The Functional Characterization of Epigenetically Related lncRNAs Involved in Dysregulated CeRNA–CeRNA Networks Across Eight Cancer Types

• Main Authors: Dahua Xu, Liqiang Wang, Sainan Pang, Meng Cao, Wenxiang Wang, Xiaorong Yu, Zhizhou Xu, Jiankai Xu, Hong Wang, Jianping Lu, Kongning Li
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2021-06-01
• Series: Frontiers in Cell and Developmental Biology
• Subjects: pan-cancer; dysregulated ceRNA; epigenetically related lncRNA; diagnostic; prognosis
• Online Access: https://www.frontiersin.org/articles/10.3389/fcell.2021.649755/full

Numerous studies have demonstrated that lncRNAs could compete with other RNAs to bind miRNAs, as competing endogenous RNAs (ceRNAs), to regulate each other. On the other hand, ceRNAs were found to be recurrently dysregulated in cancer status. However, limited studies considered the upstream epigenetic regulatory factors that disrupted the normal competing mechanism. In the present study, we constructed the lncRNA-associated dysregulated ceRNA networks across eight cancer types. lncRNAs in the individual dysregulated network and pan-cancer core dysregulated ceRNA subnetwork were found to play more important roles than mRNAs. Integrating lncRNA methylation profiles, we identified 49 epigenetically related (ER) lncRNAs involved in the dysregulated ceRNA networks, including 18 epigenetically activated (EA) lncRNAs, 18 epigenetically silenced (ES) lncRNAs, and 13 rewired ER lncRNAs across eight cancer types. Furthermore, we evaluated the epigenetic regulating patterns of these lncRNAs and screened nine pan-cancer ER lncRNAs (six EA and three ES lncRNAs). The nine lncRNAs were found to regulate the cancer hallmarks by competing with mRNAs. Moreover, we found that integrating the expression and methylation profiles of the nine lncRNAs could predict cancer incidence in eight cancer types robustly and the cancer outcome of several cancer types. These results provide an improved understanding of methylation regulation to ceRNA and offer novel potential molecular therapeutic targets for the diagnosis and prognosis across different cancer types.