Screening Of DNA Methylation Diagnostic Biomarkers Form Multiple Candidate Genes In Gastrointestinal Cancer And Its Function Analyses

Objective: The purpose of this study is to screen the differential DNA methylated genes from candidate genes by detecting DNA methylation levels of multiple candidate genes in Gastrointestinal Cancer patients;and to reveal function of DNA methylation in Gastrointestinal Cancer patients through clinical data and Bioinformatic analysis;and to set up an effective diagnostic model in established Han Gastrointestinal Cancer patients.Methods: The quantitative methylation-specific polymerase chain reaction(q MSP)was used to detect the methylation levels of 26 candidate genes in the tissues of Gastrointestinal Cancer patients.Wilcoxon Signed Ranks Test was used to test the methylation differences in tumor tissues and paired non-tumor tissues.The methylation status and its m RNA levels date of candidate genes obtained from the c Bio Portal website were used to evaluate the relationship between DNA methylation and m RNA levels.Receiver operating characteristic(ROC)analysis showed the diagnostic value of candidate genes’ methylation levels and multi-genes diagnosis model.Dual-luciferase reporter assay was conducted to evaluate the activity of genes’ promoter region.Results: 1.The promoter methylation levels of CCL2、CCND1、EED、EPAS1、FANCF、HOXA9、IL10、KDM1A、LEPR、MTHFR、PER3、PRMT6、SMYD3 in cancer tissues of colorectal cancer are significantly lower than that in paired non-tumor tissues;The methylation of TNFRSF10 C in colorectal cancer tumor tissues is significantly higher than that of paired non-tumor tissues.2.TCGA colorectal cancer database analysis finds that CCND1,EED,EPAS1,FANCF,HOXA9,IL10,KDM1 A,MTHFR,PER3,PRMT6,SMYD3 and TNFRSF10 C methylation levels are negatively correlated with their m RNA expression.3.ROC analysis finds that IL10 methylation levels have a good diagnostic value in colorectal cancer patients,with an AUC of 0.72,a sensitivity of 77.7% and a specificity of60.7%.The multiple genes prediction model(CCL2,CCND1,EPAS1,IL10,PER3 and TNFRSF10C)shows better diagnostic value,with an AUC of 0.85,a sensitivity of 95.0% and a specificity of 62.3%.4.The promoter methylation levels of ALOX12,EPAS1,FANCF,HIF1 A,IL10,KDM1 A,KLF5,LEPR,MTAP,PTPN11,PRMT6,RBL2 and TNFAIP3 in cancer tissues of gastric cancer are significantly lower than that in paired non-tumor tissues;The methylation of CXCL12,EBF3,GPX3,HHIP,HOXA9 and TLR2 in gastric cancer tumor tissues are significantly higher than that of paired non-tumor tissues.5.TCGA gastric cancer database analysis finds that AXOL12,CXCL12,EPAS1,FANCF,GPX3,HIF1 A,HHIP,HOXA9,IL10,KDM1 A,KLF5,PTPN11,LEPR,PRMT6,RBL2 and TNFAIP3 methylation levels are negatively correlated with their m RNA expression.6.ROC analysis finds that IL10 methylation levels have a good diagnostic value in gastric cancer patients,with an AUC of 0.72,a sensitivity of 76.3% and a specificity of 59.2%.The multiple genes prediction model(CXCL12,EPAS1,IL10,KDM1 A,LEPR and MTAP)shows better diagnostic value,with an AUC of 0.87,a sensitivity of 88.6% and a specificity of 71.1%.7.The results of dual-luciferase reporter assay show that long fragments of the promoter regions of CCL2,EED,FANCF,KDM1 A,LEPR,MTHFR and TNFRSF10 C can initiate the expression of the reporter gene,indicating that CCL2,EED,FANCF,KDM1 A,LEPR,MTHFR and TNFRSF10 C promoter fragment has the function of regulating gene expression.Conclusions: 1.Abnornal methylation of multiple genes occurs in both colorectal cancer and gastric cancer patients with the Han nationality.Bioinformatics analysis finds that the methylation of the candidate genes was negatively correlated with the expression data.Combined with the results of dual-luciferase reporter assay,the candidate gene may affect the expression level through methylation modification,and then be related to the mechanism of gastrointestinal cancer development.2.In this study,abnormal methylation of IL10 promoter was firstly found in both colorectal cancer and gastric cancer,and its abnormal methylation can be used as a good diagnostic biomarker.At the same time,this study established the good diagnostic model for Gastrointestinal Cancer patients.The model shows that abnormal methylation data of gastrointestinal cancer may be a good source for establishing a cancer diagnostic model,and provide clues for the further study of multi-genes diagnosis for cancer in the future.