(151f) A Dynamic Analysis of Insulin Signaling and Its Feedback Mechanisms: A Discrete Modeling Approach

Wu, M., Michigan State University
Yang, X., Michigan State University
Chan, C., Michigan State Uiversity

A major challenge in systems biology is to understand the dynamic behavior of biological signaling systems, which depends on the elements of the system and their inter-relationships. Computational modeling plays an integral part in the study of network dynamics and uncovering the underlying mechanisms. However, it is difficult to acquire quantitative measurements of all the biochemical and kinetic parameters of the signaling pathways. Thus, an alternative modeling method is needed that effectively integrates pathway information and explores the basic dynamics of the signaling network, in the absence of detail kinetic parameters.

Here we applied a discrete dynamic modeling approach using three-state logic variables to describe the protein activity levels, and transition functions in a signaling network. The model takes into account cell-cell variations on the protein activities and reaction rates, and thereby can simulate a large population of cells. The simulation results provided patterns of dynamic activation for each component in the network, despite the absence of detailed kinetic parameters and uncertainties within the signaling process (i.e., cell-cell variation in reaction timing or protein activity levels).

Our dynamic model of the insulin signaling system in liver cells provides a proof-of-concept application of the proposed methodology. Research in our group has shown that the double-stranded RNA-dependent protein kinase (PKR), which is affected by insulin, is involved in regulating insulin signaling through a feedback mechanism. As a downstream target of insulin receptor substrate 1 (IRS1), PKR induces the inhibitory serine phosphorylation of IRS1, thereby suppressing its tyrosine phosphorylation. In fact, the dynamic behavior of the insulin signaling network is tuned by a variety of feedback pathways, many of which have the potential to cross-talk with PKR, but the contribution of each is unclear. Given the complexity of insulin signaling, it is inefficient to experimentally test all possible interactions in the network to confirm their functionality. Our discrete dynamic model provides an in silico model framework that integrates potential interactions and assesses the contributions of the various interactions on the dynamic behavior of the signaling network. Simulations with the model generated testable hypothesis on the response of the network upon perturbation, which were experimentally evaluated to identify the pathways that actually are functioning in the particular liver cell system. Our modeling study in combination with the experimental results, elucidated the feedbacks involved, which suggested the insulin network is robustly designed.

Thus, our discrete modeling methodology, in spite of its simplicity, is able to reproduce the dynamic profiles, provide testable hypotheses, and help to analyze the major characteristics of the insulin signaling network. Such a modeling framework that relies solely on the network architecture can be easily extended to other dynamic network systems and serves as a basis to guide model-based experiments and potentially more detailed inquiry into the regulatory mechanisms of biological networks.