Comprehensive characterization of lncRNA-mRNA related ceRNA network across 12 major cancers

• Published in: Oncotarget Vol. 7; no. 39; pp. 64148 - 64167
• Main Authors: Zhang, Yunpeng, Xu, Yanjun, Feng, Li, Li, Feng, Sun, Zeguo, Wu, Tan, Shi, Xinrui, Li, Jing, Li, Xia
• Format: Journal Article
• Language: English
• Published: United States Impact Journals LLC 27.09.2016
• Subjects: Binding Sites; Binding, Competitive; Biomarkers, Tumor - genetics; Biomarkers, Tumor - metabolism; Case-Control Studies; Computational Biology; Databases, Genetic; Gene Expression Profiling; Gene Expression Regulation, Neoplastic; Gene Regulatory Networks; Humans; MicroRNAs - genetics; MicroRNAs - metabolism; Neoplasms - genetics; Neoplasms - metabolism; Neoplasms - pathology; Research Paper; Response Elements; RNA, Long Noncoding - genetics; RNA, Long Noncoding - metabolism; Systems Integration; Transcriptome; lncRNA; pan-cancer; ceRNA network
• ISSN: 1949-2553; 1949-2553
• DOI: 10.18632/oncotarget.11637

Recent studies indicate that long noncoding RNAs (lncRNAs) can act as competing endogenous RNAs (ceRNAs) to indirectly regulate mRNAs through shared microRNAs, which represents a novel layer of RNA crosstalk and plays critical roles in the development of tumor. However, the global regulation landscape and characterization of these lncRNA related ceRNA crosstalk in cancers is still largely unknown. Here, we systematically characterized the lncRNA related ceRNA interactions across 12 major cancers and the normal physiological states by integrating multidimensional molecule profiles of more than 5000 samples. Our study suggest the large difference of ceRNA regulation between normal and tumor states and the higher similarity across similar tissue origin of tumors. The ceRNA related molecules have more conserved features in tumor networks and they play critical roles in both the normal and tumorigenesis processes. Besides, lncRNAs in the pan-cancer ceRNA network may be potential biomarkers of tumor. By exploring hub lncRNAs, we found that these conserved key lncRNAs dominate variable tumor hallmark processes across pan-cancers. Network dynamic analysis highlights the critical roles of ceRNA regulation in tumorigenesis. By analyzing conserved ceRNA interactions, we found that miRNA mediate ceRNA regulation showed different patterns across pan-cancer; while analyzing the cancer specific ceRNA interactions reveal that lncRNAs synergistically regulated tumor driver genes of cancer hallmarks. Finally, we found that ceRNA modules have the potential to predict patient survival. Overall, our study systematically dissected the lncRNA related ceRNA networks in pan-cancer that shed new light on understanding the molecular mechanism of tumorigenesis.