Integrated rational and evolutionary engineering of genome-reduced Pseudomonas putida strains promotes synthetic formate assimilation

Formate is a promising, water-soluble C1 feedstock for biotechnology that can be efficiently produced from CO₂—but formatotrophy has been engineered in only a few industrially-relevant microbial hosts. We addressed the challenge of expanding the feedstock range of bacterial hosts by adopting Pseudomonas putida as a robust platform for synthetic formate assimilation. Here, the metabolism of a genome-reduced variant of P. putida was radically rewired to establish synthetic auxotrophies that could be functionally complemented by expressing components of the reductive glycine (rGly) pathway. We adopted a modular engineering approach, dividing C1 assimilation in segments composed of both heterologous activities (sourced from Methylobacterium extorquens) and native biochemical reactions. Modular expression of rGly pathway elements enabled growth on formate as carbon source and acetate (predominantly for energy supply), and adaptive laboratory evolution of two lineages of engineered P. putida formatotrophs lead to doubling times of ca. 15 h. We likewise identified emergent metabolic features for assimilation of C1 units in these evolved P. putida populations. Taken together, our results consolidate the landscape of useful microbial platforms that can be implemented for C1-based biotechnological production towards a formate bioeconomy.     

Model-guided dynamic control of essential metabolic nodes boosts acetyl-coenzyme A–dependent bioproduction in rewired Pseudomonas putida

Pseudomonas putida is evolutionarily endowed with features relevant for bioproduction, especially under harsh operating conditions. The rich metabolic versatility of this species, however, comes at the price of limited formation of acetyl-coenzyme A (CoA) from sugar substrates. Since acetyl-CoA is a key metabolic precursor for a number of added-value products, in this work we deployed an in silico-guided rewiring program of central carbon metabolism for upgrading P. putida as a host for acetyl-CoA–dependent bioproduction. An updated kinetic model, integrating fluxomics and metabolomics datasets in addition to manually-curated information of enzyme mechanisms, identified targets that would lead to increased acetyl-CoA levels. Based on these predictions, a set of plasmids based on clustered regularly interspaced short palindromic repeats (CRISPR) and dead CRISPR-associated protein 9 (dCas9) was constructed to silence genes by CRISPR interference (CRISPRi). Dynamic reduction of gene expression of two key targets (gltA, encoding citrate synthase, and the essential accA gene, encoding subunit A of the acetyl-CoA carboxylase complex) mediated an 8-fold increase in the acetyl-CoA content of rewired P. putida. Poly(3-hydroxybutyrate) (PHB) was adopted as a proxy of acetyl-CoA availability, and two synthetic pathways were engineered for biopolymer accumulation. By including cell morphology as an extra target for the CRISPRi approach, fully rewired P. putida strains programmed for PHB accumulation had a 5-fold increase in PHB titers in bioreactor cultures using glucose. Thus, the strategy described herein allowed for rationally redirecting metabolic fluxes in P. putida from central metabolism towards product biosynthesis—especially relevant when deletion of essential pathways is not an option.     

Mixing and matching methylotrophic enzymes to design a novel methanol utilization pathway in E. coli

One-carbon (C1) compounds, such as methanol, have recently gained attention as alternative low-cost and non-food feedstocks for microbial bioprocesses. Considerable research efforts are thus currently focused on the generation of synthetic methylotrophs by transferring methanol assimilation pathways into established bacterial production hosts. In this study, we used an iterative combination of dry and wet approaches to design, implement and optimize this metabolic trait in the most common chassis, E. coli. Through in silico modelling, we designed a new route that “mixed and matched” two methylotrophic enzymes: a bacterial methanol dehydrogenase (Mdh) and a dihydroxyacetone synthase (Das) from yeast. To identify the best combination of enzymes to introduce into E. coli, we built a library of 266 pathway variants containing different combinations of Mdh and Das homologues and screened it using high-throughput ¹³C-labeling experiments. The highest level of incorporation of methanol into central metabolism intermediates (e.g. 22% into the PEP), was obtained using a variant composed of a Mdh from A. gerneri and a codon-optimized version of P. angusta Das. Finally, the activity of this new synthetic pathway was further improved by engineering strategic metabolic targets identified using omics and modelling approaches. The final synthetic strain had 1.5 to 5.9 times higher methanol assimilation in intracellular metabolites and proteinogenic amino acids than the starting strain did. Broadening the repertoire of methanol assimilation pathways is one step further toward synthetic methylotrophy in E. coli.     

Metabolic engineering of Pseudomonas putida for production of docosahexaenoic acid based on a myxobacterial PUFA synthase

Long-chain polyunsaturated fatty acids (LC-PUFAs) can be produced de novo via polyketide synthase-like enzymes known as PUFA synthases, which are encoded by pfa biosynthetic gene clusters originally discovered from marine microorganisms. Recently similar gene clusters were detected and characterized in terrestrial myxobacteria revealing several striking differences. As the identified myxobacterial producers are difficult to handle genetically and grow very slowly we aimed to establish heterologous expression platforms for myxobacterial PUFA synthases. Here we report the heterologous expression of the pfa gene cluster from Aetherobacter fasciculatus (SBSr002) in the phylogenetically distant model host bacteria Escherichia coli and Pseudomonas putida. The latter host turned out to be the more promising PUFA producer revealing higher production rates of n-6 docosapentaenoic acid (DPA) and docosahexaenoic acid (DHA). After several rounds of genetic engineering of expression plasmids combined with metabolic engineering of P. putida, DHA production yields were eventually increased more than threefold. Additionally, we applied synthetic biology approaches to redesign and construct artificial versions of the A. fasciculatus pfa gene cluster, which to the best of our knowledge represents the first example of a polyketide-like biosynthetic gene cluster modulated and synthesized for P. putida. Combination with the engineering efforts described above led to a further increase in LC-PUFA production yields. The established production platform based on synthetic DNA now sets the stage for flexible engineering of the complex PUFA synthase.     

Optimizing metabolic pathways by screening for feasible synthetic reactions

Background

Reconstruction of genome-scale metabolic networks has resulted in models capable of reproducing experimentally observed biomass yield/growth rates and predicting the effect of alterations in metabolism for biotechnological applications. The existing studies rely on modifying the metabolic network of an investigated organism by removing or inserting reactions taken either from evolutionary similar organisms or from databases of biochemical reactions (e.g., KEGG). A potential disadvantage of these knowledge-driven approaches is that the result is biased towards known reactions, as such approaches do not account for the possibility of including novel enzymes, together with the reactions they catalyze.

Results

Here, we explore the alternative of increasing biomass yield in three model organisms, namely Bacillus subtilis, Escherichia coli, and Hordeum vulgare, by applying small, chemically feasible network modifications. We use the predicted and experimentally confirmed growth rates of the wild-type networks as reference values and determine the effect of inserting mass-balanced, thermodynamically feasible reactions on predictions of growth rate by using flux balance analysis.

Conclusions

While many replacements of existing reactions naturally lead to a decrease or complete loss of biomass production ability, in all three investigated organisms we find feasible modifications which facilitate a significant increase in this biological function. We focus on modifications with feasible chemical properties and a significant increase in biomass yield. The results demonstrate that small modifications are sufficient to substantially alter biomass yield in the three organisms. The method can be used to predict the effect of targeted modifications on the yield of any set of metabolites (e.g., ethanol), thus providing a computational framework for synthetic metabolic engineering.     

Engineering synthetic bacterial consortia for enhanced desulfurization and revalorization of oil sulfur compounds

The 4S pathway is the most studied bioprocess for the removal of the recalcitrant sulfur of aromatic heterocycles present in fuels. It consists of three sequential functional units, encoded by the dszABCD genes, through which the model compound dibenzothiophene (DBT) is transformed into the sulfur-free 2-hydroxybiphenyl (2HBP) molecule. In this work, a set of synthetic dsz cassettes were implanted in Pseudomonas putida KT2440, a model bacterial “chassis” for metabolic engineering studies. The complete dszB1A1C1-D1 cassette behaved as an attractive alternative — to the previously constructed recombinant dsz cassettes — for the conversion of DBT into 2HBP. Refactoring the 4S pathway by the use of synthetic dsz modules encoding individual 4S pathway reactions revealed unanticipated traits, e.g., the 4S intermediate 2HBP-sulfinate (HBPS) behaves as an inhibitor of the Dsz monooxygenases, and once secreted from the cells it cannot be further taken up. That issue should be addressed for the rational design of more efficient biocatalysts for DBT bioconversions. In this sense, the construction of synthetic bacterial consortia to compartmentalize the 4S pathway into different cell factories for individual optimization was shown to enhance the conversion of DBT into 2HBP, overcome the inhibition of the Dsz enzymes by the 4S intermediates, and enable efficient production of unattainable high added value intermediates, e.g., HBPS, that are difficult to obtain using the current monocultures.     

In-silico-driven metabolic engineering of Pseudomonas putida for enhanced production of poly-hydroxyalkanoates

Here, we present systems metabolic engineering driven by in-silico modeling to tailor Pseudomonas putida for synthesis of medium chain length PHAs on glucose. Using physiological properties of the parent wild type as constraints, elementary flux mode analysis of a large-scale model of the metabolism of P. putida was used to predict genetic targets for strain engineering. Among a set of priority ranked targets, glucose dehydrogenase (encoded by gcd) was predicted as most promising deletion target. The mutant P. putida Δgcd, generated on basis of the computational design, exhibited 100% increased PHA accumulation as compared to the parent wild type, maintained a high specific growth rate and exhibited an almost unaffected gene expression profile, which excluded detrimental side effects of the modification. A second mutant strain, P. putida Δpgl, that lacked 6-phosphogluconolactonase, exhibited a substantially decreased PHA synthesis, as was also predicted by the model. The production potential of P. putida Δgcd was assessed in batch bioreactors. The novel strain showed an increase of the PHA yield (+80%), the PHA titer (+100%) and cellular PHA content (+50%) and revealed almost unaffected growth and diminished by-product formation. It was thus found superior in all relevant criteria towards industrial production. Beyond the contribution to more efficient PHA production processes at reduced costs that might replace petrochemical plastics in the future, the study illustrates the power of computational prediction to tailor microbial strains for enhanced biosynthesis of added-value compounds.     

Metabolic network reconstruction,growth characterization and 13C-metabolic flux analysis of the extremophile Thermus thermophilus HB8

Thermus thermophilus is an extremely thermophilic bacterium with significant biotechnological potential. In this work, we have characterized aerobic growth characteristics of T. thermophilus HB8 at temperatures between 50 and 85 °C, constructed a metabolic network model of its central carbon metabolism and validated the model using ¹³C-metabolic flux analysis (¹³C–MFA). First, cells were grown in batch cultures in custom constructed mini-bioreactors at different temperatures to determine optimal growth conditions. The optimal temperature for T. thermophilus grown on defined medium with glucose was 81 °C. The maximum growth rate was 0.25 h⁻¹. Between 50 and 81 °C the growth rate increased by 7-fold and the temperature dependence was described well by an Arrhenius model with an activation energy of 47 kJ/mol. Next, we performed a ¹³C-labeling experiment with [1,2-¹³C] glucose as the tracer and calculated intracellular metabolic fluxes using ¹³C–MFA. The results provided support for the constructed network model and highlighted several interesting characteristics of T. thermophilus metabolism. We found that T. thermophilus largely uses glycolysis and TCA cycle to produce biosynthetic precursors, ATP and reducing equivalents needed for cells growth. Consistent with its proposed metabolic network model, we did not detect any oxidative pentose phosphate pathway flux or Entner-Doudoroff pathway activity. The biomass precursors erythrose-4-phosphate and ribose-5-phosphate were produced via the non-oxidative pentose phosphate pathway, and largely via transketolase, with little contribution from transaldolase. The high biomass yield on glucose that was measured experimentally was also confirmed independently by ¹³C–MFA. The results presented here provide a solid foundation for future studies of T. thermophilus and its metabolic engineering applications.     

Engineering adipic acid metabolism in Pseudomonas putida

Bio-upcycling of plastics is an upcoming alternative approach for the valorization of diverse polymer waste streams that are too contaminated for traditional recycling technologies. Adipic acid and other medium-chain-length dicarboxylates are key components of many plastics including polyamides, polyesters, and polyurethanes. This study endows Pseudomonas putida KT2440 with efficient metabolism of these dicarboxylates. The dcaAKIJP genes from Acinetobacter baylyi, encoding initial uptake and activation steps for dicarboxylates, were heterologously expressed. Genomic integration of these dca genes proved to be a key factor in efficient and reliable expression. In spite of this, adaptive laboratory evolution was needed to connect these initial steps to the native metabolism of P. putida, thereby enabling growth on adipate as sole carbon source. Genome sequencing of evolved strains revealed a central role of a paa gene cluster, which encodes parts of the phenylacetate metabolic degradation pathway with parallels to adipate metabolism. Fast growth required the additional disruption of the regulator-encoding psrA, which upregulates redundant β-oxidation genes. This knowledge enabled the rational reverse engineering of a strain that can not only use adipate, but also other medium-chain-length dicarboxylates like suberate and sebacate. The reverse engineered strain grows on adipate with a rate of 0.35 ± 0.01 h⁻¹, reaching a final biomass yield of 0.27 ± 0.00 g_CDW g_adipate⁻¹. In a nitrogen-limited medium this strain produced polyhydroxyalkanoates from adipate up to 25% of its CDW. This proves its applicability for the upcycling of mixtures of polymers made from fossile resources into biodegradable counterparts.     

Reconfiguration of metabolic fluxes in Pseudomonas putida as a response to sub-lethal oxidative stress

As a frequent inhabitant of sites polluted with toxic chemicals, the soil bacterium and plant-root colonizer Pseudomonas putida can tolerate high levels of endogenous and exogenous oxidative stress. Yet, the ultimate reason of such phenotypic property remains largely unknown. To shed light on this question, metabolic network-wide routes for NADPH generation—the metabolic currency that fuels redox-stress quenching mechanisms—were inspected when P. putida KT2440 was challenged with a sub-lethal H₂O₂ dose as a proxy of oxidative conditions. ¹³C-tracer experiments, metabolomics, and flux analysis, together with the assessment of physiological parameters and measurement of enzymatic activities, revealed a substantial flux reconfiguration in oxidative environments. In particular, periplasmic glucose processing was rerouted to cytoplasmic oxidation, and the cyclic operation of the pentose phosphate pathway led to significant NADPH-forming fluxes, exceeding biosynthetic demands by ~50%. The resulting NADPH surplus, in turn, fueled the glutathione system for H₂O₂ reduction. These properties not only account for the tolerance of P. putida to environmental insults—some of which end up in the formation of reactive oxygen species—but they also highlight the value of this bacterial host as a platform for environmental bioremediation and metabolic engineering.Subject terms: Bacteriology, Environmental sciences     

Engineering an anaerobic metabolic regime in Pseudomonas putida KT2440 for the anoxic biodegradation of 1,3-dichloroprop-1-ene

Pseudomonas putida KT2440, a microbial cell factory of reference for industrial whole-cell biocatalysis, is unable to support biochemical reactions that occur under anoxic conditions, limiting its utility for a large number of relevant biotransformations. Unlike (facultative) anaerobes, P. putida resorts to NADH oxidation via an oxic respiratory chain and completely lacks a true fermentation metabolism. Therefore, it cannot achieve the correct balances of energy and redox couples (i.e., ATP/ADP and NADH/NAD⁺) that are required to sustain an O₂-free lifestyle. To overcome this state of affairs, the acetate kinase (ackA) gene of the facultative anaerobe Escherichia coli and the pyruvate decarboxylase (pdc) and alcohol dehydrogenase II (adhB) genes of the aerotolerant Zymomonas mobilis were knocked-in to a wild-type P. putida strain. Biochemical and genetic assays showed that conditional expression of the entire enzyme set allowed the engineered bacteria to adopt an anoxic regime that maintained considerable metabolic activity. The resulting strain was exploited as a host for the heterologous expression of a 1,3-dichloroprop-1-ene degradation pathway recruited from Pseudomonas pavonaceae 170, enabling the recombinants to degrade this recalcitrant chlorinated compound anoxically. These results underscore the value of P. putida as a versatile agent for biotransformations able to function at progressively lower redox statuses.     

Metabolism of the fast-growing bacterium Vibrio natriegens elucidated by 13C metabolic flux analysis

Vibrio natriegens is a fast-growing, non-pathogenic bacterium that is being considered as the next-generation workhorse for the biotechnology industry. However, little is known about the metabolism of this organism which is limiting our ability to apply rational metabolic engineering strategies. To address this critical gap in current knowledge, here we have performed a comprehensive analysis of V. natriegens metabolism. We constructed a detailed model of V. natriegens core metabolism, measured the biomass composition, and performed high-resolution ¹³C metabolic flux analysis (¹³C-MFA) to estimate intracellular fluxes using parallel labeling experiments with the optimal tracers [1,2−¹³C]glucose and [1,6−¹³C]glucose. During exponential growth in glucose minimal medium, V. natriegens had a growth rate of 1.70 1/h (doubling time of 24 min) and a glucose uptake rate of 3.90 g/g/h, which is more than two 2-fold faster than E. coli, although slower than the fast-growing thermophile Geobacillus LC300. ¹³C-MFA revealed that the core metabolism of V. natriegens is similar to that of E. coli, with the main difference being a 33% lower normalized flux through the oxidative pentose phosphate pathway. Quantitative analysis of co-factor balances provided additional insights into the energy and redox metabolism of V. natriegens. Taken together, the results presented in this study provide valuable new information about the physiology of V. natriegens and establish a solid foundation for future metabolic engineering efforts with this promising microorganism.     

Parallel labeling experiments validate Clostridium acetobutylicum metabolic network model for 13C metabolic flux analysis

In this work, we provide new insights into the metabolism of Clostridium acetobutylicum ATCC 824 obtained using a systematic approach for quantifying fluxes based on parallel labeling experiments and ¹³C-metabolic flux analysis (¹³C-MFA). Here, cells were grown in parallel cultures with [1-¹³C]glucose and [U-¹³C]glucose as tracers and ¹³C-MFA was used to quantify intracellular metabolic fluxes. Several metabolic network models were compared: an initial model based on current knowledge, and extended network models that included additional reactions that improved the fits of experimental data. While the initial network model did not produce a statistically acceptable fit of ¹³C-labeling data, an extended network model with five additional reactions was able to fit all data with 292 redundant measurements. The model was subsequently trimmed to produce a minimal network model of C. acetobutylicum for ¹³C-MFA, which could still reproduce all of the experimental data. The flux results provided valuable new insights into the metabolism of C. acetobutylicum. First, we found that TCA cycle was effectively incomplete, as there was no measurable flux between α-ketoglutarate and succinyl-CoA, succinate and fumarate, and malate and oxaloacetate. Second, an active pathway was identified from pyruvate to fumarate via aspartate. Third, we found that isoleucine was produced exclusively through the citramalate synthase pathway in C. acetobutylicum and that CAC3174 was likely responsible for citramalate synthase activity. These model predictions were confirmed in several follow-up tracer experiments. The validated metabolic network model established in this study can be used in future investigations for unbiased ¹³C-flux measurements in C. acetobutylicum.     

The interplay of the EIIA component of the nitrogen-related phosphotransferase system (PTS) of Pseudomonas putida with pyruvate dehydrogenase

Background

Pseudomonas putida KT2440 is endowed with a variant of the phosphoenolpyruvate-carbohydrate phosphotransferase system (PTS^Ntr), which is not related to sugar transport but believed to rule the metabolic balance of carbon vs. nitrogen. The metabolic targets of such a system are largely unknown.

Methods

Dielectric breakdown of P. putida cells grown in rich medium revealed the presence of forms of the EIIA^Ntr (PtsN) component of PTS^Ntr, which were strongly associated to other cytoplasmic proteins. To investigate such intracellular partners of EIIA^Ntr, a soluble protein extract of bacteria bearing an E epitope tagged version of PtsN was immunoprecipitated with a monoclonal anti-E antibody and the pulled-down proteins identified by mass spectrometry.

Results

The E1 subunit of the pyruvate dehydrogenase (PDH) complex, the product of the aceE gene, was identified as a major interaction partner of EIIA^Ntr. To examine the effect of EIIA^Ntr on PDH, the enzyme activity was measured in extracts of isogenic ptsN⁺/ptsN⁻P. putida strains and the role of phosphorylation was determined. Expression of PtsN and AceE proteins fused to different fluorescent moieties and confocal laser microscopy indicated a significant co-localization of the two proteins in the bacterial cytoplasm.

Conclusion

EIIA^Ntr down-regulates PDH activity. Both genetic and biochemical evidence revealed that the non-phosphorylated form of PtsN is the protein species that inhibits PDH.

General significance

EIIA^Ntr takes part in the node of C metabolism that checks the flux of carbon from carbohydrates into the Krebs cycle by means of direct protein–protein interactions with AceE. This type of control might connect metabolism to many other cellular functions. This article is part of a Special Issue entitled: Systems Biology of Microorganisms.     

Lipoate metabolism in Pseudomonas putida LP: thiolsulfinates of lipoate and bisnorlipoate.

Lipoate thiolsulfinate and two bisnorlipoate thiolsulfinates, as well as the previously identified products of β-oxidation (bisnorlipoate, tetranorlipoate, and β-hydroxybisnorlipoate), were isolated and identified as catabolites of [¹⁴C]lipoate from cultures of Pseudomonas putida LP, an organism capable of growth on lipoic acid as a sole source of carbon and sulfur. The newly identified metabolites were characterized by ion-exchange and paper chromatography and infrared, ultraviolet, and mass spectroscopies. Comparison of the isolated catabolites with synthetic standards implies that the lipoic thiolsulfinate isolated is the S-1 monoxide of 1,2-dithiolane-3-valeric acid; one bisnorlipoic thiolsulfinate isolated is the S-1 monoxide, the other apparently the S-2 monoxide. Metabolic studies with P. putida show that lipoate thiolsulfinate is taken up by this microorganism in an energy-dependent process, but less readily than lipoate; lipoate thiolsulfinate supports oxygen consumption in short-term experiments but does not support growth. These results are interpreted as meaning that the thiolsulfinates are “dead-end” metabolites, not intermediates in the sulfur metabolism of this organism. Lipoate thiolsulfinate is not detectably β-oxidized to bisnorlipoate thiolsulfinate under the usual culture conditions.