Minimization Of Metabolic Adjustment Research Articles

Understanding the adaptive laboratory evolution of multiple stress-resistant yeast strains by genome-scale modeling.

Stress responses triggered by external exposures in adaptive laboratory evolution studies alter the ordinary behavior of cells, and the identification of the differences between the starting and the evolved strains would provide ideal strategies to obtain the desired strains. Metabolic networks are one of the most useful tools to analyze data for this purpose. This study integrates differential expression profiles of multiple Saccharomyces cerevisiae strains that have evolved in eight different stress conditions (ethanol, caffeine, coniferyl aldehyde, iron, nickel, phenylethanol, and silver) and enzyme kinetics into a genome-scale metabolic model of yeast, following a new enhanced method. Flux balance analysis, flux variability analysis, robustness, phenotype phase plane, minimization of metabolic adjustment, survivability, sensitivity analyses, and random sampling are conducted to identify the most common and divergent points within strains. Results were examined both individually and comparatively, and the target reactions, metabolites, and enzymes were identified. Our results showed that the models reconstructed by our methodology were able to simulate experimental conditions where efficient protein allocation was the main goal for survival under stressful conditions, and most of the metabolic changes in the adaptation process mainly arose from the differences in the metabolic reactions of energy maintenance (through coenzyme-A and FAD utilization), cell division (folate requirement of DNA synthesis), and cell wall formation (through sterol and ergosterol biosynthesis).

In silico gene knockout prediction using a hybrid of Bat algorithm and minimization of metabolic adjustment.

Microorganisms commonly produce many high-demand industrial products like fuels, food, vitamins, and other chemicals. Microbial strains are the strains of microorganisms, which can be optimized to improve their technological properties through metabolic engineering. Metabolic engineering is the process of overcoming cellular regulation in order to achieve a desired product or to generate a new product that the host cells do not usually need to produce. The prediction of genetic manipulations such as gene knockout is part of metabolic engineering. Gene knockout can be used to optimize the microbial strains, such as to maximize the production rate of chemicals of interest. Metabolic and genetic engineering is important in producing the chemicals of interest as, without them, the product yields of many microorganisms are normally low. As a result, the aim of this paper is to propose a combination of the Bat algorithm and the minimization of metabolic adjustment (BATMOMA) to predict which genes to knock out in order to increase the succinate and lactate production rates in Escherichia coli (E. coli).

Constraint‐based metabolic targets for the improved production of heterologous compounds across molecular classification

The natural products 6‐deoxyerythronolide B (6dEB), erythromycin D, yersiniabactin (Ybt), and salicylate 2‐O‐β‐d‐glucoside (SAG), representing a range of primary and secondary metabolites generated through heterologous microbial biosynthesis, were analyzed using computational metabolic engineering for the purpose of predicting improved production. Specifically, flux balance analysis allowed for the comprehensive screening of medium components and the determination of single gene deletions that resulted in improved product titers for the target compounds. Outcomes included the identification of amino acids and alternative carbon sources capable of culture medium supplementation for increased cellular production. Separately, Minimization of Metabolic Adjustment and OptForce were used to identify single gene deletion and overexpression targets, respectively, for improvements to the aforementioned biosynthetic schemes. The computational engineering predictions thus provide a starting point for experimental implementation with the goal of improving metabolic carbon flow to the compounds presented in this study, each of which possesses valuable bioactivity. © 2018 American Institute of Chemical Engineers AIChE J, 64: 4208–4217, 2018

PO-267 PHGDH and PSAT confer metabolic vulnerability to IDH2-driven reprogramming in breast cancer

IntroductionReprogrammed energy metabolism is one among the distinguished cancer hallmarks that unravels the dysregulated bioenergetics of cancer cells. Isocitrate dehydrogenase 2 (IDH2), a mitochondrial metabolic enzyme has been reported to be overexpressed in several cancers, yet the mechanisms of wildtype IDH2-induced transformation are still obscure. We previously identified IDH2, as a novel poor prognosis marker of breast cancer progression. In agreement, transcriptomics (TCGA) and proteomics data of breast cancer clinical samples showed similar expression changes of increased wildtype IDH2 from luminal, to triple-negative and to HER-2 positive, which correlated with tumour aggressiveness.Material and methodsWe performed intracellular metabolite tracing with heavy labelled glucose (13C6) and glutamine (13C5) followed by ultra-high performance liquid chromatography separation and mass spectrometric analysis on the Q-Exactive Plus instrument. Genome-scale metabolic modelling analysis was performed to identify the IDH2-dependent metabolic genes using Minimization of Metabolic Adjustment (MOMA) algorithm.Results and discussionsOverexpression of IDH2 increased serine biosynthesis, increased glycolytic rate and oxidative stress resistance. Genome-scale metabolic modelling of breast cancer tumour data identified dependence on serine biosynthesis as one of the metabolic vulnerability of IDH2 overexpressing cells. CRISPR-Cas9 mediated knockout of phosphoglycerate dehydrogenase (PHGDH) and phosphoserine aminotransferase (PSAT) specifically inhibited IDH2 overexpressing cells, even in the presence of exogenous serine and glycine, and reduced the glycolysis rate and glucose uptake. Furthermore, loss of serine synthesis pathway hampered the flux from glycolysis, into the TCA cycle. Moreover the metabolic perturbations of synthetic dosage lethal partners, PHGDH and PSAT sensitised the IDH2 overexpressing cells to oxidative stress.ConclusionThis study reveals WT-IDH2 as a critical metabolic regulator in breast cancer that supports the pro-tumorigenic Warburg effect. These findings may contribute to translate the metabolic dependency to improve targeted therapeutics.

Predicting the accumulation of storage compounds by Rhodococcus jostii RHA1 in the feast-famine growth cycles using genome-scale flux balance analysis.

Feast-famine cycles in biological wastewater resource recovery systems select for bacterial species that accumulate intracellular storage compounds such as poly-β-hydroxybutyrate (PHB), glycogen, and triacylglycerols (TAG). These species survive better the famine phase and resume rapid substrate uptake at the beginning of the feast phase faster than microorganisms unable to accumulate storage. However, ecophysiological conditions favouring the accumulation of either storage compounds remain to be clarified, and predictive capabilities need to be developed to eventually rationally design reactors producing these compounds. Using a genome-scale metabolic modelling approach, the storage metabolism of Rhodococcus jostii RHA1 was investigated for steady-state feast-famine cycles on glucose and acetate as the sole carbon sources. R. jostii RHA1 is capable of accumulating the three storage compounds (PHB, TAG, and glycogen) simultaneously. According to the experimental observations, when glucose was the substrate, feast phase chemical oxygen demand (COD) accumulation was similar for the three storage compounds; when acetate was the substrate, however, PHB accumulation was 3 times higher than TAG accumulation and essentially no glycogen was accumulated. These results were simulated using the genome-scale metabolic model of R. jostii RHA1 (iMT1174) by means of flux balance analysis (FBA) to determine the objective functions capable of predicting these behaviours. Maximization of the growth rate was set as the main objective function, while minimization of total reaction fluxes and minimization of metabolic adjustment (environmental MOMA) were considered as the sub-objective functions. The environmental MOMA sub-objective performed better than the minimization of total reaction fluxes sub-objective function at predicting the mixture of storage compounds accumulated. Additional experiments with 13C-labelled bicarbonate (HCO3−) found that the fluxes through the central metabolism reactions during the feast phases were similar to the ones during the famine phases on acetate due to similarity in the carbon sources in the feast and famine phases (i.e., acetate and poly-β-hydroxybutyrate, respectively); this suggests that the environmental MOMA sub-objective function could be used to analyze successive environmental conditions such as the feast and famine cycles while the metabolically similar carbon sources are taken up by microorganisms.

An integrated computational and experimental study uncovers FUT9 as a metabolic driver of colorectal cancer.

Metabolic alterations play an important role in cancer and yet, few metabolic cancer driver genes are known. Here we perform a combined genomic and metabolic modeling analysis searching for metabolic drivers of colorectal cancer. Our analysis predicts FUT9, which catalyzes the biosynthesis of Ley glycolipids, as a driver of advanced‐stage colon cancer. Experimental testing reveals FUT9's complex dual role; while its knockdown enhances proliferation and migration in monolayers, it suppresses colon cancer cells expansion in tumorspheres and inhibits tumor development in a mouse xenograft models. These results suggest that FUT9's inhibition may attenuate tumor‐initiating cells (TICs) that are known to dominate tumorspheres and early tumor growth, but promote bulk tumor cells. In agreement, we find that FUT9 silencing decreases the expression of the colorectal cancer TIC marker CD44 and the level of the OCT4 transcription factor, which is known to support cancer stemness. Beyond its current application, this work presents a novel genomic and metabolic modeling computational approach that can facilitate the systematic discovery of metabolic driver genes in other types of cancer.

A genome-scale dynamic flux balance analysis model of Streptomyces tsukubaensis NRRL18488 to predict the targets for increasing FK506 production

FK506 is a 23-membered polyketide macrolide with a good immunosuppressant activity for the treatment of autoimmune diseases. However, the low fermentation yield is a bottleneck to limit its further industrialization and clinical applications. To address this issue, a genome-scale dynamic flux balance analysis (GS-DFBA) model was developed to seek for the bottlenecks of FK506 biosynthesis. After validating under different cultural conditions, lots of targets were predicted successfully with help of Minimization of Metabolic Adjustment (MOMA). Specifically, many uncover targets that heretofore had remained hidden in the complex interconnectedness of central metabolism had also shown significant effects on FK506 accumulation. To validate the model predictions, four identified targets (gcdh, tktB, msdh and ask gene) were engineered in Streptomyces tsukubaensis NRRL18488. All of the engineered strains showed a higher FK506 production, compared with the parent strain. Finally, the best strain HT-△gcdh-tktB/msdh/ask with gcdh-deletion and tktB-, msdh- and ask-overexpression could produce 126.61±4.66mg/L FK506, 129.8% higher than that of the wild-type strain (55.1±3.83mg/L). Results demonstrated that the GS-DFBA model could provide an effective approach in engineering S. tsukubaensis NRRL18488 for FK506 overproduction. This strategy developed here could also be extended to titer improvement of other products.

A genome-scale Escherichia coli kinetic metabolic model k-ecoli457 satisfying flux data for multiple mutant strains.

Kinetic models of metabolism at a genome scale that faithfully recapitulate the effect of multiple genetic interventions would be transformative in our ability to reliably design novel overproducing microbial strains. Here, we introduce k-ecoli457, a genome-scale kinetic model of Escherichia coli metabolism that satisfies fluxomic data for wild-type and 25 mutant strains under different substrates and growth conditions. The k-ecoli457 model contains 457 model reactions, 337 metabolites and 295 substrate-level regulatory interactions. Parameterization is carried out using a genetic algorithm by simultaneously imposing all available fluxomic data (about 30 measured fluxes per mutant). The Pearson correlation coefficient between experimental data and predicted product yields for 320 engineered strains spanning 24 product metabolites is 0.84. This is substantially higher than that using flux balance analysis, minimization of metabolic adjustment or maximization of product yield exhibiting systematic errors with correlation coefficients of, respectively, 0.18, 0.37 and 0.47 (k-ecoli457 is available for download at http://www.maranasgroup.com).

Reconstruction and analysis of the genome-scale metabolic model of schizochytrium limacinum SR21 for docosahexaenoic acid production.

BackgroundSchizochytrium limacinum SR21 is a potential industrial strain for docosahexaenoic acid (DHA) production that contains more than 30–40 % DHA among its total fatty acids.MethodsTo resolve the DHA biosynthesis mechanism and improve DHA production at a systematic level, a genomescale metabolic model (GSMM), named iCY1170_DHA, which contains 1769 reactions, 1659 metabolites, and 1170 genes, was reconstructed.ResultsBased on genome annotation results and literature reports, a new DHA synthesis pathway based on a polyketide synthase (PKS) system was detected in S. limacinum. Similarly to conventional fatty acid synthesis, the biosynthesis of DHA via PKS requires abundant acetyl-CoA and NADPH. The in silico addition of malate and citrate led to increases of 24.5 % and 37.1 % in DHA production, respectively. Moreover, based on the results predicted by the model, six amino acids were shown to improve DHA production by experiment. Finally, 30 genes were identified as potential targets for DHA over-production using a Minimization of Metabolic Adjustment algorithm.ConclusionsThe reconstructed GSMM, iCY1170_DHA, could be used to elucidate the mechanism by which DHA is synthesized in S. limacinum and predict the requirements of abundant acetyl-CoA and NADPH for DHA production as well as the enhanced yields achieved via supplementation with six amino acids, malate, and citrate.Electronic supplementary materialThe online version of this article (doi:10.1186/s12864-015-2042-y) contains supplementary material, which is available to authorized users.

Model-driven intracellular redox status modulation for increasing isobutanol production in Escherichia coli.

BackgroundFew strains have been found to produce isobutanol naturally. For building a high performance isobutanol-producing strain, rebalancing redox status of the cell was very crucial through systematic investigation of redox cofactors metabolism. Then, the metabolic model provided a powerful tool for the rational modulation of the redox status.ResultsFirstly, a starting isobutanol-producing E. coli strain LA02 was engineered with only 2.7 g/L isobutanol produced. Then, the genome-scale metabolic modeling was specially carried out for the redox cofactor metabolism of the strain LA02 by combining flux balance analysis and minimization of metabolic adjustment, and the GAPD reaction catalyzed by the glyceraldehyde-3-phosphate dehydrogenase was predicted as the key target for redox status improvement. Under guidance of the metabolic model prediction, a gapN-encoding NADP+ dependent glyceraldehyde-3-phosphate dehydrogenase pathway was constructed and then fine-tuned using five constitutive promoters. The best strain LA09 was obtained with the strongest promoter BBa_J23100. The NADPH/NADP + ratios of strain LA09 reached 0.67 at exponential phase and 0.64 at stationary phase. The redox modulations resulted in the decrease production of ethanol and lactate by 17.5 and 51.7% to 1.32 and 6.08 g/L, respectively. Therefore, the isobutanol titer was increased by 221% to 8.68 g/L.ConclusionsThis research has achieved rational redox status improvement of isobutanol-producing strain under guidance of the prediction and modeling of the genome-scale metabolic model of isobutanol-producing E. coli strain with the aid of synthetic promoters. Therefore, the production of isobutanol was dramatically increased by 2.21-fold from 2.7 to 8.68 g/L. Moreover, the developed model-driven method special for redox cofactor metabolism was of very helpful to the redox status modulation of other bio-products.Electronic supplementary materialThe online version of this article (doi:10.1186/s13068-015-0291-2) contains supplementary material, which is available to authorized users.

Reconstruction and analysis of a genome-scale metabolic model of the oleaginous fungus Mortierella alpina.

BackgroundMortierella alpina is an oleaginous fungus used in the industrial scale production of arachidonic acid (ARA). In order to investigate the metabolic characteristics at a systems level and to explore potential strategies for enhanced lipid production, a genome-scale metabolic model of M. alpina was reconstructed.ResultsThis model included 1106 genes, 1854 reactions and 1732 metabolites. On minimal growth medium, 86 genes were identified as essential, whereas 49 essential genes were identified on yeast extract medium. A series of sequential desaturase and elongase catalysed steps are involved in the synthesis of polyunsaturated fatty acids (PUFAs) from acetyl-CoA precursors, with concomitant NADPH consumption, and these steps were investigated in this study. Oxygen is known to affect the degree of unsaturation of PUFAs, and robustness analysis determined that an oxygen uptake rate of 2.0 mmol gDW−1 h−1 was optimal for ARA accumulation. The flux of 53 reactions involving NADPH was significantly altered at different ARA levels. Of these, malic enzyme (ME) was confirmed as a key component in ARA production and NADPH generation. When using minimization of metabolic adjustment, a knock-out of ME led to a 38.28% decrease in ARA production.ConclusionsThe simulation results confirmed the model as a useful tool for future research on the metabolism of PUFAs.Electronic supplementary materialThe online version of this article (doi:10.1186/s12918-014-0137-8) contains supplementary material, which is available to authorized users.

Designing metabolic engineering strategies with genome-scale metabolic flux modeling

Designing metabolic engineering strategies with genome-scale metabolic flux modeling Jiun Y Yen,1,2 Imen Tanniche,1 Amanda K Fisher,1–3 Glenda E Gillaspy,2 David R Bevan,2,3 Ryan S Senger,1 1Department of Biological Systems Engineering, 2Department of Biochemistry, 3Genomics, Bioinformatics, and Computational Biology Interdisciplinary Program, Virginia Tech, Blacksburg, VA, USA Abstract: New in silico tools that make use of genome-scale metabolic flux modeling are improving the design of metabolic engineering strategies. This review highlights the latest developments in this area, explains the interface between these in silico tools and the experimental implementation tools of metabolic engineers, and provides a way forward so that in silico predictions can better mimic reality and more experimental methods can be considered in simulation studies. The several methodologies for solving genome-scale models (eg, flux balance analysis [FBA], parsimonious FBA, flux variability analysis, and minimization of metabolic adjustment) all have unique advantages and applications. There are two basic approaches to designing metabolic engineering strategies in silico, and both have demonstrated success in the literature. The first involves: 1) making a genetic manipulation in a model; 2) testing for improved performance through simulation; and 3) iterating the process. The second approach has been used in more recently designed in silico tools and involves: 1) comparing metabolic flux profiles of a wild-type and ideally engineered state and 2) designing engineering strategies based on the differences in these flux profiles. Improvements in genome-scale modeling are anticipated in areas such as the inclusion of all relevant cellular machinery, the ability to understand and anticipate the results of combinatorial enrichment experiments, and constructing dynamic and flexible biomass equations that can respond to environmental and genetic manipulations. Keywords: genome-scale modeling, genome-scale modeling, flux balance analysis, flux variability analysis, minimization of metabolic adjustment, metabolic bottleneck, pathway optimization

A hybrid of ant colony optimization and minimization of metabolic adjustment to improve the production of succinic acid in Escherichia coli

This paper presents a study on gene knockout strategies to identify candidate genes to be knocked out for improving the production of succinic acid in Escherichia coli. Succinic acid is widely used as a precursor for many chemicals, for example production of antibiotics, therapeutic proteins and food. However, the chemical syntheses of succinic acid using the traditional methods usually result in the production that is far below their theoretical maximums. In silico gene knockout strategies are commonly implemented to delete the gene in E. coli to overcome this problem. In this paper, a hybrid of Ant Colony Optimization (ACO) and Minimization of Metabolic Adjustment (MoMA) is proposed to identify gene knockout strategies to improve the production of succinic acid in E. coli. As a result, the hybrid algorithm generated a list of knockout genes, succinic acid production rate and growth rate for E. coli after gene knockout. The results of the hybrid algorithm were compared with the previous methods, OptKnock and MOMAKnock. It was found that the hybrid algorithm performed better than OptKnock and MOMAKnock in terms of the production rate. The information from the results produced from the hybrid algorithm can be used in wet laboratory experiments to increase the production of succinic acid in E. coli.

Metabolic modeling for multi-objective optimization of ethanol production in a Synechocystis mutant

Cyanobacteria have potential to produce drop-in bio-fuels such as ethanol via photoautotrophic metabolism. Although model cyanobacterial strains have been engineered to produce such products, systematic metabolic engineering studies to identify optimal strains for the same have not been performed. In this work, we identify optimal ethanol producing mutants corresponding to appropriate gene deletions that result in a suitable redirection in the carbon flux. In particular, we systematically simulate exhaustive single and double gene deletions considering a genome scale metabolic model of a mutant strain of the unicellular cyanobacterium Synechocystis species strain PCC 6803. Various optimizationbased metabolic modeling techniques, such as flux balance analysis(FBA), method of minimization of metabolic adjustment (MOMA) and regulatory on/off minimization (ROOM) were used for this analysis. For single gene deletion MOMA simulations, the Pareto front with biomass and ethanol fluxes as the two objectives to be maximized was obtained and analyzed. Points on the Pareto front represent maximal utilization of resources constrained by substrate uptake thereby representing an optimal trade-off between the two fluxes. Pareto analysis was also performed for double gene deletion MOMA and single and double gene deletion ROOM simulations. Based on these analyses, two mutants, with combined gene deletions in ethanol and purine metabolism pathways, were identified as promising candidates for ethanol production. The relevant genes were adk, pta and ackA. An ethanol productivity of approximately 0.15mmol/(gDWh) was predicted for these mutants which appears to be reasonable based on experimentally reported values in literature for other strains.

Computational evaluation of Synechococcus sp. PCC 7002 metabolism for chemical production

Cyanobacteria are ideal metabolic engineering platforms for carbon-neutral biotechnology because they directly convert CO2 to a range of valuable products. In this study, we present a computational assessment of biochemical production in Synechococcus sp. PCC 7002 (Synechococcus 7002), a fast growing cyanobacterium whose genome has been sequenced, and for which genetic modification methods have been developed. We evaluated the maximum theoretical yields (mol product per mol CO2 or mol photon) of producing various chemicals under photoautotrophic and dark conditions using a genome-scale metabolic model of Synechococcus 7002. We found that the yields were lower under dark conditions, compared to photoautotrophic conditions, due to the limited amount of energy and reductant generated from glycogen. We also examined the effects of photon and CO2 limitations on chemical production under photoautotrophic conditions. In addition, using various computational methods such as minimization of metabolic adjustment (MOMA), relative metabolic change (RELATCH), and OptORF, we identified gene-knockout mutants that are predicted to improve chemical production under photoautotrophic and/or dark anoxic conditions. These computational results are useful for metabolic engineering of cyanobacteria to synthesize value-added products.

Adaptive bi-level programming for optimal gene knockouts for targeted overproduction under phenotypic constraints

BackgroundOptimization procedures to identify gene knockouts for targeted biochemical overproduction have been widely in use in modern metabolic engineering. Flux balance analysis (FBA) framework has provided conceptual simplifications for genome-scale dynamic analysis at steady states. Based on FBA, many current optimization methods for targeted bio-productions have been developed under the maximum cell growth assumption. The optimization problem to derive gene knockout strategies recently has been formulated as a bi-level programming problem in OptKnock for maximum targeted bio-productions with maximum growth rates. However, it has been shown that knockout mutants in fact reach the steady states with the minimization of metabolic adjustment (MOMA) from the corresponding wild-type strains instead of having maximal growth rates after genetic or metabolic intervention. In this work, we propose a new bi-level computational framework--MOMAKnock--which can derive robust knockout strategies under the MOMA flux distribution approximation.MethodsIn this new bi-level optimization framework, we aim to maximize the production of targeted chemicals by identifying candidate knockout genes or reactions under phenotypic constraints approximated by the MOMA assumption. Hence, the targeted chemical production is the primary objective of MOMAKnock while the MOMA assumption is formulated as the inner problem of constraining the knockout metabolic flux to be as close as possible to the steady-state phenotypes of wide-type strains. As this new inner problem becomes a quadratic programming problem, a novel adaptive piecewise linearization algorithm is developed in this paper to obtain the exact optimal solution to this new bi-level integer quadratic programming problem for MOMAKnock.ResultsOur new MOMAKnock model and the adaptive piecewise linearization solution algorithm are tested with a small E. coli core metabolic network and a large-scale iAF1260 E. coli metabolic network. The derived knockout strategies are compared with those from OptKnock. Our preliminary experimental results show that MOMAKnock can provide improved targeted productions with more robust knockout strategies.

Comparison of Resilience against Genetic Perturbations in Microbial Metabolism

In metabolic networks, microorganisms are able to survive genetic and environmental perturbations by redirecting the flux from one pathway to another. Hence, it is important to understand how design principles of metabolic networks confer this resilience to the organism. We used Flux Balance Analysis (FBA) and Minimization of Metabolic Adjustment (MOMA) in order to map the lethal gene deletions of two genome scale models of microbial metabolism - E. coli iAF1260 under aerobic growth and Synechocystis PCC6803 under heterotrophic and autotrophic growth - onto each other. Remarkably more than 50% of cyanobacterial lethal knockouts were non-lethal in bacteria. Comparisons of the pathways involved identifying some key subsystems of the bacterial metabolism and the electron flow machinery of cyanobacteria as being primarily responsible for conferring this resilience. We also used FBA to study epistasis in the cyanobacterial network, i.e. aggravating or buffering co-operativity in double gene deletions as compared to corresponding pair of single deletions. In accordance with previous results, heterotrophic and aerobic bacterial metabolism had both aggravating and buffering interactions between genes. Surprisingly however the autotrophic environment displayed a dominance of aggravating interactions with monochromatic interactions between subsystems. We also found that epistasis in an autotrophic organism was highly sensitive to light conditions, and under high light all interactions were aggravating. Our results suggest that the cyanobacterial metabolism is less resilient against genetic perturbations as compared to E. Coli, possibly due to greater specialization under autotrophic conditions.

Metabolic analyses elucidate non-trivial gene targets for amplifying dihydroartemisinic acid production in yeast.

Synthetic biology enables metabolic engineering of industrial microbes to synthesize value-added molecules. In this, a major challenge is the efficient redirection of carbon to the desired metabolic pathways. Pinpointing strategies toward this goal requires an in-depth investigation of the metabolic landscape of the organism, particularly primary metabolism, to identify precursor and cofactor availability for the target compound. The potent antimalarial therapeutic artemisinin and its precursors are promising candidate molecules for production in microbial hosts. Recent advances have demonstrated the production of artemisinin precursors in engineered yeast strains as an alternative to extraction from plants. We report the application of in silico and in vivo metabolic pathway analyses to identify metabolic engineering targets to improve the yield of the direct artemisinin precursor dihydroartemisinic acid (DHA) in yeast. First, in silico extreme pathway (ExPa) analysis identified NADPH-malic enzyme and the oxidative pentose phosphate pathway (PPP) as mechanisms to meet NADPH demand for DHA synthesis. Next, we compared key DHA-synthesizing ExPas to the metabolic flux distributions obtained from in vivo 13C metabolic flux analysis of a DHA-synthesizing strain. This comparison revealed that knocking out ethanol synthesis and overexpressing glucose-6-phosphate dehydrogenase in the oxidative PPP (gene YNL241C) or the NADPH-malic enzyme ME2 (YKL029C) are vital steps toward overproducing DHA. Finally, we employed in silico flux balance analysis and minimization of metabolic adjustment on a yeast genome-scale model to identify gene knockouts for improving DHA yields. The best strategy involved knockout of an oxaloacetate transporter (YKL120W) and an aspartate aminotransferase (YKL106W), and was predicted to improve DHA yields by 70-fold. Collectively, our work elucidates multiple non-trivial metabolic engineering strategies for improving DHA yield in yeast.

Genome-Scale Reconstruction and Analysis of the Metabolic Network in the Hyperthermophilic Archaeon Sulfolobus Solfataricus

We describe the reconstruction of a genome-scale metabolic model of the crenarchaeon Sulfolobus solfataricus, a hyperthermoacidophilic microorganism. It grows in terrestrial volcanic hot springs with growth occurring at pH 2–4 (optimum 3.5) and a temperature of 75–80°C (optimum 80°C). The genome of Sulfolobus solfataricus P2 contains 2,992,245 bp on a single circular chromosome and encodes 2,977 proteins and a number of RNAs. The network comprises 718 metabolic and 58 transport/exchange reactions and 705 unique metabolites, based on the annotated genome and available biochemical data. Using the model in conjunction with constraint-based methods, we simulated the metabolic fluxes induced by different environmental and genetic conditions. The predictions were compared to experimental measurements and phenotypes of S. solfataricus. Furthermore, the performance of the network for 35 different carbon sources known for S. solfataricus from the literature was simulated. Comparing the growth on different carbon sources revealed that glycerol is the carbon source with the highest biomass flux per imported carbon atom (75% higher than glucose). Experimental data was also used to fit the model to phenotypic observations. In addition to the commonly known heterotrophic growth of S. solfataricus, the crenarchaeon is also able to grow autotrophically using the hydroxypropionate-hydroxybutyrate cycle for bicarbonate fixation. We integrated this pathway into our model and compared bicarbonate fixation with growth on glucose as sole carbon source. Finally, we tested the robustness of the metabolism with respect to gene deletions using the method of Minimization of Metabolic Adjustment (MOMA), which predicted that 18% of all possible single gene deletions would be lethal for the organism.

Prediction of dynamic behavior of mutant strains from limited wild-type data

Metabolic engineering is the field of introducing genetic changes in organisms so as to modify their function towards synthesizing new products of high impact to society. However, engineered cells frequently have impaired growth rates thus seriously limiting the rate at which such products are made. The problem is attributable to inadequate understanding of how a metabolic network functions in a dynamic sense. Predictions of mutant strain behavior in the past have been based on steady state theories such as flux balance analysis (FBA), minimization of metabolic adjustment (MOMA), and regulatory on/off minimization (ROOM). Such predictions are restricted to product yields and cannot address productivity, which is of focal interest to applications. We demonstrate that our framework (Song and Ramkrishna, 2010; Song and Ramkrishna, 2011), based on a “cybernetic” view of metabolic systems, makes predictions of the dynamic behavior of mutant strains of Escherichia coli from a limited amount of data obtained from the wild-type. Dynamic frameworks must necessarily address the issue of metabolic regulation, which the cybernetic approach does by postulating that metabolism is an optimal dynamic response of the organism to the environment in driving reactions towards ensuring survival. The predictions made in this paper are without parallel in the literature and lay the foundation for rational metabolic engineering.