Back

Longitudinal single cell RNA-sequencing of Pik3caH1047R-driven mammary tumorigenesis reveals coordinated transformation of epithelial and fibroblast transcriptional states

Le, J. T.; Kim, E. K.; Srivastava, V. T.; Nakagawa, R.; Phong, K. T.; Childree, G.; Moser, B. A.; Goga, A.; Gartner, Z. J.

2026-08-04 cancer biology
10.64898/2026.08.01.742267 bioRxiv
Show abstract

Breast cancer is a heterogeneous disease in which a single oncogenic driver can give rise to divergent tumor phenotypes. How oncogenic mutations generate epithelial state plasticity and coordinately remodel the surrounding tissue remains incompletely understood. Here, we applied longitudinal single cell RNA-sequencing to trace the mammary landscape during Pik3caH1047R-driven tumor progression in the mouse. We identify an expansion of the epithelial transcriptional state space, in which luminal cells lose lineage fidelity and activate ciliated, basal, and squamous-like gene expression programs. While oncogene-expressing cells lose features of luminal identity, they retain expression of hormone-sensing genes such as Esr1, Pgr, and Foxa1. These transcriptional states are established early, and the transition to overt tumors is marked by the emergence of cancer-associated fibroblasts rather than new epithelial states. We identify a Postn+ fibroblast population enriched at the epithelial interface as a candidate progenitor of cancer-associated fibroblasts. Postn+ fibroblasts express an ECM-remodeling program and display altered epithelial crosstalk in oncogenic glands. Altogether, Pik3caH1047R activation initiates a tissue-level process beginning with epithelial lineage infidelity, followed by an altered stromal microenvironment, which together mark tumor initiation.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.