Pan-cancer analysis identifies YTHDF2 as an immunotherapeutic and prognostic biomarker

Abstract

Background: N⁶-methyladenosine (m6A) modification is a dynamic and reversible post-transcriptional RNA modification prevalent in eukaryotic cells. YT521-B homology domain family 2 (YTHDF2) has been identified as a member of m6A reader protein involving in many vital biological processes, whereas its role and functional mechanisms in cancers remain unclear. Methods: Bioinformatics analyses were performed on multiple databases including GTEx, TCGA, GEO and Cbioportal to explore the connection between YTHDF2 expression and its genomic changes including tumor mutation burden, microsatellite instability and mismatch repair in 33 different cancer types. We also investigated the association of YTHDF2 expression with prognosis, immune infiltration, tumor microenvironment, immune checkpoints and chemokines. Besides, the correlation of YTHDF2 expression with copy number variation and promoter methylation was also studied in tumors compared with normal tissues. At last, we analyzed the protein-protein interacting network and related genes of YTHDF2 to enrich its potential functional mechanism in tumor development and progression. Real-time qPCR was used to verify the expression of YTHDF2-related genes in colorectal cancer cells, and immunohistochemical staining was adopted to verify immune infiltration in tissue sections from 51 hepatocellular carcinoma patients. Results: YTHDF2 was overexpressed in a majority of tumor types and associated with their poor overall survival, progression-free interval, and disease-specific survival. The correlation of YTHDF2 expression with tumor mutation burden, microsatellite instability and mismatch repair was also detected in most of the tumor types. Moreover, YTHDF2 might participate in the immune regulation through influencing the expression of immune checkpoint genes and the infiltration of immunocytes in tumor microenvironment. Notably, we demonstrated a positive correlation between YTHDF2 expression and the infiltration of CD8⁺ T cells and macrophages in many tumors, and it was verified in 51 clinic hepatocellular carcinoma tissues. In addition, the involvement of YTHDF2 in "Spliceosome" and "RNA degradation" were two potential functional mechanisms underlying its influence on tumor progression. The regulation of YTHDF2 on predicted genes has been verified in CRC cells. Conclusion: YTHDF2 might be a new therapeutic target and a potential biomarker of cancer immune evasion and poor prognosis.

Keywords: YTHDF2; immune cell infiltration; immunotherapy; prognosis; tumor microenvironment.

FIGURE 1

Differential expression of YTHDF2 across cancers. (A) The expression of YTHDF2 in tumor and adjacent normal tissues using the TCGA data. (B) The differential expression of YTHDF2 in tumor and normal tissues of 27 cancer types using the combined data of TCGA and GTEx. (C) The expression of YTHDF2 in tumors and paired adjacent normal tissues. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 2

The mutation of YTHDF2 in cancers analyzed using TCGA data. (A) The mutation frequency and types of YTHDF2 in different cancer types. (B) The mutation information of YTHDF2 in different cancers. (C) The correlation between YTHDF2 expression and CNV. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3

YTHDF2 expression is associated with overall survival time (OS). (A) Forest plot of OS associations in 34 cancer types. (B) Kaplan-Meier analysis of the association between YTHDF2 expression and OS.

FIGURE 4

The predictive potential of YTHDF2 on prognosis. (A) ROC analysis demonstrates the predictive potential of YTHDF2 on prognosis in CHOL, KICH, LGG, READ, LIHC, KIRP, and ACC. (B) Time-dependent ROC analysis demonstrates the predictive potential of YTHDF2 on 1-year, 3-year, and 5-year survival rate of patients with KICH, LGG, LIH, and KIRC.

FIGURE 5

The correlation between YTHDF2 expression and immuneScore, stromalscore and estimatescore in GBM, UCEC and ACC. (A) the correlation between YTHDF2 expression and Immunescore in UCEC, TGCT, and ACC. (B) The correlation between YTHDF2 expression and stromalscore in UCEC, TGCT and ACC. (C) The correlation between YTHDF2 expression and estimatescore in GBM, UCEC and ACC.

FIGURE 6

The correlation between YTHDF2 expression and immune infiltration in different cancer types. (A) The correlation between YTHDF2 expression and immune infiltration in the top three tumors. (B) Immunohistochemical images of hepatocellular carcinoma show intra-tissue characteristics of CD3⁺CD8⁺ T cells/CD68 macrophages with high and low expression of YTHDF2. (C) The correlation analysis of YTHDF2 expression with CD3⁺CD8⁺ T cells and CD68 macrophage infiltration in hepatocellular carcinoma validation cohort.

FIGURE 7

The correlation between YTHDF2 expression and pan-cancer immune checkpoint genes. *p < 0.05; **p < 0.01; ***p < 0.001.

FIGURE 8

The correlation between YTHDF2 expression and the MMR (A), TMB (B), and MSI (C). *p < 0.05; **p < 0.01; ***p < 0.001.

FIGURE 9

The protein-protein interactions network, Go enrichment analysis and KEGG pathway analysis. (A) The YTHDF2 interacting proteins that screened by the online STRING tools. (B) The top 100 YTHDF2-related genes in TCGA database, among which the top five targeting genes are screened out. (C) The heatmap showed the expression of the top five YTHDF2 targeting genes in different cancer types. (D) The cross analysis of YTHDF2-interacting and YTHDF2-related genes. (E) GO enrichment analysis based on the YTHDF2-interacting and YTHDF2-related genes. (F) The KEGG pathway analysis based on YTHDF2-interacting and YTHDF2-related genes.