Abstract
BACKGROUND: Tumor heterogeneity impacts invasive behaviors, treatment response, and clinical outcomes in triple-negative breast cancer (TNBC). However, this heterogeneity remains incompletely characterized. This study aims to utilize multi-scale data to investigate inter-tumoral heterogeneity and identify potential TNBC biomarkers. METHODS: Single-cell RNA expression profiles were analyzed using R packages. Specifically, the infercnv, Pyscenic, GeneNMF, SCP, Vector, CellChat, and hdWGCNA packages were employed to identify malignant cells and characterize heterogeneity in transcription factors, metaprograms, lineage evolution, developmental trajectories, cell-cell interactions, and co-expression networks. Bulk RNA datasets were incorporated to assess the prognostic value of cell clusters and candidate genes. G Protein Subunit Alpha 15 (GNA15) expression was determined via reverse transcription-quantitative PCR (RT-qPCR) and immunohistochemistry. Cell functional assays were performed to evaluate proliferation, migration, and invasion capabilities. RESULTS: A total of 14,335 malignant cells were isolated from epithelial cells across 15 single-cell RNA samples. Six tumor cell clusters were identified, which exhibited distinct prognoses, biological functions, driver transcription factors, and co-expression networks. Notably, the S2 cluster demonstrated association with multiple malignancy-related pathways and inferior survival outcomes. GNA15 emerged as the S2 cluster hub gene. In vitro experiments confirmed that GNA15 knockdown significantly attenuated proliferation, migration, and invasion in TNBC cell lines. CONCLUSIONS: Our study comprehensively delineated TNBC tumor cell heterogeneity and established the critical role of GNA15 in TNBC progression. These findings enhance the understanding of TNBC heterogeneity and provide a theoretical foundation for TNBC treatment.