Identification of Five Hub Genes as Key Prognostic Biomarkers in Liver Cancer via Integrated Bioinformatics Analysis

Abstract

Liver cancer is one of the most common cancers and the top leading cause of cancer death globally. However, the molecular mechanisms of liver tumorigenesis and progression remain unclear. In the current study, we investigated the hub genes and the potential molecular pathways through which these genes contribute to liver cancer onset and development. The weighted gene co-expression network analysis (WCGNA) was performed on the main data attained from the GEO (Gene Expression Omnibus) database. The Cancer Genome Atlas (TCGA) dataset was used to evaluate the association between prognosis and these hub genes. The expression of genes from the black module was found to be significantly related to liver cancer. Based on the results of protein-protein interaction, gene co-expression network, and survival analyses, DNA topoisomerase II alpha (TOP2A), ribonucleotide reductase regulatory subunit M2 (RRM2), never in mitosis-related kinase 2 (NEK2), cyclin-dependent kinase 1 (CDK1), and cyclin B1 (CCNB1) were identified as the hub genes. Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathway enrichment analyses showed that the differentially expressed genes (DEGs) were enriched in the immune-associated pathways. These hub genes were further screened and validated using statistical and functional analyses. Additionally, the TOP2A, RRM2, NEK2, CDK1, and CCNB1 proteins were overexpressed in tumor liver tissues as compared to normal liver tissues according to the Human Protein Atlas database and previous studies. Our results suggest the potential use of TOP2A, RRM2, NEK2, CDK1, and CCNB1 as prognostic biomarkers in liver cancer.

Keywords: CCNB1; CDK1; NEK2; RRM2; TOP2A; hub genes; liver cancer; prognostic biomarker.

Figure 1

The flow chart of data collection, processing, analysis, and validation. HCC: hepatocellular carcinoma; DEmRNAs: differentially expressed mRNAs; PPI: protein–protein interaction; GSE: gene expression data; GO: gene ontology; KEGG: Kyoto Encyclopedia of Genes and Genomes; WGCNA: weight gene co-expression network analysis; MM: module membership; GS: gene significant; K.in: intramodular connectivity.

Figure 2

Co−expression network of the top ten genes of the black module: (A) each node in the co−expression network denotes a gene; nodes with red signify the hub genes, and the middle line represents the link between genes. (B) The connection strength of the top ten genes.

Figure 3

GO (A) and KEGG (B) function enrichment analysis. The abscissa signifies the number of genes enriched in the Figure.

Figure 4

Survival analysis of five potential hub genes obtained from gene co-expression and PPI networks and three other genes of the top ten genes with the highest intramodular connectivity (RFC4, KIF14, and PRIM1) in HCC. Overall survival of the hub genes in HCC is based on Kaplan–Meier plotter. The horizontal axis represents the time to event (in days). The patients were allocated into the high-risk and low-risk groups and assigned a color. The red line designates the samples with low risk, and the green line represents the samples with high risk. p < 0.05 indicates a statistically significant difference in mortality between groups. HR: hazard ratio of the two groups.

Figure 5

Immunohistochemistry of the five potential hub genes in liver cancer (HCC) and normal tissues from the Human Protein Atlas (HPA) database and previous studies [2,33]. Protein levels of (A) TOP2A in HCC tissue; (B) TOP2A in normal liver tissue; (C) CDK1 in HCC tissue; (D) CDK1 in normal liver tissue; (E) CCNB1 in HCC tissue; (F) CCNB1 in normal liver tissue.

Figure 6

The status of the genes expression (I) and methylation (II) of the hub genes in liver cancer. The expression and promoter methylation pattern of (A) TOP2A, (B) RRM2, (C) CCNB1, (D) CDK1, and (E) NEK2 in the primary liver tumors (n = 377) as compared to the normal samples (n = 50) using TCGA samples.