| BackgroundColorectal cancer (CRC) is one of the most common cancer worldwide,1.36million new CRC cases and about690thousand CRC related deaths were reported in2012. The CRC incidence rate of China has risen rapidly in late20th century due to the change of people lifestyles. Although the rising of CRC incidence rate has slowed down, CRC remained an important public health problem in China.Long non-coding RNA (lncRNA) is a group of RNA which could not coding proteins and exceed200nucleotides in length. Recent studies provide evidence that some lncRNAs are differently expressed in cancers. Profiling microarrays has been applied to explore the expression profile of lncRNA in cancer, and some cancer relating lncRNAs were discovered. However, following problems has remained to be resolved. First, the profile of lncRNA in colorectal cancer was hardly studied before. Also, it is difficult to explain the mechanism of the lncRNAs which reported as potentially functional in cancers by profiling microarray study. Besides, gene-gene interaction was not taken into consideration in most of former cancer related lncRNA selection strategies.In current study, we conducted a colorectal cancer lncRNA and mRNA expression profiling test, and selected most differently expressing lncRNA as well as crucial lncRNAs in lncRNA-mRNA co-expression network, respectively. And then we validated the expression level of selected colorectal cancer relating lncRNAs with Real time quantitative PCR.Materials and Methods96Tumor tissues and paired normal tissues were obtained from CRC patients who had undergone curative surgery at Department of General Surgery,1st people’s hospital of Hangzhou, from May,2013to December,2013, and6pairs of the96tissue pairs were selected for profiling microarray test. Top10of these differently expressed lncRNAs were selected for further validation. Besides, lncRNA-mRNA co-expression network was established with expression data from the microarray, co-expression modules were defined with clustering method. Gene ontology (GO) and pathway analysis were carried out to detect potential functional modules, and crucial lncRNAs of the modules were picked for further validation. These potential colorectal cancer related lncRNAs were further validated by Real-time quantitative RT-PCR in90colorectal cancer tissues and paired normal tissues except the6tissue pairs for profiling microarray test. The relationship between these lncRNAs and colorectal cancer, tumor size, tumor size, differentiation level, clinical stage were analyzed. ROC comparisons were also conducted to assess the value of these lncRNAs in colorectal cancer diagnosis.ResultsAmong the30586lncRNAs in the microarray,556lncRNAs were found up-regulated in tumor tissues, and1040lncRNAs were found to be down-regulated. Among the26109mRNAs in the microarray,889mRNAs were found up-regulated in tumor tissues, and977lncRNAs were found to be down-regulated. Top10differently expressed lncRNA selected for further validation were as follows:CCAT1, UCA1, RP5-881L22.5, NOS2P3, BC005081, AK055386, AC078941.1, RP4-800J21.3, RP11-628E19.3, RP11-384P7.7.The fitness index between the lncRNA-mRNA co-expression network and the theoretic scale-free network was0.813.9co-expression modules were detected by clustering method.4of these modules were found potential functional in carcinogenesis after GO enrichment analysis. Terms such as Cell cycle, cell migration, contractile fiber, cytoskeleton, chromosome were enriched with protein coding genes in these modules. KEGG pathway analysis showed that3carcinogenesis related pathways, Cell cycle pathway, Focal adhesion pathway, Oocyte meiosis related pathway, were enriched in2modules. The number of crucial molecules in modules were54,19,14,21, respectively. RP11-58A12.3, UBE2Q2P1, SENP3-EIF4A1, AC010226.4-2, CTC-454M9.1were finally picked as crucial lncRNA in these modules and further validated.Expression level of the15lncRNA selected from the expression profile were further examined in90tissue pairs using quantitative RT-PCR. CCAT1, UCA1, RP5-881L22.5were found up-regulated in tumor tissues compared with paired normal tissue, and lower expression level of AK055386, AC078941.1, RP4-800J21.3, RP11-628E19.3, RP11-384P7.7, RP11-58A12.3, UBE2Q2P1, SENP3-EIF4A1, AC010226.4-2and CTC-454M9.1in tumor tissues were observed (p<0.05). CCAT1and RP5-881L22.5were up-regulated in colon cancer tissues when compared with paired normal tissues, while no significant expression difference was observed in rectal cancer tissues. UCA1was up-regulated in rectal cancer tissues when compared with paired normal tissues, while no significant expression difference was observed in colon cancer tissues. Expression level of CCAT1was relatively higher in early-stage colorectal cancer when comparing with paired normal tissues, and expression level of RP5-881L22.5was higher in late-stage colorectal cancer when comparing with paired normal tissues. UCA1and RP5-881L22.5were found highly expressed in colorectal cancer with mid/high differentiation level, while expressing difference between colorectal cancer with low differentiation level and paired normal tissue was not significant. ROC comparison showed that the average area under ROC curve of the down-regulated lncRNAs were larger than those of up-regulated lncRNAs while diagnosing colorectal cancer. There was no significant difference between AUC for first component of two screening strategies.ConclusionThe current study come up with following conclusion:(1) Long non-coding RNA CCAT1, UCA1, RP5-881L22.5were significantly up-regulated in colorectal cancer, and AK055386, AC078941.1, RP4-800J21.3, RP11-628E19.3, RP11-384P7.7, RP11-58A12.3, UBE2Q2P1, SENP3-EIF4A1, AC010226.4-2and CTC-454M9.1were significantly down-regulated in colorectal cancer, they may be associated with colorectal cancer genesis.(2) CCAT1was specially up-regulated in colon cancer and early stage cancer, UCA1was specially up-regulated in rectal cancer and colorectal cancer with mid/high differentiation level, RP5-881L22.5was specially up-regulated in colon cancer and colorectal cancer with mid/high differentiation level.(3) The diagnostic value of up-regulated lncRNAs were lower than those of down-regulated lncRNAs when diagnosing colorectal cancer. There was no significant difference between diagnostic value for first component of two screening strategies. |
PDF Full Text Request