Integration of metabolome data with metabolic networks reveals reporter reactions

Abstract

Interpreting quantitative metabolome data is a difficult task owing to the high connectivity in metabolic networks and inherent interdependency between enzymatic regulation, metabolite levels and fluxes. Here we present a hypothesis-driven algorithm for the integration of such data with metabolic network topology. The algorithm thus enables identification of reporter reactions, which are reactions where there are significant coordinated changes in the level of surrounding metabolites following environmental/genetic perturbations. Applicability of the algorithm is demonstrated by using data from Saccharomyces cerevisiae. The algorithm includes preprocessing of a genome-scale yeast model such that the fraction of measured metabolites within the model is enhanced, and thus it is possible to map significant alterations associated with a perturbation even though a small fraction of the complete metabolome is measured. By combining the results with transcriptome data, we further show that it is possible to infer whether the reactions are hierarchically or metabolically regulated. Hereby, the reported approach represents an attempt to map different layers of regulation within metabolic networks through combination of metabolome and transcriptome data.

Figure 1

Reporter reaction algorithm to identify differential reaction significance by integrating metabolome data with metabolic networks. Quantitative metabolome data obtained from perturbation experiments are interpreted in terms of significance of change, and mapped onto the stoichiometric network, which is represented as bi-partite undirected graph, to identify reporter reactions.

Figure 2

Example pathway structures based on Z-scores of reactions, which demonstrate the metabolomic response of the selected reactions in the reported case studies, namely, the effect of an altered redox metabolism and aerobic/anaerobic growth. The dashed lines correspond to the cutoff of 1.28 (P=0.10). See text for detailed discussion. (A) Glutamate catabolic pathway. (B) Glyoxylate metabolism. (C) Glutamate dehydrogenation reactions. (D) TCA cycle. (E) Oxaloacetate metabolism.

Figure 3

Magnitude of the regulation for the reactions of the metabolic network, ENZSUB3, at the hierarchical and metabolic levels for the effect of VHG fermentation media on laboratory (CEN.PK113-7D) and industrial (RS) strains. Z-scores calculated based on gene expression changes (zGE) and based on changes in the surrounding metabolites (zRE) are shown. Red means a positive Z-score and green means a negative Z-score indicating that the regulation is insignificant. Reactions were color-coded with respect to their Z-scores using z=1.28 (P=0.10) as the cutoff value to decide on the corresponding regulation type. Yellow denotes hierarchical regulation, black metabolical regulation, violet mixed regulation and white statistically insignificant score for both types.