Single-cell RNA-seq reveals breast cancer heterogeneity and identifies TCP1 as a therapeutic target in breast cancer.

Breast cancer is composed of diverse cell populations, and this intratumoral heterogeneity profoundly affects clinical behavior. Here, we leveraged single cell RNA sequencing (scRNA-seq) of 68 breast cancer specimens to dissect tumor heterogeneity at high resolution. Unsupervised clustering identified all major cell types of the tumor microenvironment (TME)-including malignant epithelial cells, fibroblasts, T cells, macrophages, endothelial cells, and others-with striking variability in their proportions across molecular subtypes. For example, a BRCA1-mutant triple-negative breast cancer (TNBC) sample showed dense immune infiltration, whereas an estrogen receptor (ER)-positive tumor was mostly epithelial, consistent with known subtype differences in immunogenicity. We applied inference of copy number variations (inferCNV) to distinguish malignant epithelial cells, identifying ~90,000 tumor cells with significant copy-number aberrations enriched for cancer hallmark pathways. Re-clustering of these malignant cells revealed five discrete subpopulations. Notably, a KRT17-positive subcluster displayed the highest stemness score and a distinctive ETS-family transcription factor (ERG) regulon, suggesting a stem-like phenotype. Using The Cancer Genome Atlas (TCGA) cohort, we found that genes upregulated in this KRT17+ subpopulation, particularly NFKBIA, PDLIM4, and TCP1 stratified patient survival. An 8-gene risk signature derived from the KRT17 program segregated patients into high- and low-risk groups with markedly different outcomes. High-risk tumors were characterized by an immunosuppressive TME enriched in M2-like macrophages, whereas low-risk tumors more often harbored lymphocyte-predominant infiltrates. Focusing on TCP1, a chaperonin subunit upregulated in high-risk tumors, we demonstrate that TCP1 knockdown in breast cancer cell lines substantially impairs cell migration (~50% reduction in wound closure) and invasion (P < 0.01). These findings reveal functionally distinct malignant cell states within breast cancer and identify TCP1 as a promising therapeutic target to disrupt aggressive, stem-like tumor cell programs, ultimately guiding more personalized treatment strategies.