Predicting the prognosis of hepatocellular carcinoma based on genes related to polyamine metabolism

Background Hepatocellular carcinoma (HCC) is a highly prevalent malignant tumor worldwide. Evidence showed that polyamine metabolism plays a crucial part in the regulation of cancer onset and development, however, its clinical significance in HCC remains unclear. Methods Bulk RNA sequencing (RNA-seq) and single-cell RNA sequencing (scRNA-seq) data of HCC were collected from public databases. Polyamine metabolism-related genes (PMRGs) were obtained from the MSigDB database. The molecular subtypes of HCC were classified by ConsensusClusterPlus package, and differentially expressed genes (DEGs) of the molecular subtypes were identified by the limma package, followed by enrichment analysis with clusterProfiler package. Univariate Cox and Lasso Cox regression analyses were performed to screen core genes, construct risk model, and develop a nomogram integrating clinical characteristics for survival prediction. The obtained biomarkers were validated using in vitro experiments (CCK8, wound healing, and Transwell assay). The Tumor Immune Estimation Resource (TIMER), MCP-counter, and Cell Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT) methods were employed for immune cell infiltration analysis. Finally, drug sensitivity of the HCC samples was analyzed with the oncoPredict package. Results This study identified two molecular subtypes (C1 and C2), with C2 demonstrating a more favorable prognosis. Glucose-6-phosphate dehydrogenase ( G6PD ), alcohol dehydrogenase 4 ( ADH4 ), S100 calcium binding protein A9 ( S100A9 ), aldo-keto reductase family 1 member B15 ( AKR1B15 ) were predicted as the biomarkers for HCC. Cell experiment results showed that the expressions of G6PD , AKR1B15 , and S100A9 were all notably elevated in HuH-7 cells. Moreover, the loss of G6PD gene expression reduced the viability, migratory, and invasive capabilities of HCC cells. Patients with a high RiskScore had a lower survival rate than those with a low RiskScore. Scores of immune cells such as Tregs and M0 macrophages were higher in the high-risk group, and 13 drugs were found to be significantly linked to the RiskScore. Single-cell analysis showed that G6PD and S100A9 were high-expressed mainly in hematopoietic progenitor cells (HPCs) and macrophages. Conclusion In conclusion, this study screened four key genes based on PMRGs and constructed a risk model to effectively predict the prognosis of HCC, providing novel potential targets and theoretical basis for the molecular subtyping and individualized treatment of HCC.