M. Combining the FBA model with a high-level dynamic model of

M. Combining the FBA model with a high-level dynamic model of plant metabolism allowed them to predict changes in metabolism over time, including the transition between a biomass-producing sink jasp.12117 state and a fructan-remobilizing source state in the stem late in the plant’s life cycle. The whole-leaf model presented here occupies an intermediate position between prior C4 models, with single mesophyll and bundle sheath cells, and multi-organ whole-plant models such as [81]. It represents the first attempt to model spatial variations in metabolic state within a single organ, allowing the study of developmental transitions in leaf metabolism by incorporating data from more and less differentiated cells at a single point in time, PG-1016548 mechanism of action rather than modeling development dynamically. Other interacting cell models incorporate a priori qualitative differences in the metabolic capabilities of their components (e.g., leaf, stem, and seed, or neurons and astrocytes). In contrast in the work presented here, in order to allow the metabolic differences between any two adjacent points to be purely quantitative, the same metabolic network must be used for all points. This simplifies the process of model creation but implies that meaningful predictions of spatial variation depend entirely on the BX795MedChemExpress BX795 integration of (spatially resolved) experimental data. The ability of the model to capture the experimentally observed shift from sink to source tissue along the developmental gradient based on RNA-seq and enzyme activity measurements shows that this may be done successfully with high-resolution -omics data and careful model construction.Methods Reconstruction processA local copy of CornCyc 4.0 [26] was obtained from the Plant Metabolic Network and a draft metabolic model was created using the MetaFlux module of Pathway Tools 17.0 [51]. ThePLOS ONE | DOI:10.1371/journal.pone.0151722 March 18,17 /Multiscale Metabolic Modeling of C4 Plantsresulting model, including reaction reversibility information, was converted to SBML format and iteratively revised, as described in detail in S1 Appendix, until all desired biomass components could be produced under both heterotrophic and photosynthetic conditions and realistic mitochondrial respiration and photorespiration could operate. An overall biomass reaction was adapted from iRS1563 [22] with minor modifications to components and stoichiometry, as detailed in S1 Appendix. To allow calculations with flexible biomass composition, j.jebo.2013.04.005 individual sink reactions were added for most species participating in the biomass reaction, as well as several relevant species (including chlorophyll) not originally included in the iRS1563 biomass equation, for which synthesis pathways were identified in CornCyc. Core metabolic pathways were assigned appropriately to subcellular compartments (e.g., the TCA cycle and mitochondrial electron transport chain to the mitochondrion; the light reactions of photosynthesis, the Calvin cycle, and some reactions of the C4 cycle to the chloroplast; and some reactions of the photorespiratory pathway to the peroxisome) and the intracellular transport reactions necessary for their operation were added. The model was thoroughly tested for consistency and conservation violations, confirming that no species could be created without net mass input or destroyed without net mass output (except species representing light, which can be consumed to drive futile cycles.) The base metabolic model iEB5204 is provided in SBML format.M. Combining the FBA model with a high-level dynamic model of plant metabolism allowed them to predict changes in metabolism over time, including the transition between a biomass-producing sink jasp.12117 state and a fructan-remobilizing source state in the stem late in the plant’s life cycle. The whole-leaf model presented here occupies an intermediate position between prior C4 models, with single mesophyll and bundle sheath cells, and multi-organ whole-plant models such as [81]. It represents the first attempt to model spatial variations in metabolic state within a single organ, allowing the study of developmental transitions in leaf metabolism by incorporating data from more and less differentiated cells at a single point in time, rather than modeling development dynamically. Other interacting cell models incorporate a priori qualitative differences in the metabolic capabilities of their components (e.g., leaf, stem, and seed, or neurons and astrocytes). In contrast in the work presented here, in order to allow the metabolic differences between any two adjacent points to be purely quantitative, the same metabolic network must be used for all points. This simplifies the process of model creation but implies that meaningful predictions of spatial variation depend entirely on the integration of (spatially resolved) experimental data. The ability of the model to capture the experimentally observed shift from sink to source tissue along the developmental gradient based on RNA-seq and enzyme activity measurements shows that this may be done successfully with high-resolution -omics data and careful model construction.Methods Reconstruction processA local copy of CornCyc 4.0 [26] was obtained from the Plant Metabolic Network and a draft metabolic model was created using the MetaFlux module of Pathway Tools 17.0 [51]. ThePLOS ONE | DOI:10.1371/journal.pone.0151722 March 18,17 /Multiscale Metabolic Modeling of C4 Plantsresulting model, including reaction reversibility information, was converted to SBML format and iteratively revised, as described in detail in S1 Appendix, until all desired biomass components could be produced under both heterotrophic and photosynthetic conditions and realistic mitochondrial respiration and photorespiration could operate. An overall biomass reaction was adapted from iRS1563 [22] with minor modifications to components and stoichiometry, as detailed in S1 Appendix. To allow calculations with flexible biomass composition, j.jebo.2013.04.005 individual sink reactions were added for most species participating in the biomass reaction, as well as several relevant species (including chlorophyll) not originally included in the iRS1563 biomass equation, for which synthesis pathways were identified in CornCyc. Core metabolic pathways were assigned appropriately to subcellular compartments (e.g., the TCA cycle and mitochondrial electron transport chain to the mitochondrion; the light reactions of photosynthesis, the Calvin cycle, and some reactions of the C4 cycle to the chloroplast; and some reactions of the photorespiratory pathway to the peroxisome) and the intracellular transport reactions necessary for their operation were added. The model was thoroughly tested for consistency and conservation violations, confirming that no species could be created without net mass input or destroyed without net mass output (except species representing light, which can be consumed to drive futile cycles.) The base metabolic model iEB5204 is provided in SBML format.