Research On Effect And Mechanisms Of MIR-183 And MIR-218 In Esophageal Squamous Cell Cancer Carcinogenesis

Background and Objectives:Esophageal cancer is one of the most common digestive tract malignancies with most new cases and deaths occurring in China. Of the two histological types of esophageal cancer, the most common is squamous cell carcinoma (ESCC). There is great need to elucidate the molecular mechanisms of esophageal carcinogenesis, and to find biomarkers for early detection and diagnosis. Therefore, in this study, we aim to (1) identify ESCC-related miRNAs using microRNA microarray and verify the potential miRNAs in a larger population samples; (2) explore the biological functions, target genes and downstream signaling pathways of miR-183 and miR-218 in ESCC cell lines; (3) investigate the regulatory mechanism of miR-218; (4) provide clues to clarify the pathogenesis of ESCC and to find biomarkers for early screening.Methods:1.138 patients diagnosed as primary ESCC were recruited with informed consent. Three pairs of ESCC tumor tissues and paired non-tumor tissues were prepared for miRNA microarray, and miRNA candidates were validated by real-time RT-PCR in larger population samples. The relationship between differentially expressed miRNAs and the risk of ESCC were assessed for profiling ESCC-related miRNAs. Further, the correlation between differentially expressed miRNAs and smoking, alcohol consumption, water type was analyzed according to the environmental epidemiological survey.2. The exogenous miR-183 mimic and inhibitor were transfected into ESCC cell line EC9706. Transfection efficiency was determined by quantitative RT-PCR. Gain and loss of function analysis were performed in EC9706, including cell growth and proliferation by EdU assay, cell apoptosis by Annexin-V FITC&PI assay, and cell cycle distribution by flow cytometry, and cell invasive ability by Transwell assay. Target genes of miR-183 were predicted by bioinformatics and mRNA microarray. Cumulative distribution function method (CDF) was applied to analyze the expression difference on target genes and non-target gene predicted by bioinformatics in the mRNA microarray. Quantitative RT-PCR and western blot were performed to detect target association of miR-183 on PDCD4 at mRNA and protein levels, respectively. Association of miR-183 and PDCD4 mRNA in population samples were assessed by quantitative RT-PCR. Wild type and mutant type of PDCD4 3’UTR were constructed for dual luciferase reporter assay to verify the bonafide target gene. Expression vector pcDNA3.1-PDCD4 containing and missing PDCD4 3’UTR were constructed to analyze effect of PDCD4 and miR-183 on cell apoptosis through restoring PDCD4 expression.3. CpG islands in miR-218 promoter were predicted by online software. Methylation status of miR-218 CpG islands in ESCC cell lines were assessed by bisulfite sequencing PCR (BSP) and nested methylation specific PCR (n-MSP). Demethylation treatment by 5-Aza-CdR was appilied to analyze recovery of miR-218. Further, methylation status of miR-218 in tissue samples was determined by n-MSP. Associaton of miR-218 expression and methylation status in tissues were assessed as well. The consistency of miR-218 and the host genes SLIT2 SLIT3 were detected by quantitative RT-PCR. miR-218 mimic was transfected into ESCC cell line EC 109. Gain of function analysis was performed to explore the tumor suppressive role of miR-218 in ESCC. Potential targets were predicted by bioinformatics, and verified by quantitative RT-PCR and western blot. Dual-luciferase reporter assay by constructing wild type and mutant type of ROBO1 3’UTR vectors were applied to verify the target association. Expression vector pcDNA3.1-ROBO1 containing and missing ROBO1 3’UTR were constructed to analyze effect of ROBO1 and miR-218 on cell proliferation through restoring ROBO1 expression.4. Plasma samples of ESCC patients and matched healthy control were collected to detect expression of miR-183 and miR-218 by quantitative RT-PCR. Logistic regression analysis was applied to analyze the association between aberrant expression of miR-183、miR-218 and the risk of ESCC. Diagnostic value of plasma miR-183 and miR-218 were determined by receiver operating characteristic curves (ROC) curves.Results:1. Identification of Differentially Expressed miRNAs in ESCC TissuesThe Agilent miRNA microarray analysis identified a total of 15 miRNAs that could distinguish malignant esophageal cancer lesions from adjacent non-tumor tissues, of which 7 miRNAs were up-regulated and 8 miRNAs were down-regulated. To confirm the microarray results, quantitative RT-PCR analysis in larger population samples was performed. Result showed that 7 miRNAs showed the consistency with the microarray (miR-139-5p, miR-218, miR-338-3p, miR-21-3p, miR-574-5p, miR-183, miR-601), which confirmed the reliability of microarray data. miR-183 was up-regulated in ESCC tumor tissues, while miR-218 was down-regulated in tumor tissues. In addition, down-regulation of miR-139-5p^ miR-338-3p and up-regulation of miR-21-3p、miR-574-5p、miR-183、miR-601 significantly increased the risk of ESCC by univariate regression analysis. Further multiple regression analysis revealed the aberrant expression of miR-338-3p, miR-139-5p, miR-574-5p and miR-601 increased the risk of ESCC. Moreover, we found miR-21-3p was significantly increased in heavy drinking patients.2. The Regulatory Role of miR-183 in ESCC Carcinogenesis2.1 Biological Function of miR-183 in ESCC Cell Line EC9706After treated with miR-183 mimic and inhibitor for 48h, expression level of miR-183 was significantly increased (p< 0.001) and decreased (p< 0.001), respectively. Results of cell apoptosis by Annexin V-FITC&PI assay showed that miR-183 inhibited early apoptosis of EC9706 cells (3.31±1.20 vs.7.43±1.01, p< 0.001), as well as late apoptosis (5.83±0.29 vs.8.77±1.59, p< 0.01). The inhibition effect of miR-183 on apoptosis was rescued when the expression level of miR-183 was reduced via miR-183 inhibitor [early apoptosis:(12.39±1.89 vs.6.56±1.87,p< 0.001); late apoptosis:(9.53±1.58 vs.7.36±2.42,p< 0.05)]. Results of EdU assay showed that proliferative cell rate in miR-183 mimic treated group was more than negative control (p< 0.05). Results of cell cycle analysis by flow cytometry showed that the percentage of cells at G1 phase was significantly decreased in miR-183 mimic transfected cells compared with negative control (54.87±1.135 vs.56.89±0.82, p< 0.05).2.2 Target Gene of miR-183 and Regulatory Mechanism in ESCCTwo pairs of EC9706 samples treated with miR-183 mimic or negative control were prepared for mRNA expression profiling analysis to find potential target genes of miR-183. Microarray profiling revealed 100 up-regulated genes and 83 down-regulated genes, of which HLA-DRB5、ITGB1、MICB、PSEN2 and PDCD4 were the top 5 down-regulated genes. The identified altered genes were annotated in the GO analysis for biological function classification and in the pathway analysis for elucidation of whole chains of events in miR-183 over-expressed cells. It turned out that the most significant enriched GO term in molecular function is the cyclin dependent protein kinase regulator activity, suggesting a critical role in cell cycle progression. Most altered genes were involved in tumor-associated pathways by KEGG pathway analysis. Cumulative distribution function analysis showed that expression of target gene was significantly lower than that of non-target (p< 0.001). Combined results of the bioinformatics including TargetScan and miRDB with results of mRNA microarray profiling, PDCD4 was selected as the candidate target gene. Results of RT-PCR showed that miR-183 rarely regulated PDCD4 at mRNA levels. Western blot analysis confirmed that miR-183 significantly reduced the protein level of PDCD4. Further, wild type and mutant type of PDCD4 3’UTR were constructed, which confirmed that luciferase activity in cells co-transfected with miR-183 mimic and wild type of PDCD4 3’UTR was 47% of that in cells co-transfected with negative control and mutant type of PDCD4 3’UTR (p< 0.01). In addition, a significant negative correlation between miR-183 and PDCD4 mRNA was found in ESCC tumor tissues and paired non-tumor tissues by spearman correlation analysis(r=-0.189, p< 0.05). Results of PDCD4 restore assay showed that no significant difference in cell apoptosis when cells were transfected with PDCD4 expression vectors containing or missing 3’UTR. However, cell apoptosis rates were significantly decreased when cells co-transfected with miR-183 mimic and PDCD4 expression vector containing 3’UTR (p< 0.01).3. The Regulatory Role of miR-218 in ESCC Carcinogenesis3.1 Methylation Status of miR-218 Gene in ESCCCpG islands in the promoter regions of miR-218 (miR-218-1 and miR-218-2) shared with host genes are detected by CpG Island Searcher Software (Available online:http://www. uscnorris.com/cpgislands2/cpg.aspx). Dense CpG islands are present at the position (-760 to-212) upstream to the transcription start site (TSS) in miR-218-1, and at the position (-407 to +117) to the TSS in miR-218-2. Results of bisulfite sequencing (BSP) and nested methylation specific PCR (n-MSP) showed that both of miR-218-1 and miR-218-2 were CpG-methylated in ESCC cell lines. The expression level of miR-218 in ESCC cell lines EC 109 and EC9706 were recovered accordingly by 2.53-and 2.40-fold respectively after 5-Aza-CdR treatment, while there is no difference in a human normal esophageal epithelial cell Het-IA. Results of MSP showed that miR-218-1 was found fully CpG-methylated in both EC9706 and EC 109, while unmethylated in Het-IA. CpG methylation of miR-218-1 frequently occurred in 34 cancer tissues; while only 14 paired non-tumor tissues were methylated (p< 0.05). However, there is no difference in miR-218-2 methylation status between tumor tissues and paired non-tumor tissues. Meanwhile, the tissue samples were grouped according to methylation status. A significant downregulation of miR-218 in miR-218-1 methylation group (including full methylation and semi-methylation) than that in unmethylation group was detected (p< 0.01). However, there is no such correlation in miR-218-2. Moreover, expression of SLIT2 in tumor tissues was 38.5% of that in paired non-tumor tissues (p< 0.01). While, no significant difference was found in tumor tissues and paired non-tumor tissues for SLIT3. In addition, expression miR-218 was correlated with SLIT2 and SLIT3 (p< 0.001), which suggested miR-218 may co-transcribe with its host genes.3.2 Biological Function of miR-218 in ESCC Cell Line EC109Gain of function analysis was performed via miR-218 mimic for 48h. Transfection efficiency was detected by qRT-PCR in cells treated with miR-218 mimic and negative control (p< 0.001). In cytological experiments, cells treated with miR-218 mimic displayed decreased proliferative ability measured by EdU assay (53.45%±1.20% vs.73.61%±3.21%,p < 0.001). Cell cycle assay was performed to investigate the mechanisms of miR-218 on cell growth. It turned out that cells treated with miR-218 mimic was found significantly arrested at the G1 phase (p< 0.01) and decreased at G2 phase (p< 0.01). However, no statistical difference was found in cell apoptosis.3.3 Target Gene of miR-218 and Regulatory Mechanism in ESCCPotential miR-218 targets were predicted using bioinformatics including miRDB, TargetScan and Pictar. ROBO1 was identified by all these prediction programs. Results of RT-PCR showed that expression of ROBO1 mRNA in cells treated with miR-218 mimic were decreased (fold change= 3.41) compared with negative control. Moreover, expression level of ROBO1 protein in miR-218 treated cells was accordingly decreased compared with negative control using western blot. Results of dual-luciferase reporter assay showed that luciferase activity of cells co-transfected with 3’UTR wild-type and miR-218 mimic was 55.7%of that in cells co-transfected with 3’UTR mutant-type and negative control, indicating that miR-218 directly regulates ROBO1 by binding to 3’UTR of ROBO1. Furthermore, it turned out that miR-218 expression was negatively correlated with ROBO1 mRNA expression in tissue samples (r= -0.258,p< 0.05). ROBO1 restore assay showed that cell proliferative ability was significantly decreased when cells co-transfected with miR-218 mimic and ROBO1 expression vector containing 3’UTR (p< 0.01). 4 Diagnostic value of Plasma miR-183 and miR-218 in ESCC ScreeningResults of RT-PCR showed that expression of plasma miR-183 was up-regulated in ESCC patients than healthy subjects, while plasma miR-218 was significantly down-regulated in ESCC patients. Meanwhile, reduced plasma miR-218 was correlated with the increasing risk of ESCC (OR=0.787, CI=0.685-0.903,p< 0.001). The diagnostic accuracy of miR-218 was measured by ROC, and their corresponding AUCs were 0.651.Conclusions:1. In this study, a total of seven differentially expressed miRNAs were identified between ESCC tumor tissues and paired non-tumor tissues, of which three were down-regulated and four were up-regulated. Expression of miR-218 was decreased in tumor tissues, while miR-183 was increased in tumor tissues.2. miR-183 enhanced cell proliferation, inhibited apoptosis, and accelerated G1/S transition in ESCC cell line EC9706 by directly targeting PDCD4. The inhibitory effect on apoptosis of miR-183 was exerted by binding to sequences in the 3’UTR of PDCD4.3. miR-218, especially miR-218-1, was aberrantly CpG hypermethylated in ESCC, which is responsible for miR-218 silencing. miR-218 suppressed cell proliferation and arrested cells at G1 phase of EC 109 by targeting ROBO1. It is confirmed that the roie of miR-218 on cell proliferation was exerted by binding to sequences in the 3’UTR of ROBO1.4. The trend of plasma miR-218 and miR-183 expression level was similar to those in tissue samples, which may serve as potential biomarkers of ESCC. miR-218 could distinguish ESCC patients from healthy subjects, which is an effective supplement to existing methods for early diagnosis of ESCC and provides new clues for establishment of prevention and control strategy.