Exploring the molecular mechanism of hepatitis virus

IntroductionHepatic malignant tumors can be divided into two major categories: primary and secondary. Primary hepatocellular carcinoma is one of the common malignant tumors in China with a high mortality rate, ranking third in the sequence of deaths from malignant tumors after stomach and esophagus (1). Hepatocellular carcinoma (HCC) is the major subtype of primary hepatocellular carcinoma. In 2020, there were more than 910,000 new HCC cases and 830,000 deaths, which has become a serious public health problem (2). The latest research results have shown that the occurrence of HCC is mainly related to hepatitis B virus (HBV) and hepatitis C virus (HCV) infection (3). At the same time, excessive drinking and smoking are also related to the occurrence of HCC (4).Systematic understanding of the virus involvement in the occurrence, development and metastasis of HCC is conducive to the early diagnosis and accurate treatment of patients. In theory, the persistent inflammation of hepatocytes caused by viral infection promotes the formation of hepatic fibrosis, which is the basis for the development of HCC (5). Chronic infection caused by HBV to the human body, and chronic hepatitis B, compensated cirrhosis, and decompensated cirrhosis to the onset of HCC are the main pathways for the development of HBV-associated HCC (HBV-HCC) (6). It has been found that the HBx protein carried by HBV can regulate the PI3K-Akt pathway of host hepatocytes, and then activate and release of excessive TGF-β to participate in the occurrence of liver fibrosis (7). TGF-β is a key cytokine involved in fibrogenesis and can be specifically activated by PI3K/Akt (8), which may be an important factor in the induction of HCC by HBV and HCV (9, 10). In addition, studies have found that abnormal PI3K/AKT/mTOR signaling pathway is closely related to HCC resistance (11). Some studies have demonstrated that N6-methyladenylate (m6A) modification is involved in the progression of hepatitis B virus-related liver fibrosis by regulating the infiltration of immune cells (12), and at the same time, HBx carried by HBV can interact with the methylase METTL3, which is closely related to the development of HCC (13).Despite the deepening understanding of the etiology of HCC, the available diagnosis and treatment plan has little effect. Microarray sequencing technology has been applied to genome detection for in-depth exploration of the viral carcinogenicity and tumor development mechanism. However, results from single microarray or low sample size analyses are difficult to gain more recognition. In this study, a prospective method was used that differed from conventional experiments, the expression profiles of HCC (GSE87630), HBV-HCC (GSE55092), and HCV-HCC (GSE19665) from the GEO database were integrated. Differentially expressed genes (DEGs) were identified in HCC, HBV-HCC, and HCV-HCC, compared to normal liver tissue. DEG was analyzed for gene set enrichment using “h.all. v7.4” and in addition, three types of HCC were analyzed for tumor immune invasion using CIBERSORT. To further analyze the molecular mechanisms of the involvement of the two hepatitis viruses in the development of HCC, an ExpressAnalyst was used to combine the three data sets. Compare with HCC, and in contrast to HCC, the DEG of HBV-HCC and HCV-HCC were identified, and two co-homology genes were crossed to explore the molecular basis of viral involvement in HCC. Six sub-networks of key molecular interactions were obtained using the MCODE method, and gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG), analyses were performed using the DAVID database. In addition, the hub genes were obtained by analyzing the sub-networks using six algorithms, and the correlation was explored between those genes and immune-infiltrating cells using Pearson correlation analysis. A separate cohort (GSE121248 and GSE69715) was integrated with HCC (GSE87630) to verify the stability of hub gene expression. Meanwhile, the protein levels of those genes were verified in the human protein map (HPA). The prognostic value of hub genes was verified on the UALCAN data analysis platform, and the receiver operating characteristic (ROC) curve was used to analyze the diagnostic value of hub genes in distinguishing HCC tissues (HBV-HCC and HCV-HCC) from normal liver tissues. Transcription factor and m6A methylation predictions were performed using hTFtarget and m6A2Target and validated in microarrays for hub gene. Therapeutic drugs related to genes were explored and developed by STITCH, and effect intensity analysis was performed on computer simulation software (Schrodinger). The aim of this study is to provide new insights into the pathogenesis of viral involvement in HCC and to identify new prognostic markers and precise drug targets through comprehensive analysis.MethodsData extraction, processing and consolidationAll data sets (Table 1) were from GEO (https://www.ncbi.nlm.nih.gov/geo) database, were log transformed and normalized. The differentially expressed genes (DEGs) between hepatocellular carcinoma (HCC) tissues and normal liver tissues were screened out by using “limma” function. The screening criteria for DEGs were adjusted P value < 0.05 and LogFC≥1. Volcanic map of gene distribution by using that ggplot2 function. Batch effects were removed from the dataset using “combat” function by the online platform ExpressAnalyst (https://dev.expressanalyst.ca/ExpressAnalyst/).Table 1 The characteristics of the datasets used in the present study.GSEA and immune infiltration analysisTo clarify the gene effects caused by DEGs of HCC, the R package “GSEA” was utilized to obtain the GSEA enrichment scores of hallmark pathways (h.all.v7.4.entrez) (14). HCC immune infiltration analysis was performed on the dataset using the “CIBERSORT” method on the Sangerbox3.0 platform (http://vip.sangerbox.com). Due to the small sample size of GSE19665 for HCV-HCC and the potential for large deviations in analysis results, an external independent dataset, GSE69715, was selected for immune invasion analysis.Analysis of protein-protein interaction network and functional enrichmentInteraction between intersecting gene were analyzed using a STRING database (15). The MCODE functional module was used to cluster the genes and construct a gene sub-network in cytoscape (version 3.8.3). The DAVID 2021 was used for analyzing GO and KEGG pathway of module gene (16). P 75%, 25-75%, or < 25%).Survival analysis and ROC curve drawingThe survival analysis was analyzed by using the UALCAN. Survival analysis was performed by Kaplan-Meier and log-rank test. The expression levels of hub genes were applied for ROC analysis to estimate their diagnostic significance to distinguish between HCC and normal in two independent external sets (GSE121248 and GSE69715) and internal sets (GSE55092 and GSE19665). The area under curve (AUC) > 0.5 was considered to have diagnostic value.Transcriptional factor and m6A enzyme prediction analysisIn order to find the molecules that regulate the hub genes, hTFtarget database was used to predict the TFs of the hub genes and validated the expression levels in the data set (20). Because of the extensive presence of m6A enzyme modification after RNA transcription, m6A enzyme prediction (21) and validation on the hub genes were performed, and constructed a network diagram based on the interaction relationship.Drug screening based on hub genesThe sequence of hub gene was obtained from NCBI and homology modeling was performed using Swiss Mode (22). Schrodinger Glide and IFD modules were used to molecular dock the hub gene structure with the best score with drug molecules. The lower binding energy of drug to target (hub gene) indicates more stable binding. Docking mode diagrams are drawn using LigPlus (23).ResultsDGE identification and GSEA in HCCWe compared the gene expression in Hepatocellular carcinoma (HCC) with that in the normal tissue, and obtained 1158 DEGs, among which 819 genes were down-regulated, accounting for more
Read More