation in TET family genes across pan-cancer
We delved into the methylation levels of TET genes across various cancers to decipher their epigenetic regulation.Our findings revealed heterogeneity in the methylation levels of TET family genes among different tumors.Specifically, TET1 gene exhibited hypermethylation in most cancers, with hypomethylation observed solely in LIHC.Conversely, TET2 and TET3 genes displayed hypomethylation in the majority of tumors, except for instances of hypermethylation in THCA, COAD, and KIRP for TET2 (Figure 3A).Correlation analysis between methylation and mRNA expression unveiled a negative correlation between the expression of these genes and their methylation levels across pan-cancer (Figure 3B; Table 2).Furthermore, survival analysis indicated that methylation of TET1 was linked to a lower survival rate in ACC and a higher survival rate in UVM; methylation of TET2 was associated with increased survival rates in ACC, KIRP, and PCPG; while methylation of TET3 correlated with decreased survival rates in SARC and BLCA (Figure 3C, Supplementary Table S2).These observations suggest that the methylation levels of TET genes are primarily associated with the survival outcomes of a limited number of tumors.
Given the significance of TET3, we modulated its expression in the NCI-N87 gastric cancer cell line.Overexpression of TET3 led to increased phosphorylation levels of AKT (Ser 308 and Ser 473) and mTOR (Ser2448), suppressed P21 expression, and enhanced HIF1α and c-Myc protein expression, whereas silencing TET3 exhibited the opposite effect (Figure 4B).
Sunitinib, a selective multi-target tyrosine kinase inhibitor, was employed to assess whether TET3 affects tumor behavior via the RTK pathway.We tested the therapeutic sensitivity of three different pathological subtypes of lung cancer cell lines (A549, SKMES1, NCIH460) to sunitinib.The results showed that A549 and SKMES1 cell lines with relatively high expression of TET3 exhibited higher sensitivity to sunitinib (Figure 4C).Similarly, we also tested the therapeutic sensitivity of gastric cancer cell lines (NCI-N87, BGC-823, HGC-27) with different degrees of differentiation to sunitinib.The results revealed that those with higher TET3 expression exhibited increased sensitivity to the drug (Figure 4D).These findings underscore the involvement of TET3 in regulating the PI3K/AKT and RTK pathways during tumor progression.

Influence of TET gene family on pan-cancer survival and prognosis
We conducted an analysis to assess the correlation between TET gene expression levels and the survival outcomes of patients across 33 types of tumors using data from the TCGA database.Kaplan-Meier survival curve analysis revealed that high expression of TET1 in ACC, KIRP, LIHC, SARC, and STAD was associated with shorter patient survival, while low expression in LGG and THYM indicated reduced survival.Similarly, high expression of  S3).
Furthermore, we evaluated the relationship between TET expression and cancer prognosis (Figure 6).COX regression analysis across 33 cancer types revealed that high expression of TET1 in ACC, BLCA, CESC, LIHC, PCPG, and SARC was indicative of poor prognosis, while in LGG, THYM, and UVM, it suggested a favorable prognosis.Conversely, in KIRC, high expression of TET2 was associated with a favorable prognosis.Notably, high expression of TET3 in ACC alone indicated a poor prognosis among tumor patients.

Correlation between TET family genes expression and immune subtypes in pancancer and single tumors
To elucidate the association between TETs and immune responses, we conducted a pan-cancer analysis correlating TETs expression with immune subtypes using the TCGA database.As depicted in Figure 7A, TETs expression exhibited a significant positive correlation with pan-cancer immune subtypes (p < 0.05): notably, TET1 and TET2 were markedly upregulated in immune subtype C5, whereas their expression remained stable in other subtypes; TET3 displayed variable expression levels across immune subtypes.To further delineate the specific immune subtype associations with TETs, we performed KS test correlation analysis in eight common tumors.Results indicated significant correlations between TETs expression and immune subtypes in most tumors (p < 0.05), with TET1 consistently exhibiting the lowest expression across all subtypes, while TET3 demonstrated high expression levels (Figure 7B).In STAD, LIHC, and LUAD, TETs expression showed significant positive correlations with various immune responses.Notably, TET1 displayed highest expression in STAD's C3 subtype and minimal expression in C2, while TET2 was consistently upregulated across all subtypes.Conversely, in LIHC, TET1 and TET3 exhibited decreasing expression trends across subtypes, with TET1 almost absent in C6, and TET2 lowest in C2.In LUAD, TET1 expression remained stable, while TET2 peaked in C3 and declined in C4, and TET3 expression was highest in C1 and lowest in C4.However,    Moreover, the therapeutic efficacy of radiotherapy, chemotherapy, and immunotherapy is closely linked to the tumor stem cell index.Figures 8D, E indicates that TET1 expression is positively correlated with RNAss in LGG and TGCT, while negatively correlated with KIRC, PAAD, and STAD.TET2 levels are negatively correlated with RNAss in CHOL, KIRP, and THYM.Conversely, TET3 levels are negatively correlated in KICH, KIRP, THCA, and THYM, and positively correlated in BLCA, BRCA, and PRAD.Additionally, TET1 expression is positively correlated with DNAss in GBM and TGCT, but negatively correlated with LGG, OV, and THYM.TET2 shows a positive correlation in TGCT and a negative correlation in OV, while TET3 exhibits a negative correlation in OV.
Next, we conducted a detailed analysis of the tumor microenvironment and stem cell index of eight common clinical cancers (Figure 9).The results showed that for tumor microenvironment-related scores, TET1 was significantly positively correlated with the three scores of PAAD, and    Analysis of TET family genes expression and patient prognosis using univariate KM risk proportional regression model.

TET family genes expression was correlated with clinical stages across various cancers, indicating their potential as prognostic markers
These findings underscore the close relationship between TET gene expression and both the tumor microenvironment and tumor stemness potential.Further analysis revealed distinct correlations between TET gene expression and clinical stages across eight common tumor types.As shown in Figure 10, there was no significant difference between the expression level of TETs genes and clinical stage of tumors in COAD and READ.Notably, TET2 and TET3 exhibited significant correlations with clinical stages in LIHC, with elevated expression in Stage III and reduced expression in Stage IV.Similarly, in LUSC, TET1 and TET3 expression levels were notably associated with clinical stages, showing upregulation in Stage III and downregulation in Stage IV.Furthermore, TET2 expression levels were significantly linked to clinical staging in STAD and LUAD, with varying expressions across different stages.In PAAD, TET3 expression levels exhibited significant correlation with clinical characteristics, displaying the lowest expression in Stage III and the highest expression in Stage IV.These results maybe highlight the distinctive correlations between TET gene expression and clinical stages across various tumors.

TET family genes levels were analyzed for their impact on drug sensitivity
Using the CellMiner ™ database, we conducted an analysis to explore potential correlations between gene expression levels of TET family members and drug sensitivity across various human tumors (Figure 11A, Supplementary Table S4).We identified the top 16 chemotherapy-sensitive drugs exhibiting the highest correlation coefficients with TETs genes.Notably, the expression level of TET1 displayed significant positive correlations with drug sensitivity to Arsenic trioxide, Fenretinide, Dimethylaminophenhenolide, Daunorubicin, Homoharringtonine, Imatinib, Testolactone, Pipappererone, and Lomustine.Conversely, TET2 expression was positively correlated with drug sensitivity to Fulvestrant and Raloxifene but significantly negatively correlated with Vemurafenib and Dabrafenib.Moreover, the expression level of TET3 exhibited significant positive correlations with drug sensitivity to Lapatinib, AZD-9291, and (+)-JQ1.Impact of TET family genes on pan-cancer prognosis through COX regression analysis.
Frontiers in Pharmacology frontiersin.org18 Furthermore, we assessed the therapeutic sensitivity of three distinct pathological subtypes of lung cancer cell lines (A549, SKMES1, NCIH460) to Lapatinib, AZD-9291, and (+)-JQ1.The findings indicated that the therapeutic efficacy of these drugs on tumor cells positively correlated with the expression level of TET3 (Figure 11B).

Silencing TET3 effectively suppresses the malignant behavior of tumor cells
Following our comprehensive analysis spanning various cancer types, TET3 has emerged as a prime candidate for targeted therapeutic intervention.In subsequent experiments, where we selectively silenced TET3 expression in four distinct tumor cell lines (BGC-823, HepG2, A549, TPC-1) (Figure 12A), we observed a profound attenuation of multiple malignant phenotypes.Specifically, the inhibition of TET3 led to a notable reduction in clone formation capacity (Figure 12B), as well as a significant impairment in cell migration and invasion abilities (Figures 12C, D).Furthermore, TET3 silencing induced cell cycle arrest (Figure 12E), resulting in decreased proliferation rates, while simultaneously promoting apoptosis (Figure 12F).These compelling findings underscore the pivotal oncogenic role of TET3 across various tumor contexts and highlight the potential of TET3targeted therapies as a promising avenue for effective cancer treatment.
In our analysis of transcriptome data from 33 tumors in the UCSC Xena database, we found distinct expression patterns among TET genes across different cancer types.Specifically, TET1 showed TET family members and stem cell index in eight common clinical tumors.
Frontiers in Pharmacology frontiersin.org21 notable overexpression in epithelioid squamous cell carcinoma (HNSC and LUSC) and hepatobiliary duct tissue (CHOL and LIHC), while consistently exhibiting low expression in various kidney-derived tumors (KICH, KIRC, and KIRP).Conversely, TET2 was predominantly expressed in adenocarcinomas (BRCA, COAD, READ, and THCA).In contrast, TET3 demonstrated widespread high expression across most tumors, with the exception of KICH.These expression patterns may hold significance for clinical diagnostics and the development of targeted therapies.Certain SNPs within genes can directly impact protein structure or expression levels, potentially influencing tumor genetic mechanisms (Kuhlen et al., 2019).In this study, we conducted a systematic analysis of somatic cell variations in TET gene family members.We observed a high frequency of Single Nucleotide Variants (SNV) and found that the main type of copy number variant (CNV) was heterozygous amplification and deletion.Correlation analysis revealed positive associations between TET2 expression and CNV in CHOL and ACC, while TET2 in LAML and TET3 in THMY showed negative correlations with CNV.Identifying pathogenic CNV and interpreting their clinical significance will be crucial for future research, despite the challenges associated with this endeavor.
In general, the extent of whole-genome hypomethylation in tumor cells correlates closely with disease progression, tumor size, and malignancy (Esteller and Herman, 2002).DNA methylation holds significant value in assessing tumor malignancy and guiding targeted drug selection.The TET family genes, pivotal in modulating the methylation levels of numerous genes, contribute to the complexity of gene expression regulation due to their own methylation status.Although these mechanisms enhance gene regulation accuracy, they pose challenges for researchers.Despite the negative correlation between TET family genes expression and methylation levels in pan-cancer, unique patterns were observed, such as high methylation of TET1 in most analyzed cancers (except LIHC) and low methylation of TET2 and TET3 in most tumors.Further investigation is required to elucidate whether this differential expression pattern suggests functional compensation among TET family genes.Notably, the state of DNA methylation evolves during tumor progression, necessitating dynamic and cautious interpretation of its clinical significance (Miyamoto and Ushijima, 2003).
Our analysis of TET family genes in pan-cancer revealed their involvement in various signaling pathways.Notably, TET1 was found to predominantly activate the PI3K/AKT pathway.Prior studies have highlighted TET1's role in driving the proliferation of insulin-dependent endometrial cancer by enhancing G proteincoupled estrogen receptor expression and PI3K/AKT pathway activation (Xie et al., 2017).Moreover, research by Huang elucidated TET1's impact on stem cell development through modulation of the Wnt and PI3K-Akt pathways, while TET2 deficiency in mice led to a progressive reduction in spermatogonia stem cells (Huang et al., 2020).TET2 plays a pivotal role in cell cycle regulation and DNA damage responses.Studies by Zhong have revealed that 5 mC oxidation is cell-cycle dependent, occurring primarily during the S and G2/M phases.Notably, TET2 depletion diminishes the observed elevation in 5hmC, indicating its dependence on TET2, particularly in response to idarubicin stimulation, a topoisomerase II inhibitor (Panigrahi et al., 2015).Additionally, Chen found that SMAD nuclear interacting protein one recruits TET2 to regulate c-MYC target genes and the cellular DNA damage response (Chen et al., 2018).TET3, a pivotal enzyme, showcases its versatility within cellular processes, including the cell cycle, apoptosis, hormone AR regulation, and the DNA damage response.Research led by Jiang and colleagues revealed that upon DNA damage, ATR kinase activation leads to TET3 phosphorylation in mammalian cells.This phosphorylation enhances DNA demethylation and the accumulation of 5-hydroxymethylcytosine, underlining TET3's essential role in DNA repair and cell survival (Jiang et al., 2017).In the landscape of cancer, the expression levels of the TET family offer prognostic insights.Elevated TET1 expression correlates with worse outcomes in solid organ tumors such as ACC, KIRP, LIHC, and SARC, yet it signifies better prognoses in LGG and THYM cancers.Similarly, increased TET2 expression suggests poorer outcomes in tumors of the female reproductive system, including OV and UCS, but predicts favorable outcomes in KIRC.Notably, high TET3 expression is linked to unfavorable prognoses in ACC but indicates better survival rates in THCA.These observations underscore the multifaceted roles that TET enzymes play in the vast expanse of cancer prognosis.
Diving deeper into the complex interplay between TET enzyme expression and immune subtypes, this pan-cancer study unveils their pivotal influence within the tumor immune microenvironment.The findings suggest that TET1 and TET2, especially upregulated in the C5 immune subtype known for its inflammatory profile, might influence immune evasion or activation.TET3, with its variable expression, appears to affect a broad range of immune responses across tumors.The distinct expression patterns of TET enzymes across different cancers and immune subtypes underscore their potential in modulating tumor immunity, progression, and therapy response.This research not only sheds light on the complex interactions between TET enzymes and the immune system but also emphasizes the importance of TET genes as biomarkers and therapeutic targets in cancer.It advances our understanding of the tumor immune microenvironment, paving the way for novel immunotherapeutic strategies.
Building on this understanding, the study further explores the relationship between TET expression and the TME, a critical factor in tumor survival, immune evasion, and drug resistance (Arneth, 2019).For instance, augmenting TET1 expression may improve the TME in PAAD while potentially exacerbating it in GBM, LGG, and TGCT.Insights from Li suggest that TET1 could inhibit epithelialmesenchymal transition and increase PAAD cells' sensitivity to chemotherapy agents like 5FU and gemcitabine (Li et al., 2020).However, the impact of TET1 overexpression on GBM, LGG, and TGCT TME remains unexplored.Secondly, TET2 expression restoration may improve LAML's TME but worsen that of BLCA and GBM.Cimmino et al. found that TET2 restoration might reverse aberrant hematopoietic stem and progenitor cell selfrenewal in vitro and in vivo (Cimmino et al., 2017).Similarly, the effect of inducing TET2 expression on BLCA and GBM TME remains unclear.Lastly, inducing TET3 expression could enhance KICH, KIRC, and LAML TME while deteriorating BLCA, CESC, ESCA, GBM, LUSC, and UCEC TME.This comprehensive analysis sheds further light on the nuanced roles of TET enzymes in the TME, highlighting their potential as biomarkers and therapeutic targets in the ongoing battle against cancer.
Further investigation into the tumor stem cell index revealed a positive correlation between heightened TET1 expression and increased levels of ribonucleic acid synthesis (RNAss) and deoxyribonucleic acid synthesis (DNAss) in testicular germ cell tumors (TGCT).Notably, Benešová et al. observed a marked increase in TET1 dioxygenase expression in most seminomas, suggesting its potential utility as a marker for seminomas and mixed germ cell tumors (Benešová et al., 2017).These findings underscore the potential of TET1 as a pivotal indicator for evaluating stemness maintenance and chemoradiotherapy resistance in TGCT tumor stem cells.
The examination of TET family genes expression concerning the clinical staging of common tumors reveals nuanced patterns.While TET1 demonstrates only limited correlation with tumor staging, TET2 exhibits significant associations in STAD, LIHC, and LUAD (p < 0.01).Similarly, TET3 shows correlations with staging in PAAD, LIHC, and LUSC (p < 0.05).These findings are in line with the research by Liu, indicating decreased genomic 5hmC and 5 fC contents in early LIHC stages, with further reductions in late stages.Moreover, the significantly positive correlations among the expression levels of TET2 in para-tumor tissues were generally attenuated or even disappeared in LIHC tumor tissues (Liu et al., 2019).Moreover, Sajadian underscore the impaired expression and activity of TET2 and TET3 in hepatocellular carcinoma, further validating our analysis (Sajadian et al., 2015).These conclusions are close to our analysis results.
Effective cancer treatment relies on understanding drug sensitivity, a cornerstone of personalized therapy and precision medicine advancement (Chaudhry and Asselin, 2009).Yet, due to individual variations and disparate drug responses, optimal resource utilization remains a challenge (Panczyk, 2014).Therefore, investigating molecules influencing drug reactions is essential for refining treatment strategies.Our study delves into the interplay between TET family genes expression and drug sensitivity, yielding significant insights.Notably, we find that TET genes expression levels correlate with the efficacy of specific drugs, highlighting the importance of assessing TET1, TET2, and TET3 expression for informed clinical drug selection.
Our study delved into the intricate role of TET family genes across various aspects of cancer biology.We identified TET1, TET2, and TET3 as pivotal players influencing tumor progression, prognosis, immune response, tumor microenvironment, and drug sensitivity.Notably, the analysis reveals that each TET family member, including TET3, displays unique expression patterns in at least ten detected tumor types.This heterogeneity suggests that TET3 may play distinct roles in different cancer subtypes.Furthermore, the finding that TET3 genes display hypomethylation in most cancers, which correlates closely with patient prognosis, highlights its potential involvement in cancer progression and metastasis.The association between TET3 expression and various cancer-related factors, such as pathway activity, tumor microenvironment, stemness score, immune subtype, clinical staging, and drug sensitivity, further underscores its relevance as a potential therapeutic target.The results from molecular biology and cytology experiments validating the potential role of TET3 in tumor progression strengthen this argument.
In summary, studying TET3 in different cancers is highly relevant due to its potential as a therapeutic target.The comprehensive pan-cancer analysis presented in the manuscript provides a foundation for future research aimed at developing targeted therapies that may improve cancer treatment outcomes.

Conclusion
The comprehensive analysis of TET family genes in pan-cancer unveiled their multifaceted roles.Through transcriptome data analysis, distinct expression patterns were observed across various tumor types, indicating their potential diagnostic significance.Moreover, correlation with prognosis highlighted their prognostic value, while associations with the tumor microenvironment and drug sensitivity underscored their therapeutic implications.These findings suggest that TET genes could serve as valuable targets for precision medicine approaches in cancer treatment.

Data download preparation
Based on data obtained from the UCSC Xena database, RNA-Seq and clinical data for 33 tumor types prefixed with "GDC TCGA" were downloaded.This includes "HTSeq FPKM (n = 151) GDC Hub" data for gene expression RNAseq, as well as "Phenotype (n = 697) GDC Hub (Clinical Traits)" and "Survival Data (n = 626) GDC Hub" under the Phenotype category.TCGA pan-cancer (PANCAN) data, such as "Immune subtype" under Phenotype and "Stemness score (DNA methylation based) pan-cancer Atlas Hub" and "Stemness score (RNA based) pan-cancer Atlas Hub" under Signatures, was also retrieved.Additionally, drug sensitivity information was obtained from the CellMiner ™ database (https:// discover.nci.mih.gov/cellminer/home.do).

Differential analysis of gene expression
Utilize Perl software to extract, transform, and integrate the 'HTSeq FPKM (n = 151) GDC Hub' data obtained from the Gene expression RNAseq item.Generate boxplots to illustrate the expression profiles of TET family genes across diverse tumor samples.Next, filter samples with a minimum of five normal controls per tumor type and create gene expression boxplots accordingly.Conduct expression difference analysis of TET family genes in different cancer types using the 'Wilcox.test' method, with statistical significance levels indicated by '*', '**', and '***' for p < 0.05, <0.01, and <0.001, respectively.Utilize the R package 'Pheatmap' to generate a heatmap based on the resulting p-values.Finally, examine the correlation between genes within the TET family using the R package 'Coreplot'.

Somatic mutation analysis
Data on single nucleotide variant (SNV) and copy number variant (CNV) for 33 tumor types were retrieved from the TCGA database through the Xena Functional Genomics Explorer (https://xenabrowser.net/datapages/).
The SNV data encompasses various non-silent mutations, including Missense_Mutation, Nonsense_Mutation, Frame_Shift_Del, Splice_Site, Frame_Shift_Ins, In_Frame_Del, In_Frame_Ins, Translation_Start_Site, and Multi_Hit.The SNV mutation frequency (%) for each gene coding region is calculated as the number of mutation samples divided by the total number of cancer samples.Finally, SNV landscape maps were generated using Maftools.
For CNV analysis, we classified CNV into two types: homozygous and heterozygous, representing amplifications and deletions, respectively.We then computed the percentage statistics for each CNV subtype using GISTIC processed CNV data.Next, we investigated the correlation between CNV and mRNA expression levels.Genes with CNV exceeding 5% were identified, and their association with TET expression was explored.Utilizing the method described by Tyagi (Tyagi et al., 2024), we assessed the correlation between mRNA expression and CNV percentage samples using Pearson's product-moment correlation coefficient, with p-values corrected for false discovery rate (FDR).

Methylation analysis
To perform methylation analysis, we initially obtained DNA methylation data from the UCSC Xena database.We focused on 14 tumor types with a minimum of 10 paired samples of tumor and normal tissues.Differential methylation between tumor and normal samples was assessed using Student's t-test, with p-values adjusted for false discovery rate (FDR).Significance was determined at FDR <0.05.Subsequently, we integrated the methylation data with TET gene expression data.Spearman correlation coefficient was computed to evaluate the correlation between methylation levels and gene expression.
Further analysis involved categorizing the median methylation level of genes into two groups.This categorization was based on the threshold defined by the median methylation level.Cox regression analysis was then conducted to estimate the hazard ratio (HR) of gene methylation, considering covariates specific to each cancer type.A Cox coefficient >0 indicated worse survival rates for the high methylation group, hence defining it as high-risk, while a Cox coefficient ≤0 indicated low-risk.Additionally, we compared the distribution of the two methylation groups using the log-rank test to assess their association with clinical outcomes.A significance level of p < 0.05 was considered statistically significant in these comparisons.
Pathway activity scores are calculated by summing the relative protein levels of positive regulatory components and subtracting those of negative regulatory components within each pathway.Following this, employing methodologies outlined in prior studies, such as those by Akbani and Ye.(Akbani et al., 2014), pathway activity scores (PAS) are derived.A higher PAS in one group compared to another suggests an activating effect of certain genes on the pathway, while a lower PAS indicates an inhibitory effect.

Survival and prognostic analysis
We performed expression survival analysis by integrating mRNA expression data of TETs genes with clinical survival data across 33 cancer types, utilizing the Survival Data (n = 626) GDC Hub from the UCSC Xena database.Employing the Kaplan-Meier method and log-rank test (p < 0.05), tumor samples were stratified into high and low expression groups based on median gene expression levels.
Subsequently, we applied a univariate Cox proportional hazards regression model to investigate the relationship between TET family genes expression and patient prognosis across various cancers.Finally, the results were visualized using forest plots generated with the "survival" and "Forestplot" R packages.scratch closure or the rate of migration, typically by measuring the remaining scratch width using image analysis software.This assay provides insights into cell migration dynamics and can be used to assess the effects of various factors on cell motility.
The Transwell assay, cells are seeded into the upper chamber of a Transwell insert, while the lower chamber is filled with medium containing chemoattractant.The cells are allowed to migrate or invade through the porous membrane of the insert towards the lower chamber for a specified period of time.After incubation, nonmigratory or non-invasive cells on the upper surface of the membrane are removed, while cells that have migrated or invaded to the lower surface are fixed, stained, and counted under a microscope.The number of migrated or invaded cells is quantified to assess the migratory or invasive capacity of the cells in response to different experimental conditions or treatments.
Flow cytometry analysis of the cell cycle and apoptosis, cells are typically harvested, fixed, and permeabilized to preserve their structural integrity.For cell cycle analysis, the fixed cells are treated with DNA intercalating dyes, such as propidium iodide (PI), to stain DNA.The stained cells are then subjected to flow cytometry analysis to measure the DNA content, allowing for the identification of cells in different phases of the cell cycle.Conversely, for apoptosis analysis, cells are stained with fluorescent dyes that selectively bind to apoptotic cells, such as Annexin V and PI.The stained cells are then analyzed by flow cytometry to quantify the percentage of apoptotic cells based on their fluorescence properties.By comparing treated samples to untreated controls, the effects of various treatments or experimental conditions on the cell cycle progression and apoptosis induction can be assessed.

Statistics methods
Each validation experiment included three replicates and was repeated thrice for reliability.Data analysis and graphing were performed using GraphPad software v.5.01, with results displayed as mean ± SEM.Student's t-test compared two independent groups, while One-way ANOVA assessed multiple groups.Statistical significance was set at p < 0.05, denoted as *p < 0.05, **p < 0.01, ***p < 0.001, and ns for no significant difference.

FIGURE 1
FIGURE 1 Expression and differential analysis of TET family genes in pan-cancer.(A) TET1, TET2, and TET3 expression was assessed across pan-cancer samples.(B-D) Differences in TET1-3 expression between pan-cancer and normal tissues were analyzed.(E) A heatmap visualized expression disparities among tumors using Wilcox analysis.(F) The expression correlation among TET family genes was analyzed in pan-cancer.(G-I) TET3 protein expression was evaluated across various cancer cell types (G), lung tumor subtypes (H), and gastric cancer cell differentiation levels (I).Data are shown as mean ± SD, with statistical significance denoted as *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 2
FIGURE 2 Somatic mutations in TET family genes were analyzed.(A) Single Nucleotide Variants (SNV) frequencies of TETs across cancers are presented in a table, indicating mutated gene counts per cancer type.'0' indicates that there was no mutation in the gene coding region, and no number indicates there was no mutation in any region of the gene.(B) SNV variant types of TETs illustrate mutation distribution and categorization.(C) copy number variant (CNV) distribution across 33 cancers is depicted in pie charts, showing combined heterozygous/homozygous CNV proportions for each gene in each cancer.Hete Amp = heterozygous amplification; Hete Del = heterozygous deletion; Homo Amp = homozygous amplification; Homo Del = homozygous deletion; None = no CNV.(D, E) Heterozygous and homozygous CNV profiles display the percentage of CNV, including amplification and deletion, for each gene in each cancer.Only genes with >5% CNV in a given cancer are shown as a point on figure.(F) CNV correlation with mRNA expression is depicted through Person's correlation analysis, with point size indicating statistical significance.FDR, false discovery rate.

FIGURE 3
FIGURE 3Methylation analysis of TET family genes in pan-cancer.(A) Methylation levels of TETs genes were examined to assess epigenetic regulation.(B) Correlation between methylation and mRNA expression of TETs genes in pan-cancer was analyzed.Then, association of survival rate with methylation level of TETs genes was investigated (C).

FIGURE 4
FIGURE 4 Pathway activity analysis of TET family genes in pan-cancer.(A) Combined percentage of TET genes' effect on pathway activity is depicted.Red indicates activation, while blue indicates inhibition.Numbers in the table represent the percentage of pathway activity on TETs.Red represents activate effect and blue represents inhibition.(B) Impact of TET3 on protein expression levels of key signaling pathway members was assessed via Western blot.(C, D) Therapeutic sensitivity of lung and gastric cancer cells to Sunitinib, a multi-target tyrosine kinase inhibitor, was tested to elucidate the role of TET3 in regulating the RTK pathway as shown in Figure 4A.

FIGURE 7
FIGURE 7 Correlation between expression of TET family members and immune subtypes.(A) Analysis of the relationship between TET family genes expression and immune subtypes across pan-cancer, based on the TCGA database.(B) Further analysis of the correlation between immune subtypes of eight common clinical tumors and TETs expression levels.Data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 8 TET
FIGURE 8 TET family members and tumor microenvironment in pan-cancer.(A-C) Association between TET family genes expression and tumor microenvironment parameters including tumor matrix score, immune score, and tumor purity score in pan-cancer.(D, E) Relationship between TET family genes expression and RNAs and DNAs in pan-cancer.

FIGURE 10
FIGURE 10Relationship between TET gene family expression and clinical characteristics in eight common tumors.

FIGURE 11 Correlation
FIGURE 11    Correlation analysis between TET gene family expression and drug sensitivity.(A) The potential correlation between gene expression levels and drug sensitivity of TET family members were analyzed based on the CellMiner™ database.(B) We tested the therapeutic sensitivity of three different pathological subtypes of lung cancer cell lines to Lapatinib, AZD-9291, and (+) -JQ1.

TABLE 1
Differential expression of TET family members in various tumors.The single nucleotide variants index of TETs in the tumor genome.

TABLE 2 (
Continued) The single nucleotide variants index of TETs in the tumor genome.

TABLE 2 (
Continued) The single nucleotide variants index of TETs in the tumor genome.

TABLE 3
The copy number variant index of TETs in the tumor genome.

TABLE 3 (
Continued) The copy number variant index of TETs in the tumor genome.

TABLE 3 (
Continued) The copy number variant index of TETs in the tumor genome.
COAD and LUSC immune subtypes showed significant correlations only with TET1 and TET3 expression.TET1 was highly expressed across all COAD immune subtypes, while TET3 was least expressed in C3.In LUSC, TET1 displayed minimal expression in C3 and C4, while TET3 was lowest in C6 and stable in other subtypes.In summary, TET1 exhibited consistently low expression across

TABLE 4
The correlation between CNV and their mRNA expression.

TABLE 4 (
Continued) The correlation between CNV and their mRNA expression.

TABLE 4 (
Continued) The correlation between CNV and their mRNA expression.

TABLE 5
Signaling pathways that are significantly correlated with TETs.