Genetic characteristics and prognostic implications of m1A regulators in pancreatic cancer

Abstract

Studies have identified the methylation of N1 adenosine (m1A), an RNA modification, playing an important role in the progression of the tumorigenesis. The present study aimed to analyze the genetic characteristics and prognostic value of m1A regulators in pancreatic cancer. In the present study, data on gene mutations, single-nucleotide variants (SNVs), and copy number variation (CNV) were obtained from 363 patients with pancreatic cancer in the Cancer Genome Atlas (TCGA) database, and survival analysis was performed using the logarithmic rank test and Cox regression model. The chi-squared test was used to examine the relationship between the changes in m1A regulatory factors and clinicopathological characteristics. And we used ICGC database to verify the reliability of prognostic markers. The results show that changes in m1A-regulating genes are related to clinical stage and that the expression of some m1A-regulating genes is positively correlated with CNV. In addition, the low expression of the 'eraser' gene ALKBH1 is related to the poor prognosis of patients with pancreatic cancer, and its expression level has important clinical significance for patients with pancreatic adenocarcinoma (PAAD). Mechanistically, ALKBH1 may participate in the occurrence and development of pancreatic cancer through mTOR and ErbB signaling pathway. The expression of m1A-regulating genes can be used as a prognostic marker for pancreatic cancer. These findings provide valuable clues for us to understand the epigenetics of m1A in pancreatic cancer.

Keywords: N1-methyladenosine; Pancreatic cancer; Prognostic marker; RNA modification.

Figure 1. Mutations and CNVs of m1A-regulating genes in PAAD

(A) Statistics of mutation frequencies of m1A-regulating genes with different functions in PAAD. (B) The mutation site of YTHDC1. (C) The relationship between eight functionally altered genes and patient survival. (D) CNV statistics of m1A-regulating genes in PAAD.

Figure 2. Correlation between the CNV and mRNA expression levels of ten m1A-regulating genes

(A–I) The significant relationship between the CNV and expression levels of nine m1A-regulating genes. (J) There was no significant relationship between the CNV and expression level of m1A-regulating genes.

Figure 3. Association between m1A-regulating genes and survival in patients with pancreatic cancer

(A) The relationship between different clinical stages and patient prognosis. (B) Clustering heat map of m1A-regulating genes and cases of different stages.

Figure 4. Expression levels of ten m1A-regulating genes in cases of different clinical stages

(A and B) Significant expression of two m1A-regulating genes in cases of different clinical stages. (C–J) Eight m1A-regulating genes are not significantly differentially expressed in cases of different clinical stages.

Figure 5. Survival and AUC in PAAD patients and GSEA results of different expression levels of ALKBH1

(A) Survival curve of multivariate Cox regression. (B) AUC curve of multivariate Cox regression. (C) The relationship between ALKBH1 expression and patient prognosis. (D) Risk analysis of the ALKBH1 gene. (E) AUC of the ALKBH1 gene. (F–H) GSEA results of ALKBH1 expression. (F) histone methyltransferase activity. (G) RNA polyadenylation. (H) peptide N-acetyltransferase activity.

Figure 6. Validation of the relationship between the ALKBH1 gene and survival in the data set

(A) Risk analysis of the ALKBH1 gene in the validation data set. (B) AUC of the ALKBH1 gene in the validation data set. (C) The relationship between ALKBH1 expression and patient prognosis. (D) Pearson correlation between m1A regulators and KEGG pathways. Relationship of m1A regulator genes with mTOR and ErbB expression. The correlogram shows that major m1A regulator genes have a significant correlation with the mTOR and ErbB signaling pathways. Red represents positive correlations, while blue represents negative correlations; *P<0.05.