Proteomic Analysis Identifies NDUFS1 and ATP5O as Novel Markers for Survival Outcome in Prostate Cancer

SIMPLE SUMMARY: Due to the heterogeneity of prostate cancer (PCa), it is still difficult to provide risk stratification. Metabolic changes in PCa tissue have been described during tumor progression at genetic and transcriptomic level, but these have not yet clearly contributed to improved diagnosis and therapy. The aim of our study was to identify novel markers for aggressive prostate cancer in a proteomics-derived dataset by immunohistochemical analysis and correlation with transcriptomic data. Here, we provide potential new markers—NDUFS1 and ATP5O—for risk stratification in PCa. Additionally, we reveal for the first time a concordant increase of NDUFS1/ATP5O of mRNA expression in transcriptomic datasets and at protein level. ABSTRACT: We aimed to identify novel markers for aggressive prostate cancer in a STAT3-low proteomics-derived dataset of mitochondrial proteins by immunohistochemical analysis and correlation with transcriptomic data and biochemical recurrence in a STAT3 independent PCa cohort. Formalin-fixed paraffin-embedded tissue (FFPE) sample selection for proteomic analysis and tissue-microarray (TMA) generation was conducted from a cohort of PCa patients. Retrospective data analysis was performed with the same cohort. 153 proteins differentially expressed between STAT3-low and STAT3-high samples were identified. Out of these, 46 proteins were associated with mitochondrial processes including oxidative phosphorylation (OXPHOS), and 45 proteins were upregulated, including NDUFS1/ATP5O. In a STAT3 independent PCa cohort, high expression of NDUFS1/ATP5O was confirmed by immunocytochemistry (IHC) and was significantly associated with earlier biochemical recurrence (BCR). mRNA expression levels for these two genes were significantly higher in intra-epithelial neoplasia and in PCa compared to benign prostate glands. NDUFS1/ATP5O levels are increased both at the mRNA and protein level in aggressive PCa. Our results provide evidence that NDUFS1/ATP5O could be used to identify high-risk PCa patients.