MicroRNA Expression Analysis In Colorectal Carcinoma And Preliminary Functional Studies Of MiR-17

Background and objective:Colorectal carcinoma (CRC) is the third most common cancer in western countries and one of the most common cancers in China. Though its mortality rate has decreased slightly over the past three decades because of increased early diagnosis and advances in treatments, its morbility has increased in recent years in China. The carcinogenesis of CRC involves multiple alterations of oncogenes and tumor suppressor genes, such as inactivating APC and TP53 and oncogenically mutating RAS and BRAF. However, few of these genes are helpful for early diagnosis, and the molecular mechanisms underlying development and progression of CRC remain poorly understood. Therefore, further understanding its molecular mechanisms and identifying both early diagnostic markers and novel therapeutic targets are of great clinical value.MicroRNAs (miRNAs) are small non-coding RNAs that are 18 to 25 nucleotides long and that post-transcriptionally inhibit many genes by binding to the 3’ untranslated region (3’UTR) of target mRNAs. Recently, many miRNAs were found to be differentially expressed in many human cancers. These miRNAs could regulate cell differentiation, proliferation, apoptosis and angiogenesis by targeting the genes involved in the development and progression of human cancers, and could represent the alteration in biological characteristics. Moreover, the aberrantly expressed miRNAs could serve as molecular biomarker for diagnosis, treatment and prognostic prediction. Therefore, exploring the miRNA expression profiling and the role of differentially expressed miRNAs in colorectal carcinoma(CRC) would provide new insight into the molecular mechanisms of CRC.In our study, we profiled CRC tissue samples by miRNA microarray, hen screened out the miRNAs aberrantly expressed in CRC. Furthermore, the role and its mechanism of miR-17 in the colorectal carcinogenesis were examined to obtain more evidence whether miR-17 could serve as a new biomarker for diagnosis, treatment and prognostic prediction of CRC.Methods:1. miRNA expression profiling in colorectal carcinomaOligonucleotide microarray analysis from Affymetrix company was used to detect miRNA expression in 4 CRC tissues and adjacent normal tissues. The microarray contained 1320 probes in triplicate that corresponded to 988 human including 122 predicted miRNAs,627 mouse, and 350 rat mature miRNAs, which were from the miRNA Registry. Then SAM、PCA、CLUSTER and TREEVIEW were used to screen out the aberrantly expressed miRNAs for further validation by qRT-PCR in 21 CRC tissues and corresponding adjacent normal tissues.2. The effect of miR-17 on the biological behaviors of CRC cellsThe miR-17 inhibitor was transfected into Lovo cells to establish Lovo/miR-17 inhibitor cells. The RNA from the cells was isolated by the one-step Trizol method and then subjected to quantitative Real Time PCR (qRT-PCR) for detecting the miR-17 expression. MTT assay was engineered to detect the effect of miR-17 inhibition on cell proliferation, and Flow Cytometry was used to examine the cell cycle progression.3. Exploring the possible molecular mechanism underlying miR-17-mediated effect on promoting proliferation and cell cycle progression in CRC cells.Potential miRNA targets were predicted and analyzed with three publicly available algorithms:PicTar, TargetScan Human 5.1 and miRanda. Considering the effect of miR-17 on cell proliferation and cell cycle progression, the target genes predicted by at least two programs were taken into consideration. As a result, RND3 was selected for Dual-Luciferase report assay.The expression of miR-17 and RND3 was detected in 19 CRC tissues by qRT-PCR. The correlation between them was analyzed using Spearman’s Rank correlation. In addition, the RND3 was examined by qRT-PCR and Western blot in Lovo/NC cells and Lovo/miR-17 inhibitor cells.The expression of RND3 was examined in specimens of 41 CRC,42 adenoma and 49 nomal colorectal mucosae by immunohistochemical (S-P) method. Kruskal-Wallis was performed for IHC staining and Mann-Whitney U-test was used for multiple comparison tests.The level of RND3 was validated by qRT-PCR and Western blots in Lovo and HCT116 (wtp53) cells after they were transfected with either siRNA for RND3, miR-17 inhibitor, or both. After transfection, the cells grown in normal media for one week while the MTT assay was performed daily or collected for FACS analysis.ResultsThe main results and findings are as follows:1. miRNA expression profiles of colorectal carcinomaMicroarray analysis of miRNAs from four pairs of CRC and adjacent normal tissues identified 44 miRNAs differentially expressed in CRC. Of these,19 miRNAs had increased expression and 25 had reduced expression in the CRC tissues compared with the normal tissues. As shown in Unsupervised Hierarchical Clustering, these miRNAs could distinguish CRC from adjacent normal tissues. Further, PCA indicated 19 differentially expressed miRNAs that contributed to the differences between cancer and normal tissues that included five members of miR-17 family.2. Identifying preferential overexpression of four members of the miR-17 family in CRCThe expression of all four of these miRNAs, miR-17, miR-106a, miR-18a, and miR-18b, was further confirmed in 21 pairs of CRC and normal clinical tissue sample by qRT-PCR. Using nonparametric test, we comfirmed that all of these four miRNAs were significantly higher expressed in CRC than in normal clinical tissue sample (P= 0.011 for miR-17, P=0.002 for miR-18a, P=0.002 for miR-18b and P=0.013 for miR-106a).3. Inhibiting miR-17 reduced the proliferation of Lovo cells and induced Go/G1 arrestWe focused on the role of miR-17 in CRC by transfecting Lovo cells with a siRNA inhibitor of miR-17. The transfection efficiency was determined by qRT-PCR, and the results showed that miR-17 expression is significantly lower in cells transfected with miR-17 inhibitor than in NC. By MTT, we found that Cells transfected with the siRNA miR-17 inhibitor had lower levels of cell proliferation than the NC or mock control cells (F=964.521, P<0.001). In addition, the Lovo cells transfected with the miR-17 inhibitor had a higher percentage of cells in the G1-phase than those transfected with NC (P<0.01). These results indicated that miR-17 could enhance proliferation and promote cell cycle progression.4. Target gene prediction of miR-17 and validating the target of miR-17 by dual-luciferase reporter assayTargetScan Human 5.1, miRanda and Pic Tar were used for target gene prediction of miR-17. As a result, RND3 was predicted to be a potential target of miR-17 by computer-based sequence analysis by both TargetScan Human 5.1 and miRanda. We sought to validate the regulation of RND3 by miR-17 through a luciferase reporter assay that could verify the direct interaction between miR-17 and RND3-3’UTR, and found that overexpressing miR-17 reduced luciferase activity from the reporter vector containing the 3’UTR of RND3. Moreover, mutating the putative binding site on RND3 3’UTR abrogated this repression, supporting the direct interaction of miR-17 with RND3.5. MiR-17 expression negatively correlated with RND3 expressionThe mRNA expression levels of both miR-17 and RND3 were analyzed by qRT-PCR in 19 samples from patients with CRC. Our result showed the mRNA level of RND3 was negatively correlated to that of miR-17 (R=-0.537, P=0.018). On the other hand, cells transfected with miR-17 had increased RND3 expression of RND3 at the mRNA and protein levels compared with Mock control and NC transfected control cells (P<0.001). This result suggests that miR-17 can directly regulate the expression of endogenous RND3.6. The expression of RND3 in CRC, adenoma and adjacent normal tissues.We evaluated the expression of RND3 in 132 samples, including 41 CRC tissues, 42 adenoma tissues, and 49 adjacent normal tissues. RND3 was expressed heterogeneously among the normal colorectal mucosae, adenoma and colorectal carcinoma (χ2=15.343, P<0.01). Furthermore, the CRC tissues had lower RND3 expression than adjacent normal tissues and adenoma tissues (P<0.01). However, there was no significant difference between normal colorectal mucosae and adenoma.7. The effect of miR-17 on promoting cell proliferation and cell cycle progression is mediated by RND3The level of RND3 was validated by qRT-PCR and Western blots in Lovo and HCT116 (wtp53) cells after they were transfected for 48 hours with either siRNA for RND3, miR-17 inhibitor, or both. Our results confirmed that RND3 expression decreased significantly in the cells transfected with RND3 siRNA (P<0.001), and increased in cells transfected with miR-17 inhibitor (P<0.001). When cotransfecting the cells with RND3 siRNA and miR-17 inhibitor, RND3 expression had no significant change.The MTT assay found that cell proliferation was inhibited by inhibitor for miR-17 and enhanced by siRNA for RND3 (F=1374.362, P< 0.001 for Lovo cells; F= 1172.702, P< 0.001 for HCT116 cells). Moreover, the reduced proliferation due to inhibiting miR-17 was attenuated after cells were cotransfected with siRNA for RND3. This result indicated that RND3 reversed the effect of miR-17 in CRC cells that promotes proliferation.Lovo cells were transfected with miR-17 inhibitor for 72 h and then collected for FACS analysis. The results showed that inhibiting RND3 promoted progress of the cell cycle and inhibiting miR-17 induced arrest of the cell cycle in the Go/Gi phase compared with NC (F=403.640, P<0.01). The Lovo cells cotransfected with both the miR-17 inhibitor and siRNA for RND3 had no significant change, suggesting that inhibiting RND3 reversed the arrest of the cell cycle in Go/G1 that is caused by inhibiting miR-17.Conclusion1. MiR-17 acts as a gene that regulates proliferation and cell cycle progression.2. RND3 functions to inhibit cell proliferation and cell cycle progression in CRC, suggesting that RND3 may inhibit tumor growth in the development and progression of CRC.3. MiR-17 could promote cell proliferation and cell cycle progression in CRC by targeting RND3.