Construction and validation of a two-gene signature based on SUMOylation regulatory genes in non-small cell lung cancer patients

Background Post-translational modification plays an important role in the occurrence and development of various tumors. However, few researches were focusing on the SUMOylation regulatory genes as tumor biomarkers to predict the survival for specific patients. Here, we constructed and validated a two-gene signature to predict the overall survival (OS) of non-small cell lung cancer (NSCLC) patients. Methods The datasets analyzed in this study were downloaded from TCGA and GEO databases. The least absolute shrinkage and selection operator (LASSO) Cox regression was used to construct the two-gene signature. Gene set enrichment analysis (GSEA) and Gene Ontology (GO) was used to identify hub pathways associated with risk genes. The CCK-8 assay, cell cycle analysis, and transwell assay was used to validate the function of risk genes in NSCLC cell lines. Results Firstly, most of the SUMOylation regulatory genes were highly expressed in various tumors through the R package 'limma' in the TCGA database. Secondly, our study found that the two gene signature constructed by LASSO regression analysis, as an independent prognostic factor, could predict the OS in both the TCGA training cohort and GEO validation cohorts (GSE68465, GSE37745, and GSE30219). Furthermore, functional enrichment analysis suggests that high-risk patients defined by the risk score system were associated with the malignant phenomenon, such as DNA replication, cell cycle regulation, p53 signaling pathway. Finally, the results of the CCK-8 assay, cell cycle analysis, and transwell assay demonstrated that the two risk genes, SAE1 and UBA2, could promote proliferation and migration in non-small cell lung cancer cells. Conclusions The two-gene signature constructed in our study could predict the OS and may provide valuable clinical guidance for the treatment of NSCLC patients.

Automated summary

A two-gene signature was constructed and validated to predict the overall survival of NSCLC patients. The study found that SUMOylation regulatory genes were highly expressed in various tumors. The two-gene signature, constructed using LASSO regression analysis, was able to predict patient survival in both the training and validation cohorts. Functional enrichment analysis revealed that high-risk patients were associated with malignant phenomena. Experimental results demonstrated that the two risk genes promoted proliferation and migration in NSCLC cells.