Mitochondrial fission and stemness in prostate cancer

cancer stem-like cells (CSCs) capable of both selfreproducing and generating a progeny of non-stem tumor cells with various degrees of differentiation and proliferative capacity [1]. CSCs contribute actively to intra-tumor heterogeneity, disease progression, metastatic spread and treatment failure [2]. In prostate cancer, CSCs, which are unaffected by standard androgen receptor (AR) targeting therapies, contribute to the emergence of aggressive and treatment resistant tumors [3]. To improve this scenario, new agents and treatment strategies are needed to eliminate CSCs along with bulk tumor cells. In our study [4], we investigated the ability of inhibitors of bromodomain and extraterminal domain (BET) proteins to affect CSCs in prostate cancer models. BET proteins, like Bromodomain containing 4 (BRD4), recognize acetylated lysine residues on histone and nonhistone proteins enabling the formation of transcription activating complexes at specific genomic sites [5]. BET protein inhibitors (BETi) block the formation of coactivator complexes and inhibit transcription in a cellcontext specific manner [5]. BETi have anticancer activity and are currently evaluated in several clinical trials in many tumor types, including prostate cancer. However, whether BETi treatment affects prostate CSCs was not known. To address this issue, we compared the effects of BETi and BRD4 knockdown in bulk tumor cells (BTCs) and prostate CSCs [4]. Surprisingly, we found the two cell subpopulations responded quite differently. BTCs experienced a transient growth arrest, which was more pronounced in AR-proficient cells (e.g., LNCaP cells) than in AR-negative cells (e.g., DU145). Conversely, BRD4 inhibition severely reduced tumor-sphere formation and self-renewal capacity at subsequent passages. Furthermore, BRD4 inhibition reduced tumor growth due to the consistent loss of tumor-initiating stem-like cells in vivo. These anti-CSC effects were seen both in vitro and in vivo equally in AR signaling proficient and deficient cells. Thus, inhibition of CSC expansion, rather than the transient arrest of BTCs, was driving the in vivo activity of BETi in the different xenograft models. Incidentally, these results emphasize the need of novel and diversified assays and experimental systems to fully assess the therapeutic potential of anticancer compounds. In our study, in vitro/ Editorial