Brain Res. 2026 Oct 26:150604. doi: 10.1016/j.brainres.2026.150604. Online ahead of print.
ABSTRACT
BACKGROUND: Glioblastoma (GBM) is the most common and lethal primary brain tumor. However, the dynamic evolution of myeloid cells within its microenvironment and their role in immune suppression remain unclear. This study aims to elucidate the heterogeneity and immune regulatory mechanisms of myeloid cells, particularly tumor-associated neutrophils, through single-cell transcriptomic analysis.
METHODS: We integrated single-cell transcriptomic data from 4 cases of GBM and 4 cases of low-grade glioma (LGG) to systematically analyze the myeloid cell landscape and identify neutrophil subpopulations and their differentiation trajectories. Utilizing CellChat cell-cell communication analysis, pseudotime trajectory analysis (performed on CD8 + T cells), and functional scoring, we revealed the interaction network between SRGN + neutrophils and CD8 + T cell subpopulations. Furthermore, based on the TCGA cohort, we constructed a SRGN + signature-based prognostic model to evaluate immune infiltration, immune checkpoints, TIDE immune evasion, and temozolomide sensitivity.
RESULTS: Neutrophils were specifically enriched in GBM. A SRGN + subpopulation highly expressed S100A8/A9, IL1B, and PMN-MDSC-related genes, with elevated inflammation and S100 scores. Functional scoring indicated that this subpopulation exhibited pro-inflammatory and immunoregulatory characteristics. Among CD8 + T cells, three exhausted subsets were defined: Tex_PD1, Tex_NKG2A, and Tex_TIM3. CellChat-based ligand-receptor analysis predicted preferential chemotactic signaling from SRGN + neutrophils toward Tex_PD1 cells via CCL pathways and toward Tex_TIM3 cells via CXCL pathways, as well as potential inhibitory signaling toward Tex_NKG2A and Tex_TIM3 cells via the HLA-E-NKG2A axis. A 12-gene risk model stratified GBM patients into high- and low-risk groups. High-risk patients showed shorter overall survival but greater temozolomide sensitivity (reduced IC50 predicted by oncoPredict analysis), alongside an immune phenotype of "increased T-cell infiltration yet functional exhaustion." TIDE analysis confirmed elevated dysfunction scores, reduced exclusion scores, and low MDSC phenotype scores, consistent with SRGN + neutrophils lacking classic MDSC markers.
CONCLUSION: This study systematically characterizes the dynamic evolution and immunosuppressive functions of myeloid cells, particularly SRGN + neutrophils, in GBM, suggesting a potential dual "chemotactic recruitment and selective suppression" mechanism by which they may shape the landscape of CD8 + T-cell exhaustion. The risk model constructed based on this subpopulation demonstrated prognostic stratification value within the TCGA-GBM cohort, pending validation in independent cohorts, and suggests a potential therapeutic strategy combining targeting of the HLA-E-NKG2A axis with temozolomide.
PMID:42933716 | DOI:10.1016/j.brainres.2026.150604