Spheroid-Induced Epithelial-Mesenchymal Transition Provokes Global Alterations of Breast Cancer Lipidome: A Multi-Layered Omics Analysis

Metabolic rewiring has been recognized as an important feature to the progression of cancer. However, the essential components and functions of lipid metabolic networks in breast cancer progression are not fully understood. In this study, we investigated the roles of altered lipid metabolism in the malignant phenotype of breast cancer. Using a spheroid-induced epithelial-mesenchymal transition (EMT) model, we conducted multi-layered lipidomic and transcriptomic analysis to comprehensively describe the rewiring of the breast cancer lipidome during the malignant transformation. A tremendous homeostatic disturbance of various complex lipid species including ceramide, sphingomyelin, ether-linked phosphatidylcholines, and ether-linked phosphatidylethanolamine was found in the mesenchymal state of cancer cells. Noticeably, polyunsaturated fatty acids composition in spheroid cells was significantly decreased, accordingly with the gene expression patterns observed in the transcriptomic analysis of associated regulators. For instance, the up-regulation of SCD, ACOX3, and FADS1 and the down-regulation of PTPLB, PECR, and ELOVL2 were found among other lipid metabolic regulators. Significantly, the ratio of C22:6n3 (docosahexaenoic acid, DHA) to C22:5n3 was dramatically reduced in spheroid cells analogously to the down-regulation of ELOVL2. Following mechanistic study confirmed the up-regulation of SCD and down-regulation of PTPLB, PECR, ELOVL2, and ELOVL3 in the spheroid cells. Furthermore, the depletion of ELOVL2 induced metastatic characteristics in breast cancer cells via the SREBPs axis. A subsequent large-scale analysis using 51 breast cancer cell lines demonstrated the reduced expression of ELOVL2 in basal-like phenotypes. Breast cancer patients with low ELOVL2 expression exhibited poor prognoses (HR = 0.76, CI = 0.67-0.86). Collectively, ELOVL2 expression is associated with the malignant phenotypes and appear to be a novel prognostic biomarker in breast cancer. In conclusion, the present study demonstrates that there is a global alteration of the lipid composition during EMT and suggests the down-regulation of ELOVL2 induces lipid metabolism reprogramming in breast cancer and contributes to their malignant phenotypes.