Neddylation related gene signature for prognostic biomarkers in colorectal cancer with integrated bioinformatics and experimental analysis

The role of neddylation in colorectal cancer (CRC) is increasingly recognized as a significant factor. Our study employed bioinformatics analysis to investigate the functions of neddylation-related genes in CRC. Transcriptome data from multiple public databases were integrated and analyzed using a comprehensive bioinformatics approach, including differential expression analysis, weighted gene co-expression network analysis, and machine learning tactics to identify prognostic genes. These genes were then implemented to establish a risk prediction model. Single-cell RNA sequencing (scRNA-seq) was utilized to reveal the distribution and interactions of various cell types. The characteristics of immune cell infiltration were assessed by immune infiltration analysis. Finally, the functional roles of the prognostic gene were validated through the wound healing assay, cell counting kit-8, and transwell assay. Three genes—CCNF, ANKRD13D, and PSMA7—were identified as prognostic markers. The risk prediction model built on these genes demonstrated modest predictive performance, with the area under the curve values exceeding 0.6 at 3, 5, and 7 years. Immune infiltration analysis showed that the infiltration levels of 15 immune cells, including plasmacytoid dendritic cells and eosinophil, were markedly higher in the high-risk cohort. Furthermore, scRNA-seq data indicated that M1 macrophages exhibited higher scores for the neddylation gene set in CRC. Experimental validation demonstrated that knockdown of ANKRD13D inhibits cell proliferation, migration, and invasion. This study established a preliminary prognostic risk model based on CCNF, ANKRD13D, and PSMA7, which exhibits exploratory predictive value for CRC prognosis. These results identified CCNF, ANKRD13D, and PSMA7 as potential candidate biomarkers for prognostic assessment in CRC.