Defining the nutritional input for genome-scale metabolic models: A roadmap

Community metabolic modeling of host-microbiota interactions through multi-objective optimization

The human gut microbiota comprises various microorganisms engaged in intricate interactions among themselves and with the host, affecting its health. While advancements in omics technologies have led to the inference of clear associations between microbiome composition and health conditions, we usually lack a causal and mechanistic understanding of these associations. For modeling mechanisms driving the interactions, we simulated the organism's metabolism using in silico genome-scale metabolic models (GEMs). We used multi-objective optimization to predict and explain metabolic interactions among gut microbes and an intestinal epithelial cell. We developed a score integrating model simulation results to predict the type (competition, neutralism, mutualism) and quantify the interaction between several organisms. This framework uncovered a potential cross-feeding for choline, explaining the predicted mutualism between Lactobacillus rhamnosus GG and the epithelial cell. Finally, we analyzed a five-organism ecosystem, revealing that a minimal microbiota can favor the epithelial cell's maintenance.

Metabolic model predictions enable targeted microbiome manipulation through precision prebiotics

While numerous health-beneficial interactions between host and microbiota have been identified, there is still a lack of targeted approaches for modulating these interactions. Thus, we here identify precision prebiotics that specifically modulate the abundance of a microbiome member species of interest. In the first step, we show that defining precision prebiotics by compounds that are only taken up by the target species but no other species in a community is usually not possible due to overlapping metabolic niches. Subsequently, we use metabolic modeling to identify precision prebiotics for a two-member Caenorhabditis elegans microbiome community comprising the immune-protective target species Pseudomonas lurida MYb11 and the persistent colonizer Ochrobactrum vermis MYb71. We experimentally confirm four of the predicted precision prebiotics, L-serine, L-threonine, D-mannitol, and γ-aminobutyric acid, to specifically increase the abundance of MYb11. L-serine was further assessed in vivo, leading to an increase in MYb11 abundance also in the worm host. Overall, our findings demonstrate that metabolic modeling is an effective tool for the design of precision prebiotics as an important cornerstone for future microbiome-targeted therapies.IMPORTANCEWhile various mechanisms through which the microbiome influences disease processes in the host have been identified, there are still only few approaches that allow for targeted manipulation of microbiome composition as a first step toward microbiome-based therapies. Here, we propose the concept of precision prebiotics that allow to boost the abundance of already resident health-beneficial microbial species in a microbiome. We present a constraint-based modeling pipeline to predict precision prebiotics for a minimal microbial community in the worm Caenorhabditis elegans comprising the host-beneficial Pseudomonas lurida MYb11 and the persistent colonizer Ochrobactrum vermis MYb71 with the aim to boost the growth of MYb11. Experimentally testing four of the predicted precision prebiotics, we confirm that they are specifically able to increase the abundance of MYb11 in vitro and in vivo. These results demonstrate that constraint-based modeling could be an important tool for the development of targeted microbiome-based therapies against human diseases.

Changes in Bacterial Gut Composition in Parkinson's Disease and Their Metabolic Contribution to Disease Development: A Gut Community Reconstruction Approach

Parkinson's disease (PD) is a chronic and progressive neurodegenerative disease with the major symptoms comprising loss of movement coordination (motor dysfunction) and non-motor dysfunction, including gastrointestinal symptoms. Alterations in the gut microbiota composition have been reported in PD patients vs. controls. However, it is still unclear how these compositional changes contribute to disease etiology and progression. Furthermore, most of the available studies have focused on European, Asian, and North American cohorts, but the microbiomes of PD patients in Latin America have not been characterized. To address this problem, we obtained fecal samples from Colombian participants (n = 25 controls, n = 25 PD idiopathic cases) to characterize the taxonomical community changes during disease via 16S rRNA gene sequencing. An analysis of differential composition, diversity, and personalized computational modeling was carried out, given the fecal bacterial composition and diet of each participant. We found three metabolites that differed in dietary habits between PD patients and controls: carbohydrates, trans fatty acids, and potassium. We identified six genera that changed significantly in their relative abundance between PD patients and controls, belonging to the families Lachnospiraceae, Lactobacillaceae, Verrucomicrobioaceae, Peptostreptococcaceae, and Streptococcaceae. Furthermore, personalized metabolic modeling of the gut microbiome revealed changes in the predicted production of seven metabolites (Indole, tryptophan, fructose, phenylacetic acid, myristic acid, 3-Methyl-2-oxovaleric acid, and N-Acetylneuraminic acid). These metabolites are associated with the metabolism of aromatic amino acids and their consumption in the diet. Therefore, this research suggests that each individual's diet and intestinal composition could affect host metabolism. Furthermore, these findings open the door to the study of microbiome-host interactions and allow us to contribute to personalized medicine.

Predictions of rhizosphere microbiome dynamics with a genome-informed and trait-based energy budget model

Soil microbiomes are highly diverse, and to improve their representation in biogeochemical models, microbial genome data can be leveraged to infer key functional traits. By integrating genome-inferred traits into a theory-based hierarchical framework, emergent behaviour arising from interactions of individual traits can be predicted. Here we combine theory-driven predictions of substrate uptake kinetics with a genome-informed trait-based dynamic energy budget model to predict emergent life-history traits and trade-offs in soil bacteria. When applied to a plant microbiome system, the model accurately predicted distinct substrate-acquisition strategies that aligned with observations, uncovering resource-dependent trade-offs between microbial growth rate and efficiency. For instance, inherently slower-growing microorganisms, favoured by organic acid exudation at later plant growth stages, exhibited enhanced carbon use efficiency (yield) without sacrificing growth rate (power). This insight has implications for retaining plant root-derived carbon in soils and highlights the power of data-driven, trait-based approaches for improving microbial representation in biogeochemical models.

Predicting microbial interactions with approaches based on flux balance analysis: an evaluation

BACKGROUND

Given a genome-scale metabolic model (GEM) of a microorganism and criteria for optimization, flux balance analysis (FBA) predicts the optimal growth rate and its corresponding flux distribution for a specific medium. FBA has been extended to microbial consortia and thus can be used to predict interactions by comparing in-silico growth rates for co- and monocultures. Although FBA-based methods for microbial interaction prediction are becoming popular, a systematic evaluation of their accuracy has not yet been performed.

RESULTS

Here, we evaluate the accuracy of FBA-based predictions of human and mouse gut bacterial interactions using growth data from the literature. For this, we collected 26 GEMs from the semi-curated AGORA database as well as four previously published curated GEMs. We tested the accuracy of three tools (COMETS, Microbiome Modeling Toolbox and MICOM) by comparing growth rates predicted in mono- and co-culture to growth rates extracted from the literature and also investigated the impact of different tool settings and media. We found that except for curated GEMs, predicted growth rates and their ratios (i.e. interaction strengths) do not correlate with growth rates and interaction strengths obtained from in vitro data.

CONCLUSIONS

Prediction of growth rates with FBA using semi-curated GEMs is currently not sufficiently accurate to predict interaction strengths reliably.

Unveiling metabolic pathways involved in the extreme desiccation tolerance of an Atacama cyanobacterium

Gloeocapsopsis dulcis strain AAB1 is an extremely xerotolerant cyanobacterium isolated from the Atacama Desert (i.e., the driest and oldest desert on Earth) that holds astrobiological significance due to its ability to biosynthesize compatible solutes at ultra-low water activities. We sequenced and assembled the G. dulcis genome de novo using a combination of long- and short-read sequencing, which resulted in high-quality consensus sequences of the chromosome and two plasmids. We leveraged the G. dulcis genome to generate a genome-scale metabolic model (iGd895) to simulate growth in silico. iGd895 represents, to our knowledge, the first genome-scale metabolic reconstruction developed for an extremely xerotolerant cyanobacterium. The model's predictive capability was assessed by comparing the in silico growth rate with in vitro growth rates of G. dulcis, in addition to the synthesis of trehalose. iGd895 allowed us to explore simulations of key metabolic processes such as essential pathways for water-stress tolerance, and significant alterations to reaction flux distribution and metabolic network reorganization resulting from water limitation. Our study provides insights into the potential metabolic strategies employed by G. dulcis, emphasizing the crucial roles of compatible solutes, metabolic water, energy conservation, and the precise regulation of reaction rates in their adaptation to water stress.

Transcriptome-guided metabolic network analysis reveals rearrangements of carbon flux distribution in Neisseria gonorrhoeae during neutrophil co-culture

The ability of bacterial pathogens to metabolically adapt to the environmental conditions of their hosts is critical to both colonization and invasive disease. Infection with Neisseria gonorrhoeae (the gonococcus, Gc) is characterized by the influx of neutrophils [polymorphonuclear leukocytes (PMNs)], which fail to clear the bacteria and make antimicrobial products that can exacerbate tissue damage. The inability of the human host to clear Gc infection is particularly concerning in light of the emergence of strains that are resistant to all clinically recommended antibiotics. Bacterial metabolism represents a promising target for the development of new therapeutics against Gc. Here, we generated a curated genome-scale metabolic network reconstruction (GENRE) of Gc strain FA1090. This GENRE links genetic information to metabolic phenotypes and predicts Gc biomass synthesis and energy consumption. We validated this model with published data and in new results reported here. Contextualization of this model using the transcriptional profile of Gc exposed to PMNs revealed substantial rearrangements of Gc central metabolism and induction of Gc nutrient acquisition strategies for alternate carbon source use. These features enhanced the growth of Gc in the presence of neutrophils. From these results, we conclude that the metabolic interplay between Gc and PMNs helps define infection outcomes. The use of transcriptional profiling and metabolic modeling to reveal new mechanisms by which Gc persists in the presence of PMNs uncovers unique aspects of metabolism in this fastidious bacterium, which could be targeted to block infection and thereby reduce the burden of gonorrhea in the human population. IMPORTANCE The World Health Organization designated Gc as a high-priority pathogen for research and development of new antimicrobials. Bacterial metabolism is a promising target for new antimicrobials, as metabolic enzymes are widely conserved among bacterial strains and are critical for nutrient acquisition and survival within the human host. Here we used genome-scale metabolic modeling to characterize the core metabolic pathways of this fastidious bacterium and to uncover the pathways used by Gc during culture with primary human immune cells. These analyses revealed that Gc relies on different metabolic pathways during co-culture with human neutrophils than in rich media. Conditionally essential genes emerging from these analyses were validated experimentally. These results show that metabolic adaptation in the context of innate immunity is important to Gc pathogenesis. Identifying the metabolic pathways used by Gc during infection can highlight new therapeutic targets for drug-resistant gonorrhea.

Development of a Serum-Free Medium To Aid Large-Scale Production of Mycoplasma-Based Therapies

To assist in the advancement of the large-scale production of safe Mycoplasma vaccines and other Mycoplasma-based therapies, we developed a culture medium free of animal serum and other animal components for Mycoplasma pneumoniae growth. By establishing a workflow method to systematically test different compounds and concentrations, we provide optimized formulations capable of supporting serial passaging and robust growth reaching 60 to 70% of the biomass obtained in rich medium. Global transcriptomic and proteomic analysis showed minor physiological changes upon cell culture in the animal component-free medium, supporting its suitability for the production of M. pneumoniae-based therapies. The major contributors to growth performance were found to be glucose as a carbon source, glycerol, cholesterol, and phospholipids as a source of fatty acids. Bovine serum albumin or cyclodextrin (in the animal component-free medium) were required as lipid carriers to prevent lipid toxicity. Connaught Medical Research Laboratories medium (CMRL) used to simplify medium preparation as a source of amino acids, nucleotide precursors, vitamins, and other cofactors could be substituted by cysteine. In fact, the presence of protein hydrolysates such as yeastolate or peptones was found to be essential and preferred over free amino acids, except for the cysteine. Supplementation of nucleotide precursors and vitamins is not strictly necessary in the presence of yeastolate, suggesting that this animal origin-free hydrolysate serves as an efficient source for these compounds. Finally, we adapted the serum-free medium formulation to support growth of Mycoplasma hyopneumoniae, a swine pathogen for which inactivated whole-cell vaccines are available. IMPORTANCE Mycoplasma infections have a significant negative impact on both livestock production and human health. Vaccination is often the first option to control disease and alleviate the economic impact that some Mycoplasma infections cause on milk production, weight gain, and animal health. The fastidious nutrient requirements of these bacteria, however, challenges the industrial production of attenuated or inactivated whole-cell vaccines, which depends on the use of animal serum and other animal raw materials. Apart from their clinical relevance, some Mycoplasma species have become cellular models for systems and synthetic biology, owing to the small size of their genomes and the absence of a cell wall, which offers unique opportunities for the secretion and delivery of biotherapeutics. This study proposes medium formulations free of serum and animal components with the potential of supporting large-scale production upon industrial optimization, thus contributing to the development of safe vaccines and other Mycoplasma-based therapies.

Metabolic model predictions enable targeted microbiome manipulation through precision prebiotics

The microbiome is increasingly receiving attention as an important modulator of host health and disease. However, while numerous mechanisms through which the microbiome influences its host have been identified, there is still a lack of approaches that allow to specifically modulate the abundance of individual microbes or microbial functions of interest. Moreover, current approaches for microbiome manipulation such as fecal transfers often entail a non-specific transfer of entire microbial communities with potentially unwanted side effects. To overcome this limitation, we here propose the concept of precision prebiotics that specifically modulate the abundance of a microbiome member species of interest. In a first step, we show that defining precision prebiotics by compounds that are only taken up by the target species but no other species in a community is usually not possible due to overlapping metabolic niches. Subsequently, we present a metabolic modeling network framework that allows us to define precision prebiotics for a two-member C. elegans microbiome model community comprising the immune-protective Pseudomonas lurida MYb11 and the persistent colonizer Ochrobactrum vermis MYb71. Thus, we predicted compounds that specifically boost the abundance of the host-beneficial MYb11, four of which were experimentally validated in vitro (L-serine, L-threonine, D-mannitol, and γ-aminobutyric acid). L-serine was further assessed in vivo, leading to an increase in MYb11 abundance also in the worm host. Overall, our findings demonstrate that constraint-based metabolic modeling is an effective tool for the design of precision prebiotics as an important cornerstone for future microbiome-targeted therapies.

Systematic diet composition swap in a mouse genome-scale metabolic model reveals determinants of obesogenic diet metabolism in liver cancer

Dietary nutrient availability and gene expression, together, influence tissue metabolic activity. Here, we explore whether altering dietary nutrient composition in the context of mouse liver cancer suffices to overcome chronic gene expression changes that arise from tumorigenesis and western-style diet (WD). We construct a mouse genome-scale metabolic model and estimate metabolic fluxes in liver tumors and non-tumoral tissue after computationally varying the composition of input diet. This approach, called Systematic Diet Composition Swap (SyDiCoS), revealed that, compared to a control diet, WD increases production of glycerol and succinate irrespective of specific tissue gene expression patterns. Conversely, differences in fatty acid utilization pathways between tumor and non-tumor liver are amplified with WD by both dietary carbohydrates and lipids together. Our data suggest that combined dietary component modifications may be required to normalize the distinctive metabolic patterns that underlie selective targeting of tumor metabolism.

Genome Sequence of Eubacterium limosum B2 and Evolution for Growth on a Mineral Medium with Methanol and CO2 as Sole Carbon Sources

Eubacterium limosum is an acetogen that can produce butyrate along with acetate as the main fermentation end-product from methanol, a promising C1 feedstock. Although physiological characterization of E. limosum B2 during methylotrophy was previously performed, the strain was cultured in a semi-defined medium, limiting the scope for further metabolic insights. Here, we sequenced the complete genome of the native strain and performed adaptive laboratory evolution to sustain growth on methanol mineral medium. The evolved population significantly improved its maximal growth rate by 3.45-fold. Furthermore, three clones from the evolved population were isolated on methanol mineral medium without cysteine by the addition of sodium thiosulfate. To identify mutations related to growth improvement, the whole genomes of wild-type E. limosum B2, the 10th, 25th, 50th, and 75th generations, and the three clones were sequenced. We explored the total proteomes of the native and the best evolved clone (n°2) and noticed significant differences in proteins involved in gluconeogenesis, anaplerotic reactions, and sulphate metabolism. Furthermore, a homologous recombination was found in subunit S of the type I restriction-modification system between both strains, changing the structure of the subunit, its sequence recognition and the methylome of the evolved clone. Taken together, the genomic, proteomic and methylomic data suggest a possible epigenetic mechanism of metabolic regulation.

A pipeline for the reconstruction and evaluation of context-specific human metabolic models at a large-scale

Constraint-based (CB) metabolic models provide a mathematical framework and scaffold for in silico cell metabolism analysis and manipulation. In the past decade, significant efforts have been done to model human metabolism, enabled by the increased availability of multi-omics datasets and curated genome-scale reconstructions, as well as the development of several algorithms for context-specific model (CSM) reconstruction. Although CSM reconstruction has revealed insights on the deregulated metabolism of several pathologies, the process of reconstructing representative models of human tissues still lacks benchmarks and appropriate integrated software frameworks, since many tools required for this process are still disperse across various software platforms, some of which are proprietary. In this work, we address this challenge by assembling a scalable CSM reconstruction pipeline capable of integrating transcriptomics data in CB models. We combined omics preprocessing methods inspired by previous efforts with in-house implementations of existing CSM algorithms and new model refinement and validation routines, all implemented in the Troppo Python-based open-source framework. The pipeline was validated with multi-omics datasets from the Cancer Cell Line Encyclopedia (CCLE), also including reference fluxomics measurements for the MCF7 cell line. We reconstructed over 6000 models based on the Human-GEM template model for 733 cell lines featured in the CCLE, using MCF7 models as reference to find the best parameter combinations. These reference models outperform earlier studies using the same template by comparing gene essentiality and fluxomics experiments. We also analysed the heterogeneity of breast cancer cell lines, identifying key changes in metabolism related to cancer aggressiveness. Despite the many challenges in CB modelling, we demonstrate using our pipeline that combining transcriptomics data in metabolic models can be used to investigate key metabolic shifts. Significant limitations were found on these models ability for reliable quantitative flux prediction, thus motivating further work in genome-wide phenotype prediction.

Genome-scale models of human metabolism are promising tools capable of contextualising large omics datasets within a framework that enables analysis and manipulation of metabolic phenotypes. Despite various successes in applying these methods to provide mechanistic hypotheses for deregulated metabolism in disease, there is no standardized workflow to extract these models using existing methods and the tools required to do so are mostly implemented using proprietary software. We have assembled a generic pipeline to extract and validate context-specific metabolic models using multi-omics datasets and implemented it using the troppo framework. We first validate our pipeline using MCF7 cell line models and assess their ability to predict lethal gene knockouts as well as flux activity using multi-omics data. We also demonstrate how this approach can be generalized for large-scale transcriptomics datasets and used to generate insights on the metabolic heterogeneity of cancer and relevant features for other data mining approaches. The pipeline is available as part of an open-source framework that is generic for a variety of applications.

Integrative Gene Expression and Metabolic Analysis Tool IgemRNA

Genome-scale metabolic modeling is widely used to study the impact of metabolism on the phenotype of different organisms. While substrate modeling reflects the potential distribution of carbon and other chemical elements within the model, the additional use of omics data, e.g., transcriptome, has implications when researching the genotype–phenotype responses to environmental changes. Several algorithms for transcriptome analysis using genome-scale metabolic modeling have been proposed. Still, they are restricted to specific objectives and conditions and lack flexibility, have software compatibility issues, and require advanced user skills. We classified previously published algorithms, summarized transcriptome pre-processing, integration, and analysis methods, and implemented them in the newly developed transcriptome analysis tool IgemRNA, which (1) has a user-friendly graphical interface, (2) tackles compatibility issues by combining previous data input and pre-processing algorithms in MATLAB, and (3) introduces novel algorithms for the automatic comparison of different transcriptome datasets with or without Cobra Toolbox 3.0 optimization algorithms. We used publicly available transcriptome datasets from Saccharomyces cerevisiae BY4741 and H4-S47D strains for validation. We found that IgemRNA provides a means for transcriptome and environmental data validation on biochemical network topology since the biomass function varies for different phenotypes. Our tool can detect problematic reaction constraints.

Exometabolomic Analysis of Susceptible and Multi-Drug Resistant Pseudomonas aeruginosa

Multidrug resistant (MDR) Pseudomonas aeruginosa strains have recently become one of the major public health concerns worldwide leading to difficulties in selecting appropriate antibiotic treatment. Thus, it is important to elucidate the characteristics of MDR isolates. Herein, we aimed to determine the unique exometabolome profile of P. aeruginosa clinical isolates in monocultures that comprise high resistance to multiple antibiotics, and compare the differential metabolite profiles obtained from susceptible isolates by using GC/MS. Our results showed that partial least square-discriminant analysis (PLS-DA) score plot clearly discriminated the MDR and susceptible isolates indicating the altered exometabolite profiles, and highlighted the significantly enriched levels of trehalose and glutamic acid in MDR isolates. Expression of trehalose synthase (treS) was also 1.5 fold higher in MDR isolates, relatively to susceptible isolates. Overall, our study provides insights into the distinct footprints of MDR P. aeruginosa isolates in mono-culture.

In vitro interaction network of a synthetic gut bacterial community

A key challenge in microbiome research is to predict the functionality of microbial communities based on community membership and (meta)-genomic data. As central microbiota functions are determined by bacterial community networks, it is important to gain insight into the principles that govern bacteria-bacteria interactions. Here, we focused on the growth and metabolic interactions of the Oligo-Mouse-Microbiota (OMM¹²) synthetic bacterial community, which is increasingly used as a model system in gut microbiome research. Using a bottom-up approach, we uncovered the directionality of strain-strain interactions in mono- and pairwise co-culture experiments as well as in community batch culture. Metabolic network reconstruction in combination with metabolomics analysis of bacterial culture supernatants provided insights into the metabolic potential and activity of the individual community members. Thereby, we could show that the OMM¹² interaction network is shaped by both exploitative and interference competition in vitro in nutrient-rich culture media and demonstrate how community structure can be shifted by changing the nutritional environment. In particular, Enterococcus faecalis KB1 was identified as an important driver of community composition by affecting the abundance of several other consortium members in vitro. As a result, this study gives fundamental insight into key drivers and mechanistic basis of the OMM¹² interaction network in vitro, which serves as a knowledge base for future mechanistic in vivo studies.

Protocol for hybrid flux balance, statistical, and machine learning analysis of multi-omic data from the cyanobacterium Synechococcus sp. PCC 7002

Combining a computational framework for flux balance analysis with machine learning improves the accuracy of predicting metabolic activity across conditions, while enabling mechanistic interpretation. This protocol presents a guide to condition-specific metabolic modeling that integrates regularized flux balance analysis with machine learning approaches to extract key features from transcriptomic and fluxomic data. We demonstrate the protocol as applied to Synechococcus sp. PCC 7002; we also outline how it can be adapted to any species or community with available multi-omic data. For complete details on the use and execution of this protocol, please refer to Vijayakumar et al. (2020).

DCcov: Repositioning of drugs and drug combinations for SARS-CoV-2 infected lung through constraint-based modeling

The 2019 coronavirus disease (COVID-19) became a worldwide pandemic with currently no approved effective antiviral drug. Flux balance analysis (FBA) is an efficient method to analyze metabolic networks. Here, FBA was applied on human lung cells infected with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) to reposition metabolic drugs and drug combinations against the virus replication within the host tissue. Making use of expression datasets of infected lung tissue, genome-scale COVID-19-specific metabolic models were reconstructed. Then, host-specific essential genes and gene pairs were determined through in silico knockouts that permit reducing the viral biomass production without affecting the host biomass. Key pathways that are associated with COVID-19 severity in lung tissue are related to oxidative stress, ferroptosis, and pyrimidine metabolism. By in silico screening of Food and Drug Administration (FDA)-approved drugs on the putative disease-specific essential genes and gene pairs, 85 drugs and 52 drug combinations were predicted as promising candidates for COVID-19 (https://github.com/sysbiolux/DCcov).

Addressing uncertainty in genome-scale metabolic model reconstruction and analysis

The reconstruction and analysis of genome-scale metabolic models constitutes a powerful systems biology approach, with applications ranging from basic understanding of genotype-phenotype mapping to solving biomedical and environmental problems. However, the biological insight obtained from these models is limited by multiple heterogeneous sources of uncertainty, which are often difficult to quantify. Here we review the major sources of uncertainty and survey existing approaches developed for representing and addressing them. A unified formal characterization of these uncertainties through probabilistic approaches and ensemble modeling will facilitate convergence towards consistent reconstruction pipelines, improved data integration algorithms, and more accurate assessment of predictive capacity.