Nutrition-related lipid metabolism and ferroptosis in endometrial cancer: transcriptomic profiling and hypoxia-associated gene expression

Résumé

Background Endometrial cancer is closely linked to obesity, insulin resistance and nutrition-related metabolic dysfunction. We examined lipid-synthesis and ferroptosis-related programs across tumor cell states and assessed lipogenic and ferroptosis-regulatory gene expression under hypoxia. Methods Six single-cell RNA-sequencing libraries from three primary endometrial tumors and one Visium section were analyzed alongside 545 TCGA-UCEC tumors. Lipid-metabolism, ferroptosis, hypoxia and immune scores were evaluated at cellular, spatial and bulk-tissue levels. HEC-1-A and KLE cells were cultured under normoxia or hypoxia for 24 h, with three independent cultures per condition, and gene expression was assessed by qRT-PCR. Results After quality control and doublet removal, 4,167 cells were retained and assigned to 14 annotated populations. Epithelial cells had the highest median lipid-synthesis score. Among 1,273 epithelial-lineage cells, hypoxia was weakly correlated with the ferroptosis-driver score ( r = 0.091, FDR = 0.0096), but not with defense. Across 1,347 spatial spots, epithelial and lipid-synthesis scores were positively correlated ( r = 0.191, FDR = 1.5 × 10–11). TCGA resistance scores were lower in serous than endometrioid tumors and higher in MSI-H than MSS tumors. Resistance was not independently associated with overall survival (HR = 0.83, 95% CI 0.47–1.47) and was unrelated to BMI. Under hypoxia, SREBF1, FASN and SCD expression was higher in both cell lines, together with CA9, GPX4, SLC7A11, AIFM2 and ACSL4. Conclusion Lipid-synthesis and ferroptosis-related programs varied across endometrial-cancer cell states and tissue regions. Hypoxic cell models showed concurrent changes in lipogenic and ferroptosis-regulatory transcripts. These findings identify metabolic transcriptional responses for functional investigation, without establishing a dietary effect or independent prognostic value for the resistance score.