Characterization of sphingolipid-based stress responses in the yeast Saccharomyces cerevisiae through reverse engineering

Abstract

Sphingolipids regulate numerous cell functions through the activation of specific signaling cascades. Although sphingolipids have been investigated for several decades, a detailed mechanistic and systemic understanding of their biosynthesis and utilization is still lacking. As a consequence, it is still impossible to predict with reliability how cells react and adapt to external stresses and which specific roles sphingolipids play in such stress responses. In this thesis, I develop mathematical and computational approaches that shed light on the regulatory mechanisms with which baker’s yeast responds to two types of stresses: heat and hydroxyurea. Stresses typically mandate the transition of a cellular system from its normal steady state to a different state. Thus, in the first project, I perform a theoretical state transition analysis. Based on this theoretical foundation, I propose in the second project an optimization strategy that appropriately captures the sphingolipid dynamics under 30 minutes of heat stress. This analysis reveals novel cellular response strategies, including a switch from biosynthesis to sphingolipid retrieval from cell membranes. To address the roles of these distinct ceramide variants, which differ in their fatty acyl CoA chain lengths, I propose a model that uses the previous model as boundary conditions. The model reveals interesting patterns of ceramide dynamics that are different for variants with long and very fatty acyl groups. Finally, I analyze long-term exposure of yeast cells to hydroxyurea. This analysis permits the novel identification and characterization of subtle regulatory mechanisms that are based on the cells’ distinction between ceramides with saturated or unsaturated fatty acyl groups. Taken together, this dissertation work reveals novel control mechanisms with which yeast cells coordinate complex responses to external stresses. Beyond the analysis of sphingolipids, this work demonstrates how innovative techniques of dynamic modeling and optimization can assist in the extraction of detailed information from modern metabolomics data.