Impact and mechanisms of ferroptosis in ovarian carcinoma

Abstract

Ovarian cancer (OC) is a highly lethal gynaecological malignancy due to the high recurrence rate and general resistance to apoptosis-inducing chemotherapies. Inducing non-apoptotic cell death has been emerging as a novel approach for cancer therapy. Ferroptosis is a non-apoptotic form of programmed cell death driven by iron-dependent reactive oxygen species (ROS) accumulation, leading to the oxidative damage of membrane lipids. As one of the central events of ferroptosis, lipid peroxidation involves a preferential oxidation of polyunsaturated fatty acids (PUFAs)-rich phosphatidylethanolamine (PE). Ethanolamine phosphotransferase 1 (EPT1) is a critical gene involved in the synthesis of PE via the Kennedy pathway. Accumulating evidence suggests OC may be susceptible to ferroptotic cell death. However, the impact of ferroptosis on OC and the underlying molecular mechanism have not been fully characterized.

In this study, we sought to determine the sensitivity of OC cells to ferroptosis and investigate the possibility of augmenting the anti-tumour capability of platinum-based chemotherapy with ferroptosis-inducing agents. We also explored the effect and molecular mechanism of EPT1 on ferroptosis in OC. Our results demonstrated that ferroptosis agonists erastin and RSL3 dose-dependently induced ferroptotic cell death in a series of OC cell lines hallmarked by significant reduction of glutathione level, elevation of ROS and lipid peroxidation levels. Nearly complete reversal by lipid ROS scavenger was observed. The susceptibility of OC cells was significantly associated with level of GPX4 expression but not SLC7A11 expression or histological subtypes of serous and clear cell carcinomas. erastin and RSL3 synergistically enhanced cytotoxic effect of cisplatin in cisplatin-resistant OC cells OVCA420 and KOC7C. In comparison to single cisplatin, combinatorial treatment of cisplatin and erastin (or RSL3) resulted in an integrated tumoricidal effect as inducing ferroptotic and apoptotic cell deaths, as evidenced by significant decreased glutathione level and increased levels of ROS, lipid peroxidation and cleaved Caspase 3.

A higher expression of EPT1 was found in multiple cancer tissues including OC than normal tissues and was significantly associated with poorer prognosis. Compared to EPT1-wildtype cells, EPT1 knockout established by CRISPR-Cas9 system significantly sensitized OVCA420 (serous) and KOC7C (clear cell) cells to RSL3 accompanied by significant increase of lipid peroxidation level. Re-overexpression of EPT1 restored the resistance of OVCA420 cells to RSL3. Mechanistically, knockout of EPT1 resulted in the reduced phosphorylation of AKT and downstream autophagic initiator ULK1, as well as the increased level of LC3-II in OVCA420 cells upon RSL3 exposure, thereby enhancing the autophagy-mediated ferroptosis. Transcription profiling of EPT1-wildtype and -knockout OVCA420 cells revealed the significant changes of PUFAs synthesis and glycolysis pathways, which may be involved in the modulation of ROS production and lipid peroxidation mediated by EPT1 deficiency.

In conclusion, these findings demonstrated the general susceptibility of OC cells to ferroptosis and the critical role of ferroptosis in enhancing cytotoxicity of cisplatin in cisplatin-resistant OC, highlighting the value of inducing ferroptosis as a potential strategy for OC, especially platinum-resistant tumours. The data suggested possibility of EPT1 as a potential ferroptosis regulatory gene by modulating autophagy, providing new insights into the molecular mechanisms of ferroptosis.