5-Methylcytosine methylation predicts cervical cancer prognosis, shaping immune cell infiltration

Abstract

BackgroundEpigenetics, encompassing DNA and RNA modifications, has emerged as a prominent area of research in the post-genomic era. Numerous studies have elucidated the impact of epigenetics on tumor regulation. However, the methylation patterns of 5-methylcytosine in cervical cancer as well as its role in the tumor microenvironment and immunotherapy remain poorly understood.MethodsUtilizing a comprehensive dataset encompassing samples from 306 patients with cervical cancer from The Cancer Genome Atlas and Gene Expression Omnibus repositories, we conducted an in-depth analysis to evaluate the potential association between the modification patterns of 5-methylcytosine and the infiltration of cells within the tumor microenvironment, taking into account 11 regulators of 5-methylcytosine modification. Subsequently, we employed stepwise regression and Least Absolute Shrinkage and Selection Operator Cox regression to quantify 5-methylcytosine modification patterns in patients with cervical squamous cell carcinoma and endocervical adenocarcinoma, yielding the 5-methylcytosine score. Our study explored the link between the 5-methylcytosine score and clinical characteristics as well as prognostic outcomes in patients with cervical squamous cell carcinoma and endocervical adenocarcinoma.ResultsA comprehensive analysis of 306 patients with cervical cancer revealed two distinct 5-methylcytosine modification patterns, henceforth labeled as 5-methylcytosine clusters A and B. These clusters exhibited distinct immunological profiles and biological attributes, with 5-methylcytosine cluster A exhibiting a higher degree of immune cell infiltration. Utilizing univariate Cox regression analysis, we identified 367 genes regulated by 5-methylcytosine that were significantly correlated with patient prognosis. This analysis further stratified the samples into three distinct genomic subtypes. Survival analyses indicated that patients belonging to gene cluster C exhibited more favorable survival outcomes than those belonging to gene clusters A and B. Intriguingly, most 5-methylcytosine regulatory factors had higher expression levels in gene cluster B than in gene cluster A. Gene set enrichment analysis of a single sample revealed elevated immune cell infiltration within gene cluster B, indicating a stronger immune response in this cluster. The 5-methylcytosine score feature was utilized to determine the 5-methylcytosine modification pattern in cervical cancer, revealing that patients with low 5-methylcytosine scores exhibited better survival rates, whereas those with high scores had increased mutation frequencies and better treatment responses.ConclusionsThis research underscores the key role of 5-methylcytosine modification patterns in cervical cancer. Analysis of these patterns will deepen our understanding of the cervical cancer tumor microenvironment, paving the way for the development of more refined and effective immunotherapy strategies.

Keywords: 5-methylcytosine; Cervical cancer; RNA methylation; immunotherapy; tumor microenvironment.

Figure 1.

Genetic variation of m5C regulators in cervical cancer. (a) Mutation frequency of 24 m5C regulators in 289 patients with cervical cancer from TCGA-CESE. (b) CNV frequency of the m5C regulators analyzed. (c) CNV locations mapped on 23 chromosomes from TCGA data. (d) Significant differences in m5C regulator expression between normal and tumor samples. (*p < 0.05, **p < 0.01, ***p < 0.001; ns, not significant). TCGA: The Cancer Genome Atlas; CESE: cervical squamous cell carcinoma and endocervical adenocarcinoma; CNV: copy number variation; m5C: 5-methylcytosine.

Figure 2.

Identification of distinct m5C modification patterns via analysis of m5C regulators. (a) Identification of the network of m5C regulators in CESE. (b) Heatmap of two m5C modification patterns using 24 regulators. (c) KEGG pathway analysis. (d) Survival analysis of m5C regulators in cervical cancer using TCGA and GSE30759 datasets. (e) DNMT3B expression in TET3 wild-type and mutants. CESE: cervical squamous cell carcinoma and endocervical adenocarcinoma; TGCA: The Cancer Genome Atlas; m5C: 5-methylcytosine.

Figure 3.

GSVA enrichment analysis and characteristics of tumor microenvironment cell infiltration in two m5C modification modes of cervical cancer. (a) Consensus clustering matrix at k = 2. (b) Principal component analysis between Cluster A and Cluster B scoregroups in cervical cancer. (c) GO enrichment of m5C DEGs. (d) KEGG pathways of m5C DEGs. GSVA: gene set variation; GO: Gene Ontology; DEGs: differentially expressed genes; KEGG: Kyoto Encyclopedia of Genes and Genomes; m5C: 5-methylcytosine.

Figure 4.

m5C gene patterns mediated by m5C phenotype-related DEGs. (a) Pearson correlation analysis revealed m5C regulator relationships, selecting a k = 3 clustering matrix. (b) Kaplan–Meier curves showed survival differences across m5C gene patterns. (c) Analysis of 24 m5C regulators, significance indicated by asterisks. (d) Heatmap highlights distinct m5C gene patterns. (e) Alluvial diagram showing m5C score, clusters, genes, and survival correlations and (f) m5C scores differed significantly among three gene patterns. DEGs: differentially expressed genes; m5C: 5-methylcytosine.

Figure 5.

The m5C gene signature and its role in immunotherapy. (a) Patient survival proportions vary between low- and high-m5C score subgroups. (b) Tumor m5C score was positively correlated with TMB. (c) Kaplan–Meier curves showed survival differences by m5C score. (d) Kaplan–Meier curves compared G1-2 survival with G3-4 survival. (e) The relationship between IPS and the m5C scoring system, encompassing IPS-PD1, IPS-CTLA4, and IPS-PD1/CTLA4, was examined in both low- and high-m5C score subgroups. (f) The m5C score was correlated with immune cell infiltration in the TME, indicated by blue (negative) and red (positive). TMB: tumor mutation burden; IPS: immunophenotypic scores; m5C: 5-methylcytosine.

Figure 6.

Characteristics of cervical cancer subtypes related to m5C and mutations. (a) Stratification of patients with cervical cancer by m5C score and TMB. (b) Survival analysis based on TMB. (c) Survival curve combining TMB and m5C score. (d, e) Mutation frequencies in high- and low-m5C score subgroups. TMB: tumor mutation burden; m5C: 5-methylcytosine.