Graduation Date

Fall 12-18-2020

Document Type


Degree Name

Doctor of Philosophy (PhD)


Cancer Research

First Advisor

Kay-Uwe Wagner


Breast cancers of the claudin-low subtype make up a substantial portion of triple-negative breast cancers and have stem cell-like and mesenchymal features. Although it has been recognized for some time that this breast cancer subtype is highly aggressive and difficult to treat, the molecular drivers and cellular origin of claudin-low breast cancer have been poorly defined. The lack of suitable in vivo models has prohibited the study of tumor initiation and progression of this subtype. In this work, we report two novel mouse models that, upon expression of oncogenic RAS in the mammary epithelium, develop highly metastatic triple-negative mammary tumors that belong to the claudin-low subtype. The main technological advance of these models is that we uncoupled the expression of the oncogene from differentiation-state dependent promoters, which uncovered a remarkable range of cellular plasticity that mammary tumor cells can assume during cancer progression. Specifically, when mutant KRAS is expressed under its own gene locus or another ubiquitously active promoter in luminal epithelial cells, we observed that cancer cells engage a continuous transdifferentiation program that promotes epithelial-to-mesenchymal transition (EMT), which ultimately lead to the development of undifferentiated, mesenchymal and stem cell-like mammary cancers. The longitudinal study revealed that claudin-low mammary cancers can progressively evolve from luminal cells that transition through basal-like neoplastic lesions and tumors before reaching a mesenchymal-like differentiation state. Moreover, we demonstrated that the continuous signaling of oncogenic RAS in cooperation with regulators of EMT plays a crucial role in the cellular plasticity and maintenance of the mesenchymal and stem cell characteristics of claudin-low mammary cancer cells. In summary, we have shown that claudin-low mammary cancers can arise from functionally differentiated luminal cells and, therefore, this cancer subtype does not exclusively originate from normal undifferentiated mammary stem cells as previously proposed. Moreover, we established that hyperactive RAS signaling plays an important role in claudin-low breast cancer biology and may serve as a suitable target for breast cancer prevention, diagnosis, and treatment.