Elucidating the molecular mechanisms underlying androgen-regulated lipid metabolism in prostate cancer

Abstract

Prostate cancer (PCa) is the most frequently diagnosed male cancer in Australia and a leading cause of cancer-related death. Androgen receptor (AR), a ligand-activated transcription factor, regulates growth and progression of PCa. Increased AR activity is associated with PCa development and progression to lethal castration resistant PCa (CRPC). Prostate tumours are highly reliant on lipids for growth, and one of the proposed mechanisms by which AR drives PCa progression is by regulating lipid uptake and metabolism. However, the molecular mechanisms underlying AR-regulated lipid metabolism are unclear. A better understanding of this interplay between AR signalling and lipids could identify new biomarkers and therapeutic targets, which could improve outcomes associated with this common disease. Therefore, this study aimed to elucidate mechanisms by which AR regulates lipid metabolic processes, particularly those associated with tumour growth and disease progression. By integrating cistromic and transcriptomic data, we identified Acyl-CoA Synthetase Medium Chain Family Members 1 and 3 (ACSM1 and ACSM3) as putative new AR-regulated genes in PCa. These factors are purported to activate fatty acids for their utilisation in energy production via mitochondrial beta-oxidation. Regulation of ACSM1 and ACSM3 by AR was validated using androgen and anti-androgen treatments and confirming direct binding of AR to proximal cis-regulatory elements of the genes by chromatin immunoprecipitation (ChIP). ACSM1 and ACSM3 were found to be upregulated in prostate tumours compared to non-malignant prostate tissues and expressed more highly in PCa than other cancer types. We subsequently applied metabolomics, lipidomics and functional assays to decipher the roles of ACSM1 and ACSM3 in PCa cells. Knockdown of ACSM1 and ACSM3 in PCa cells resulted in growth arrest and ATP depletion, supporting a key role for both factors in energy production from fatty acids. Furthermore, lipidomic analysis of cells showed that poly-unsaturated fatty acids accumulate in response to loss of ACSM1 and ACSM3. Metabolomics revealed that cells adapt to loss of ACSM1 and ACSM3 by switching to a glycolytic phenotype. The metabolic dysregulation induced by knockdown of ACSM1 and ACSM3 caused mitochondrial oxidative stress and subsequent lipid peroxidation, eventually resulting in cell death. Accumulation of mitochondrial reactive oxygen species was abrogated by ferrostatin-1 (an iron chelator), suggesting that cell death was due to an iron-dependent form of apoptosis termed ferroptosis. Supporting this concept, over-expression of ACSM1 and ACSM3 elicited resistance to the ferroptosis inducers Erastin and ML210. Our study has revealed a novel mechanism by which AR regulates lipid metabolism in PCa cells. Importantly, the critical role of ACSM1 and ACSM3 as key regulators of growth and protectors against ferroptosis emphasises their potential as novel therapeutic targets.