Project 1: Patterned Heterogeneity in Colon Cancer

在许多实体恶性肿瘤中，单个肿瘤内的分子和细胞异质性使我们无法鉴别

In many solid malignancies, molecular and cellular heterogeneity within a single tumor confounds our ability to understand tumorigenesis and to design effective therapies. Much effort has focused on genetic variants, the forces that lead to clonal outgrowth, and the relevance these clones have to the development of drug resistance. However it is the highly dynamic forms of non-genetic heterogeneity that are thought to enable adaptation to the rapidly changing stresses that confront the tumor as it grows and wounds the surrounding environment. Little is understood about non-genetic heterogeneity, whether it reflects a random epiphenomenon or mutualistic cooperation among cancer cells to benefit the tumor as a whole. Here we address these questions using a systems biology approach to mathematically model a pattern of non-genetic heterogeneity in xenografted colon tumors. Our studies of Wnt-β-signaling and its regulation of glycolysis have led us to discover a striking quasi-regular array of cell clusters (spots). Cells clusters are identified by high levels of Wnt (β-catenin and its target LEF1) and we identify a similar spotted pattern using markers of glycolysis. Manipulation of the levels of Wnt signaling in these xenografts changes the spotted pattern and reduces tumor growth. Whether the growth defect is functionally linked to the changes in pattern and heterogeneity is a fundamental unknown we wish to address. The overarching goal of this project therefore is to identify molecules and strategies that create pattern in this system, to ask how these strategies influence tumor growth, and to determine whether these mechanisms are involved in drug resistance. Aim 1 will use high-resolution single cell RNA-seq (scRNA-seq) and tumor imaging to build on our existing mathematical model and to explain the relationships between tumor heterogeneity, growth and invasion. Data from experiments wherein expression of candidate regulators and cell populations have been manipulated will be used to validate and refine the predictive power of our model. Work in Aim 2 will build a general model(s) that can explain how overt differences in tumor growth, heterogeneity, and metabolism arise as emergent behaviors of non-linearly interacting networks that qualitatively and quantitatively affect the Wnt pathway. In Aim 3 we use xenograft models that develop resistance to the anti-angiogenic drug bevacizumab and single cell RNA-seq approaches to examine the changes in cellular heterogeneity and patterning that accompany acquisition of resistance. Mathematical modeling will identify the most likely resistance mechanisms: mutualistic metabolic symbiosis, metabolic/population rewiring, mutualistic non-metabolic symbiosis, or none of the above. Modeling predictions of strategies that re-establish drug sensitivity will be tested via genetic engineering (CRISPR/Cas9) or small molecule drug therapies.