Modification patterns and metabolic characteristics of m6A regulators in digestive tract tumors

Abstract

M6A is essential for tumor occurrence and progression. The expression patterns of m6A regulators differ in various kinds of tumors. Transcriptomic expression statistics together with clinical data from a database were analyzed to distinguish patients with digestive tract tumors. Based on the expression patterns of diverse m6A regulators, patients were divided into several clusters. Survival analysis suggested significant differences in patient prognosis among the m6A clusters. The results showed overlapping of m6A expression patterns with energy metabolism and nucleotide metabolism. Functional analyses imply that m6A modifications in tumor cells probably drive metabolic reprogramming to sustain rapid proliferation of cancer cells. Our analysis highlights the m6A risk characterizes various kinds of metabolic features and predicts chemotherapy sensitivity in digestive tract tumors, providing evidence for m6A regulators as markers to predict patient outcomes.

Keywords: Digestive; Epigenetics; Metabolism; Tumor; m6A.

Fig. 1

Landscape of genetic variation and mRNA expression levels of 25 m6A regulators. (A–D) Expression of 25 m6A regulators between normal and tumor tissues in ESCA, STAD, COAD and READ. Tumors are in red; normal is in blue. The upper and lower ends of the box plots represent the interquartile range of values. The lines in the boxes represent medians and the dots indicate outliers. Asterisks represent statistical P values (*P < 0.05; **P < 0.01; ***P < 0.001). (E–H) Mutation frequencies of m6A regulators in ESCA, STAD, COAD and READ. Each column represents an individual patient, the bar above shows the TMB, and the numbers on the right indicate the mutation frequency of each regulator. The right bar shows the proportion of each variant type, and the stacked bars below show the conversion rate in each sample. (I–L) The circle size represents the effect of each regulator on prognosis, and the values calculated by Log-rank test range from P < 0.0001, P < 0.001, P < 0.01, P < 0.05, and P < 1. The colors in the circles represent the gene type and risk gene type of m6A. The lines connecting the regulators indicate their interactions, and the thickness indicates the strength of the correlation between the regulators. Negative correlations are marked in blue and positive correlations are marked in pink. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 2

Consensus cluster of m6A regulators based on their expression pattern. (A–D) Unsupervised clustering of 25 m6A genes in ESCA, STAD, COAD and READ. cluster, survival status, age, gender, stage, grade and other clinical features were shown. Red represents high expression of regulatory factors; blue represents low expression. (E) The survival analysis was performed for the three m6A modification patterns in ESCA. Kaplan-Meier curve with a log-rank p-value<0.001 showed a significant survival difference between the three m6A modification patterns. (F) Kaplan-Meier curve with a log-rank p-value = 0.005 in STAD. (G) Kaplan-Meier curve with a log-rank p-value = 0.022 in COAD. (H) Kaplan-Meier curve with a log-rank p-value = 0.036 in READ. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Biological behaviors in distinct m6A modification patterns. (A–D) GO analysis of differential genes enrichment between m6A clusters in ESCA, STAD, COAD and READ. (E–H) KEGG analysis of differential genes in distinct m6A modification patterns of ESCA, STAD, COAD and READ.

Fig. 4

Role of metabolism in different m6A clusters. (A–D) Abundance of each metabolism in the several m6A modification patterns in ESCA, STAD, COAD and READ. The upper and lower ends of the boxes represent the interquartile distance of the values. Lines in the boxes represent median values and dots represent outliers. Asterisks represent statistical p-values (P < 0.001, '***', P < 0.01 '**', P < 0.05 '*') (E) Expression of ALKBH5 in high and low expression group of energy metabolism. (F) Expression of HNRNPC in high and low expression group of energy metabolism. (G) Expression of HNRNPC in high and low expression group of nucleotide metabolism. (H) Expression of LRPPRC in high and low expression group of nucleotide metabolism. (I) Expression of YTHDC1 in high and low expression group of nucleotide metabolism. (J–M) Univariate analysis of seven kinds of metabolisms to identify the metabolism that significantly correlated with OS in ESCA, STAD, COAD and READ.

Fig. 5

Construction of the prognostic signature of m6A risk. (A–D) The prognostic signature of m6A risk constructed by the minimum criterion of LASSO Cox regression algorithm in ESCA, STAD, COAD and READ. (E–H) Time-dependent ROC curves for risk models. Results of test sets were shown in four kinds of tumors. (I–L) Survival analysis of risk models. Construction of survival curves for the test set by Kaplan-Meier analysis in ESCA, STAD, COAD and READ.

Fig. 6

Survival analysis were assessed bifactorally using m6A scores and seven metabolisms. (A–C) Metabolism type which showed significance among m6A clusters were chose to be showed in ESCA. (D–F) Metabolism type which showed significance among m6A clusters were chose to be showed in STAD. (G–L) Metabolism type which showed significance among m6A clusters were chose to be showed in COAD. (M–R) Metabolism type which showed significance among m6A clusters were chose to be showed in READ. The results were compared two by two according to the four cases in the lower left corner, and the P values of the compared results were in the upper right corner, and only P < 0.05 results were retained.

Fig. 7

Potency of the m6A risk score in predicting drug sensitivity. (A)The correlation of m6A score and IC50 values of chemotherapy drugs in four kind digestive tract tumors. (B–E) The IC50 data of drugs in high and low m6A risk score group (B), STAD (C), COAD (D) and READ (E). Asterisks represent statistical P values (*P < 0.05; **P < 0.01; ***P < 0.001).

Fig. S1

Copy number changes of m6A regulators.

Fig. S2

Cluster analysis of m6A regulators.

Fig. S3

GSVA enrichment analysis showing the activation status of biological pathways in different m6A modification patterns in ESCA.

Fig. S4

GSVA enrichment analysis in different m6A modification patterns in STAD.

Fig. S5

GSVA enrichment analysis in different m6A modification patterns in COAD.

Fig. S6

GSVA enrichment analysis in different m6A modification patterns in READ.

Fig. S7

Expression of m6A regulators in different groups of metabolisms.

Fig. S8

Validation of the m6A risk score in predicting patient prognostic.