Integrating bioinformatics and experiments to identify neddylation-related prognostic features in prostate adenocarcinoma

Targeting neddylation represents a promising treatment strategy for prostate adenocarcinoma (PRAD). This study integrated bioinformatics and experimental approaches to identify neddylation-related prognostic genes in PRAD, with the goal of providing new insights. Public datasets were utilized for differential expression analysis, Cox regression, and machine learning algorithms to identify prognostic genes. A prognostic risk model was developed based on the biochemical recurrence (BCR) of patients and the identified prognostic genes. The tumor microenvironment, key cell populations, and causal relationships were explored using immune infiltration, single-cell RNA sequencing (scRNA-seq), and Mendelian randomization (MR). The expression of prognostic genes was subsequently validated through reverse transcription-quantitative PCR (RT-qPCR) in PRAD cell lines and a transgenic mouse model of PRAD. Additionally, pharmacological inhibition of neddylation with MLN4924 was conducted in PRAD cells to elucidate the regulatory effects of the neddylation pathway on these candidate genes. Four prognostic genes (TOP2A, SNRNP70, NFE2L2, and CAV1) were identified, with a prognostic risk model based on these genes effectively predicting PRAD BCR. Immune infiltration analysis demonstrated significant differences in five immune cell types, including M1 macrophages. Basal epithelial cells emerged as key populations, with SNRNP70, NFE2L2, and CAV1 exhibiting distinct expression patterns between PRAD and control samples. MR analysis indicated statistically significant associations of CAV1 (odds ratio [OR] = 0.275, P = 0.037, protective factor) and NFE2L2 (OR = 5.154, P = 0.007, risk factor) with benign prostatic neoplasm. Expression profiling revealed that TOP2A and SNRNP70 were significantly upregulated, whereas NFE2L2 and CAV1 were downregulated in both PRAD cell lines and TGMAP allograft tissues. Following MLN4924-mediated neddylation inhibition, transcriptional repression of TOP2A and SNRNP70 was observed, alongside induced expression of CAV1; however, NFE2L2 mRNA levels remained stable despite notable accumulation of its encoded NRF2 protein at the post-translational level. Four prognosis-related genes (TOP2A, SNRNP70, NFE2L2, and CAV1) were identified, and a risk model was constructed, providing valuable biomarkers and potential therapeutic targets for the management of PRAD.