Integrative Single-Cell and Spatial Transcriptomic Analyses Link Arachidonic Acid Metabolic Reprogramming to...

Background/Objectives: Glioblastoma (GBM) is widely recognised as a highly aggressive form of malignant tumour arising within the central nervous system. The proneural-to-mesenchymal (PN-MES) state transition is a key process underlying malignant progression and therapeutic resistance. The present study was designed to elucidate the association between the PN-MES transition and arachidonic acid (AA) metabolic reprogramming in GBM. Methods: We integrated four public single-cell RNA-sequencing cohorts, two spatial transcriptomic cohorts, and three bulk RNA-sequencing cohorts for GBM. By combining single-cell transcriptomics, spatial transcriptomics, pseudotime trajectory inference, and cell-cell communication analysis, we systematically evaluated the cell-state dependence and spatial heterogeneity of AA metabolism in GBM. We further incorporated machine learning-based survival modelling, molecular docking, molecular dynamics simulations, and in vitro functional assays to identify and validate potential prognostic biomarkers and therapeutic targets. Results: Mesenchymal-like (MES-like) tumour cells showed the highest transcriptionally inferred AA metabolism-related score, and AA metabolism-related gene expression was closely coupled with PN-MES state transition. The peptidylprolyl isomerase A (PPIA)-basigin (BSG) signalling axis was selectively enriched in tumour cells with high expression of arachidonic acid metabolism-related genes (AAMGs), whereas virtual knockout of BSG perturbed MES- and invasion-related gene modules. Spatial transcriptomic analysis confirmed that PPIA-BSG-associated communication was enhanced within MES-like tumour niches and was coupled with inflammatory and hypoxic programmes. AA metabolism-related genes enabled robust prognostic stratification. Molecular simulation and in vitro experiments suggested that Venetoclax could bind BSG and suppress GBM cell viability. Conclusions: AA metabolism-related transcriptional reprogramming defines mesenchymal-associated malignant niches in GBM and may contribute to PN-MES-related state evolution through PPIA-BSG-mediated microenvironmental communication. AA metabolism-related features and the BSG-associated signalling axis may provide candidate directions for prognostic stratification and targeted intervention.