Ferroptosis is an iron-dependent cell death characterized by the accumulation of hydroperoxided phospholipids. Here, we report that the NUPR1 inhibitor ZZW-115 induces ROS accumulation followed by a ferroptotic cell death, which could be prevented by ferrostatin-1 (Fer-1) and ROS-scavenging agents. The ferroptotic activity can be improved by inhibiting antioxidant factors in pancreatic ductal adenocarcinoma (PDAC)- and hepatocellular carcinoma (HCC)-derived cells. In addition, ZZW-115-treatment increases the accumulation of hydroperoxided lipids in these cells. We also found that a loss of activity and strong deregulation of key enzymes involved in the GSH- and GPX-dependent antioxidant systems upon ZZW-115 treatment. These results have been validated in xenografts induced with PDAC- and HCC-derived cells in nude mice during the treatment with ZZW-115. More importantly, we demonstrate that ZZW-115-induced mitochondrial morphological changes, compatible with the ferroptotic process, as well as mitochondrial network disorganization and strong mitochondrial metabolic dysfunction, which are rescued by both Fer-1 and N-acetylcysteine (NAC). Of note, the expression of TFAM, a key regulator of mitochondrial biogenesis, is downregulated by ZZW-115. Forced expression of TFAM is able to rescue morphological and functional mitochondrial alterations, ROS production, and cell death induced by ZZW-115 or genetic inhibition of NUPR1. Altogether, these results demonstrate that the mitochondrial cell death mediated by NUPR1 inhibitor ZZW-115 is fully rescued by Fer-1 but also via TFAM complementation. In conclusion, TFAM could be considered as an antagonist of the ferroptotic cell death.

The conceptualization of a novel type of cell death, called ferroptosis, opens new avenues for the development of more efficient anti-cancer therapeutics. In this context, a full understanding of the ferroptotic pathways, the players involved, their precise role, and dispensability is prerequisite. Here, we focused on the importance of glutathione (GSH) for ferroptosis prevention in pancreatic ductal adenocarcinoma (PDAC) cells. We genetically deleted a unique, rate-limiting enzyme for GSH biosynthesis, γ-glutamylcysteine ligase (GCL), which plays a key role in tumor cell proliferation and survival. Surprisingly, although glutathione peroxidase 4 (GPx4) has been described as a guardian of ferroptosis, depletion of its substrate (GSH) led preferentially to apoptotic cell death, while classical ferroptotic markers (lipid hydroperoxides) have not been observed. Furthermore, the sensitivity of PDAC cells to the pharmacological/genetic inhibition of GPx4 revealed GSH dispensability in this context. To the best of our knowledge, this is the first time that the complete dissection of the xCT-GSH-GPx4 axis in PDAC cells has been investigated in great detail. Collectively, our results revealed the necessary role of GSH in the overall redox homeostasis of PDAC cells, as well as the dispensability of this redox-active molecule for a specific, antioxidant branch dedicated to ferroptosis prevention.

Glutathione peroxidase 4 (GPX4) has been reported as one of the major targets for ferroptosis induction, due to its pivotal role in lipid hydroperoxide removal. However, recent studies pointed toward alternative antioxidant systems in this context, such as the Coenzyme Q-FSP1 pathway. To investigate how effective these alternative pathways are in different cellular contexts, we used human colon adenocarcinoma (CRC) cells, highly resistant to GPX4 inhibition. Data obtained in the study showed that simultaneous pharmacological inhibition of GPX4 and FSP1 strongly compromised the survival of the CRC cells, which was prevented by the ferroptosis inhibitor, ferrostatin-1. Nonetheless, this could not be phenocopied by genetic deletion of FSP1, suggesting the development of resistance to ferroptosis in FSP1-KO CRC cells. Considering that CRC cells are highly glycolytic, we used CRC Warburg-incompetent cells, to investigate the role metabolism plays in this phenomenon. Indeed, the sensitivity to inhibition of both anti-ferroptotic axes (GPx4 and FSP1) was fully revealed in these cells, showing typical features of ferroptosis. Collectively, data indicate that two independent anti-ferroptotic pathways (GPX4-GSH and CoQ10-FSP1) operate within the overall physiological context of cancer cells and in some instances, their inhibition should be coupled with other metabolic modulators, such as inhibitors of glycolysis/Warburg effect.

Medulloblastoma (MB) is the most frequent solid tumor in children, localized in the brain posterior fossa. Its standard of care comprises maximal resection surgery followed by craniospinal irradiation and chemotherapy. Despite a long-term survival rate of 70%, wide disparities among patients have been observed. Relevant targets for naive and recurrent MB are urgently needed. Primary and recurrent MBs are characterized by aggressive invasion into surrounding brain tissue, active angiogenesis, and radioresistance. Integrin-alpha v beta 3 was a major driver of these features in glioblastoma. Nevertheless, such observations have not yet been reported in MB. Integrin alpha v beta 3 was found to be expressed in a subset of MB patients. We investigated the role of integrin-alpha v beta 3 using MB-derived cell lines with beta 3 -subunit depletion or overexpression both in vitro and in vivo. Radioresistant MB cell lines were generated and showed increased integrin-alpha v beta 3 expression, which correlated with increased susceptibility to pharmacological integrin-alpha v beta 3 inhibition with cilengitide, a competitive ligand mimetic. Finally, we conducted single-photon emission computed tomography (SPECT)/magnetic resonance imaging (MRI) studies on orthotopic models using a radiolabeled integrin-alpha v beta 3 ligand (99mTc-RAFT-RGD). This approach offers the prospect of a novel predictive imaging modality in MB. Altogether, our data pave the way for SPECT/MRI-based selection of a subpopulation of MB patients eligible for integrin-alpha v beta 3-directed therapies.