Production of biofuels and biochemicals: in need of an ORACLE

Learning metabolic dynamics from irregular observations by Bidirectional Time-Series State Transfer Network

Modeling microbial metabolic dynamics is important for the rational optimization of both biosynthetic systems and industrial processes to facilitate green and efficient biomanufacturing. Classical approaches utilize explicit equation systems to represent metabolic networks, enabling the quantification of pathway fluxes to identify metabolic bottlenecks. However, these white-box models, despite their diverse applications, have limitations in simulating metabolic dynamics and are intrinsically inaccurate for industrial strains that lack information on network structures and kinetic parameters. On the other hand, black-box models do not rely on prior mechanistic knowledge of strains but are built upon observed time-series trajectories of biosynthetic systems in action. In practice, these observations are typically irregular, with discontinuously observed time points across multiple independent batches, each time point potentially containing missing measurements. Learning from such irregular data remains challenging for existing approaches. To address this issue, we present the Bidirectional Time-Series State Transfer Network (BTSTN) for modeling metabolic dynamics directly from irregular observations. Using evaluation data sets derived from both ideal dynamic systems and a real-world fermentation process, we demonstrate that BTSTN accurately reconstructs dynamic behaviors and predicts future trajectories. This approach exhibits enhanced robustness against missing measurements and noise, as compared to the state-of-the-art methods.IMPORTANCEIndustrial biosynthetic systems often involve strains with unclear genetic backgrounds, posing challenges in modeling their distinct metabolic dynamics. In such scenarios, white-box models, which commonly rely on inferred networks, are thereby of limited applicability and accuracy. In contrast, black-box models, such as statistical models and neural networks, are directly fitted or learned from observed time-series trajectories of biosynthetic systems in action. These methods typically assume regular observations without missing time points or measurements. If the observations are irregular, a pre-processing step becomes necessary to obtain a fully filled data set for subsequent model training, which, at the same time, inevitably introduces errors into the resulting models. BTSTN is a novel approach that natively learns from irregular observations. This distinctive feature makes it a unique addition to the current arsenal of technologies modeling metabolic dynamics.

Bottom-up parameterization of enzyme rate constants: Reconciling inconsistent data

Kinetic models of metabolism are promising platforms for studying complex metabolic systems and designing production strains. Given the availability of enzyme kinetic data from historical experiments and machine learning estimation tools, a straightforward modeling approach is to assemble kinetic data enzyme by enzyme until a desired scale is reached. However, this type of 'bottom up' parameterization of kinetic models has been difficult due to a number of issues including gaps in kinetic parameters, the complexity of enzyme mechanisms, inconsistencies between parameters obtained from different sources, and in vitro-in vivo differences. Here, we present a computational workflow for the robust estimation of kinetic parameters for detailed mass action enzyme models while taking into account parameter uncertainty. The resulting software package, termed MASSef (the Mass Action Stoichiometry Simulation Enzyme Fitting package), can handle standard 'macroscopic' kinetic parameters, including Km, kcat, Ki, Keq, and nh, as well as diverse reaction mechanisms defined in terms of mass action reactions and 'microscopic' rate constants. We provide three enzyme case studies demonstrating that this approach can identify and reconcile inconsistent data either within in vitro experiments or between in vitro and in vivo enzyme function. We further demonstrate how parameterized enzyme modules can be used to assemble pathway-scale kinetic models consistent with in vivo behavior. This work builds on the legacy of knowledge on kinetic behavior of enzymes by enabling robust parameterization of enzyme kinetic models at scale utilizing the abundance of historical literature data and machine learning parameter estimates.

Machine learning predicts system-wide metabolic flux control in cyanobacteria

Metabolic fluxes and their control mechanisms are fundamental in cellular metabolism, offering insights for the study of biological systems and biotechnological applications. However, quantitative and predictive understanding of controlling biochemical reactions in microbial cell factories, especially at the system level, is limited. In this work, we present ARCTICA, a computational framework that integrates constraint-based modelling with machine learning tools to address this challenge. Using the model cyanobacterium Synechocystis sp. PCC 6803 as chassis, we demonstrate that ARCTICA effectively simulates global-scale metabolic flux control. Key findings are that (i) the photosynthetic bioproduction is mainly governed by enzymes within the Calvin-Benson-Bassham (CBB) cycle, rather than by those involve in the biosynthesis of the end-product, (ii) the catalytic capacity of the CBB cycle limits the photosynthetic activity and downstream pathways and (iii) ribulose-1,5-bisphosphate carboxylase/oxygenase (RuBisCO) is a major, but not the most, limiting step within the CBB cycle. Predicted metabolic reactions qualitatively align with prior experimental observations, validating our modelling approach. ARCTICA serves as a valuable pipeline for understanding cellular physiology and predicting rate-limiting steps in genome-scale metabolic networks, and thus provides guidance for bioengineering of cyanobacteria.

KETCHUP: Parameterizing of large-scale kinetic models using multiple datasets with different reference states

Large-scale kinetic models provide the computational means to dynamically link metabolic reaction fluxes to metabolite concentrations and enzyme levels while also conforming to substrate level regulation. However, the development of broadly applicable frameworks for efficiently and robustly parameterizing models remains a challenge. Challenges arise due to both the heterogeneity, paucity, and difficulty in obtaining flux and/or concentration data but also due to the computational difficulties of the underlying parameter identification problem. Both the computational demands for parameterization, degeneracy of obtained parameter solutions and interpretability of results has so far limited widespread adoption of large-scale kinetic models despite their potential. Herein, we introduce the Kinetic Estimation Tool Capturing Heterogeneous Datasets Using Pyomo (KETCHUP), a flexible parameter estimation tool that leverages a primal-dual interior-point algorithm to solve a nonlinear programming (NLP) problem that identifies a set of parameters capable of recapitulating the (non)steady-state fluxes and concentrations in wild-type and perturbed metabolic networks. KETCHUP is benchmarked against previously parameterized large-scale kinetic models demonstrating an at least an order of magnitude faster convergence than the tool K-FIT while at the same time attaining better data fits. This versatile toolbox accepts different kinetic descriptions, metabolic fluxes, enzyme levels and metabolite concentrations, under either steady-state or instationary conditions to enable robust kinetic model construction and parameterization. KETCHUP supports the SBML format and can be accessed at https://github.com/maranasgroup/KETCHUP.

Rational strain design with minimal phenotype perturbation

Devising genetic interventions for desired cellular phenotypes remains challenging regarding time and resources. Kinetic models can accelerate this task by simulating metabolic responses to genetic perturbations. However, exhaustive design evaluations with kinetic models are computationally impractical, especially when targeting multiple enzymes. Here, we introduce a framework for efficiently scouting the design space while respecting cellular physiological requirements. The framework employs mixed-integer linear programming and nonlinear simulations with large-scale nonlinear kinetic models to devise genetic interventions while accounting for the network effects of these perturbations. Importantly, it ensures the engineered strain's robustness by maintaining its phenotype close to that of the reference strain. The framework, applied to improve the anthranilate production in E. coli, devises designs for experimental implementation, including eight previously experimentally validated targets. We expect this framework to play a crucial role in future design-build-test-learn cycles, significantly expediting the strain design compared to exhaustive design enumeration.

From microbiome composition to functional engineering, one step at a time

SUMMARYCommunities of microorganisms (microbiota) are present in all habitats on Earth and are relevant for agriculture, health, and climate. Deciphering the mechanisms that determine microbiota dynamics and functioning within the context of their respective environments or hosts (the microbiomes) is crucially important. However, the sheer taxonomic, metabolic, functional, and spatial complexity of most microbiomes poses substantial challenges to advancing our knowledge of these mechanisms. While nucleic acid sequencing technologies can chart microbiota composition with high precision, we mostly lack information about the functional roles and interactions of each strain present in a given microbiome. This limits our ability to predict microbiome function in natural habitats and, in the case of dysfunction or dysbiosis, to redirect microbiomes onto stable paths. Here, we will discuss a systematic approach (dubbed the N+1/N-1 concept) to enable step-by-step dissection of microbiome assembly and functioning, as well as intervention procedures to introduce or eliminate one particular microbial strain at a time. The N+1/N-1 concept is informed by natural invasion events and selects culturable, genetically accessible microbes with well-annotated genomes to chart their proliferation or decline within defined synthetic and/or complex natural microbiota. This approach enables harnessing classical microbiological and diversity approaches, as well as omics tools and mathematical modeling to decipher the mechanisms underlying N+1/N-1 microbiota outcomes. Application of this concept further provides stepping stones and benchmarks for microbiome structure and function analyses and more complex microbiome intervention strategies.

Simulated Design-Build-Test-Learn Cycles for Consistent Comparison of Machine Learning Methods in Metabolic Engineering

Combinatorial pathway optimization is an important tool in metabolic flux optimization. Simultaneous optimization of a large number of pathway genes often leads to combinatorial explosions. Strain optimization is therefore often performed using iterative design-build-test-learn (DBTL) cycles. The aim of these cycles is to develop a product strain iteratively, every time incorporating learning from the previous cycle. Machine learning methods provide a potentially powerful tool to learn from data and propose new designs for the next DBTL cycle. However, due to the lack of a framework for consistently testing the performance of machine learning methods over multiple DBTL cycles, evaluating the effectiveness of these methods remains a challenge. In this work, we propose a mechanistic kinetic model-based framework to test and optimize machine learning for iterative combinatorial pathway optimization. Using this framework, we show that gradient boosting and random forest models outperform the other tested methods in the low-data regime. We demonstrate that these methods are robust for training set biases and experimental noise. Finally, we introduce an algorithm for recommending new designs using machine learning model predictions. We show that when the number of strains to be built is limited, starting with a large initial DBTL cycle is favorable over building the same number of strains for every cycle.

Control analysis in the identification of key enzymes driving metabolic adaptations: Towards drug target discovery

Metabolic Control Analysis (MCA) marked a turning point in understanding the design principles of metabolic network control by establishing control coefficients as a means to quantify the degree of control that an enzyme exerts on flux or metabolite concentrations. MCA has demonstrated that control of metabolic pathways is distributed among many enzymes rather than depending on a single rate-limiting step. MCA also proved that this distribution depends not only on the stoichiometric structure of the network but also on other kinetic determinants, such as the degree of saturation of the enzyme active site, the distance to thermodynamic equilibrium, and metabolite feedback regulatory loops. Consequently, predicting the alterations that occur during metabolic adaptation in response to strong changes involving a redistribution in such control distribution can be challenging. Here, using the framework provided by MCA, we illustrate how control distribution in a metabolic pathway/network depends on enzyme kinetic determinants and to what extent the redistribution of control affects our predictions on candidate enzymes suitable as targets for small molecule inhibition in the drug discovery process. Our results uncover that kinetic determinants can lead to unexpected control distribution and outcomes that cannot be predicted solely from stoichiometric determinants. We also unveil that the inference of key enzyme-drivers of an observed metabolic adaptation can be dramatically improved using mean control coefficients and ruling out those enzyme activities that are associated with low control coefficients. As the use of constraint-based stoichiometric genome-scale metabolic models (GSMMs) becomes increasingly prevalent for identifying genes/enzymes that could be potential drug targets, we anticipate that incorporating kinetic determinants and ruling out enzymes with low control coefficients into GSMM workflows will facilitate more accurate predictions and reveal novel therapeutic targets.

Omics data for sampling thermodynamically feasible kinetic models

Kinetic models are key to understanding and predicting the dynamic behaviour of metabolic systems. Traditional models require kinetic parameters which are not always available and are often estimated in vitro. Ensemble models overcome this challenge by sampling thermodynamically feasible models around a measured reference point. However, it is unclear if the convenient distributions used to generate the ensemble produce a natural distribution of model parameters and hence if the model predictions are reasonable. In this paper, we produced a detailed kinetic model for the central carbon metabolism of Escherichia coli. The model consists of 82 reactions (including 13 reactions with allosteric regulation) and 79 metabolites. To sample the model, we used metabolomic and fluxomic data from a single steady-state time point for E. coli K-12 MG1655 growing on glucose minimal M9 medium (average sampling time for 1000 models: 11.21 ± 0.14 min). Afterwards, in order to examine whether our sampled models are biologically sound, we calculated Km, Vmax and kcat for the reactions and compared them to previously published values. Finally, we used metabolic control analysis to identify enzymes with high control over the fluxes in the central carbon metabolism. Our analyses demonstrate that our platform samples thermodynamically feasible kinetic models, which are in agreement with previously published experimental results and can be used to investigate metabolic control patterns within cells. This renders it a valuable tool for the study of cellular metabolism and the design of metabolic pathways.

Optimal enzyme utilization suggests that concentrations and thermodynamics determine binding mechanisms and enzyme saturations

Deciphering the metabolic functions of organisms requires understanding the dynamic responses of living cells upon genetic and environmental perturbations, which in turn can be inferred from enzymatic activity. In this work, we investigate the optimal modes of operation for enzymes in terms of the evolutionary pressure driving them toward increased catalytic efficiency. We develop a framework using a mixed-integer formulation to assess the distribution of thermodynamic forces and enzyme states, providing detailed insights into the enzymatic mode of operation. We use this framework to explore Michaelis-Menten and random-ordered multi-substrate mechanisms. We show that optimal enzyme utilization is achieved by unique or alternative operating modes dependent on reactant concentrations. We find that in a bimolecular enzyme reaction, the random mechanism is optimal over any other ordered mechanism under physiological conditions. Our framework can investigate the optimal catalytic properties of complex enzyme mechanisms. It can further guide the directed evolution of enzymes and fill in the knowledge gaps in enzyme kinetics.

Comparative study of two Saccharomyces cerevisiae strains with kinetic models at genome-scale

The parameterization of kinetic models requires measurement of fluxes and/or metabolite levels for a base strain and a few genetic perturbations thereof. Unlike stoichiometric models that are mostly invariant to the specific strain, it remains unclear whether kinetic models constructed for different strains of the same species have similar or significantly different kinetic parameters. This important question underpins the applicability range and prediction limits of kinetic reconstructions. To this end, herein we parameterize two separate large-scale kinetic models using K-FIT with genome-wide coverage corresponding to two distinct strains of Saccharomyces cerevisiae: CEN.PK 113-7D strain (model k-sacce306-CENPK), and growth-deficient BY4741 (isogenic to S288c; model k-sacce306-BY4741). The metabolic network for each model contains 306 reactions, 230 metabolites, and 119 substrate-level regulatory interactions. The two models (for CEN.PK and BY4741) recapitulate, within one standard deviation, 77% and 75% of the fitted dataset fluxes, respectively, determined by ¹³C metabolic flux analysis for wild-type and eight single-gene knockout mutants of each strain. Strain-specific kinetic parameterization results indicate that key enzymes in the TCA cycle, glycolysis, and arginine and proline metabolism drive the metabolic differences between these two strains of S. cerevisiae. Our results suggest that although kinetic models cannot be readily used across strains as stoichiometric models, they can capture species-specific information through the kinetic parameterization process.

A dynamic kinetic model captures cell-free metabolism for improved butanol production

Cell-free systems are useful tools for prototyping metabolic pathways and optimizing the production of various bioproducts. Mechanistically-based kinetic models are uniquely suited to analyze dynamic experimental data collected from cell-free systems and provide vital qualitative insight. However, to date, dynamic kinetic models have not been applied with rigorous biological constraints or trained on adequate experimental data to the degree that they would give high confidence in predictions and broadly demonstrate the potential for widespread use of such kinetic models. In this work, we construct a large-scale dynamic model of cell-free metabolism with the goal of understanding and optimizing butanol production in a cell-free system. Using a combination of parameterization methods, the resultant model captures experimental metabolite measurements across two experimental conditions for nine metabolites at timepoints between 0 and 24 h. We present analysis of the model predictions, provide recommendations for butanol optimization, and identify the aldehyde/alcohol dehydrogenase as the primary bottleneck in butanol production. Sensitivity analysis further reveals the extent to which various parameters are constrained, and our approach for probing valid parameter ranges can be applied to other modeling efforts.

Symbolic kinetic models in python (SKiMpy): intuitive modeling of large-scale biological kinetic models

MOTIVATION

Large-scale kinetic models are an invaluable tool to understand the dynamic and adaptive responses of biological systems. The development and application of these models have been limited by the availability of computational tools to build and analyze large-scale models efficiently. The toolbox presented here provides the means to implement, parameterize and analyze large-scale kinetic models intuitively and efficiently.

RESULTS

We present a Python package (SKiMpy) bridging this gap by implementing an efficient kinetic modeling toolbox for the semiautomatic generation and analysis of large-scale kinetic models for various biological domains such as signaling, gene expression and metabolism. Furthermore, we demonstrate how this toolbox is used to parameterize kinetic models around a steady-state reference efficiently. Finally, we show how SKiMpy can implement multispecies bioreactor simulations to assess biotechnological processes.

AVAILABILITY AND IMPLEMENTATION

The software is available as a Python 3 package on GitHub: https://github.com/EPFL-LCSB/SKiMpy, along with adequate documentation.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Reconstructing Kinetic Models for Dynamical Studies of Metabolism using Generative Adversarial Networks

Kinetic models of metabolism relate metabolic fluxes, metabolite concentrations and enzyme levels through mechanistic relations, rendering them essential for understanding, predicting and optimizing the behaviour of living organisms. However, due to the lack of kinetic data, traditional kinetic modelling often yields only a few or no kinetic models with desirable dynamical properties, making the analysis unreliable and computationally inefficient. We present REKINDLE (Reconstruction of Kinetic Models using Deep Learning), a deep-learning-based framework for efficiently generating kinetic models with dynamic properties matching the ones observed in cells. We showcase REKINDLE's capabilities to navigate through the physiological states of metabolism using small numbers of data with significantly lower computational requirements. The results show that data-driven neural networks assimilate implicit kinetic knowledge and structure of metabolic networks and generate kinetic models with tailored properties and statistical diversity. We anticipate that our framework will advance our understanding of metabolism and accelerate future research in biotechnology and health.

Construction of Multiscale Genome-Scale Metabolic Models: Frameworks and Challenges

Genome-scale metabolic models (GEMs) are effective tools for metabolic engineering and have been widely used to guide cell metabolic regulation. However, the single gene–protein-reaction data type in GEMs limits the understanding of biological complexity. As a result, multiscale models that add constraints or integrate omics data based on GEMs have been developed to more accurately predict phenotype from genotype. This review summarized the recent advances in the development of multiscale GEMs, including multiconstraint, multiomic, and whole-cell models, and outlined machine learning applications in GEM construction. This review focused on the frameworks, toolkits, and algorithms for constructing multiscale GEMs. The challenges and perspectives of multiscale GEM development are also discussed.

Metabolic Engineering and Regulation of Diol Biosynthesis from Renewable Biomass in Escherichia coli

As bulk chemicals, diols have wide applications in many fields, such as clothing, biofuels, food, surfactant and cosmetics. The traditional chemical synthesis of diols consumes numerous non-renewable energy resources and leads to environmental pollution. Green biosynthesis has emerged as an alternative method to produce diols. Escherichia coli as an ideal microbial factory has been engineered to biosynthesize diols from carbon sources. Here, we comprehensively summarized the biosynthetic pathways of diols from renewable biomass in E. coli and discussed the metabolic-engineering strategies that could enhance the production of diols, including the optimization of biosynthetic pathways, improvement of cofactor supplementation, and reprogramming of the metabolic network. We then investigated the dynamic regulation by multiple control modules to balance the growth and production, so as to direct carbon sources for diol production. Finally, we proposed the challenges in the diol-biosynthesis process and suggested some potential methods to improve the diol-producing ability of the host.

Module engineering coupled with omics strategies for enhancing D-pantothenate production in Escherichia coli

Biosynthesis of D-pantothenate has been widely studied as D-pantothenate is one kind of important vitamins used in food and pharmaceuticals. However, the engineered strain for D-pantothenate production was focused solely on the main biosynthetic pathway, while other important factors such as one carbon unit were ignored. Here the systematic modular engineering on different factors coupled with omics analysis were studied in Escherichia coli for efficient D-pantothenate production. Through reinforcing the precursor pool, refactoring the one carbon unit generation pathway, optimization of reducing power and energy supply, the D-pantothenate titer reached 34.12 g/L with the yield at 0.28 g/g glucose under fed-batch fermentation in 5-L bioreactor. With a further comparative transcriptome and metabolomics studies, the addition of citrate was implemented and 45.35 g/L D-pantothenate was accumulated with a yield of 0.31 g/g glucose. The systematic modular engineering coupled with omics studies provide useful strategies for the industrial production of D-pantothenate.

Quantitative metabolic fluxes regulated by trans-omic networks

Cells change their metabolism in response to internal and external conditions by regulating the trans-omic network, which is a global biochemical network with multiple omic layers. Metabolic flux is a direct measure of the activity of a metabolic reaction that provides valuable information for understanding complex trans-omic networks. Over the past decades, techniques to determine metabolic fluxes, including 13C-metabolic flux analysis (13C-MFA), flux balance analysis (FBA), and kinetic modeling, have been developed. Recent studies that acquire quantitative metabolic flux and multi-omic data have greatly advanced the quantitative understanding and prediction of metabolism-centric trans-omic networks. In this review, we present an overview of 13C-MFA, FBA, and kinetic modeling as the main techniques to determine quantitative metabolic fluxes, and discuss their advantages and disadvantages. We also introduce case studies with the aim of understanding complex metabolism-centric trans-omic networks based on the determination of metabolic fluxes.

Multi-omics-based label-free metabolic flux inference reveals obesity-associated dysregulatory mechanisms in liver glucose metabolism

Glucose homeostasis is maintained by modulation of metabolic flux. Enzymes and metabolites regulate the involved metabolic pathways. Dysregulation of glucose homeostasis is a pathological event in obesity. Analyzing metabolic pathways and the mechanisms contributing to obesity-associated dysregulation in vivo is challenging. Here, we introduce OMELET: Omics-Based Metabolic Flux Estimation without Labeling for Extended Trans-omic Analysis. OMELET uses metabolomic, proteomic, and transcriptomic data to identify relative changes in metabolic flux, and to calculate contributions of metabolites, enzymes, and transcripts to the changes in metabolic flux. By evaluating the livers of fasting ob/ob mice, we found that increased metabolic flux through gluconeogenesis resulted primarily from increased transcripts, whereas that through the pyruvate cycle resulted from both increased transcripts and changes in substrates of metabolic enzymes. With OMELET, we identified mechanisms underlying the obesity-associated dysregulation of metabolic flux in the liver.

•

We developed OMELET to infer metabolic flux from label-free multi-omic data

•

Contributions of metabolites, enzymes, and transcripts for flux were inferred

•

Gluconeogenic flux increased in fasting ob/ob mice by increased transcripts

•

Increased pyruvate cycle fluxes were led by increased transcripts and substrates

Bacterial metabolism and pathogenesis intimate intertwining: time for metabolic modelling to come into action

We take a snapshot of the recent understanding of bacterial metabolism and the bacterial-host metabolic interplay during infection, and highlight key outcomes and challenges for the practical implementation of bacterial metabolic modelling computational tools in the pathogenesis field.

Genome Scale Modeling to Study the Metabolic Competition between Cells in the Tumor Microenvironment

The tumor's physiology emerges from the dynamic interplay of numerous cell types, such as cancer cells, immune cells and stromal cells, within the tumor microenvironment. Immune and cancer cells compete for nutrients within the tumor microenvironment, leading to a metabolic battle between these cell populations. Tumor cells can reprogram their metabolism to meet the high demand of building blocks and ATP for proliferation, and to gain an advantage over the action of immune cells. The study of the metabolic reprogramming mechanisms underlying cancer requires the quantification of metabolic fluxes which can be estimated at the genome-scale with constraint-based or kinetic modeling. Constraint-based models use a set of linear constraints to simulate steady-state metabolic fluxes, whereas kinetic models can simulate both the transient behavior and steady-state values of cellular fluxes and concentrations. The integration of cell- or tissue-specific data enables the construction of context-specific models that reflect cell-type- or tissue-specific metabolic properties. While the available modeling frameworks enable limited modeling of the metabolic crosstalk between tumor and immune cells in the tumor stroma, future developments will likely involve new hybrid kinetic/stoichiometric formulations.

Integrating systemic and molecular levels to infer key drivers sustaining metabolic adaptations

Metabolic adaptations to complex perturbations, like the response to pharmacological treatments in multifactorial diseases such as cancer, can be described through measurements of part of the fluxes and concentrations at the systemic level and individual transporter and enzyme activities at the molecular level. In the framework of Metabolic Control Analysis (MCA), ensembles of linear constraints can be built integrating these measurements at both systemic and molecular levels, which are expressed as relative differences or changes produced in the metabolic adaptation. Here, combining MCA with Linear Programming, an efficient computational strategy is developed to infer additional non-measured changes at the molecular level that are required to satisfy these constraints. An application of this strategy is illustrated by using a set of fluxes, concentrations, and differentially expressed genes that characterize the response to cyclin-dependent kinases 4 and 6 inhibition in colon cancer cells. Decreases and increases in transporter and enzyme individual activities required to reprogram the measured changes in fluxes and concentrations are compared with down-regulated and up-regulated metabolic genes to unveil those that are key molecular drivers of the metabolic response.

Deciphering the essential events in the reprogramming of metabolic networks subjected to complex perturbations, including the response to pharmacological treatments in multifactorial diseases like cancer, is crucial for the design of efficient therapies. Yet, tools to infer the molecular drivers sustaining such metabolic responses remain elusive for large metabolic networks. Here we develop an efficient computational strategy that integrates measured changes at systemic and molecular levels and combines metabolic control analysis with linear programming tools to infer key molecular drivers sustaining the metabolic adaptations to complex perturbations, such as an antitumoral drug therapy. The collective behavior is approximated using linear expressions where the adaptation of systemic concentrations and fluxes to a perturbation is described as a function of the molecular reprogramming of transport and enzyme activities. Starting from measured changes in fluxes and concentrations, we identify changes in the reprogramming of transporter and enzyme activities that are required to orchestrate the metabolic adaptation of colon cancer cells to a cell cycle inhibitor.

Active and machine learning-based approaches to rapidly enhance microbial chemical production

In order to make renewable fuels and chemicals from microbes, new methods are required to engineer microbes more intelligently. Computational approaches, to engineer strains for enhanced chemical production typically rely on detailed mechanistic models (e.g., kinetic/stoichiometric models of metabolism)-requiring many experimental datasets for their parameterization-while experimental methods may require screening large mutant libraries to explore the design space for the few mutants with desired behaviors. To address these limitations, we developed an active and machine learning approach (ActiveOpt) to intelligently guide experiments to arrive at an optimal phenotype with minimal measured datasets. ActiveOpt was applied to two separate case studies to evaluate its potential to increase valine yields and neurosporene productivity in Escherichia coli. In both the cases, ActiveOpt identified the best performing strain in fewer experiments than the case studies used. This work demonstrates that machine and active learning approaches have the potential to greatly facilitate metabolic engineering efforts to rapidly achieve its objectives.

Constraint-based metabolic control analysis for rational strain engineering

The advancements in genome editing techniques over the past years have rekindled interest in rational metabolic engineering strategies. While Metabolic Control Analysis (MCA) is a well-established method for quantifying the effects of metabolic engineering interventions on flows in metabolic networks and metabolite concentrations, it does not consider the physiological limitations of the cellular environment and metabolic engineering design constraints. We report here a constraint-based framework, Network Response Analysis (NRA), for rational genetic strain design. NRA is cast as a Mixed-Integer Linear Programming problem that integrates MCA, Thermodynamically-based Flux Analysis (TFA), biologically relevant constraints, as well as genome editing restrictions into a comprehensive platform for identifying metabolic engineering targets. We show that the NRA formulation and its core constraints are equivalent to the ones of Flux Balance Analysis (FBA) and TFA, which allows it to be used for a wide range of optimization criteria and with various physiological constraints. We also show how the parametrization and introduction of biological constraints enhance the NRA formulation compared to the classical MCA approach, and we demonstrate its features and its ability to generate multiple alternative optimal strategies given several user-defined boundaries and objectives. In summary, NRA is a sophisticated alternative to classical MCA for rational metabolic engineering that accommodates the incorporation of physiological data at metabolic flux, metabolite concentration, and enzyme expression levels.

Quantum Mechanical Methods Predict Accurate Thermodynamics of Biochemical Reactions

Thermodynamics plays a crucial role in regulating the metabolic processes in all living organisms. Accurate determination of biochemical and biophysical properties is important to understand, analyze, and synthetically design such metabolic processes for engineered systems. In this work, we extensively performed first-principles quantum mechanical calculations to assess its accuracy in estimating free energy of biochemical reactions and developed automated quantum-chemistry (QC) pipeline (https://appdev.kbase.us/narrative/45710) for the prediction of thermodynamics parameters of biochemical reactions. We benchmark the QC methods based on density functional theory (DFT) against different basis sets, solvation models, pH, and exchange-correlation functionals using the known thermodynamic properties from the NIST database. Our results show that QC calculations when combined with simple calibration yield a mean absolute error in the range of 1.60-2.27 kcal/mol for different exchange-correlation functionals, which is comparable to the error in the experimental measurements. This accuracy over a diverse set of metabolic reactions is unprecedented and near the benchmark chemical accuracy of 1 kcal/mol that is usually desired from DFT calculations.

The effects of model complexity and size on metabolic flux distribution and control: case study in Escherichia coli

Background

Significant efforts have been made in building large-scale kinetic models of cellular metabolism in the past two decades. However, most kinetic models published to date, remain focused around central carbon pathways or are built around ad hoc reduced models without clear justification on their derivation and usage. Systematic algorithms exist for reducing genome-scale metabolic reconstructions to build thermodynamically feasible and consistently reduced stoichiometric models. However, it is important to study how network complexity affects conclusions derived from large-scale kinetic models built around consistently reduced models before we can apply them to study biological systems.

Results

We reduced the iJO1366 Escherichia Coli genome-scale metabolic reconstruction systematically to build three stoichiometric models of different size. Since the reduced models are expansions around the core subsystems for which the reduction was performed, the models are nested. We present a method for scaling up the flux profile and the concentration vector reference steady-states from the smallest model to the larger ones, whilst preserving maximum equivalency. Populations of kinetic models, preserving similarity in kinetic parameters, were built around the reference steady-states and their metabolic sensitivity coefficients (MSCs) were computed. The MSCs were sensitive to the model complexity. We proposed a metric for measuring the sensitivity of MSCs to these structural changes.

Conclusions

We proposed for the first time a workflow for scaling up the size of kinetic models while preserving equivalency between the kinetic models. Using this workflow, we demonstrate that model complexity in terms of networks size has significant impact on sensitivity characteristics of kinetic models. Therefore, it is essential to account for the effects of network complexity when constructing kinetic models. The presented metric for measuring MSC sensitivity to structural changes can guide modelers and experimentalists in improving model quality and guide synthetic biology and metabolic engineering. Our proposed workflow enables the testing of the suitability of a kinetic model for answering certain study-specific questions. We argue that the model-based metabolic design targets that are common across models of different size are of higher confidence, while those that are different could be the objective of investigations for model improvement.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12859-021-04066-y.

Model reduction of genome-scale metabolic models as a basis for targeted kinetic models

Constraint-based, genome-scale metabolic models are an essential tool to guide metabolic engineering. However, they lack the detail and time dimension that kinetic models with enzyme dynamics offer. Model reduction can be used to bridge the gap between the two methods and allow for the integration of kinetic models into the Design-Built-Test-Learn cycle. Here we show that these reduced size models can be representative of the dynamics of the original model and demonstrate the automated generation and parameterisation of such models. Using these minimal models of metabolism could allow for further exploration of dynamic responses in metabolic networks.

Building kinetic models for metabolic engineering

Kinetic formalisms of metabolism link metabolic fluxes to enzyme levels, metabolite concentrations and their allosteric regulatory interactions. Though they require the identification of physiologically relevant values for numerous parameters, kinetic formalisms uniquely establish a mechanistic link across heterogeneous omics datasets and provide an overarching vantage point to effectively inform metabolic engineering strategies. Advances in computational power, gene annotation coverage, and formalism standardization have led to significant progress over the past few years. However, careful interpretation of model predictions, limited metabolic flux datasets, and assessment of parameter sensitivity remain as challenges. In this review we highlight fundamental considerations which influence model quality and prediction, advances in methodologies, and success stories of deploying kinetic models to guide metabolic engineering.

Recent advances in constraint and machine learning-based metabolic modeling by leveraging stoichiometric balances, thermodynamic feasibility and kinetic law formalisms

Understanding the governing principles behind organisms' metabolism and growth underpins their effective deployment as bioproduction chassis. A central objective of metabolic modeling is predicting how metabolism and growth are affected by both external environmental factors and internal genotypic perturbations. The fundamental concepts of reaction stoichiometry, thermodynamics, and mass action kinetics have emerged as the foundational principles of many modeling frameworks designed to describe how and why organisms allocate resources towards both growth and bioproduction. This review focuses on the latest algorithmic advancements that have integrated these foundational principles into increasingly sophisticated quantitative frameworks.

Metabolic Modelling as a Framework for Metabolomics Data Integration and Analysis

Metabolic networks are regulated to ensure the dynamic adaptation of biochemical reaction fluxes to maintain cell homeostasis and optimal metabolic fitness in response to endogenous and exogenous perturbations. To this end, metabolism is tightly controlled by dynamic and intricate regulatory mechanisms involving allostery, enzyme abundance and post-translational modifications. The study of the molecular entities involved in these complex mechanisms has been boosted by the advent of high-throughput technologies. The so-called omics enable the quantification of the different molecular entities at different system layers, connecting the genotype with the phenotype. Therefore, the study of the overall behavior of a metabolic network and the omics data integration and analysis must be approached from a holistic perspective. Due to the close relationship between metabolism and cellular phenotype, metabolic modelling has emerged as a valuable tool to decipher the underlying mechanisms governing cell phenotype. Constraint-based modelling and kinetic modelling are among the most widely used methods to study cell metabolism at different scales, ranging from cells to tissues and organisms. These approaches enable integrating metabolomic data, among others, to enhance model predictive capabilities. In this review, we describe the current state of the art in metabolic modelling and discuss future perspectives and current challenges in the field.

Large-scale kinetic metabolic models of Pseudomonas putida KT2440 for consistent design of metabolic engineering strategies

BACKGROUND

Pseudomonas putida is a promising candidate for the industrial production of biofuels and biochemicals because of its high tolerance to toxic compounds and its ability to grow on a wide variety of substrates. Engineering this organism for improved performances and predicting metabolic responses upon genetic perturbations requires reliable descriptions of its metabolism in the form of stoichiometric and kinetic models.

RESULTS

In this work, we developed kinetic models of P. putida to predict the metabolic phenotypes and design metabolic engineering interventions for the production of biochemicals. The developed kinetic models contain 775 reactions and 245 metabolites. Furthermore, we introduce here a novel set of constraints within thermodynamics-based flux analysis that allow for considering concentrations of metabolites that exist in several compartments as separate entities. We started by a gap-filling and thermodynamic curation of iJN1411, the genome-scale model of P. putida KT2440. We then systematically reduced the curated iJN1411 model, and we created three core stoichiometric models of different complexity that describe the central carbon metabolism of P. putida. Using the medium complexity core model as a scaffold, we generated populations of large-scale kinetic models for two studies. In the first study, the developed kinetic models successfully captured the experimentally observed metabolic responses to several single-gene knockouts of a wild-type strain of P. putida KT2440 growing on glucose. In the second study, we used the developed models to propose metabolic engineering interventions for improved robustness of this organism to the stress condition of increased ATP demand.

CONCLUSIONS

The study demonstrates the potential and predictive capabilities of the kinetic models that allow for rational design and optimization of recombinant P. putida strains for improved production of biofuels and biochemicals. The curated genome-scale model of P. putida together with the developed large-scale stoichiometric and kinetic models represents a significant resource for researchers in industry and academia.

Bayesian inference of metabolic kinetics from genome-scale multiomics data

Modern biological tools generate a wealth of data on metabolite and protein concentrations that can be used to help inform new strain designs. However, learning from these data to predict how a cell will respond to genetic changes, a key need for engineering, remains challenging. A promising technique for leveraging omics measurements in metabolic modeling involves the construction of kinetic descriptions of the enzymatic reactions that occur within a cell. Parameterizing these models from biological data can be computationally difficult, since methods must also quantify the uncertainty in model parameters resulting from the observed data. While the field of Bayesian inference offers a wide range of methods for efficiently estimating distributions in parameter uncertainty, such techniques are poorly suited to traditional kinetic models due to their complex rate laws and resulting nonlinear dynamics. In this paper, we employ linear-logarithmic kinetics to simplify the calculation of steady-state flux distributions and enable efficient sampling and inference methods. We demonstrate that detailed information on the posterior distribution of parameters can be obtained efficiently at a variety of problem scales, including nearly genome-scale kinetic models trained on multiomics datasets. These results allow modern Bayesian machine learning tools to be leveraged in understanding biological data and in developing new, efficient strain designs.

Uncertainty reduction in biochemical kinetic models: Enforcing desired model properties

A persistent obstacle for constructing kinetic models of metabolism is uncertainty in the kinetic properties of enzymes. Currently, available methods for building kinetic models can cope indirectly with uncertainties by integrating data from different biological levels and origins into models. In this study, we use the recently proposed computational approach iSCHRUNK (in Silico Approach to Characterization and Reduction of Uncertainty in the Kinetic Models), which combines Monte Carlo parameter sampling methods and machine learning techniques, in the context of Bayesian inference. Monte Carlo parameter sampling methods allow us to exploit synergies between different data sources and generate a population of kinetic models that are consistent with the available data and physicochemical laws. The machine learning allows us to data-mine the a priori generated kinetic parameters together with the integrated datasets and derive posterior distributions of kinetic parameters consistent with the observed physiology. In this work, we used iSCHRUNK to address a design question: can we identify which are the kinetic parameters and what are their values that give rise to a desired metabolic behavior? Such information is important for a wide variety of studies ranging from biotechnology to medicine. To illustrate the proposed methodology, we performed Metabolic Control Analysis, computed the flux control coefficients of the xylose uptake (XTR), and identified parameters that ensure a rate improvement of XTR in a glucose-xylose co-utilizing S. cerevisiae strain. Our results indicate that only three kinetic parameters need to be accurately characterized to describe the studied physiology, and ultimately to design and control the desired responses of the metabolism. This framework paves the way for a new generation of methods that will systematically integrate the wealth of available omics data and efficiently extract the information necessary for metabolic engineering and synthetic biology decisions.

Kinetic models are the most promising tool for understanding the complex dynamic behavior of living cells. The primary goal of kinetic models is to capture the properties of the metabolic networks as a whole, and thus we need large-scale models for dependable in silico analyses of metabolism. However, uncertainty in kinetic parameters impedes the development of kinetic models, and uncertainty levels increase with the model size. Tools that will address the issues with parameter uncertainty and that will be able to reduce the uncertainty propagation through the system are therefore needed. In this work, we applied a method called iSCHRUNK that combines parameter sampling and machine learning techniques to characterize the uncertainties and uncover intricate relationships between the parameters of kinetic models and the responses of the metabolic network. The proposed method allowed us to identify a small number of parameters that determine the responses in the network regardless of the values of other parameters. As a consequence, in future studies of metabolism, it will be sufficient to explore a reduced kinetic space, and more comprehensive analyses of large-scale and genome-scale metabolic networks will be computationally tractable.

Seeing the wood for the trees: a forest of methods for optimization and omic-network integration in metabolic modelling

Metabolic modelling has entered a mature phase with dozens of methods and software implementations available to the practitioner and the theoretician. It is not easy for a modeller to be able to see the wood (or the forest) for the trees. Driven by this analogy, we here present a 'forest' of principal methods used for constraint-based modelling in systems biology. This provides a tree-based view of methods available to prospective modellers, also available in interactive version at http://modellingmetabolism.net, where it will be kept updated with new methods after the publication of the present manuscript. Our updated classification of existing methods and tools highlights the most promising in the different branches, with the aim to develop a vision of how existing methods could hybridize and become more complex. We then provide the first hands-on tutorial for multi-objective optimization of metabolic models in R. We finally discuss the implementation of multi-view machine learning approaches in poly-omic integration. Throughout this work, we demonstrate the optimization of trade-offs between multiple metabolic objectives, with a focus on omic data integration through machine learning. We anticipate that the combination of a survey, a perspective on multi-view machine learning and a step-by-step R tutorial should be of interest for both the beginner and the advanced user.

Metabolic kinetic modeling provides insight into complex biological questions, but hurdles remain

Metabolic models containing kinetic information can answer unique questions about cellular metabolism that are useful to metabolic engineering. Several kinetic modeling frameworks have recently been developed or improved. In addition, techniques for systematic identification of model structure, including regulatory interactions, have been reported. Each framework has advantages and limitations, which can make it difficult to choose the most appropriate framework. Common limitations are data availability and computational time, especially in large-scale modeling efforts. However, recently developed experimental techniques, parameter identification algorithms, as well as model reduction techniques help alleviate these computational bottlenecks. Opportunities for additional improvements may come from the rich literature in catalysis and chemical networks. In all, kinetic models are positioned to make significant impact in cellular engineering.

Understanding and Engineering Distributed Biochemical Pathways in Microbial Communities

Microbiomes impact nearly every environment on Earth by modulating the molecular composition of the environment. Temporally changing environmental stimuli and spatial organization are major variables shaping the structure and function of microbiomes. The web of interactions among members of these communities and between the organisms and the environment dictates microbiome functions. Microbial interactions are major drivers of microbiomes and are modulated by spatiotemporal parameters. A mechanistic and quantitative understanding of ecological, molecular, and environmental forces shaping microbiomes could inform strategies to control microbiome dynamics and functions. Major challenges for harnessing the potential of microbiomes for diverse applications include the development of predictive modeling frameworks and tools for precise manipulation of microbiome behaviors.

Kinetic models of metabolism that consider alternative steady-state solutions of intracellular fluxes and concentrations

Large-scale kinetic models are used for designing, predicting, and understanding the metabolic responses of living cells. Kinetic models are particularly attractive for the biosynthesis of target molecules in cells as they are typically better than other types of models at capturing the complex cellular biochemistry. Using simpler stoichiometric models as scaffolds, kinetic models are built around a steady-state flux profile and a metabolite concentration vector that are typically determined via optimization. However, as the underlying optimization problem is underdetermined, even after incorporating available experimental omics data, one cannot uniquely determine the operational configuration in terms of metabolic fluxes and metabolite concentrations. As a result, some reactions can operate in either the forward or reverse direction while still agreeing with the observed physiology. Here, we analyze how the underlying uncertainty in intracellular fluxes and concentrations affects predictions of constructed kinetic models and their design in metabolic engineering and systems biology studies. To this end, we integrated the omics data of optimally grown Escherichia coli into a stoichiometric model and constructed populations of non-linear large-scale kinetic models of alternative steady-state solutions consistent with the physiology of the E. coli aerobic metabolism. We performed metabolic control analysis (MCA) on these models, highlighting that MCA-based metabolic engineering decisions are strongly affected by the selected steady state and appear to be more sensitive to concentration values rather than flux values. To incorporate this into future studies, we propose a workflow for moving towards more reliable and robust predictions that are consistent with all alternative steady-state solutions. This workflow can be applied to all kinetic models to improve the consistency and accuracy of their predictions. Additionally, we show that, irrespective of the alternative steady-state solution, increased activity of phosphofructokinase and decreased ATP maintenance requirements would improve cellular growth of optimally grown E. coli.

Discovery and Evaluation of Biosynthetic Pathways for the Production of Five Methyl Ethyl Ketone Precursors

The limited supply of fossil fuels and the establishment of new environmental policies shifted research in industry and academia toward sustainable production of the second generation of biofuels, with methyl ethyl ketone (MEK) being one promising fuel candidate. MEK is a commercially valuable petrochemical with an extensive application as a solvent. However, as of today, a sustainable and economically viable production of MEK has not yet been achieved despite several attempts of introducing biosynthetic pathways in industrial microorganisms. We used BNICE.ch as a retrobiosynthesis tool to discover all novel pathways around MEK. Out of 1325 identified compounds connecting to MEK with one reaction step, we selected 3-oxopentanoate, but-3-en-2-one, but-1-en-2-olate, butylamine, and 2-hydroxy-2-methylbutanenitrile for further study. We reconstructed 3 679 610 novel biosynthetic pathways toward these 5 compounds. We then embedded these pathways into the genome-scale model of E. coli, and a set of 18 622 were found to be the most biologically feasible ones on the basis of thermodynamics and their yields. For each novel reaction in the viable pathways, we proposed the most similar KEGG reactions, with their gene and protein sequences, as candidates for either a direct experimental implementation or as a basis for enzyme engineering. Through pathway similarity analysis we classified the pathways and identified the enzymes and precursors that were indispensable for the production of the target molecules. These retrobiosynthesis studies demonstrate the potential of BNICE.ch for discovery, systematic evaluation, and analysis of novel pathways in synthetic biology and metabolic engineering studies.

Synthesis and techno-economic assessment of microbial-based processes for terpenes production

BACKGROUND

Recent advances in metabolic engineering enable the production of chemicals from sugars through microbial bio-conversion. Terpenes have attracted substantial attention due to their relatively high prices and wide applications in different industries. To this end, we synthesize and assess processes for microbial production of terpenes.

RESULTS

To explain a counterintuitive experimental phenomenon where terpenes such as limonene (normal boiling point 176 °C) are often found to be 100% present in the vapor phase after bio-conversion (operating at only ~ 30 °C), we first analyze the vapor-liquid equilibrium for systems containing terpenes. Then, we propose alternative production configurations, which are further studied, using limonene as an example, in several case studies. Next, we perform economic assessment of the alternative processes and identify the major cost components. Finally, we extend the assessment to account for different process parameters, terpene products, ways to address terpene toxicity (microbial engineering vs. solvent use), and cellulosic biomass as a feedstock. We identify the key cost drivers to be (1) feed glucose concentration (wt%), (2) product yield (% of maximum theoretical yield) and (3) VVM (Volume of air per Volume of broth liquid per Minute, i.e., aeration rate in min^-1). The production of limonene, based on current experimental data, is found to be economically infeasible (production cost ~ 465 $/kg vs. market selling price ~ 7 $/kg), but higher glucose concentration and yield can lower the cost. Among 12 terpenes studied, limonene appears to be the most reasonable short-term target because of its large market size (~ 160 million $/year in the US) and the relatively easier to achieve break-even yield (~ 30%, assuming a 14 wt% feed glucose concentration and 0.1 min^-1 VVM).

CONCLUSIONS

The methods proposed in this work are applicable to a range of terpenes as well as other extracellular insoluble chemicals with density lower than that of water, such as fatty acids. The results provide guidance for future research in metabolic engineering toward terpenes production in terms of setting targets for key design parameters.

Formulation, construction and analysis of kinetic models of metabolism: A review of modelling frameworks

Kinetic models are critical to predict the dynamic behaviour of metabolic networks. Mechanistic kinetic models for large networks remain uncommon due to the difficulty of fitting their parameters. Recent modelling frameworks promise new ways to overcome this obstacle while retaining predictive capabilities. In this review, we present an overview of the relevant mathematical frameworks for kinetic formulation, construction and analysis. Starting with kinetic formalisms, we next review statistical methods for parameter inference, as well as recent computational frameworks applied to the construction and analysis of kinetic models. Finally, we discuss opportunities and limitations hindering the development of larger kinetic reconstructions.

redGEM: Systematic reduction and analysis of genome-scale metabolic reconstructions for development of consistent core metabolic models

Genome-scale metabolic reconstructions have proven to be valuable resources in enhancing our understanding of metabolic networks as they encapsulate all known metabolic capabilities of the organisms from genes to proteins to their functions. However the complexity of these large metabolic networks often hinders their utility in various practical applications. Although reduced models are commonly used for modeling and in integrating experimental data, they are often inconsistent across different studies and laboratories due to different criteria and detail, which can compromise transferability of the findings and also integration of experimental data from different groups. In this study, we have developed a systematic semi-automatic approach to reduce genome-scale models into core models in a consistent and logical manner focusing on the central metabolism or subsystems of interest. The method minimizes the loss of information using an approach that combines graph-based search and optimization methods. The resulting core models are shown to be able to capture key properties of the genome-scale models and preserve consistency in terms of biomass and by-product yields, flux and concentration variability and gene essentiality. The development of these “consistently-reduced” models will help to clarify and facilitate integration of different experimental data to draw new understanding that can be directly extendable to genome-scale models.

Reduced models are used commonly to understand the metabolism of organisms and to integrate experimental data for many different studies such as physiology, fluxomics and metabolomics. Without consistent or clear criteria on how these reduced models are actually developed, it is difficult to ensure that they reflect the detailed knowledge that is kept in genome scale metabolic network models (GEMs). The redGEM algorithm presented here allows us to systematically develop consistently reduced metabolic models from their genome-scale counterparts. We applied redGEM for the construction of a core model for E. coli central carbon metabolism. We constructed the core model irJO1366 based on the latest genome-scale E. coli metabolic reconstruction (iJO1366). irJO1366 contains the central carbon pathways and other immediate pathways that must be connected to them for consistency with the iJO1366. irJO1366 can be used to understand metabolism of the organism and also to provide guidance for metabolic engineering purposes. The algorithm is also designed to be modular so that heterologous reactions or pathways can be appended to the core model akin to a “plug-and-play”, synthetic biology approach. The algorithm is applicable to any compartmentalized or non-compartmentalized GEM.

A design-build-test cycle using modeling and experiments reveals interdependencies between upper glycolysis and xylose uptake in recombinant S. cerevisiae and improves predictive capabilities of large-scale kinetic models

BACKGROUND

Recent advancements in omics measurement technologies have led to an ever-increasing amount of available experimental data that necessitate systems-oriented methodologies for efficient and systematic integration of data into consistent large-scale kinetic models. These models can help us to uncover new insights into cellular physiology and also to assist in the rational design of bioreactor or fermentation processes. Optimization and Risk Analysis of Complex Living Entities (ORACLE) framework for the construction of large-scale kinetic models can be used as guidance for formulating alternative metabolic engineering strategies.

RESULTS

We used ORACLE in a metabolic engineering problem: improvement of the xylose uptake rate during mixed glucose-xylose consumption in a recombinant Saccharomyces cerevisiae strain. Using the data from bioreactor fermentations, we characterized network flux and concentration profiles representing possible physiological states of the analyzed strain. We then identified enzymes that could lead to improved flux through xylose transporters (XTR). For some of the identified enzymes, including hexokinase (HXK), we could not deduce if their control over XTR was positive or negative. We thus performed a follow-up experiment, and we found out that HXK2 deletion improves xylose uptake rate. The data from the performed experiments were then used to prune the kinetic models, and the predictions of the pruned population of kinetic models were in agreement with the experimental data collected on the HXK2-deficient S. cerevisiae strain.

CONCLUSIONS

We present a design-build-test cycle composed of modeling efforts and experiments with a glucose-xylose co-utilizing recombinant S. cerevisiae and its HXK2-deficient mutant that allowed us to uncover interdependencies between upper glycolysis and xylose uptake pathway. Through this cycle, we also obtained kinetic models with improved prediction capabilities. The present study demonstrates the potential of integrated "modeling and experiments" systems biology approaches that can be applied for diverse applications ranging from biotechnology to drug discovery.