Microarray expression profile analysis of long non-coding RNAs in pancreatic ductal adenocarcinoma.

Long non-coding RNA (lncRNA) is a variety of the human transcriptome that does not code for proteins and plays an important role in the development and progression of multiple solid malignant tumors. However, the roles of lncRNAs in the development of pancreatic ductal adenocarcinoma (PDAC) remain unknown. In this study, we investigated the expression patterns of lncRNAs in three PDAC tumor samples (T) relative to those of matched adjacent non-tumor tissues (N) via a microarray with 30,586 lncRNA probes and 26,109 mRNA probes. The lncRNA microarray revealed 27,279 lncRNAs in PDAC samples, of which 2,331 were significantly upregulated (P<0.05; T/N>2.0) and 1,641 were downregulated (P<0.05; N/T>2.0) compared with matched adjacent non-tumor samples. In addition, 19,995 mRNAs were detected, of which 1,676 were significantly upregulated (P<0.05; T/N>2.0) and 1,981 were downregulated (P<0.05; N/T>2.0). Pathway analysis indicated that 41 pathways corresponded to upregulated transcripts and 25 pathways corresponded to downregulated transcripts (P-value cut-off is 0.05). Gene ontology (GO) analysis showed that the highest enriched GOs targeted by upregulated and downregulated transcripts were tissue homeostasis. The validation results from quantitative reverse transcription polymerase chain reaction (qRT-PCR) analysis and microarray analysis were consistent. Furthermore, the expression level of long intergenic non-coding RNA HOTAIRM1 was upregulated in 12 PDAC tissues samples compared with matched adjacent non-tumor samples by qRT-PCR. The results showed that the lncRNA and mRNA expression profiles differed significantly between the PDAC tissues and their adjacent non-tumor tissues, and the revelation of an association between HOTAIRM1 expression and PDAC is especially noteworthy. These findings may provide new potential molecular markers for diagnosis and treatment of PDAC.