Expression of TET protein family in pan-cancer
We analyzed the TCGA and GTEx databases to gather information on the expression levels of TET protein family across 33 types of cancer. During our research, it was discovered that TET1 showed high levels of expression in 14 different tumors, including ACC, CHOL, DLBC, ESCA, HNSC, LGG, LIHC, LUAD, LUSC, PAAD, STAD, TGCT, THYM and USC. However, it was low expressed in BRAC, CESC, COAD, KICH, LAML, OV, PRAD, READ, THCA (Fig. 1A). Compared to normal tissues, TET2 expression showed a significant increase in 10 types of tumor tissues, including CHOL, ESCA, GBM, KIRC, LAML, LGG, PAAD, PCPG, STAD and TGCT,
while it decreased significantly in 15 types of tumor tissues, including ACC, BRAC, COAD, DLBC, LIHC, LUAD, LUSC, OV, PRAD, READ, SKCM, THCA, THYM, UCEC and UCS (Fig. 1B). Moreover, TET3 expression significantly increased in 23 types of tumor tissues, such as BLCA, BRCA, CESC, CHOL, COAD, ECSA, GBM, HNSC, KIRC, KIRP, LAML, LGG, LUSC, OV, PAAD, PCPG, PRAD, READ, STAD, TGCT, THCA, UCEC and UCS, compared to normal tissues, whereas it significantly decreased in 5 types of tumor tissues, including ACC, DLBC, KICH, LUAD and SKCM (Fig. 1C).
Figure 1. Expression analysis of TET protein family in pan-cancer. (A) TET1. (B) TET2. (C) TET3. *P < 0.05, **P < 0.01, ***P < 0.001.
Survival analysis of TET protein family in pan-cancer
In patients with ACC, KIRP, LIHC, MESO, SARC, and STAD, the Cox regression analysis showed that high TET1 expression was associated with a poor prognosis. Conversely, high TET1 expression in patients with LGG + GBM and THYM was related to a favorable prognosis (Fig. 2A). A higher overall survival is observed in patients with high levels of TET2 expression in LGG + GBM, KIRC and READ (Fig. 2B). In ACC, MESO, and SARC, high TET3 expression was associated with poor overall survival, whereas in LGG and GBM, there was a higher rate of overall survival (Fig. 2C).
Subcellular localization of TET protein family
We acquired immunofluorescence images of the TET protein family from the HPA database in order to understand their subcellular localization and gain a better understanding of this protein family. As shown in Fig. 3, the main location of TET1, TET2, and TET3 was in the nucleoplasm.
Analysis of TET protein family expression in glioma
Based on the CGGA data, our findings revealed that glioma tissues exhibit higher expression levels of TET protein family in comparison to normal tissues. Using this database, we also investigated the connection between TET protein family expression and IDH status in glioma. Our findings showed that TET protein family expression was higher in patients with an IDH-mutant compared to those with IDH-wildtype. In both IDH-wildtype and IDH-mutant glioma, the results suggested that GBM displayed a lower level of TET1 expression compared to LGG. Although the expression of TET2 and TET3 was increased in LGG compared to GBM, there is no significant difference (Fig. 4A). The results from the TCGA database are similar to those from the CGGA database. However, there is a difference in the expressions of TET2 and TET3 between LGG and GBM in IDH-wildtype glioma that has significant significance. This difference may be attributed to different sample sizes in the two databases (Fig. 4B). We found that TET1 expression is higher in glioma and decreases as the tumor grade progresses in both CGGA and TCGA databases. Consequently, we focused on studying the role of TET1 in glioma.
Expression of TET1 in Ditan data
To determine the protein expression of TET1 in glioma, immunohistochemical staining was performed. The results showed a significant increase of TET1 in glioma compared to the normal brain (Fig. 5). Additionally, the findings also showed that the expression of TET1 was higher in LGG tissues than in GBM tissues, regardless of whether the glioma was IDH-mutant or IDH-wildtype.
The relationship between TET1 expression and clinicopathological features
In both the CGGA and TCGA databases, the expression of TET1 showed no correlation with age or gender in IDH-mutant and IDH-wildtype glioma (Fig. 1S). In our Ditan data, we also found that there was no correlation between age, gender, and the expression of TET1 (Table 1).
Table 1
Association between TET1 expression and clinical characteristics in Ditan data.
| Total | TET1 expression | IHC SQS | P value |
n = 76 | Low expression | High expression | (0ཞ12) |
Age | | | | | |
<=50 years | 47 | 22 | 25 | 5.08 ± 0.42 | 0.6383 |
> 50 years | 29 | 15 | 14 | 4.76 ± 0.60 |
Gender | | | | | |
Male | 44 | 24 | 20 | 5.50 ± 0.54 | 0.1841 |
Female | 32 | 13 | 19 | 4.57 ± 0.44 |
IDH- mutant | | | | | |
LGG | 33 | 13 | 30 | 5.48 ± 0.41 | 0.0366* |
GBM | 4 | 3 | 1 | 2.75 ± 1.11 |
IDH-wildtype | | | | | |
LGG | 12 | 2 | 10 | 7.25 ± 0.94 | 0.038* |
GBM | 27 | 18 | 9 | 3.62 ± 0.56 |
* P < 0.05.
Survival prognosis of TET1
After discovering that the expression of TET1 is associated with tumor grade, we further investigated the impact of TET1 expression on patient prognosis. Based on the CGGA database, TCGA database and Ditan data shown in Fig. 6A, the results of the KM survival analysis indicated that decreased TET1 expression was associated with poor overall survival in both all-grade glioma and LGG. However, there was no correlation found between the expression of TET1 and the prognosis in GBM. Age, tumor grade, and TET1 expression were discovered to have a significant association with OS in glioma patients, according to the univariate Cox analysis. Furthermore, the multivariate Cox analysis results demonstrated that age, tumor grade, and TET1 expression served as independent prognostic factors for patients with glioma (Fig. 6B).
The association between TET1 and the infiltration of immune cells in glioma
The ssGSEA algorithm was used to evaluate the correlation between TET1 expression and 24 different immune cell types. The findings indicated that there was a positive correlation of TET1 expression with the levels of infiltration of Tcm, Tgd, pDc, T helper cells, Tem, CD8 T cells, and TFH cells. On the other hand, increased TET1 expression led to a decrease in macrophages, neutrophils, eosinophils, cytotoxic cells, T cells, aDC, iDC, NK CD56 dim cells, NK cells, Th17 cells, Th2 cells, B cells, mast cells, Th1 cells, and CD56 bright cells (Fig. 7A). To evaluate the association between TET1 levels and the immune response, we utilized the ESTIMATE algorithm. This algorithm provided ImmuneScore, StromalScore, and ESTIMATEScore values by utilizing information obtained from the TCGA database. In glioma, a significant negative correlation was discovered between the expression of TET1 and the ImmuneScore, StromalScore, and ESTIMATEScore (Fig. 7B).
GO and KEGG functional enrichment analysis
To understand the potential roles of TET1, GO enrichment analysis was performed to investigate the connection between TET1 and signaling pathways. In the BP category terms, the GO enrichment analysis indicated that the TET1 gene was primarily associated with cell-cell adhesion, leukocyte migration, regulation of leukocyte cell-cell adhesion, and leukocyte cell-cell adhesion. In the group of CC terms, the TET1 showed significant enrichment in the collagen-containing extracellular matrix, external side of the plasma membrane, endoplasmic reticulum lumen, and vesicle lumen. The TET1 also showed significant enrichment in signaling receptor activator activity, receptor-ligand activity, extracellular matrix structural constituent, and immune receptor activity in MF terms. These results show that TET1 is mainly present in immune system-related signaling pathways (Fig. 8A). Furthermore, according to the KEGG analysis, TET1 was predominantly found to be enriched in cytokine-cytokine receptor interaction, virtual protein interaction with cytokine and cytokine receiver, hematopoietic cell lineage, systematic lupus erythematosus, and Staphylococcus aureus infection (Fig. 8B). The research results confirmed the significant role of TET1 in the immune system, as it was discovered to be responsible for controlling immune factors.
Drug sensitivity analysis
Using the IC50, we optimized chemotherapeutic drugs for glioma. Patients with lower IC50 values were more sensitive to the drug. The IC50 values for NPK76-II-72-1, NSC-207895, FK866, and vorinostat were lower in the TET1 high-expression group suggesting that patients with TET1 overexpression were more receptive to these drugs.
However, the low TET1 expression group was more susceptible to Temozolomide, and Trametinib (Fig. 9).