Key points are not available for this paper at this time.
BACKGROUND: Glioblastoma (GBM) is the most aggressive adult brain tumor, marked by intratumoral heterogeneity and therapy resistance. Metabolic reprogramming through histone lactylation has been linked to tumor progression and immune suppression. However, the spatial and single-cell landscape of lactylation in GBM and its prognostic significance remain poorly understood. METHODS: We employed a multi-omics approach integrating bulk RNA sequencing, single-cell RNA sequencing (scRNA-seq), and spatial transcriptomics to investigate lactylation-related signatures in GBM. Differential expression and pathway analyses were performed using GEO and TCGA datasets. Cell clustering, SCENIC transcriptional network inference, CellChat intercellular communication modeling, and pseudotime analysis were conducted. A prognostic risk model was constructed using LASSO-Cox regression based on lactylation-associated genes. Experimental validation was performed using western blotting, immunohistochemistry, and functional assays in GBM cell lines. RESULTS: Lactylation-related genes were significantly upregulated in GBM and associated with poor prognosis and immunosuppressive tumor microenvironments. Single-cell analysis revealed high-lactylation malignant subpopulations enriched in hypoxic tumor cores, exhibiting metabolic reprogramming and enhanced immune evasion. Spatial transcriptomics confirmed the localization of S100A6-high-lactylation GBM cells in aggressive tumor regions. A nine-gene lactylation-based risk model stratified patients into high- and low-risk groups with significantly different survival outcomes (AUC: 0.77-0.87). Experimental knockdown of S100A6 reduced GBM cell proliferation, migration, and invasion. CONCLUSIONS: Lactylation defines distinct tumor cell clusters in GBM that are spatially localized, metabolically reprogrammed, and immunosuppressive. The S100A6-associated lactylation signature serves as a robust prognostic biomarker and potential therapeutic target in GBM.
Han et al. (Sat,) studied this question.