生物
腺癌
小RNA
肿瘤微环境
转录组
计算生物学
肺癌
基因表达调控
聚腺苷酸
癌症研究
基因表达
生物信息学
基因
遗传学
癌症
肿瘤科
医学
作者
Yuchu Zhang,Libing Shen,Qili Shi,Guofang Zhao,Fajiu Wang
标识
DOI:10.3389/fgene.2021.645360
摘要
Background Alternative polyadenylation (APA) is a pervasive posttranscriptional mechanism regulating gene expression. However, the specific dysregulation of APA events and its potential biological or clinical significance in lung adenocarcinoma (LUAD) remain unclear. Methods Here, we collected RNA-Seq data from two independent datasets: GSE40419 ( n = 146) and The Cancer Genome Atlas (TCGA) LUAD ( n = 542). The DaPars algorithm was employed to characterize the APA profiles in tumor and normal samples. Spearman correlation was used to assess the effects of APA regulators on 3′ UTR changes in tumors. The Cox proportional hazard model was used to identify clinically relevant APA events and regulators. We stratified 512 patients with LUAD in the TCGA cohort through consensus clustering based on the expression of APA factors. Findings We identified remarkably consistent alternative 3′ UTR isoforms between the two cohorts, most of which were shortened in LUAD. Our analyses further suggested that aberrant usage of proximal polyA sites resulted in escape from miRNA binding, thus increasing gene expression. Notably, we found that the 3′ UTR lengths of the mRNA transcriptome were correlated with the expression levels of APA factors. We further identified that CPSF2 and CPEB3 may serve as key regulators in both datasets. Finally, four LUAD subtypes according to different APA factor expression patterns displayed distinct clinical results and oncogenic features related to tumor microenvironment including immune, metabolic, and hypoxic status. Interpretation Our analyses characterize the APA profiles among patients with LUAD and identify two key regulators for APA events in LUAD, CPSF2 and CPEB3, which could serve as the potential prognostic genes in LUAD.
科研通智能强力驱动
Strongly Powered by AbleSci AI