Single-cell transcriptomic profiling uncovers key molecular signatures in glioma pathogenesis.

BACKGROUND

Glioma represents the most prevalent and lethal primary malignant tumor of the central nervous system, characterized by remarkable cellular heterogeneity and poor prognosis. Comprehensive characterization of glioma at single-cell resolution is essential for identifying novel therapeutic targets and improving patient outcomes.

METHODS

We performed comprehensive single-cell RNA sequencing (scRNA-seq) analysis on glioma samples obtained from the Gene Expression Omnibus (GEO) database. Advanced computational approaches including principal component analysis (PCA), uniform manifold approximation and projection (UMAP), t-distributed stochastic neighbor embedding (t-SNE), and differential expression analysis were employed to characterize cellular heterogeneity. Gene co-expression network construction and pathway enrichment analysis were conducted to identify functional modules. Candidate biomarkers were validated using quantitative real-time PCR (qRT-PCR) and enzyme-linked immunosorbent assay (ELISA) in GL261 glioma cells, C8-D1A normal glial cells, and intervertebral disc nucleus pulposus cells.

RESULTS

Quality control analysis revealed high-quality single-cell data with median gene counts of 5,437 and UMI counts of 14,207 per cell. A total of 4,753 highly variable genes were identified, and 22 distinct cellular clusters were delineated using the Louvain algorithm. PCA loading analysis identified key contributing genes including cell cycle regulators (CDK1, TOP2A, BIRC5), immune-related genes (C1QA, C1QB, HLA-DRB6), and neural lineage markers (MOBP, MAG, GJB1). Cell type annotation identified seven major populations: astrocytes, oligodendrocytes, microglia, neural stem cells, OPC/immature neurons, pericytes, and T cells. Differential expression analysis uncovered 847 upregulated and 652 downregulated genes (|log2FC| > 1, adjusted P < 0.05). Gene co-expression network analysis revealed five major functional modules centered on hub genes including CLEC12A, CLU, AQP4, and S100A16. Pathway enrichment demonstrated significant involvement of cell cycle, Notch signaling, MAPK pathway, and neurogenesis. Experimental validation confirmed that Plp1 was significantly downregulated in GL261 cells (0.38 ± 0.05-fold, P < 0.01), while Fth1 (2.15 ± 0.28-fold, P < 0.001) and Gm42418 (5.67 ± 0.52-fold, P < 0.001) were markedly upregulated compared to normal glial cells.

CONCLUSION

This comprehensive single-cell transcriptomic analysis successfully characterized the cellular heterogeneity of glioma, identifying distinct cell populations and molecular signatures. PLP1, FTH1, and GM42418 were validated as potential molecular biomarkers, providing novel insights into glioma pathogenesis and potential therapeutic targets.