Predicting stress response and improved protein overproduction in Bacillus subtilis

Genome-scale metabolic models (GEMs) are widely used in systems biology to investigate metabolism and predict perturbation responses. Automatic GEM reconstruction tools generate GEMs with different properties and predictive capacities for the same organism. Since different models can excel at different tasks, combining them can increase metabolic network certainty and enhance model performance. Here, we introduce GEMsembler, a Python package designed to compare cross-tool GEMs, track the origin of model features, and build consensus models containing any subset of the input models. GEMsembler provides comprehensive analysis functionality, including identification and visualization of biosynthesis pathways, growth assessment, and an agreement-based curation workflow. GEMsembler-curated consensus models built from four Lactiplantibacillus plantarum and Escherichia coli automatically reconstructed models outperform the gold-standard models in auxotrophy and gene essentiality predictions. Optimizing gene-protein-reaction (GPR) combinations from consensus models improves gene essentiality predictions, even in the manually curated gold-standard models. GEMsembler explains model performance by highlighting relevant metabolic pathways and GPR alternatives, informing experiments to resolve model uncertainty. Thus, GEMsembler facilitates building more accurate and biologically informed metabolic models for systems biology applications.IMPORTANCEGenome-scale metabolic models (GEMs) capture our knowledge of cellular metabolism as encoded in the genome, enabling us to describe and predict how cells function under different conditions. While several automated tools can generate these models directly from genome data, the resulting models often contain gaps and uncertainties, highlighting areas where our metabolic knowledge is incomplete. Here, we introduce a new tool called GEMsembler, which integrates GEMs constructed by different methods, evaluate model uncertainty, and build consensus models, harnessing the unique features of each approach. These consensus models more accurately reflect experimentally observed metabolic traits, such as nutrient requirements and condition-specific gene essentiality. GEMsembler facilitates comprehensive analysis of model structure and function, helping to pinpoint knowledge gaps and prioritize experiments to address them. By synthesizing information from diverse sources, GEMsembler accelerates the development of more reliable and biologically meaningful models, advancing research in metabolic engineering, pathogen biology, and microbial community studies.

BackgroundClostridium difficile is a bacterium which can infect various animal species, including humans. Infection with this bacterium is a leading healthcare-associated illness. A better understanding of this organism and the relationship between its genotype and phenotype is essential to the search for an effective treatment. Genome-scale metabolic models contain all known biochemical reactions of a microorganism and can be used to investigate this relationship.ResultsWe present icdf834, an updated metabolic network of C. difficile that builds on iMLTC806cdf and features 1227 reactions, 834 genes, and 807 metabolites. We used this metabolic network to reconstruct the metabolic landscape of this bacterium. The standard metabolic model cannot account for changes in the bacterial metabolism in response to different environmental conditions. To account for this limitation, we also integrated transcriptomic data, which details the gene expression of the bacterium in a wide array of environments. Importantly, to bridge the gap between gene expression levels and protein abundance, we accounted for the synonymous codon usage bias of the bacterium in the model. To our knowledge, this is the first time codon usage has been quantified and integrated into a metabolic model. The metabolic fluxes were defined as a function of protein abundance. To determine potential therapeutic targets using the model, we conducted gene essentiality and metabolic pathway sensitivity analyses and calculated flux control coefficients. We obtained 92.3% accuracy in predicting gene essentiality when compared to experimental data for C. difficile R20291 (ribotype 027) homologs. We validated our context-specific metabolic models using sensitivity and robustness analyses and compared model predictions with literature on C. difficile. The model predicts interesting facets of the bacterium’s metabolism, such as changes in the bacterium’s growth in response to different environmental conditions.ConclusionsAfter an extensive validation process, we used icdf834 to obtain state-of-the-art predictions of therapeutic targets for C. difficile. We show how context-specific metabolic models augmented with codon usage information can be a beneficial resource for better understanding C. difficile and for identifying novel therapeutic targets. We remark that our approach can be applied to investigate and treat against other pathogens.

Metabolic modeling provides the means to define metabolic processes at a systems level; however, genome-scale metabolic models often remain incomplete in their description of metabolic networks and may include reactions that are experimentally unverified. This shortcoming is exacerbated in reconstructed models of newly isolated algal species, as there may be little to no biochemical evidence available for the metabolism of such isolates. The phenotype microarray (PM) technology (Biolog, Hayward, CA, USA) provides an efficient, high-throughput method to functionally define cellular metabolic activities in response to a large array of entry metabolites. The platform can experimentally verify many of the unverified reactions in a network model as well as identify missing or new reactions in the reconstructed metabolic model. The PM technology has been used for metabolic phenotyping of non-photosynthetic bacteria and fungi, but it has not been reported for the phenotyping of microalgae. Here, we introduce the use of PM assays in a systematic way to the study of microalgae, applying it specifically to the green microalgal model species Chlamydomonas reinhardtii. The results obtained in this study validate a number of existing annotated metabolic reactions and identify a number of novel and unexpected metabolites. The obtained information was used to expand and refine the existing COBRA-based C. reinhardtii metabolic network model iRC1080. Over 254 reactions were added to the network, and the effects of these additions on flux distribution within the network are described. The novel reactions include the support of metabolism by a number of d-amino acids, l-dipeptides, and l-tripeptides as nitrogen sources, as well as support of cellular respiration by cysteamine-S-phosphate as a phosphorus source. The protocol developed here can be used as a foundation to functionally profile other microalgae such as known microalgae mutants and novel isolates.

Sphingopyxis granuli strain TFA is an α-proteobacterium that belongs to the sphingomonads, a group of bacteria well-known for its degradative capabilities and oligotrophic metabolism. Strain TFA is the only bacterium in which the mineralisation of the aromatic pollutant tetralin has been completely characterized at biochemical, genetic, and regulatory levels and the first Sphingopyxis characterised as facultative anaerobe. Here we report additional metabolic features of this α-proteobacterium using metabolic modelling and the functional integration of genomic and transcriptomic data. The genome-scale metabolic model (GEM) of strain TFA, which has been manually curated, includes information on 743 genes, 1114 metabolites and 1397 reactions. This represents the largest metabolic model for a member of the Sphingomonadales order thus far. The predictive potential of this model was validated against experimentally calculated growth rates on different carbon sources and under different growth conditions, including both aerobic and anaerobic metabolisms. Moreover, new carbon and nitrogen sources were predicted and experimentally validated. The constructed metabolic model was used as a platform for the incorporation of transcriptomic data, generating a more robust and accurate model. In silico flux analysis under different metabolic scenarios highlighted the key role of the glyoxylate cycle in the central metabolism of strain TFA.

Here, we describe SR7, a dual-function antisense RNA encoded on the Bacillus subtilis chromosome. This RNA was earlier described as SigB-dependent regulatory RNA S1136 and reported to reduce the amount of the small ribosomal subunit under ethanol stress. We found that the 5ʹ portion of SR7 encodes a small protein composed of 39 amino acids which we designated SR7P. It is translated from a 185 nt SigB-dependent mRNA under five different stress conditions and a longer SigB-independent RNA constitutively. About three-fold higher amounts of SR7P were detected in B. subtilis cells exposed to salt, ethanol, acid or heat stress. Co-elution experiments with SR7PC-FLAG and Far-Western blotting demonstrated that SR7P interacts with the glycolytic enzyme enolase. Enolase is a scaffolding component of the B. subtilis degradosome where it interacts with RNase Y and phosphofructokinase PfkA. We found that SR7P increases the amount of RNase Y bound to enolase without affecting PfkA. RNA does not bridge the SR7P-enolase-RNase Y interaction. In vitro-degradation assays with the known RNase Y substrates yitJ and rpsO mRNA revealed enhanced enzymatic activity of enolase-bound RNase Y in the presence of SR7P. Northern blots showed a major effect of enolase and a minor effect of SR7P on the half-life of rpsO mRNA indicating a fine-tuning role of SR7P in RNA degradation.

The genome-scale metabolic model (GEM), a mathematical representation of whole-cell metabolism, enables in silico high-throughput simulations of metabolic flux distribution under different conditions. As a mechanistic and quantitative framework at the system level, GEMs have been applied to investigate microbial interactions and provide guidelines for the rational design of synthetic microbial communities that can perform the desired functions and provide novel strategies for bioremediation. Here, we review different types of microbial interactions and their research progress in environmental remediation. The detailed process of GEM reconstruction and the current main automatic reconstruction programs as well as knowledge databases are summarized. Exciting recent application studies combining metabolic models, omics, and machine learning approaches are also presented. The combination of these methods improves the predictive ability and broadens the applications of GEMs. However, extensive work is still needed to deeply understand and arbitrarily use GEMs to describe the metabolic interactions of microbial communities and apply them in pollutant biodegradation. Finally, an in-depth discussion of the current challenges and limitations of metabolic models provides us with an outlook for their future development in environmental science, especially synthetic microbiology.

Generating condition-specific metabolic models by mapping gene expression data to genome-scale metabolic models (GEMs) is a routine approach to elucidate disease mechanisms from a metabolic perspective. On the other hand, integrating variants that perturb enzyme functionality from the same RNA-seq data may enhance GEM accuracy, offering insights into genome-wide metabolic pathology. Our study pioneers the extraction of both transcriptomic and genomic data from the same RNA-seq data to reconstruct personalized metabolic models. We map genes with significantly higher load of pathogenic variants in Alzheimer’s disease (AD) onto a human GEM together with the gene expression data. Comparative analysis of the resulting personalized patient metabolic models with the control models shows enhanced accuracy in detecting AD-associated metabolic pathways compared to the case where only expression data is mapped on the GEM. Besides, several otherwise would-be missed pathways are annotated in AD by considering the effect of genomic variants.

The analysis of the dynamic behaviour of genome-scale models of metabolism (GEMs) currently presents considerable challenges because of the difficulties of simulating such large and complex networks. Bacterial GEMs can comprise about 5000 reactions and metabolites, and encode a huge variety of growth conditions; such models cannot be used without sophisticated tool support. This article is intended to aid modellers, both specialist and non-specialist in computerized methods, to identify and apply a suitable combination of tools for the dynamic behaviour analysis of large-scale metabolic designs. We describe a methodology and related workflow based on publicly available tools to profile and analyse whole-genome-scale biochemical models. We use an efficient approximative stochastic simulation method to overcome problems associated with the dynamic simulation of GEMs. In addition, we apply simulative model checking using temporal logic property libraries, clustering and data analysis, over time series of reaction rates and metabolite concentrations. We extend this to consider the evolution of reaction-oriented properties of subnets over time, including dead subnets and functional subsystems. This enables the generation of abstract views of the behaviour of these models, which can be large—up to whole genome in size—and therefore impractical to analyse informally by eye. We demonstrate our methodology by applying it to a reduced model of the whole-genome metabolism of Escherichia coli K-12 under different growth conditions. The overall context of our work is in the area of model-based design methods for metabolic engineering and synthetic biology.

Over the last years, several genome-scale metabolic models (GEMs) and kinetic models of Escherichia coli were published. Their predictive performance varies according to the evaluation metric considered, the computational simulation methods used, and the type/quality of experimental data available. However, the GEM approach is often not compared with the kinetic modeling framework. Also, the different genome-scale reconstruction versions and simulation methods of mutant phenotypes are usually not validated to predict intracellular fluxes using large experimental datasets. Here, we intended to (i) systematically evaluate the prediction performance of three E. coli GEMs (iJR904, iAF1260, and iJO1366) available in the literature according to predictive growth metrics (intracellular flux distribution); (ii) assess the reliability of a E. coli GEM in the prediction of gene knockout phenotypes when different simulation methods (parsimonious flux balance analysis, Minimization of Metabolic Adjustment, linear version of MoMA, Regulatory on/off minimization, and Minimization of Metabolites Balance) are used; and finally (iii) investigate the flux distribution predictive power of the constrained-based modeling approach (selected stoichiometric GEM) and compare it with the kinetic modeling approach (two published kinetic models) for E. coli central metabolism, in order to assess their accuracy. Results show that the phenotype predictions were not significantly sensitive to the metabolic models, although the GEM iAF1260 was more accurate in the prediction of central carbon fluxes at low dilution rates. Furthermore, we observed that the choice of the appropriate simulation method of mutant phenotypes depends on the biological question to be addressed. In terms of the two modeling approaches, none outperformed the other for all the tested scenarios. © 2018 American Institute of Chemical Engineers Biotechnol. Prog., 34:1344-1354, 2018.

Chrysanthemic acid is an unconventional monoterpene moiety of the natural pesticide pyrethrins with notable anti‐insect activity, making its industrial biosynthesis a promising avenue for sustainable agriculture. Here, an E. coli cell factory is designed and build for highly efficient chrysanthemic acid production guided by Genome‐scale metabolic models (GEM). The biosynthetic pathway is reconstructed and simulated the metabolic changes caused by exogenous modules are simulated. A key metabolic branch point catalyzed by ispA is identified by this model, and inhibiting its expression using synthetic small RNA redirected the metabolic flux, resulting in the titers of precursor chrysanthemol and chrysanthemic acid increasing by 162% and 59%, respectively. The effect of the expression level of downstream dehydrogenases on chrysanthemic acid titer is also predicated using GEM, and the further optimization of copy number for dehydrogenase genes led to a notably 570% increase in chrysanthemic acid titer experimentally. By integrating the debranching strategy with copy number optimization, a record chrysanthemic acid titer 141.78 mg L−1 is achieved in a bioreactor. The work seamlessly integrated in silico modeling optimization with wet‐lab practices that significantly enhance target metabolite titer through metabolic network engineering, offering a new route for constructing efficient cell factories for natural bioproducts.

Constraint-based modeling can mechanistically simulate the behavior of a biochemical system, permitting hypothesis generation, experimental design and interpretation of experimental data, with numerous applications, especially the modeling of metabolism. Given a generic model, several methods have been developed to extract a context-specific, genome-scale metabolic model by incorporating information used to identify metabolic processes and gene activities in each context. However, the existing model extraction algorithms are unable to ensure that a context-specific model is thermodynamically flux consistent. Here we introduce XomicsToModel, a semiautomated pipeline that integrates bibliomic, transcriptomic, proteomic and metabolomic data with a generic genome-scale metabolic reconstruction, or model, to extract a context-specific, genome-scale metabolic model that is stoichiometrically, thermodynamically and flux consistent. One of the key advantages of the XomicsToModel pipeline is its ability to seamlessly incorporate omics data into metabolic reconstructions, ensuring not only mechanistic accuracy but also physicochemical consistency. This functionality enables more accurate metabolic simulations and predictions across different biological contexts, enhancing its utility in diverse research fields, including systems biology, drug development and personalized medicine. The XomicsToModel pipeline is exemplified for extraction of a specific metabolic model from a generic metabolic model; it enables omics data integration and extraction of physicochemically consistent mechanistic models from any generic biochemical network. It can be implemented by anyone who has basic MATLAB programming skills and the fundamentals of constraint-based modeling.

Genome-scale metabolic models (GSMs) provide a comprehensive representation of cellular metabolism. GSMs provide a mechanistic link between cellular genotypes and metabolic phenotypes, and are thus widely used to analyze metabolism at the systems level. GSMs consist of hundreds or thousands of reactions. They have thus largely been analyzed with computationally efficient constraint-based methods such as flux-balance analysis, limiting their scope and phenotype prediction accuracy. Computationally more demanding but potentially more informative methods, such as kinetic and dynamic modeling, are currently limited to small or medium-sized models. Thus, it is desirable to achieve unbiased stoichiometric reductions of large-scale metabolic models to small, coarse-grained model representations that capture significant metabolic modules. Here, we review published automated and semiautomated methods used for large-scale metabolic model reduction. The top-down methods discussed provide minimal networks that retain a set of user-protected phenotypes, but may reduce the model's metabolic and phenotypic versatility. In contrast, the two bottom-up approaches reviewed retain a more unbiased set of phenotypes; at the same time, these methods require the partitioning of the GSM into metabolic subsystems by the user, and make strong assumptions on the subsystems' connections and their states, respectively.

In silico encounters: harnessing metabolic modelling to understand plant–microbe interactions

BackgroundA genome-scale metabolic network model (GEM) is a mathematical representation of an organism’s metabolism. Today, GEMs are popular tools for computationally simulating the biotechnological processes and for predicting biochemical properties of (engineered) strains.ObjectivesIn the present study, we have evaluated the predictive power of two GEMs, namely iBsu1103 (for Bacillus subtilis 168) and iMZ1055 (for Bacillus megaterium WSH002).Materials and MethodsFor comparing the predictive power of Bacillus subtilis and Bacillus megaterium GEMs, experimental data were obtained from previous wet-lab studies included in PubMed. By using these data, we set the environmental, stoichiometric and thermodynamic constraints on the models, and FBA is performed to predict the biomass production rate, and the values of other fluxes. For simulating experimental conditions in this study, COBRA toolbox was used.ResultsBy using the wealth of data in the literature, we evaluated the accuracy of in silico simulations of these GEMs. Our results suggest that there are some errors in these two models which make them unreliable for predicting the biochemical capabilities of these species. The inconsistencies between experimental and computational data are even greater where B. subtilis and B. megaterium do not have similar phenotypes.ConclusionsOur analysis suggests that literature-based improvement of genome-scale metabolic network models of the two Bacillus species is essential if these models are to be successfully applied in biotechnology and metabolic engineering.

Diabetic Kidney Disease (DKD) is one of the most common and serious complications of Type II Diabetes Mellitus (T2DM). Previous work measured plasma metabolites in DKD patients versus T2DM controls without DKD. DKD patients showed elevated hydroxyl‐ and dicarboxyl‐acylcarnitines. These metabolites are produced by omega‐oxidation (ωOx), a microsomal process that is known to be upregulated during mitochondrial dysregulation and inborn errors of metabolism. Briefly, dicarboxylic fatty acids from ωOx can be converted to dicarboxyl‐acylcarnitines or transported to the peroxisome for breakdown through peroxisomal β‐oxidation. ωOx is a poorly‐studied pathway that has never been subjected to computational simulations, nor included in detail in genome‐scale computational models of human metabolism.We developed a computational model of ωOx and integrated it with a previously published metabolic flux model of interplay between three tissue types (liver, muscle, adipose) in diabetes. We then performed metabolic flux simulations on this extended model using Flux‐Balance Analysis (FBA). Monte‐Carlo Flux sampling techniques identified cellular reactions with flux levels that were differentially regulated by allowing/blocking ωOx, or by allowing/blocking efflux of peroxisomal beta‐oxidation intermediates. Forcing the model to utilize ωOx caused consistent flux changes in nitrogen processing and the urea cycle (for muscle and liver), suggesting altered levels of urea and ammonia in plasma. While other pathways were also affected to different degrees, the effect was less consistent than with the urea cycle.In conclusion, we provide the first metabolic model that incorporates ωOx and peroxisomal beta‐oxidation in a multi‐tissue setting. Simulating a model of diabetes with ωOx identified links between mitochondrial dysfunction, fatty acid processing and nitrogen metabolism. Flux sampling simulations are useful for generating experimental hypotheses and for shedding light on systems‐level effects, but experimental validation is required before inferring causality. The importance of urea as an osmolyte and metabolite motivates further study of this relationship, which could have important implications for renal function.Support or Funding Information1. National Research Foundation (NRF), Prime Minister's Office, Singapore, under its CREATE programme, Singapore‐MIT Alliance for Research and Technology (SMART) BioSystems and Micromechanics (BioSyM) IRG2. Duke‐NUS SRP Phase 2 Research Block GrantThis abstract is from the Experimental Biology 2018 Meeting. There is no full text article associated with this abstract published in The FASEB Journal.