Identifying the transcription factors that control the conversion to EMT and metastasis in breast cancer.
It is the ability to metastasize that ultimately makes breast cancer a fatal disease. Metastatic cells are often characterized as having undergone an epithelial to mesenchymal transition (EMT). EMT is a common feature of both embryonic development and invasive tumors where epithelial cells dedifferentiate to a more fibroblast-like state and regain the ability to invade, migrate, and/or proliferate in an uncontrolled fashion. Mayny studies have sought to define the genes and signaling pathways that underlie the conversion to EMT and metastasis in breast cancer. Much less emphasis has been placed on identifying the transcription factors that ultimately control this process.
We have recently developed a new model for EMT in human breast cancer involving the transcription factor CCAAT/Enhancer Binding Protein (C/EBP)beta. C/EBPbeta is critical for growth and differentiation of the mammary gland. Increased mammary epithelial cell proliferation, migration, and branching during puberty or early pregnancy and differentiation at late pregnancy are severely impaired in C/EBPbeta null mice which fail to lactate. 3 isoforms of C/EBPbeta can be produced in cells via alternative translation initiation at 3 in-frame methionines. C/EBPbeta-1 and beta-2 are transactivators, and differ by only 23 N-terminal amino acids present in beta-1 but not beta-2. C/EBPbeta-3, lacks the N-terminal half of C/EBPbeta including the transactivation domain, and therefore represses transcription. C/EBPbeta-1 is the only isoform present in normal tissue from reduction mammoplasty. However, 70% of invasive surgical primary breast tumor samples have acquired a high level of C/EBPbeta-2 expression, and C/EBPbeta-2 is the only transactivator isoform expressed in breast cancer cell lines.
Although it was first assumed that C/EBPbeta-1 and ?2 would be functionally redundant transactivators because of their extensive similarity, their different expression patterns suggest otherwise. In fact, MCF10A normal human mammary epithelial cells overexpressing C/EBPbeta-2, but not C/EBPbeta-1, undergo EMT and acquire an invasive phenotype. MCF10A C/EBPbeta-2 cells are anchorage-independent, form foci in soft agar, show loss of junctional E cadherin localization, exhibit cytoskeletal reorganization with actin stress fibers typical of motile fibroblasts, express vimentin, and are invasive in vitro.
From these and other studies we propose that C/EBPbeta-1 and -2 govern different phases of mammary gland development. C/EBPbeta-1 may be required for terminal differentiation during late pregnancy and lactation (likely activating milk protein genes), whereas ductal epithelial outgrowth and invasion through the stromal fat pad during puberty is the dominion of C/EBPbeta-2. Abberant C/EBPbeta-2 expression during cancer progression may activate a genetic program of motility and invasion in breast tumor cells.
Currently, we are extending our studies into animal models. We are evaluating if expression of C/EBPbeta-2 in breast cancer cell lines that are not invasive (MCF7, BT20) will cause the cells to undergo EMT and metastasize once implanted as xenografts in the mammary gland. We have also generated mice carrying an MMTV-driven C/EBPbeta -2 transgene; virgin females exhibit precocious, hyperplastic mammary gland development whereas multiparous females develop tumors. We will continue to study these animals to determine if females show accelerated development of metastatic carcinoma when crossed with other mouse models of breast cancer. Understanding the transcription factors responsible for metastatic capability, of which C/EBPbeta-2 is an enticing candidate, will accelerate the design of molecularly-targeted therapies capable of slowing or halting the often fatal spread of breast cancer to secondary sites.