m5C Regulator-Mediated Methylation Modification Patterns and Tumor Microenvironment Infiltration Characterization in Papillary Thyroid Carcinoma

Abstract

Recently, immune response modulation at the epigenetic level is illustrated in studies, but the possible function of RNA 5-methylcytosine (m⁵C) modification in cell infiltration within the tumor microenvironment (TME) is still unclear. Three different m⁵C modification patterns were identified, and high differentiation degree was observed in the cell infiltration features within TME under the above three identified patterns. A low m⁵C-score, which was reflected in the activated immunity, predicted the relatively favorable prognostic outcome. A small amount of effective immune infiltration was seen in the high m⁵C-score subtype, indicating the dismal patient survival. Our study constructed a diagnostic model using the 10 signature genes highly related to the m⁵C-score, discovered that the model exhibited high diagnostic accuracy for PTC, and screened out five potential drugs for PTC based on this m5C-score model. m⁵C modification exerts an important part in forming the TME complexity and diversity. It is valuable to evaluate the m⁵C modification patterns in single tumors, so as to enhance our understanding towards the infiltration characterization in TME.

Keywords: 5-methylcytosine (m5C) modification; immune infiltration; papillary thyroid carcinoma (PTC); subtype; tumor microenvironment (TME).

Figure 1

The nine-regulator-mediated m⁵C methylation modification patterns. (A) PCA analysis based on nine m⁵C regulators. (B, C) Consensus clustering was adopted for identifying the different m⁵C modification patterns. (D) The expression levels of 9 m⁵C regulators in different m⁵C modification patterns. (E) Survival analysis of the three subtypes. ***statistical significance.

Figure 2

TME-infiltrating cell features in different m⁵C modification patterns. (A) Biological behaviors in the different m⁵C modification patterns were conducted by GSVA. (B) Levels of stromal scores in different m⁵C modification patterns. (C) The levels of infiltration of 22 immune cells in different m⁵C modification patterns. (D) The expression levels of 37 immune checkpoints in different m⁵C modification patterns. *, **, ***, **** statistic difference at different levels; ns, no significance.

Figure 3

The distribution of clinical features (gender, stage, and age) of samples in the m⁵C-cluster 1–3.

Figure 4

m⁵C gene signature establishment along with functional annotation. (A) KEGG enrichment analysis of 690 DEGs. (B) Levels of stromal scores in three m⁵C gene-cluster subtypes. (C) The levels of infiltration of 22 immune cells in three m⁵C gene-cluster subtypes. *, **, ***, **** statistic difference at different levels; ns, no significance.

Figure 5

Clinical features and transcriptome traits of the m⁵C-associated phenotypes. (A) Alluvial diagram showing the changes of m⁵C modification patterns, gender, age, gene cluster, and the m⁵C-score. (B) DEGs between high and low m⁵C-score samples. (C, D) Differences in DFS (C) and OS (D) between high and low m⁵C-score samples. (E) Relationship between the m⁵C-score value and the score of CD3+CD4+/CD3+CD8+ cells of the peripheral blood samples of 24 GBM patients. The m⁵C-Score value was negatively associated with the ratio of CD3+CD4+/CD3+CD8+ cells. (F) Relationship between the m⁵C-score value and the percentage of CD4+CD25+ Tregs in peripheral blood samples of the 24 GBM patients. The m⁵C-score value was positively related with the percentage of CD4+CD25+ Tregs.

Figure 6

The influences of the m⁵C-score and various immunocyte (resting CD4⁺ memory T cells, CD8⁺ T cells, activated NK cells, and monocytes) infiltration levels on the prognosis for PTC patients.

Figure 7

Construction and verification of the m⁵C-score-based PTC diagnostic model. (A) Comparison of classification results of TCGA-PTC samples by diagnostic model constructed based on 10 signature genes and the m⁵C-score. (B) Accuracy of classification of TCGA samples by a diagnostic model constructed based on 10 signature genes. (C–E) Accuracy of classification of samples in GSE29265 (C), GSE33630 (D), and GSE65144 (E) by a diagnostic model constructed based on 10 signature genes.

Figure 8

Five potential drugs based on the PTC m5C-score model could impair growth of PTC cells. (A) Histogram showing the viability of PTC cells with or without five potential drugs (cephaeline, emetine, anisomycin, ouabain, and thapsigargin) for 48 h at 20 μM. (B) BALB/c mice were subcutaneously injected with PTC cells. After 5 days, the nude mice were treated with cephaeline, emetine, anisomycin, ouabain, or thapsigargin (100 mg/kg daily, intraperitoneal injection). Tumor weights were measured after 6 weeks (n = 5 mice/group). ***statistical significance.