Single-cell transcriptomes of dissecting the intra-tumoral heterogeneity of breast cancer microenvironment

Objective To investigate the mechanism by which heterogeneity in breast cancer developed and acted in single-cell transcriptomes. Methods The composition of breast cancer based on the single-cell transcriptomes of 54,055 high-quality cells from clinical specimens of 4 malignant and 4 non-malignant patients were investigated. Results We identified six common expression programs and six subtype-specific expression programs form malignant epithelial cells. The expression program of malignant epithelial cells exhibited activated EMT (Epithelial Mesenchymal Transition) in TME, which might indicate EMT intervention have a general therapeutic effect on various subtypes. Gene set enrichment analysis (GSEA) based on the top 50 highly NMF (non-negative matrix factorization) score genes in each program depicted the distinct function of each program in breast cancer progression. Moreover, we revealed the profound cellular heterogeneity of myeloid cell lineages in breast cancer. In macrophages, two mainly tumor associated macrophages (TAMs), TAM1 and TAM2, were also detected and the highly variable genes in TAM2 were strongly enriched in IFN-α and IFN-γ. The changes of lipid metabolism pathways in macrophages are closely related to the microenvironment of breast cancer. Conclusion We constructed a comprehensive single-cell transcriptome atlas of 54,055 cells from 4 malignant and 4 nonmalignant patients, providing insights into the mechanisms underlying breast cancer progression and the development of potential therapeutic strategies in breast cancer.