Aberrant expression of multiple glycolytic enzyme genes is significantly associated with disease progression and survival outcomes in prostate cancers.

Prostate cancer is the leading cause of cancer death after lung cancer in men. Recent studies showed that aberrant metabolic pathways are involved in prostate cancer development and progression. In this study, we performed a systemic analysis of glycolytic enzyme gene expression using the TCGA-PRAD RNAseq dataset. Our analysis revealed that among 25 genes, only four genes (HK2/GPI/PFKL/PGAM5) were significantly upregulated while nine genes (HK1/GCK/PFKM/PFKP/ALDOC/PGK1/PGAM1/ENO2/PKM) were downregulated in primary prostate cancer tissues compared to benign compartments. Among these 13 altered genes, four genes (ENO2/ALDOC/GPI/GCK) exhibited strong diagnostic potential in distinguishing malignant and benign tissues. Meanwhile, GPI expression exerted as a prognostic factor of progression-free and disease-specific survival. PFKL and PGAM5 gene expressions were associated with AR signaling scores in castration-resistant patients, and AR-targeted therapy suppressed their expression. In LuCap35 xenograft tumors, PFKL and PGAM5 expression was significantly reduced after animal castration, confirming the AR dependency. Conversely, GCK/PKLR genes were significantly associated with neuroendocrinal progression, representing two novel neuroendocrinal biomarkers for prostate cancer. In conclusion, our results suggest that GPI expression is a strong prognostic factor for prostate cancer progression and survival while GCK/PKLR are two novel biomarkers of prostate cancer progression to neuroendocrinal status.