Integrative single-cell and spatial transcriptomics analysis reveals a baicalein-responsive 10-gene signature for non-small cell lung cancer.

Journal: Translational oncology
Published Date:

Abstract

BACKGROUND: Non‑small cell lung cancer (NSCLC) remains a leading cause of cancer‑related mortality worldwide. Baicalein, a natural flavonoid, has shown anti‑cancer activity but its molecular targets and cell‑type‑specific effects in the tumor microenvironment (TME) are incompletely understood. METHODS: We integrated network pharmacology, single‑cell RNA‑seq, bulk transcriptomics, spatial transcriptomics, machine learning, and in vitro experiments to identify baicalein‑responsive genes in NSCLC. RESULTS: Single‑cell analysis resolved 21 cell populations, revealing that baicalein targets were enriched in a 32‑gene core set. Consensus clustering defined three molecular subtypes (cluster 1-cluster 3) with distinct immune infiltration; cluster 1 showed an immune‑cold phenotype with upregulation of cell cycle and metabolic pathways, while cluster 3 was immune‑hot. Machine learning selected a 10‑gene signature (TOP2A, CDH1, CCNB1, SATB2, CA9, HMGB2, MB, NQO1, AURKB, CCNB2) with high diagnostic accuracy. SHAP analysis identified TOP2A as the most influential contributor. Spatial transcriptomics confirmed significantly elevated expression of all ten genes in tumor versus adjacent/normal tissues. Cell‑cell communication analysis highlighted enhanced MIF pathway signaling in the TME, with epithelial cells and SPP1+ TAMs acting as key senders. Molecular docking showed strong binding affinities (ΔG ≤ -8.5 kcal/mol) between baicalein and AURKB, MB, TOP2A, and CCNB2, and molecular dynamics simulations further confirmed the stability of these complexes. In vitro, baicalein (30 μM, 24 h) differentially modulated signature gene expression in PC‑9 and A549 cells and suppressed viability in a dose‑ and time‑dependent manner. CONCLUSIONS: This integrative study provides a comprehensive single‑cell and spatial atlas of baicalein‑responsive genes in NSCLC, identifies a robust diagnostic signature, and offers mechanistic insights for developing baicalein as a potential therapeutic agent.

Authors

Keywords

No keywords available for this article.