Abstract A53: Bayesian Network Inference Modeling Reveals Novel Regulators of Cell Cycle Progression and Survival

Molecular networks governing responses to targeted therapies in cancer cells are complex dynamic systems that demonstrate non-intuitive behaviors. We applied a novel computational strategy to decipher causal relationships between signaling network components. Gene networks were constructed from global gene expression profiles to model G1-S cell cycle arrest as a consequence of inhibition of MEK, a key component of the oncogenic RAF-MEK-ERK signal transduction pathway. Simulation of causally driven models generated a ranked list of 12 testable hypotheses, which confirmed role of known cell cycle regulators (CCND1, CCNE2, and CDC25A) and identified novel drivers of G1-S transition (IER2, TRIB1, C14ORF133). Validation confirmed 11 out 12 predictions. Second model which inferred causal relationships between genes lead to discovery of TRIB1 (tribbles homolog of Drosophila) as a regulator of NFkB-dependent gene expression. TRIB1 also mediates cross-talk to the NFκB pathway by enhancing p100 and IκBα-phosphorylation and degradation thus regulating cell survival and expression of cytokines crucial for tumor-stroma interaction. In agreement with this, knock-down of TRIB1 results in induction of apoptosis and sensitizes cells to cell killing induced by the death-receptor agonist TRAIL via upregulation of TRAIL receptor, DR5. Thus TRIB1 is a central regulator of cell cycle and survival and represents a potential therapeutic target. Therefore, applying REFS (Reverse Engineering Forward Simulation) to analyze gene expression profiles maximizes our ability to extract knowledge from experimental data.