Unraveling the basis of cell type specificity in signal transduction pathways using perturbation-based modeling and multi-omics analyses

A primary goal of cancer systems biology is to build models of cell regulatory systems that can predict the impact of genetic changes that cause cancer and that can be used to design therapeutic strategies. Unfortunately, the great majority of current signaling models are mostly descriptive and lack the ability to predict responses across different cell types or the impact of gene copy number alterations. To address this critical capability gap, we have developed a measurement-modeling-perturbation approach to identify pathway elements that are responsible for cell type-specific differences in the response of cells to growth factors and cytokines. Our initial studies have focused on MCF10A and HS578T cells, which display distinct sensitivities and response dynamics to epidermal growth factor (EGF) despite both being mammary epithelial cells with similar levels of EGF receptors (EGFR). To understand the basis of this difference, we first built a model of MCF10A cells and parameterized it with data from that cell type. The response profile of this model was then used to predict the response profile of HS578T cells. Differences between model predictions and actual HS578T cell responses were used to design additional experiments to determine the mechanistic basis. These experiments used perturbations (both chemical and genetic) of potential mechanisms to see which best explained the results. The impact of some of these perturbations was evaluated at multiple molecular levels using targeted proteomics and multiplexed phospho-proteomics. This allowed us to see how the perturbation altered the overall system and which parts of the models needed to be changed to accommodate the experimental observations. This information was then used to update the MCF10A model so that it could better predict responses across cell types. Currently, key differences between the two cell types can be at least partially explained by differences in the expression level of specific pathway proteins. In addition, the strength of feedback loops are key determinants in cell type-specific signaling dynamics. The ultimate goal of these studies is to build a general model of signaling pathways that can be easily modified to work across different cell types.