Bioinformatics Analysis Of Gene Expression Profile And Function Of Hub Genes In Liver Cancer

Part I Application of Weighted Gene Coexpression Network(WGCNA)in Gene Expression Profile of Liver CancerObjective We performed WGCNA and other bioinformatics methods to explore the relationship between gene expression profile and clinical features.Method Liver cancer cohort data and clinical features were from NCBI-GEO database.GSE1452 was as the training set and GSE6764,GSE76427,GSE54236were as the testing sets.Raw microarray data of GSE14520 were calculated by robust multiarray average(RMA)background correction and log2 transformed and normalized by quantile normalization.The“limma”(linear models for microarray data)R package was used to screen the DEGs between liver cancer tumor tissues and non-tumor tissues and the“WGCNA”package in R was used to construct the coexpression network for the DEGs.Then,the highest correlation in the Module-feature.Based on the gene significance and module menmbership,the function“networkScreening”in the“WGCNA”package and PPI network were applied to identify the hub genes.The hub genes were used to construct the proportional hazards model,then the prognosis of model was predict.The clinical stage of hub genes were validated by the testing set of GSE6764 and the prognosis of model was validated by the testing sets of GSE76427 and GSE54236.Results A total of 3670 DEGs form the training set of GSE1452 were selected for constructing the WGCNA.We identified 12 module and red module was the most correlation with the PRMS(R~2=-0.74),also this module in tumor size(R~2=-0.21),multinodular(R~2=-0.16)and TNM.stage(R~2=-0.31)has largest the correlation coefficient between the modules.After screening,we we obtained six hub genes:ABAT,AGXT,ALDH6A1,CYP4A11,DAO,EHHADH.In the testing set of GSE6764,we found that the higher the tumor stage,the lower the expression of these6 genes(P<0.05);We constructed 6-gene signature of liver cancer by COX model,the AUC of ROC of model is 0.704 and the C-index ofmodel is 0.694.The prognosis of model was anlysised,the survival rate of high riskgroup was significantly higher than that of low risk group(Log-rank P<0.05);The testing sets of GSE76427 and GSE54236 were validated the prognosis of 6-gene signature model.Conclusions We performed WGCNA and other bioinformatics methods to explore the relationship between gene expression profile and clinical features.We identified 6 hub genes(ABAT?AGXT?ALDH6A1?CYP4A11?DAO?EHHADH)from 1significance module.6-gene signature of liver cancer by COX model was constructed and might be provide a potential biomarker for the prognosis of liver cancer.Part II Assocation between The Expression of ABAT Gene and Prognosis in Liver CancerObjective To investigate the association of the expression of the ABAT gene with clinical features and prognosis in samples with liver cancer.Method Liver cancer cohort data and clinical features of GSE14520 were from NCBI-GEO database.Totally 214 samples of liver cancer as well as their clinical features enrolled in the study.Liver cancer cohort data and clinical features from TCGA database was used to validate the prognosis.First,the online tool of Oncomine and GEPIA were used to measure the expression of the 6 genes from the part I and construct the ABAT(probe: 209459_s_at)gene as the study object.Then,the samples of liver cancer were divided into high-expression ABAT group(n=107)and low-expression ABAT group(n=107)according to the mdian expression of ABAT gene(9.26).The clinical features and prognosis were compared between the 2 groups and gene set enrichment analysis(GSEA)was used to identify the potential biological functions associated with ABAT.Results Our study demonstrated that the samples of liver cancer had a significantly lower expression of ABAT gene than normal liver samples.The high expression group of ABAT gene was associated with lower level of AFP,a better of TNM.stage,a lower of PRMS(P<0.05;The survival analysis showed that the samples with low ABAT gene expression group had a significantly poorer prognosis(P<0.05)and the TCGA data also was validated the prognosis of ABAT gene.The mulitivariate COX analysis showed the expression of ABAT gene was independent factor for the prognosis of liver cancer(P<0.05);GSEA showed that ABAT gene may be associated with Peroxisome,Wnt signaling pathway,Cysteine and methionine metabolism,Lysine degradation,Tyrosine metabolism,Ppar signaling pathway ppar,Glycine serine and threonine metabolism.Conclusions The samples of liver cancer had a significantly lower expression of ABAT gene than normal liver samples and is an independent factor for prognosis of liver cancer.Patients with high expression of ABAT gene have a better prognosisPart III Expression and Function of ABAT Gene in Hepatoma CellObjective To investigate the expression of the ABAT gene in hepatoma cell and its potential mechanism.Method The expression of ABAT gene in hepatoma cell was evaluated by q RT-PCR and western blot assays;Then,ABAT gene interference cell was constructed;The effect of ABAT gene silcecing on cell proliferation,cell cycle and apoptosis were detected by CCK-8 and flow cytometry assays;Clonal formation,invasion and migration were assessed by clonal formation,transwell assays;The expression levsls of Wnt/?-catenin signaling pathway and cleaved caspase3 was detected by western blot assays.Results Our study demonstrated that downregulation of ABATm RNA and ABAT protein in hepatoma cell was compared to L02 cell and ABAT m RNA and ABAT protein of Hep G2 and SMMC-7721 were higher than other two hepatoma cells.Then,Hep G2 and SMMC-7721 cells were selected as the subsequent experimental;#2si RNA was used to construct the ABAT gene interfering cell;Interfering with ABAT gene prompoted the cell proliferation,anti-apoptotic ability,colony formation,invasion and migration;Interfering with ABAT gene up-regulated the levels of ?-catenin?c-Myc and down-regulated level of cleaved caspse3.Conclusions The expression of ABAT gene is down-regulated in hepatoma cells,which may affect the proliferation,apoptosis,invasion and metastasis in hepatoma cells that may be related to the regulation of cell cycle,apoptosis and Wnt/?-catenin signaling pathway.