Systems genetics analysis of tumor evolution in the mouse

Abstract The major goal of our research has been to develop a comprehensive mouse model system for studying cancer evolution from the cell of origin, through early pre-neoplasia, to malignant and widely disseminated disease. We also wished to encompass the critical roles played by environmental mutagens, tumor promoters, and variable host genetic background, to mimic the major factors that contribute to cancer complexity in humans. The value of these models is emphasized by recent observations of abundant mutations in normal human tissues, suggesting that tumor promoters may be essential, rate-limiting factors for clonal selection and early tumor growth. Using a mouse skin model, we identified stem cells in the upper hair follicle as the major target cells of origin in which initiating mutations occur. Using a novel method involving lineage tracing of the fate of these stem cells within tumors, together with gene expression network analysis, we identified a previously unappreciated stem cell hierarchy in tumor development, in which Lgr6+ stem cells give rise to progeny expressing a range of other “cancer stem cell” markers. The initiated Lgr6+ cells face a binary decision and either undergo self renewal and proliferate, or engage a differentiation program leading to quiescence and multi-lineage stem cell plasticity. Single cell RNASeq and functional analysis have identified a “transition point” in benign and malignant tumors that controls the balance between this proliferation/DNA damage cell state, and the quiescence/lineage plasticity state. In our published work we proposed that a minor splice variant of the Kras gene, the most commonly mutated oncogene in human cancers, may control the balance between these cell states. Inhibition of the minor Kras4A isoform led to increased expression of DNA damage/ proliferation genes, and induced vulnerability to drugs that kill dividing cells. Alternatively, inhibition of the proliferative cell state using drugs that kill dividing cells, or by genetic inhibition of the Kras4B isoform, led to increased expression of the quiescence/lineage plasticity state. Our data recall classical studies of normal wound healing and the cell cycle, which led to identification of the “restriction point” , i.e. a binary decision point between alternative cell fates linked to self renewal or to differentiation. We propose that in tumors, Kras splicing is a regulator of this transition point, and that mutations in Kras abrogate critical pathways linked to wound healing responses involving the Cxcl12/Sdf1 and Cxcr4 signaling pathways. Our systems genetics approach has revealed a conceptual framework that links Kras splicing with the classical cell cycle restriction point, stem cell fate decisions, wound healing, and cancer.