Comprehensive analysis of long non-coding RNA and mRNA expression profile in rectal cancer

Abstract Background: Rectal cancer (RC) is a malignant tumor that seriously threatens human health. Long non-coding RNAs (lncRNAs) play a vital role in tumor regulation. Nevertheless, their exact expression features and functions remain obscure, and therefore was the aim of the current study. Methods: We utilized the Affymetrix human GeneChip to screen differentially expressed profiles of lncRNAs and mRNAs from the cancer tissues and matched paracancer tissues of 6 RC patients. Gene Ontology (GO) and pathway enrichment analyses identified crucial functions and pathways of the aberrantly expressed mRNAs. We used quantitative real-time polymerase chain reaction to verify the significant expression differences of 11 candidate lncRNAs between the cancer and paracancer tissues. LncRNA-mRNA coexpression networks were built by calculating the Pearson correlation value to identify significant correlation pairs. Online bioinformatics tools GEPIA2, ONCOMINE, and PROGgeneV2 were used to mine the expression and prognosis of three crucial mRNAs and six verified lncRNAs. Competing endogenous RNA networks were constructed by predicting microRNA response elements and calculating free energy. Results: We found 1658 differentially expressed lncRNAs (778 up-regulated and 880 down-regulated) and 1783 aberrantly expressed mRNAs (909 up-regulated and 874 down-regulated). GO and pathway enrichment analyses revealed the vital functions of the differentially expressed mRNAs, including cell proliferation, cell migration, angiogenesis, and cellular response to zinc ion. The canonical signaling pathways mainly included the interleukin-17, cell cycle, Wnt, and mineral absorption signaling pathways. Six lncRNAs including AC017002.2 ( P = 0.039), cancer susceptibility 19 (CASC19) ( P = 0.021), LINC00152 ( P = 0.013), NONHSAT058834 ( P = 0.007), NONHSAT007692 ( P = 0.045), and ENST00000415991.1 ( P = 0.045) showed significant differences in expression levels between the cancer tissue and paracancer tissue groups. AC017002.2, NONHSAT058834, NONHSAT007692, and ENST00000415991.1 have not yet been reported in RC. The crucial mRNAs myelocytomatosis viral oncogene (MYC), transforming growth factor beta induced (TGFBI), and solute carrier family 7 member 5 (SLC7A5) were selected. AC017002.2 and LINC00152 were positively correlated with MYC, TGFBI, and cytochrome P450 family 2 sub-family B member 6 (All r > 0.900, P < 0.050). NONHSAT058834 was positively associated with MYC ( r = 0.930, P < 0.001), and CASC19 was positively correlated with SLC7A5 ( r = 0.922, P < 0.001). Conclusion: This study offers convincing evidence of differentially expressed lncRNAs and mRNAs as potential biomarkers in RC.