OBJECTIVETo detect and analyze the characteristic miRNAs profile of anal fistula and explore their possible target genes and potential clinical significance.

METHODSThe anal mucosa close to the hemorrhoids were collected from three patients undergoing fistulectomy and hemorrhoidectomy (fistula group) as well as three patients receiving only hemorroidectomy(hemorrhoids group), matching with fistula group in age, gender and body weight. miRNA microarray was used to compare the expression of 1 285 human miRNAs of the anal mucosa between two groups. Cluster analysis was adopted to analyze the accumulation of the differentially expressed miRNAs(P<0.05, fold≥2.0 or ≤0.5) and their target genes were predicted with 10 softwares such as DIANAmT, miRanda, miRDB, miRWalk etc. Comprehensive scoring was performed to identify genes with highest predictive score. Gene ontology (GO) concentration technique was used to analyze the target gene-associated biological process. Immunohistochemistry was used to examine protein expression of genes with the highest score.

RESULTSAmong 1285 miRNAs in fistula group, 13 miRNAs were differentially expressed with those in hemorrhoid group, including 2 of up-regulation and 11 of down-regulation. Paired t test showed that in fistula group, miRNA-3609 up-regulation was 5.98 folds(P=0.0231) and miR-181a-2-3p down-regulation was 0.13 folds(P=0.0067) compared to those in hemorrhoid group, which had the greatest differential expression. Cluster analysis suggested that up-regulated miR-3609 and miR-6086 had similar change trend in both groups. Among 11 down-regulated miRNAs, miR-125bp-1-3p and miR-548q had similar expression and other 9 miRNAs had similar expression as well, including miR-1185-1-3p, miR-532-3p, miR-1233-5p, miR-769-5p, miR-149-5p, miR-99b-3p, miR-141-3p, miR-138-5p, and miR-181a-2-3p. Target gene prediction analysis of above 13 genes showed that 7 miRNAs(53.8%) were eligible to predict their potential target genes, yielding totally 104 possible target genes. The rest of 6 miRNAs(46.2%) failed to predict any target gene. The highest score in prediction of target gene was chitinase 1(ChIT1) and its corresponding differential miRNA was miR-769-5p(r=-0.94286, P=0.0167). Gene ontology analysis showed that the most associated biological process related with these 104 target genes was keratinization, immune response and signal transduction. Immunohistochemistry revealed ChiT1 expression of anal mucosa in fistula group was significantly higher compared to hemorrhoid group(P<0.01).

CONCLUSIONSThere is a characteristic miRNAs profile in anal fistula patients, which may play a role in the occurrence and development of anal fistula.