The genome-scale metabolic model for the purple non-sulfur bacterium Rhodopseudomonas palustris Bis A53 accurately predicts phenotypes under chemoheterotrophic, chemoautotrophic, photoheterotrophic, and photoautotrophic growth conditions

Abstract

The purple non-sulfur bacterium Rhodopseudomonas palustris is recognized as a critical microorganism in the nitrogen and carbon cycle and one of the most common members in wastewater treatment communities. This bacterium is metabolically extremely versatile. It is capable of heterotrophic growth under aerobic and anaerobic conditions, but also able to grow photoautotrophically as well as mixotrophically. Therefore R. palustris can adapt to multiple environments and establish commensal relationships with other organisms, expressing various enzymes supporting degradation of amino acids, carbohydrates, nucleotides, and complex polymers. Moreover, R. palustris can degrade a wide range of pollutants under anaerobic conditions, e.g., aromatic compounds such as benzoate and caffeate, enabling it to thrive in chemically contaminated environments. However, many metabolic mechanisms employed by R. palustris to breakdown and assimilate different carbon and nitrogen sources under chemoheterotrophic or photoheterotrophic conditions remain unknown. Systems biology approaches, such as metabolic modeling, have been employed extensively to unravel complex mechanisms of metabolism. Previously, metabolic models have been reconstructed to study selected capabilities of R. palustris under limited experimental conditions. Here, we developed a comprehensive metabolic model (M-model) for R. palustris Bis A53 (iDT1294) consisting of 2,721 reactions, 2,123 metabolites, and comprising 1,294 genes. We validated the model using high-throughput phenotypic, physiological, and kinetic data, testing over 350 growth conditions. iDT1294 achieved a prediction accuracy of 90% for growth with various carbon and nitrogen sources and close to 80% for assimilation of aromatic compounds. Moreover, the M-model accurately predicts dynamic changes of growth and substrate consumption rates over time under nine chemoheterotrophic conditions and demonstrated high precision in predicting metabolic changes between photoheterotrophic and photoautotrophic conditions. This comprehensive M-model will help to elucidate metabolic processes associated with the assimilation of multiple carbon and nitrogen sources, anoxygenic photosynthesis, aromatic compound degradation, as well as production of molecular hydrogen and polyhydroxybutyrate.

Fig 1. Workflow to build a metabolic model of R. palustris BisA53 using a semiautomatic approach.

A A draft model was reconstructed using optimized BLAST parameters (e-value, query length, and identity percentage) from three template models present in BiGG (Escherichia coli K-12 substr. MG1655, Synechocystis sp. PCC 6803, and Synechococcus elongatus PCC 7942). NCBI reference sequence annotation was employed in GPR associations. The RAVEN and COBRA packages for MATLAB were deployed in the reconstruction stage. B The resulting optimized draft model and constituents of the Biomass Objective Function (BOF) were manually curated. Protein, RNA and DNA metabolites of the BOF were calculated based on genomic and proteomic data, meanwhile the rest of the metabolic requirements were estimated based on experimental evidence previously published. Disconnected metabolites were properly integrated into metabolic pathways using bioinformatics databases. The iDT1278 model (A. vinelandii) worked as a template for nitrogen fixation, hydrogen production, and PHB biosynthesis reactions using BLASTp homology. The resulting draft model of the BOF and dead-ends curation contained 2,298 reactions, 1,918 metabolites, and 1,515 genes (200 exogenous genes). C Four detailed subsystems were manually added to the model to reflect specific metabolic capabilities of R. palustris BisA53: aromatic compound degradation under anaerobic conditions, nitrogen fixation and denitrification (with partial denitrification), PHB production and pigment metabolism (including carotenoid and bacteriochlorophyll of PNSB). For the aromatic compounds biosynthesis pathways, the reactions and metabolites involved were carefully added based on enough experimental evidence and curated information of the pathways from the bioinformatics databases (See Methods). D The resulting model was validated using experimental data retrieved from the literature and growth experiments performed in this study. The iterative model refinement process included manual curation, gap-filling, and curation under chemoheterotrophic, chemoautotrophic, photoautotrophic, and photoheterotrophic conditions changing the oxygen requirements depending on the experimental environments. Ultimately, the model was validated using kinetics data to compare the prediction capabilities of the M-model using growth rates and substrate concentrations for specific time points. The final model, containing 2,721 reactions, 2,123 metabolites, and 1,294 genes, predicted growth with 90% accuracy for carbon and nitrogen substrates.

Fig 2. Model properties and prediction capabilities of iDT1294.

A A general comparison was executed among the three principal template models (Escherichia coli K-12 substr. MG1655, iML1515, Synechocystis sp. PCC 6803, iJN678 and Synechococcus elongatus PCC 7942, iJB785) and iDT1294 reactions. The four models share 341 core metabolic reactions. B Six principal model versions were generated from the initial reconstruction process to the final validated model. Across the different stages, the model increased the number of reactions, metabolites, and RPE genes, while the number of exogenous genes was reduced to zero. C iDT1294 accuracy and phenotypic predictions capabilities were evaluated under the four main metabolic states of this versatile organism, testing over 350 experimental conditions. D R. palustris Bis A53 model contains 549 unique reactions related to aromatic compounds metabolism, PHB production, carotenoids, pigments and bacteriochlorophylls biosynthesis specifically synthetized by PNSB as well as nitrogen fixation and denitrification. Reactions (2,721 in total) were distributed in 17 subsystems representing the entire metabolism of R. palustris Bis A53.

Fig 3. Comparison of the high-throughput growth phenotypic experimental and simulated results across the three principal metabolic models.

R. palustris Bis A53 was cultivated in 190 carbon and 95 nitrogen sources under monoculture conditions for 96 hours. Subsequently, the three GEMs (iDT1294, iRpa940, and iAN1128) were properly constrained to simulate all the Biolog conditions. Each heat map compares every carbon (PM1 and PM2) and nitrogen (PM3) source’s experimental result against the simulation growth output per metabolic model. Growth results were classified into two possible results: Growth (purple) and No Growth (light pink). Additionally, we determined growth by measuring optical density (OD600, see methods) after 96 hrs, as indicated in blue to the right of each Biolog plate set. The substrates studied were sorted in descending order based on growth values. Statistical parameters to determine the global accuracy and prediction capabilities of the three models were calculated.

Fig 4. Comparison of dynamic flux balance analysis with experimental data.

The set of plots shows the dynamic prediction capabilities of iDT1294 using nine different growth conditions. For each plot, the experimental time points (blue dots) were adjusted to the exponential phase, and time points in the stationary phase were removed. Later, dFBA was executed for iDT1294 to determine the growth rates and consumption of substrates across time. R2 and log₁₀P parameters were determined for iDT1294 (orange and star) to identify how the model predictions fit to the experimental results.