Identification and development of long non-coding RNA associated regulatory network in pancreatic adenocarcinoma

Abstract

Background and Aims: Pancreatic adenocarcinoma (PAAD) is the most lethal cancer type around the world. With the in-depth exploration of the function of long non-coding RNAs (lncRNAs), the competing endogenous RNA (ceRNA) mechanism has shown its potential to partially reveal the pathogenesis of PAAD. This study aimed to construct a lncRNA-associated ceRNA network and explore ceRNA regulatory axes with experimental and prognostic value in PAAD. Methods: First, we applied differential expression analysis in the TCGA_PAAD data-set. Then, interaction analysis and survival analysis in multiple RNA interaction databases were conducted to construct a ceRNA network. Finally, a potential regulatory axis was validated using clinical samples and cell lines by quantitative realtime PCR (qRT-PCR). Results: A ceRNA network comprising 13 lncRNAs, 96 miRNAs, and 30 mRNAs was successfully constructed. Survival analysis further narrowed this network to five lncRNAs, three miRNAs, and seven mRNAs, which were significantly associated with patients’ overall survival. A potential regulatory axis CASC8-miR-129-5p-TOB1 was further experimentally validated. The expression of these genes was associated with clinicopathological factors and their expression trend was consistent with ceRNA mechanism. Specifically, knockdown of lncRNA-CASC8 led to the overexpression of miR-129-5p and down-regulation of TOB1, while overexpression of CASC8 showed opposite effects. Conclusion: This novel ceRNA regulatory network could provide new insight into the pathogenesis of PAAD. The new regulatory axis CASC8-miR-129-5p-TOB1 might serve as a potential therapeutic target for patients.