A continuous ferroptosis-to-apoptosis landscape identifies ferroptosis biomarkers and repressors for cancer therapy

Ferroptosis, a cell death pathway driven by lipid peroxidation, is implicated in several human diseases including cancer. Although ferroptosis is entwined with apoptosis, it currently has no characteristic biomarkers or gene signature. In this project we established a continuous phenotypic gradient between ferroptosis and apoptosis and coupled it to transcriptomic and metabolomic landscapes. Omics and mechanistic studies revealed that ferroptosis was associated with enhanced MITF activity, lysosomal function, glutaminolysis, TCA cycle, and activation of ATF4, while its transition into apoptosis was attributed to enhanced ER-stress, ATF4/CHOP switch and PE/PC metabolic shift. The gradual ferroptosis-to-apoptosis transcriptomic landscape was used to generate a unique, unbiased transcriptomic predictor, the Gradient Gene Set (GGS), which classified ferroptosis and apoptosis with high accuracy. Further GGS optimization and datasets analysis of various ferroptotic and apoptotic responses revealed highly specific ferroptosis biomarkers, which were robustly validated in vitro and in vivo. A subset of the GGS was correlated with poor prognosis in breast cancer patients and in PDX models and was found to consist of different ferroptosis repressors, including PDAP1, whose knockdown suppressed breast tumor growth in a mouse model. Collectively, this study highlights molecular switches between ferroptosis and apoptosis, uncovering a highly prognostic predictive ferroptosis gene signature, pointing to a few ferroptosis repressors for targeted breast cancer therapy. Overall design: In a previous project we identified a synthetic lethal drug combination, JQ1 (BRD4 inhibitor) and BTZ (proteasome inhibitor). We discovered that the combination of JQ1 (100nM) and BTZ (2nM) induces ferroptosis, while incresing the concentration of BTZ to 6nM (and maintining JQ1 at 100 nM) causes the combination to induce apoptosis. To study the underlying mechanism of these combinations, we treated two triple negative breast cancer cells, MDA-MB-468 and HCC70, with the different combinations (JB2: JQ1 100nM+BTZ 2nM; JB3: JQ1 100nM+BTZ 3nM; JB6: JQ1 100nM+BTZ 6nM). We also treated the cells with the single treatments (J: JQ1 100nM; B2: BTZ 2nM; B3: BTZ 3nM; B6: BTZ 6nM). DMSO was used as control (DMSO concentration is simillar in all samples). MDA-MB-468 cells were treated in 2 time points (24h and 6h), while HCC70 were treated with 24h. For comparisons to other ferroptosis inducers, MDA-MB-468 and HCC70 were also treated with the ferroptosis inducers Erastin and RSL3. We added another cell line, HCC38, treated with Erastin. Two independent repeats were done.