Modeling Physiological Responses to Stress: Metabolism of Fasting and Acetaminophen-Induced Liver Damage in the Laboratory Rat

Metabolism is an integral component of our response to physiological stress. We need to convert nutrients to energy, make the requisite macromolecular components, and use these components, to function normally and handle varying physiological stress. Capturing these physiological responses using metabolic network modeling is instrumental for obtaining scientific insights from transcriptomic and metabolomic data. Here, we studied coupled liver and kidney metabolism in the laboratory rat under mild stress conditions, i.e., during short-term fasting and during the initial period of acute acetaminophen toxicity before the onset of cellular damage. We used a genome-scale network reconstruction of rat liver and kidney metabolism, together with a modeling approach constrained by stress-induced gene expression data, central carbon fluxes derived from isotope tracer techniques, and physiological flux bounds. We gauged the performance of the approach by comparing metabolite levels secreted in plasma and urine predicted by the model with those obtained from non-targeted metabolic profiling. In the case of short-term fasting, we correctly accounted for glucose/glycogen metabolism in the liver, using a modeling approach based on the metabolic-network structure coupled only with measured central carbon fluxes. For acetaminophen—a known hepatotoxicant—we further incorporated changes in liver and kidney gene expression into multiple modeling frameworks to study the resultant changes in endogenous metabolism. A comparison of the modeled results with the global metabolic profiling data revealed that, our approach satisfactorily predicted altered plasma metabolite levels as early as 5 h after exposure to 2 g/kg of acetaminophen, and that, after 10 h of exposure the predictions significantly improved. We achieved a 70% correspondence between predicted and measured changes of metabolite levels in plasma and urine. The coupled multi-tissue modeling framework of in vivo metabolism provides both mechanistic insights and a capability to identify plasma and urine metabolites as early markers of toxicant-induced organ damage.