HMGA2 regulates GPX4 expression and ferroptosis in prostate cancer cells

Prostate cancer (PCa) remains a significant global health burden and an increase in oxidative stress is associated with cancer progression. High Mobility Group A2 (HMGA2), a chromatin architectural protein, increases oxidative stress and promotes sensitivity to ferroptosis inducers, however, the mechanism is unknown. We investigated the role of HMGA2 in GPX4 regulation and the impact on cellular responses to oxidative stress and ferroptosis sensitivity. We conducted UALCAN database analysis, western blot analysis, and lipid peroxidation assays to determine the relationship between HMGA2 and GPX4 and the levels of lipid reactive oxygen species in a panel of PCa cell lines, including an enzalutamide-resistant cancer cell line (C4-2B MDVR). Our results show an inverse relationship between HMGA2 and GPX4 expression with high HMGA2 and low GPX4 expression associated with higher Gleason score and lower survival probability in prostate adenocarcinoma (PRAD) patients, while low/moderate HMGA2 expression is positively associated with increased GPX4 expression and higher survival probability. Cell lines showed a moderately negative but not statistically significant correlation between HMGA2 and GPX4 expression, however, PC3 and DU145 PCa cells display higher lipid peroxides concomitant with higher endogenous levels of HMGA2 and low GPX4. Overexpression of wild-type HMGA2 in LNCaP and 22Rv1 cells leads to higher HMGA2 expression compared to Neo control and is associated with higher SLC7A11 and GPX4 expression, while interestingly truncated HMGA2 overexpression in LNCaP and 22Rv1 cells coincides with higher HMGA2 and reduced GPX4 expression, leading to increased lipid peroxides and susceptibility to ferroptosis. Overexpression of wild-type and truncated HMGA2 in 22Rv1 cells increases SLC7A11 mRNA yet differing GPX4 protein expression suggests posttranslational regulation of GPX4. Moreover, enzalutamide-resistant C4-2B MDVR cells display higher HMGA2 levels compared to C4-2B cells, as well as sensitivity to RSL3 ferroptosis inducer, which is partially reversed by ferroptosis inhibitor, ferrostatin-1. Interestingly, GPX4 expression is higher in C4-2B MDVR cells compared to C4-2B, and HMGA2 knockdown further increases its expression but does not significantly alter its susceptibility to ferroptosis. In conclusion, our study shows that HMGA2 regulation of GPX4 expression is complex and truncated HMGA2 downregulates GPX4 and increases lipid peroxides. Moreover, HMGA2-expressing cells including enzalutamide-resistant cells are susceptible to RSL-3-induced ferroptosis. Thus, ferroptosis sensitivity offers promising insights for the development of targeted therapeutic interventions for aggressive PCa.