Exploration and Validation of Key Genes and Immune Infiltration in Alcoholic Hepatitis

Abstract

Yingna Mei,1 Yitao Zheng,2 Xingfen Zhang,1 Ting Hu,1 Xiaoqing Zheng,1 Cui Zhang,3 Qinzhi Deng1 1Department of Liver Disease, Ningbo No.2 Hospital, Ningbo, People’s Republic of China; 2Department of Pancreatic and Gastrointestinal Surgery Division, Ningbo No. 2 Hospital, Ningbo, People’s Republic of China; 3Department of Cardiology, Renmin Hospital of Wuhan University Cardiovascular Research Institute, Wuhan University; Hubei Key Laboratory of Cardiology, Wuhan, 430060, People’s Republic of ChinaCorrespondence: Qinzhi Deng, Department of Liver Disease, Ningbo No.2 Hospital, Ningbo, People’s Republic of China, Email [email protected] Cui Zhang, Department of Cardiology, Renmin Hospital of Wuhan University, Wuhan, People’s Republic of China, Email [email protected]: Apart from alcohol abstinence and glucocorticoids, there is still no effective treatment to improve alcoholic hepatitis, and the specific mechanism of its pathogenesis is still unclear.Methods: We screened the differential genes in GEO alcoholic hepatitis database by differential analysis and screened the eQTL genes that have causal relationship with alcoholic hepatitis by Mendelian randomization analysis. The intersection of differential genes and eQTL genes was used to obtain candidate genes. The candidate genes were then screened out by machine learning, and their expression was further verified in clinical patients and mice with alcoholic hepatitis. Based on key genes, pathway analysis via single-gene GSEA analysis and immune microenvironment analysis via ssGSEA analysis were conducted to explore the relationship between key genes and immune microenvironment. Finally, regulatory relationship between key genes and immune cells was explored based on cell experiments.Results: Based on the GEO Alcoholic hepatitis database (GSE28619 and GSE142530) and Mendelian randomization of eQTL genes, we obtained 17 candidate genes. We then obtained two key genes (CXCL8 and CTNNA1) through lasso and random forest tree algorithms. CXCL8 and CTNNA1 were highly expressed in the alcoholic hepatitis group, which were verified in clinical patients and mice. Through single-gene GSEA analysis, two key genes were identified to be enriched in the antigen presentation pathway. At the same time, the alcoholic hepatitis group had obvious immune infiltration disorder, and two key genes were correlated with immune environment via correlation analysis. B cells and NKT cells exhibited the highest correlation with key genes. In alcoholic hepatitis mice, liver infiltration of B cells and NKT cells was verified. Through cell experiments, ethanol exposure increased CTNNA1 and CXCL8 expression in NKT and B cells, enhancing inflammatory cytokine release and suppressing IgG production, respectively. Silencing CXCL8 and CTNNA1 reversed these effects.Conclusion: These results suggested that CXCL8 and CTNNA1 were potential biomarkers for alcoholic hepatitis, and might be new targets for the treatment of alcoholic hepatitis.Keywords: alcoholic hepatitis, immune environment, Mendelian randomization analysis, machine learning, CXCL8, CTNNA1

Keywords

• alcoholic hepatitis
• immune environment
• mendelian randomization analysis
• machine learning
• CXCL8
• CTNNA1