Gene Expression Patterns And The Regulatory Mechanism Prediction In Male Liver Cancer

Liver cancer is the second leading cause of cancer death in China.With the process of aging and the increasing environmental risk factor exposures,the incidence of liver cancer remains high.Liver cancer is characterized by a higher risk in males.It is estimated that the incidence of liver cancer in males is about three times than that of in females.Lifestyles,such as drinking and smoking,can not account for all the liver cancer risks in males.The currently researches about liver cancer in males mainly focus on alcohol drinking,Hepatitis B Virus(HBV),sex hormone receptors,and the interactions among them.Sex hormone receptors are considered to be the most definite cause of the high risks in male liver cancer.Androgen receptor(AR)plays a cancer promoting role,while estrogen receptor(ER)lowers the risks of liver cancer.HBV may interact with sex hormone receptors and jointly promote tumorigenesis and development.The interaction of drinking and HBV also enhance the carcinogenic ability of HBV.The current researches on the gender disparity in liver cancer mainly focus on the following two aspects: 1.explore the population characters and risk factors of liver cancer in males by mean of epidemiology.2.study the mechanism of a specific molecule or pathways in male liver cancer by means of molecular biology.In the present study,we showed the comprehensive gene expression patterns in male liver cancer based on the analysis of mRNA expression data.We evaluated the clinical value of our funding by survival analysis.Also we tried to predict the regulatory mechanisms for the gene expression patten in males.AimsTo show the specific gene expression patterns contributing to the high risks of liver cancer in males using bioinformatics.We also intend to predict the possible regulatory mechanisms of the male specific gene expression patterns based on miRNA,methylation and sex hormone receptors.Additionally,we aimed to show the clinical value of the specific gene expression patterns in male liver cancer and to develop a mathematical model to predict the prognostic of the male liver cancer patients.Methods1.the sources of data sets: the major data were derived from the the Cancer Genome Atlas(TCGA).We extracted the RNA-seq,miRNA,methylation,and clinical information of liver cancer form TCGA.The validation dataset was RNA-seq data of cancer and adjacent tissues from male liver cancer patients.2.Sample selection: sample selection was conducted using frequency pairing and propensity score matching methods.3.Differential expression analysis was conducted to extract male specific high and low expressed genes.The same differential expression analysis was also conducted in females to extract female specific high and low expressed genes.4.We use the survival analysis to investigate the relationship between the male specific enriched pathways and prognostic of liver cancer patients.5.Validation: the repeatability of the fundings in TCGA were veritifed in the validation dataset.6.We constructed cancer risk prognostic model in males to predict the survival time with several genes.Spike-and-Slab Lasso Cox model was applied in the male specific genes to construct the prognostic model.7.Differential expression analysis was performed on miRNA data to extract the gender specific high and low expression miRNAs in males and females.The TARGETSCAN and miRANDA databases were used to predict miRNA target genes.8.The Weighted Gene Co-Expression Network Analysis(WGCNA)was used to extract the module with similar expression patterns in males.We conducted gene ontology enrichment analysis in the modules to determine their functions.Co-expression networks were constructed to search for the hub genes in the modules.9.Methylation analysis was conducted in males to obtain the differential methylated probes(DMP)and differentially methylated regions(EMR).The methylation results were integrated with differential expressed genes and co-expression networks to evaluate the possible role of methylation in carcinogenesis10.The binding sites of androgen receptor(AR)and estrogen receptor(ER)in the promoter regions of the genes were predicted using transcription factor recognition motifs.We excluded the inaccessible genes using DNase I hypersensitive sites sequencing(DNaseSeq)and obtained the potential target genes of AR and ER.The functions of AR and ER were evaluated in the carcinogenesis of male liver cancer.Result1.Through mRNA differential expression analysis,we extracted 633 male specific high expressed genes and 852 low expressed genes.We also obtained 713 female specific high expressed genes,and 339 low expressed genes.The male specific high expressed genes were enriched in ribonucleoprotein complex biogenesis,ncRNA processing,ribosome biogenesis,and regulation of cellular response to heat.The male specific low expressed genes were enriched in immune related pathways.The female specific high expressed genes were enriched in cellular potassium ion transport related pathways and cell-cell adhesion pathways,whereas no pathway was enriched in female specific low expressed genes.2.A lower expression in male specific low expressed genes enriched pathways indicated an unfavorable prognostic.while a higher expression in male specific high expressed genes enriched pathways also indicated an unfavorable prognostic.3.The prognostic model included 29 genes,among which 9 genes were low expressed and 20 genes were high expressed in males.The model performed well in predicting the prognostic in male liver cancer patients.4.A total of 46 male specific miRNAs were extracted,among which 34 were specifically high expressed in males and 34 were specifically low expressed in males.The interaction between mRNA and miRNA were identified in 18 miRNAs,of which 14 miRNAs were high expressed in males.The miRNAs were predicted to interact with 322 male specific low expressed genes,which were enriched in immune-related pathways.5.A total of 12 modules were found in the WGCNA.The module enriched by immuneralated pathways was negative correlated with the modules about RNA processinng,methylation,steroid metabolic process.6.The methylation analysis identified 174551 DMPs,among which 33444 located in the promoter regions.miRNAs and LINCRNA genes possessing 3 or more DMPs in promoter regions were found in those genes.We also identified 1506 DMRs,which covered 97 male specific low expressed genes.7.Among the male specific expressed genes,80 genes might be regulated by AR,40 genes might be regulated by ER.The targets genes of AR and ER were associated with the RNA processing and RNA modification.Conclusions1.Immunosuppression was the specific gene expression pattern in male liver cancer.2.The pathways enriched by male specific expressed genes were associated with the prognosis of the male liver cancer patients.3.The specific expressed genesin males could be used to predict the prognosis of male patients with liver cancer.4.miRNAs might be the potential regulator of immunosuppression by suppressing the expression of immune-related genes in males.5.Methylation might indirectly regulate the immune function by regulating the RNA modification process,as well as directly regulat the immune function by DNA methylation.6.AR and ER,acting as transcription factors,might engage in RNA processing,consequently regulate the male specific expressed genes.