Aims: Unique in the broader category of drug-resistant cells, persister cancer cells (PSs) acquire their tolerance to compounds through reversible, chromatin-mediated changes, allowing them to ‘persist’ in the face of cancer therapeutic agents. PSs are implicated in minimal residual disease from which cancer relapse occurs, and given their established sensitivity to ferroptosis, PSs present a critical point through which identification and targeting of drug-resistant cancers may be possible. Ferroptosis sensitivity in drug-resistant cancers may be caused by the attainment of the persister state, or it may merely be correlative with this state and due instead to extended inhibition of oncogenic signaling or the induction of chemotherapy stress. Nonetheless, ferroptosis sensitivity has emerged as a common phenotype across multiple PS and drug-resistant cancer cell types. Identifying biomarkers for and drivers of ferroptosis sensitivity in drug-resistant and PS cells is therefore a high priority.

Methods: We derived PS cells from the lung carcinoma cell line PC9 (PS^PC9), performed transcriptomic analysis, and subsequently lipidomics on the PC9/PS^PC9 system. Additionally, we reverted PS^PC9 cells to the ferroptosis-resistant parental state (PC9^{PS -> PC9}) and assessed the resulting lipid changes. We generated two additional PS-like cell models: PS-like prostate carcinoma (PS^LNCaP) from LNCaP cells and PS-like fibrosarcoma (PS^HT1080) from HT1080 cells, with lipidomics analysis. Finally, we performed a mitochondrial elimination assay and assessed its effect on ferroptosis sensitivity.

Results: We observed enrichment of lipid and sugar metabolism gene expression in PS^PC9; lipidomics revealed enrichment within PS^PC9 for ferroptosis-driving diPUFA phospholipids (diPUFA-PL), as well as polyunsaturated free fatty acids (PUFA FFAs). Upon PS^PC9 reversion to the ferroptosis-resistant parental state (PC9^{PS -> PC9}), this lipid signature reverted. The LNCaP and HT1080 PS-like models individually showed features consistent with PS, including an increased labile-iron pool, reversibility, and enhanced ferroptosis sensitivity, and had lipid features consistent with those in PS^PC9. Finally, mitochondrial elimination partially abrogated ferroptosis sensitivity and altered the PS lipid profile.

Conclusion: In summary, lipidomic changes dependent on the presence of mitochondria are key to the ferroptosis sensitivity of drug-tolerant persister cancer cells.

Aims: Unique in the broader category of drug-resistant cells, persister cancer cells (PSs) acquire their tolerance to compounds through reversible, chromatin-mediated changes, allowing them to ‘persist’ in the face of cancer therapeutic agents. PSs are implicated in minimal residual disease from which cancer relapse occurs, and given their established sensitivity to ferroptosis, PSs present a critical point through which identification and targeting of drug-resistant cancers may be possible. Ferroptosis sensitivity in drug-resistant cancers may be caused by the attainment of the persister state, or it may merely be correlative with this state and due instead to extended inhibition of oncogenic signaling or the induction of chemotherapy stress. Nonetheless, ferroptosis sensitivity has emerged as a common phenotype across multiple PS and drug-resistant cancer cell types. Identifying biomarkers for and drivers of ferroptosis sensitivity in drug-resistant and PS cells is therefore a high priority.

Methods: We derived PS cells from the lung carcinoma cell line PC9 (PS^PC9), performed transcriptomic analysis, and subsequently lipidomics on the PC9/PS^PC9 system. Additionally, we reverted PS^PC9 cells to the ferroptosis-resistant parental state (PC9^{PS -> PC9}) and assessed the resulting lipid changes. We generated two additional PS-like cell models: PS-like prostate carcinoma (PS^LNCaP) from LNCaP cells and PS-like fibrosarcoma (PS^HT1080) from HT1080 cells, with lipidomics analysis. Finally, we performed a mitochondrial elimination assay and assessed its effect on ferroptosis sensitivity.

Results: We observed enrichment of lipid and sugar metabolism gene expression in PS^PC9; lipidomics revealed enrichment within PS^PC9 for ferroptosis-driving diPUFA phospholipids (diPUFA-PL), as well as polyunsaturated free fatty acids (PUFA FFAs). Upon PS^PC9 reversion to the ferroptosis-resistant parental state (PC9^{PS -> PC9}), this lipid signature reverted. The LNCaP and HT1080 PS-like models individually showed features consistent with PS, including an increased labile-iron pool, reversibility, and enhanced ferroptosis sensitivity, and had lipid features consistent with those in PS^PC9. Finally, mitochondrial elimination partially abrogated ferroptosis sensitivity and altered the PS lipid profile.

Conclusion: In summary, lipidomic changes dependent on the presence of mitochondria are key to the ferroptosis sensitivity of drug-tolerant persister cancer cells.