Modeling of liver citric acid cycle and gluconeogenesis based on 13C mass isotopomer distribution analysis of intermediates

Abstract

We have developed and implemented a model that can predict the positional isotopomer distribution of various hepatic metabolites labeled with [U-13C3]lactate and/or [U-13C3]pyruvate for given relative flux rates through the citric acid cycle and gluconeogenesis reactions. Our model includes (i) isotopic exchange between α-ketoglutarate and glutamate, (ii) a reversible isocitrate dehydrogenase reaction, (iii) an active ATP-citrate lyase, and (iv) aspartate and malate shuttles with separate cytosolic and mitochondrial pools for oxaloacetate, malate, and fumarate. A parameter estimation routine fit the mass isotopomer distribution of selected metabolites measured by gas chromatography-mass spectrometry to the model predicted distributions. We fit measured mass isotopomer distributions of phosphoenol-pyruvate, citrate, α-ketoglutarate, glutamate, and pyruvate isolated from fasted rat livers perfused with [U-13C3]lactate + [U-13C3]pyruvate. This fitting yielded rates which we express relative to that of pyruvate carboxylase: citric acid cycle represented by the irreversible α-ketoglutarate dehydrogenase = 0.32; citrate synthase = 0.64; reversal of isocitrate dehydrogenase = 0.52; citrate lyase = 0.33, aspartate shuttle = 0.24, and malate shuttle = 0.44. Rates calculated for the cytosolic and mitochondrial fumarate and malate dehydrogenase reactions are subject to uncertainties as indicated by identifiability analyses. Previous forms of our model that did not include pyruvate kinase, exchange of α-ketoglutarate with glutamate, reversibility of isocitrate dehydrogenase, and/or ATP-citrate lyase activity were not as successful at predicting our measured values. This model offers a general tool for studying the regulation of the citric acid cycle and gluconeogenesis and can be readily modified for any 13C-labeled lactate or pyruvate substrate.