Mapping condition-dependent regulation of metabolism in yeast through genome-scale modeling

Abstract

Background: The genome-scale metabolic model of Saccharomyces cerevisiae, first presented in 2003, was the first genome-scale network reconstruction for a eukaryotic organism. Since then continuous efforts have been made in order to improve and expand the yeast metabolic network.

Results: Here we present iTO977, a comprehensive genome-scale metabolic model that contains more reactions, metabolites and genes than previous models. The model was constructed based on two earlier reconstructions, namely iIN800 and the consensus network, and then improved and expanded using gap-filling methods and by introducing new reactions and pathways based on studies of the literature and databases. The model was shown to perform well both for growth simulations in different media and gene essentiality analysis for single and double knock-outs. Further, the model was used as a scaffold for integrating transcriptomics, and flux data from four different conditions in order to identify transcriptionally controlled reactions, i.e. reactions that change both in flux and transcription between the compared conditions.

Conclusion: We present a new yeast model that represents a comprehensive up-to-date collection of knowledge on yeast metabolism. The model was used for simulating the yeast metabolism under four different growth conditions and experimental data from these four conditions was integrated to the model. The model together with experimental data is a useful tool to identify condition-dependent changes of metabolism between different environmental conditions.

Figure 1

The reconstruction process for iTO977 and its comparison with Yeast 5. (A) Yeast genome-scale metabolic model pedigree. The iTO977 model (red) was reconstructed with the iIN800 model and the consensus network (blue) as starting points. The two starting point reconstructions were merged and further expanded by adding new pathways and reactions from databases and literature. (B) A comparison of the genes included in iTO977 and Yeast 5.

Figure 2

Model validation by comparing in silico prediction of the specific growth rate with experimental data. Growth phenotypes were collected from literature and compared to simulated values for chemostat cultivations at four different conditions, nitrogen limited aerobic (green) and anaerobic (red), carbon limited aerobic (blue) and anaerobic (white).

Figure 3

Model validation by comparing simulation of single and double gene knock-out mutant growth to experimental observations. (A) Single gene knock-out mutants cultivated in minimal media. (B) Single gene knock-out mutants cultivated in rich YPD media. (C) Double gene knock-out mutants cultivated in rich YPD media. TP = true positive, TN = true negative, FP = false positive and FN = false negative.

Figure 4

Transcriptionally controlled reactions (reactions where the change in flux correlate with the change in expression of the involved genes) when comparing carbon limited with nitrogen limited growth and comparing aerobic and anaerobic growth. Red color corresponds to an up-regulation both in flux and expression and blue color corresponds to down-regulation.

Figure 5

Transcriptionally controlled reactions in glycolysis and around the FPP branch point. (A) Three glycolytic reactions are considered to be transcriptionally controlled in all four conditions. The color indicates up- or down- regulation between two conditions, in flux and in expression. Only the reactions that are considered as transcriptionally controlled are shown. White color means that there was no significant change between the two conditions in that particular comparison. The four comparisons that were made was Anaerobic – Aerobic under nitrogen limitation (upper left corner), Anaerobic – Aerobic under carbon limitation (lower left corner), C-limited – N-limited under anaerobic conditions (upper right corner) and C-limited – N-limited under aerobic conditions (lower right corner). (B) Transcriptionally controlled reactions in the ergosterol biosynthesis pathway.