Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Jul 22;7:63.
doi: 10.1186/1752-0509-7-63.

Bridging the Gap Between Gene Expression and Metabolic Phenotype via Kinetic Models

Affiliations
Free PMC article

Bridging the Gap Between Gene Expression and Metabolic Phenotype via Kinetic Models

Francisco G Vital-Lopez et al. BMC Syst Biol. .
Free PMC article

Abstract

Background: Despite the close association between gene expression and metabolism, experimental evidence shows that gene expression levels alone cannot predict metabolic phenotypes, indicating a knowledge gap in our understanding of how these processes are connected. Here, we present a method that integrates transcriptome, fluxome, and metabolome data using kinetic models to create a mechanistic link between gene expression and metabolism.

Results: We developed a modeling framework to construct kinetic models that connect the transcriptional and metabolic responses of a cell to exogenous perturbations. The framework allowed us to avoid extensive experimental characterization, literature mining, and optimization problems by estimating most model parameters directly from fluxome and transcriptome data. We applied the framework to investigate how gene expression changes led to observed phenotypic alterations of Saccharomyces cerevisiae treated with weak organic acids (i.e., acetate, benzoate, propionate, or sorbate) and the histidine synthesis inhibitor 3-aminotriazole under steady-state conditions. We found that the transcriptional response led to alterations in yeast metabolism that mimicked measured metabolic fluxes and concentration changes. Further analyses generated mechanistic insights of how S. cerevisiae responds to these stresses. In particular, these results suggest that S. cerevisiae uses different regulation strategies for responding to these insults: regulation of two reactions accounted for most of the tolerance to the four weak organic acids, whereas the response to 3-aminotriazole was distributed among multiple reactions. Moreover, we observed that the magnitude of the gene expression changes was not directly correlated with their effect on the ability of S. cerevisiae to grow under these treatments. In addition, we identified another potential mechanism of action of 3-aminotriazole associated with the depletion of tetrahydrofolate.

Conclusions: Our simulation results show that the modeling framework provided an accurate mechanistic link between gene expression and cellular metabolism. The proposed method allowed us to integrate transcriptome, fluxome, and metabolome data to determine and interpret important features of the physiological response of yeast to stresses. Importantly, given its flexibility and robustness, our approach can be applied to investigate the transcriptional-metabolic response in other cellular systems of medical and industrial relevance.

Figures

Figure 1
Figure 1
Construction of large-scale kinetic models using commonly available information and data. The method starts with the automatic translation of a metabolic network reconstruction into a generic kinetic model, which is parameterized using the metabolic profile for a reference condition (ref). The kinetic model is parameterized to simulate other conditions using gene expression profiles and tuned using the metabolic profiles for the conditions of interest (cnd). The tuned model can be used to perform different model-based analyses. C denotes a diagonal matrix with elements equal to the absolute metabolite concentrations, c represents the vector of normalized metabolite concentrations and ċ denotes its time derivative, S denotes the stoichiometric matrix of the metabolic network reconstruction, r represents the vector of reaction rates, v denotes the flux distribution, g represents the vector of gene expression ratios, and p represents a vector of other condition-specific model parameters.
Figure 2
Figure 2
Metabolic network of the central carbon metabolism and amino acids synthesis pathways of S. cerevisiae. The network includes glycolysis, the pentose phosphate pathway (PPP), the citric acid cycle, and pathways for the synthesis of biomass precursors (i.e., amino acids, carbohydrates, lipids, and RNA). The figure shows a simplified diagram of the network. The actual network has 75 metabolites and 125 reactions.
Figure 3
Figure 3
Predicted metabolic response of the wild-type culture treated with 3-aminotriazole. (A) Predicted metabolic fluxes plotted against experimental values. (B) Predicted concentration of free amino acids plotted against experimental values. The experimental data were taken from Moxley et al. [25].
Figure 4
Figure 4
Contribution of gene expression changes of individual reactions to 3-aminotriazole (3-AT) treatment tolerance. (A) Definition of the metric used to compare the effects of gene expression changes of individual reactions. The normalized tolerance change (NTC) for reaction i was defined as the average of the changes (Δtol+i and Δtol-i) in the maximum 3-AT inhibition level (khis) tolerated by S. cerevisiae in simulations where only the gene expression data (GED) of reaction i (gi) are considered (No GED+gi) or excluded (GED-gi). The average was normalized by the difference in the maximum khis (Δtol) in simulations with (GED) and without (No GED) gene expression changes for all reactions. In these simulations, ksergly was varied linearly with khis such that when khis=1.0 then ksergly=1.0 and when khis=4.87×10-1 then ksergly=2.27×10-2. The values khis=1.0 and khis=0.0 represent no inhibition and complete inhibition, respectively. (B) Top ten reactions with the larger NTC magnitude. EC denotes the Enzyme Commission number. For lumped reactions (L), only the EC number for the first step is shown. (C) Overall gene expression changes for the ten reactions shown in B.
Figure 5
Figure 5
Antimicrobial effect of weak organic acids (WOAs) and resistance mechanisms of S. cerevisiae. At low extracellular pH, WOAs are mainly in their undissociated form, which can diffuse through the cellular membrane. The WOAs dissociate in the cytosol and the cell responds by upregulating transporter proteins, such as Pma1 and Pdr12, to secrete protons and carboxylate anions (XCOO-), respectively, to avoid toxicity.
Figure 6
Figure 6
Predicted metabolic response of S. cerevisiae to different weak organic acids. Both panels show the sum of squared errors (SSE) of the predicted exchange fluxes and biomass yield normalized using the SSE between the experimental values for the reference and the corresponding treated culture. (A) Response of the treated cultures predicted using the corresponding gene expression data (GED) or assuming no gene expression changes (No GED). The predictions of the treated cultures using gene expression data correspond to the data in Table 3. (B) Predicted response of cultures under the reference condition with the expression level of the untreated culture (No GED) or with the gene expression levels of the treated cultures (GED). For simulations in (B), we set the extracellular glucose and biomass concentrations to the experimental reference values.
Figure 7
Figure 7
Biomass concentration as a function of the weak organic acid (WOA) uptake rate. The curves were constructed by using the model to simulate increasing WOA uptake rates using the experimental gene expression data (GED), assuming no gene expression changes (No GED), and by extrapolating the gene expression changes (i.e., gene expression ratios on a logarithmic scale were multiplied by 2.0). The uptake rate and biomass concentration were normalized using the uptake rate of acetic acid and biomass concentration under the reference condition, respectively. Vertical dashed lines indicate the WOA uptake rate under the treatment conditions.
Figure 8
Figure 8
Contribution of gene expression changes of individual reactions to weak organic acid (WOA) treatment tolerance. (A) Definition of the metric used to compare the effects of gene expression changes of individual reactions. The normalized uptake change (NUC) for reaction i is defined as the average of the changes in the WOA uptake rate (Δwur+i and Δwur-i) that reduced biomass to 5.0% of the reference value in simulations where only the gene expression data (GED) of reaction i (gi) are considered (No GED+gi) or excluded (GED-gi). The average was normalized by the difference in the WOA uptake rate that reduced the biomass to 5.0% of the reference value in simulations with (GED) and without (No GED) gene expression changes for all reactions. (B) Normalized uptake changes for each WOA. We only show the reactions with 10 higher contributions. EC denotes the Enzyme Commission number. The y-axis shows only the EC number for the first step of lumped reactions (L). Note that the two most influential reactions, in bold font, were EC 2.7.1.1 and EC 4.1.1.1, for all cases.

Similar articles

See all similar articles

Cited by 6 articles

See all "Cited by" articles

References

    1. Feist AM, Herrgard MJ, Thiele I, Reed JL, Palsson BO. Reconstruction of biochemical networks in microorganisms. Nat Rev Microbiol. 2009;7:129–143. - PMC - PubMed
    1. Fang X, Wallqvist A, Reifman J. Modeling synergistic drug inhibition of Mycobacterium tuberculosis growth in murine macrophages. Mol Biosyst. 2011;7:2622–2636. doi: 10.1039/c1mb05106g. - DOI - PubMed
    1. Becker SA, Feist AM, Mo ML, Hannum G, Palsson BO, Herrgard MJ. Quantitative prediction of cellular metabolism with constraint-based models: the COBRA Toolbox. Nat Protoc. 2007;2:727–738. doi: 10.1038/nprot.2007.99. - DOI - PubMed
    1. Pharkya P, Burgard AP, Maranas CD. OptStrain: a computational framework for redesign of microbial production systems. Genome Res. 2004;14:2367–2376. doi: 10.1101/gr.2872004. - DOI - PMC - PubMed
    1. Liu L, Agren R, Bordel S, Nielsen J. Use of genome-scale metabolic models for understanding microbial physiology. FEBS Lett. 2010;584:2556–2564. doi: 10.1016/j.febslet.2010.04.052. - DOI - PubMed

MeSH terms

Feedback