| Objective:(1)To construct miRNA expression profiles in tumor tissues and adjacent normal tissues (ANT) from patients with colorectal adenocarcinoma (CAC). (2)To construct mRNA expression profiles in tumor tissues and ANT from patients with CAC. (3)Validate and pick out the differently expressed miRNAs by performing bioinformatic analysis. (4)Validate the differently expressed miRNAs by conducting quantitive Real-Time Polymerase Chain Reaction (qRT-PCR), and evaluate the ability of miR-133b to distiguish CAC tumor and normal tissues. (5)Explore the potential function of miR-133b by performing bioinformatic analysis.Methods:(1)14 CAC patients provided tumor tissue and ANT, which were pooled into four samples that went through miRNA sequencing. Edger was used to analyze the expression differences, the definition of significantly differently expression miRNA were:Fold Change (FC)≥2 or FC<0.5 (|log2FC|≥1), P<0.05 and False Discovery Rate (FDR) P<0.01. (2)The above mentioned samples also went through mRNA sequencing. The definition of significantly differently expression mRNA were:|log2FC|≥1, P<0.05 and FDR-P<0.01. (3)We searched the Gene Expression Omnibus (GEO) database to download datasets related with CAC miRNA expression. Student’s t-test and followed FDR multitest were used to compare the expression levels of miRNA between tumor tissue and ANT. Next, the successfully validated miRNAs were picked out. (4) The expression levels of hsa-miR-133b, hsa-miR-133a, hsa-miR-139-5p and hsa-miR-140-3p, normalizing by reference gene RNAU44, were compared by Paired Stuent’s t-test between tumor tissue and ANT in 111 CAC patients. Receiver Operating Characteristic Curve (ROC) was then used to assess the ablility of miRNA to distinguish tumor tissues and ANT. (5) Potential targets of miRNA were combinely predicted by miRBase, Targetscan and miRDB. We downloaded the level 3 data of mRNA and miRNA from the Cancer Genome Atlas (TCGA) database. The correlations between miR-133b and its potential targets were tested by Pearson Correlation test. The miRNA-mRNA interaction exit if the Pearson coefficient less than-0.1, P less than 0.05 and FDR-P less than 0.10.Results:(1)The expression levels of 53 and 47 miRNAs were decreased and increased consistanlty in comparasions of tumor tissues and ANT, respectively. (2)The expression levels of 1375 and 1425 mRNAs were increased and decreased consistanlty in comparasions of tumor and normal tissues, respectively. (3)GSE35982 and GSE38389 were recruited. Ten miRNAs were sucessfuly validated; they were miR-1246, miR-129-5p, miR-133a, miR-133b, miR-139-5p, miR-140-3p, miR-188-5p, miR-301b, miR-552 and miR-592. (4)According to Paired Student’s T-testt, miR-133b was significalty and differently expressed between tumor tissue and ANT (P=0.0003); while miR-139-5p showed marginally significance (P=0.061). Unlucky, miR-133a and miR-140-3p did not reach the significant threshold (P=0.121 and 0.539, respcetively). ROC was then used to asses miR-133b to distinguish tumor tissues and ANT, yielding area under the curve of 76.68%. (5) 191 of predicted target genes were found in the TCGA levels 3 mRNA data. SFXN2 (Pearson correlation coefficient=-0.139,P=0.007, FDR-P=0.059) and ACAT2 (Pearson correlation coefficient=-0.137, P=0.008, FDR-P=0.061) were found to be negativley related with miR-133b.Conclusions:By performing RNA sequencing, bioinformatic analysis, and experimental validation, we found that hsa-miR-113b was expressed significantly different between CAC tumor tissues and ANT. It may influence the development of CAC by targeting SFXN2 and ACAT2. The result warrants further study... |
PDF Full Text Request