Recent whole-genome sequence analyses of human breast cancer samples have provided us with a comprehensive view of the mutational landscape of this cancer type. A critical step following these massive efforts is to identify and validate the driver mutations and to understand how they contribute to breast tumorigenesis. Many of the significantly mutated genes identified from these sequencing efforts are already known to play roles in breast tumorigenesis when mutated, such as TP53, PIK3CA, GATA3 and CDH1. In addition to these known mutations, several novel recurrent mutations have also emerged; among these, somatic point mutations and deletions in genes encoding a transcription factor, RUNX1, as well as its non-DNA-binding partner, CBFbeta, are of particular interests. Both RUNX1 and CBFbeta play central roles in hematopoiesis and translocations and mutations involving their coding genes are frequently found in human leukemias. To understand the role of Runx1 in mammary epithelial cell (MEC) development and tumorigenesis, we characterized its expression in MECs by immunohistochemistry and found that although Runx1 is expressed in both basal/myoepithelial cells and luminal MECs in virgin or parous non-pregnant mammary glands (MGs), it is only expressed in basal/myoepithelial cells during pregnancy and lactation. We then performed a conditional knockout study of Runx1 in MGs using MMTV-Cre or K14-Cre. During lactation, we found almost all conditional knockout females failed to nurse their pups, due to milk stasis. As Runx1 is only expressed in myoepithelial cells at this stage, we reasoned that Runx1 might control a normal function of myoepithelial cells during lactation. In virgin females, although Runx1-null MGs appeared phenotypically normal, Matrigel-based colony assay of sorted Runx1-null luminal MECs showed that they formed slightly more and significantly larger 3-D colonies compared to wild type luminal cells. Microarray analysis of sorted luminal cells further revealed that in Runx1-null luminal MECs, many genes in a previously defined luminal progenitor signature are upregulated, whereas genes in a mature luminal cell signature are largely downregulated. Since the phenotype of Runx1-null luminal cells in virgin MGs resembles that of Gata3-null luminal cells, we hypothesize that Runx1 and Gata3 may work together in luminal MECs. In support of this, we found in breast cancer cell lines, RUNX1 and GATA3 physically interact and by gel filtration chromatography, we found that RUNX1 and GATA3 co-migrated and appeared to be in the same multi-protein complexes. Overall, we conclude that Runx1 plays distinct roles in different subsets of MECs. In particular in ductal luminal MECs, Runx1 and Gata3 work together to control the mature luminal cell fate and RUNX1 mutations identified in human breast tumors may contribute to breast tumorigenesis by blocking maturation of mammary luminal cells.

Citation Format: Maaike van Bragt, Ying Xie, Zhe Li. Runx1 controls maturation of luminal mammary epithelial cells. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr LB-116. doi:10.1158/1538-7445.AM2013-LB-116