Modeling the dynamics of Let-7-coupled gene regulatory networks linking cell proliferation to malignant transformation

The microRNA Let-7 controls the expression of proteins that belong to two distinct gene regulatory networks, namely a cyclin-dependent kinases (Cdks) network driving the cell cycle and a cell transformation network which can undergo an epigenetic switch between a non-transformed and a malignant transformed cell state.\n\nUsing mathematical modeling and transcriptomic data analysis, we here investigate how Let-7 controls the cdk-dependent cell cycle network, and how it couples the latter with the transformation network. We also determine whether the two networks can be combined into a larger entity that impacts on cancer progression.\n\nOur analysis shows that the switch from a quiescent to a cycling state depends on the relative levels of Let-7 and several cell cycle activators. Numerical simulations further indicate that the Let-7-coupled cell cycle and transformation networks control each other, and our model identifies key players for this mutual control. Transcriptomic data analysis from the The Cancer Genome Atlas (TCGA) suggest that the two networks are activated in cancer, in particular in gastrointestinal cancers, and that the activation levels vary significantly among patients affected with a same cancer type. Our mathematical model, when applied to a heterogeneous cell population, suggests that heterogeneity among tumors results from stochastic switches between a non-transformed cell state with low proliferative capability and a transformed cell state with high proliferative property. The model further predicts that Let-7 may reduce tumor heterogeneity by decreasing the occurrence of stochastic switches towards a transformed, proliferative cell state.\n\nIn conclusion, we identified the key components responsible for the qualitative dynamics of two GRNs interconnected by Let-7. The two GRNs are heterogeneously involved in several cancers, thereby stressing the need to consider patients specific GRN characteristics to optimize therapeutic strategies.