Integrative bioinformatics analysis and experimental validation of key biomarkers driving the progression of cirrhotic portal hypertension.

BACKGROUND

Portal hypertension is a driving factor of cirrhosis complications, but the specific molecular mechanism of portal hypertension in cirrhosis remains unclear. The aim of this study was to identify hub genes for predicting persistent progression of portal hypertension in patients with liver cirrhosis.

METHODS

Related microarray datasets were obtained from the Gene Expression Omnibus database. Weighted gene co-expression network analysis and differential expression genes analysis were used to identify the correlation sets of genes. In addition, protein-protein interaction networks and machine learning algorithms were conducted to screen center of candidate genes. To validate the diagnostic effect of hub genes, receiver operating characteristic curves were utilized in another dataset that is publicly accessible. Furthermore, the CIBERSORT algorithm was employed to investigate the immune infiltration levels of 22 immune cells and their connection to hub gene markers. Immunohistochemistry and reverse transcription quantitative polymerase chain reaction (RT-qPCR) were conducted to validate novel hub genes in clinical specimens.

RESULTS

We obtained 671 differentially expressed genes and 11 module genes related to cirrhotic portal hypertension. Two candidate genes namely oncoprotein-induced transcript 3 protein (OIT3) and lysyl oxidase like protein 1 (LOXL1) were identified as biomarkers. RT-qPCR and immunohistochemistry (IHC) verified the expression of LOXL1 and OIT3 at mRNA and protein levels in liver tissue.

CONCLUSIONS

OIT3 and LOXL1 were identified as potential novel targets for the diagnosis and treatment of cirrhotic portal hypertension (CPH).