Immunol Res. 2024 Jun 18.
Breast cancer remains the most common malignant carcinoma among women globally and is resistant to several therapeutic agents. There is a need for novel targets to improve the prognosis of patients with breast cancer. Bioinformatics analyses were conducted to explore potentially relevant prognostic genes in breast cancer using The Cancer Genome Atlas (TCGA) and The Gene Expression Omnibus (GEO) databases. Gene subtypes were categorized by machine learning algorithms. The machine learning-related breast cancer (MLBC) score was evaluated through principal component analysis (PCA) of clinical patients' pathological statuses and subtypes. Immune cell infiltration was analyzed using the xCell and CIBERSORT algorithms. Kyoto Encyclopedia of Genes and Genomes enrichment analysis elucidated regulatory pathways related to speedy/RINGO cell cycle regulator family member C (SPDYC) in breast cancer. The biological functions and lipid metabolic status of breast cancer cell lines were validated via quantitative real-time polymerase chain reaction (RT‒qPCR) assays, western blotting, CCK-8 assays, PI‒Annexin V fluorescence staining, transwell assays, wound healing assays, and Oil Red O staining. Key differentially expressed genes (DEGs) in breast cancer from the TCGA and GEO databases were screened and utilized to establish the MLBC score. Moreover, the MLBC score we established was negatively correlated with poor prognosis in breast cancer patients. Furthermore, the impacts of SPDYC on the tumor immune microenvironment and lipid metabolism in breast cancer were revealed and validated. SPDYC is closely related to activated dendritic cells and macrophages and is simultaneously correlated with the immune checkpoints CD47, cytotoxic T lymphocyte antigen-4 (CTLA-4), and poliovirus receptor (PVR). SPDYC strongly correlated with C-C motif chemokine ligand 7 (CCL7), a chemokine that influences breast cancer patient prognosis. A significant relationship was discovered between key genes involved in lipid metabolism and SPDYC, such as ELOVL fatty acid elongase 2 (ELOVL2), malic enzyme 1 (ME1), and squalene epoxidase (SQLE). Potent inhibitors targeting SPDYC in breast cancer were also discovered, including JNK inhibitor VIII, AICAR, and JW-7-52-1. Downregulation of SPDYC expression in vitro decreased proliferation, increased the apoptotic rate, decreased migration, and reduced lipid droplets. SPDYC possibly influences the tumor immune microenvironment and regulates lipid metabolism in breast cancer. Hence, this study identified SPDYC as a pivotal biomarker for developing therapeutic strategies for breast cancer.
Keywords: Breast cancer; Chemotherapy; Lipid metabolism; Machine learning; SPDYC; Tumor immune microenvironment