Identification of novel pharmacological targets linked to retinoid metabolism for the treatment of advanced prostate cancer

Prostate cancer represents the second most frequently diagnosed malignancy among men and the fifth leading cause of cancer-associated death worldwide. Although prostate cancer patients live substantially longer, metastatic and castration-resistant prostate cancers are currently difficult to treat. Therefore, there is an urgent need to develop novel pharmacological strategies. Metabolic reprogramming is widely recognized as a hallmark of cancer, and the aberrant metabolic phenotype of proliferating cancer cells provides potential therapeutic opportunities. This project aims to identify new metabolic gene alterations capable of affecting prostate cancer progression and, therefore, potentially targetable with existing or innovative therapeutics. To this end, RNA-sequencing and polysome profiling was performed on prostate specimens isolated from five different genetically engineered mouse models (GEMMs) of prostate cancer. Within the top-ranked upregulated genes, cellular retinol-binding proteins, essential chaperones of retinol homeostasis encoded by Rbp1 and Rbp2, were identified. Also, many other retinol-associated genes, such as Aldh1a2, Lrat, and Rdh5, resulted differentially expressed in most prostate cancer GEMMs, suggesting that tumor cells engage in an aberrant retinol metabolic program. To further investigate these findings, tumorigenic and non-transformed human prostatic epithelial cells were assessed for sensitivity to retinol and its bioactive form all-trans retinoic acid (atRA) and evaluated for RNA and protein expression of critical genes associated with retinoid metabolism. Interestingly, non-tumorigenic cells resulted considerably more sensitive to retinol and atRA, while neoplastic cells were almost insensitive. mRNA and protein levels of RALDHs, key enzymes responsible for retinol conversion into RA, decrease dramatically in most prostate cancer epithelial cells, possibly preventing RA production. Consistently, lipidomics analysis revealed profound changes in retinoid metabolism, with a decrease of retinoid storage and an increase of free retinol in prostate cancer cells. Furthermore, this aberrant metabolic phenotype in retinoid homeostasis was also revealed by in vivo experiments. In this direction, immunohistochemistry analyses were performed on prostate specimens isolated from both PTENpc-/- and wild-type mice to evaluate the expression levels of RALDHs enzymes in normal and transformed tissues. Surprisingly, RALDH1, RALDH2, and RALDH3 expression levels were higher in prostatic intraepithelial neoplasia lesions when compared to normal glands, suggesting an involvement of the tumor microenvironment in prostate cancer-associated retinoid metabolism rewiring. These promising, but still preliminary results, might enable the identification of new retinoid-associated metabolic liabilities, paving the way for the discovery of original therapeutic approaches.