Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2021 Nov 22;10(12):3518–3526. doi: 10.1021/acssynbio.1c00440

Channeling Anabolic Side Products toward the Production of Nonessential Metabolites: Stable Malate Production in Synechocystis sp. PCC6803

Beatrice Battaglino †,‡, Wei Du §, Cristina Pagliano †, Joeri A Jongbloets §, Angela Re ‡, Guido Saracco †, Filipe Branco dos Santos §,*
PMCID: PMC8689693  PMID: 34808039

Abstract

graphic file with name sb1c00440_0006.jpg

Powered by (sun)light to oxidize water, cyanobacteria can directly convert atmospheric CO2 into valuable carbon-based compounds and meanwhile release O2 to the atmosphere. As such, cyanobacteria are promising candidates to be developed as microbial cell factories for the production of chemicals. Nevertheless, similar to other microbial cell factories, engineered cyanobacteria may suffer from production instability. The alignment of product formation with microbial fitness is a valid strategy to tackle this issue. We have described previously the “FRUITS” algorithm for the identification of metabolites suitable to be coupled to growth (i.e., side products in anabolic reactions) in the model cyanobacterium Synechocystis. sp PCC6803. However, the list of candidate metabolites identified using this algorithm can be somewhat limiting, due to the inherent structure of metabolic networks. Here, we aim at broadening the spectrum of candidate compounds beyond the ones predicted by FRUITS, through the conversion of a growth-coupled metabolite to downstream metabolites via thermodynamically favored conversions. We showcase the feasibility of this approach for malate production using fumarate as the growth-coupled substrate in Synechocystis mutants. A final titer of ∼1.2 mM was achieved for malate during photoautotrophic batch cultivations. Under prolonged continuous cultivation, the most efficient malate-producing strain can maintain its productivity for at least 45 generations, sharply contrasting with other producing Synechocystis strains engineered with classical approaches. Our study also opens a new possibility for extending the stable production concept to derivatives of growth-coupled metabolites, increasing the list of suitable target compounds.

Keywords: cyanobacteria, strain stability, growth-coupled production, malate production, thermodynamically favored reactions

Introduction

Cyanobacteria are promising microbial hosts that can be designed as cell factories to enable the development of sustainable production processes. Energized by (sun)light through their photosynthetic systems, engineered cyanobacteria can directly convert inorganic carbon dioxide into a number of chemicals such as alcohols and derived fuels (ethanol, butanol),1,2 chemical building blocks (isoprene, lactate, polyhydroxybutyrate),3−5 and fine chemicals.6,7

Classical strain improvement aims to rewire cellular metabolism to enhance the production of valuable native compounds or to endow cells with the ability to newly produce non-native ones.8,9 However, the diversion of carbon fluxes and energy toward the production of compounds of interest implemented by the introduction and/or modification of biosynthetic pathways interferes with cellular functionalities that are subject to multilevel (e.g., proteome-, metabolic-, population-, and environmental) constraints.10−14 The host cells could therefore experience notable pressure to opportunely reallocate resources between biomass precursors and target compound formation.15−18 Strong evolutionary pressure favors revertant cells with impaired production, challenging the feasibility and sustainability of the industrial exploitation of cyanobacteria as cell factories.19,20 Although negative results are seldom shown, the issue of production instability in cyanobacteria engineered to obtain valuable products such as ethylene, lactic acid, isopropanol, mannitol, and fatty acids has been reported in a handful of reports.3,21−23

Developing cyanobacterial-producing strains in biotechnological setups requires minimizing the negative impact of instability raised by the loss of fitness of engineered strains. Therefore, one approach is to make the growth of the engineered strain dependent on the synthesis of the compound that is to be produced (i.e., coupling the product forming pathway to microbial growth). In this way, the selection acting on engineered strains during their evolutionary trajectories allows producing strains to outcompete the nonproducing ones. This growth-coupled production has been implemented computationally using the FRUITS algorithm—that “Finds Reactions Usable in Tapping Side-products”—and experimentally validated in Synechocystis sp. PCC6803 (hereafter, Synechocystis) engineered to independently accumulate either acetate24 or fumarate.25 The biotechnologically relevant compounds obtainable through this approach are defined as the side products of anabolic pathways. Those side products can be accumulated by disrupting endogenous enzymatic reactions responsible for reintroducing them back into metabolism within a resource-efficient usage framework. In summary, the compounds resulting from this approach are growth-coupled since they are obligate side products of the formation of essential precursor metabolites. However, the number of candidates for implementing this growth-coupled strategy is limited to the metabolites that meet such criteria under the conditions for which it is determined using FRUITS. For Synechocystis, using model iJN67826 constrained to simulate photoautotrophic growth, this is restricted to nine metabolites.24

The metabolic strategy shown in the present study broadens the aforementioned limitations inherent to the FRUITS approach. Based on a combination of metabolic modeling and genetic engineering solutions, we show how we could exploit a growth-coupled metabolite previously identified in Synechocystis(25) to achieve the production of a metabolite positioned immediately downstream of the growth-coupled product itself. Specifically, we report the first stable malate-producing strain of Synechocystis obtained by exploiting the stoichiometrically growth-coupled metabolite fumarate as a supply for the accumulation of malate, the direct subproduct of fumarate in Synechocystis.

Malate is a C4-dicarboxylic acid with a wide range of industrial applications since it is used as an acidulant in the feed and food industry,27 building block for bioplastic and resin production and a component of drug delivery systems.28 Given the manifold applications and the increasing market demand, alternative and sustainable production systems for malate are highly advisable, for instance, in a photoautotrophic host.29

In this study, we created the markerless deletion Synechocystis mutants with impaired malate consumption pathways and further improved the production by overexpressing the enzyme that converts fumarate into malate, namely, fumarase C (FumC). The rationale behind the latter is that cells would then accumulate the target product malate because (i) the reactions that consume malate have been deleted and (ii) the formation of its direct precursor, fumarate, has been previously shown to be a strict requirement for cell growth.25 Moreover, we validated the stability of the malate production of the engineered strains over prolonged continuous cultivation. The strains obtained by implementing our new metabolic engineering strategy represent the first report of stable cyanobacterial production of malate via the conversion of the closest upstream growth-coupled compound, fumarate. It provides a starting point for the application of this novel engineering strategy to the next stages of industrial exploitation.

Results and Discussion

In Silico Simulation for Malate Production

In Synechocystis under photoautotrophic conditions, malate is exclusively produced from the precursor fumarate. The latter is mainly produced as a side product of urea and purine anabolism, and it is subsequently recycled via anaplerotic reactions. First, the FumC enzyme catalyzes the conversion of fumarate into malate (Figure 1a). Then, malate is converted either into oxaloacetate by malate dehydrogenase (Mdh) or into pyruvate by the malic enzyme (Me) through oxidative decarboxylation (Figure 1a). It has previously been demonstrated that fumarate can be produced in a growth-coupled manner, as also highlighted by the correlation existing between fumarate productivity and the growth rate of the engineered strain.25 Considering that for Synechocystis fumarate is a nonessential metabolite that can be produced in a growth-coupled fashion, it is conceivable that channeling fumarate toward the accumulation of its first derivative metabolite, malate, would not cause major fitness impairments. This hypothesis is supported by Flux Variability Analysis (FVA)30 of the genome-scale metabolic model (GSMM) of Synechocystis iJN678, which predicted that if enzymatic reactions are responsible for the consumption of malate are deleted, either the latter or fumarate would then be accumulated without affecting the growth capabilities. In particular, according to FVA, the single knockout of me or mdh genes would neither enhance malate accumulation nor affect Synechocystis growth (Tables 1 and 2), since one could take over the function of each other in simulations. On the other hand, the double deletion of the two target genes would induce the accumulation of malate up to a total amount of 0.860 mM gDW–1, with a small negative effect on cell growth (Table 2). The reaction catalyzed by the FumC is reversible and, according to a biochemical thermodynamic analysis performed through eQuilibrator,31 the conversion of fumarate into malate is allowed until the equilibrium of the reaction is reached, which corresponds to a molar ratio malate:fumarate of 4:1 under standard conditions (Figure 1b). For this reason, a residual amount of fumarate is always to be expected.

Figure 1.

Figure 1

Open in a new tab

Malate metabolism in Synechocystis and mutant construction. (a) Schematic representation of malate-producing and consuming pathways, as reported by the genome-scale metabolic model (GSMM).26 The genes deleted in this study are indicated in red, the overexpressed gene is indicated in green. (b) Thermodynamic analysis of the reaction catalyzed by the FumC. Calculations were obtained using eQuilibrator31 for the reaction fumarate + H2O ↔ malate. The dashed lines indicate the equilibrium point of the reaction (ΔGr = 0). (c) PCR confirmation of the strains constructed for markerless deletion of me and mdh. With the primers on each side of the upstream and downstream homologous region (H1 and H2 of ∼1 kb each), a markerless construct gave a PCR product of 1.5 and 1.8 kb for Δme and Δmdh, respectively. (d) PCR confirmation of the fumC overexpression strain. The integration of fumC gene in the neutral site slr0168 gave a PCR product of 4 kb. The genomic DNA of the WT and pBB1 were used as negative and positive controls, respectively.

Table 1. List of Plasmids and Strains Used in This Study.

plasmids and strains description reference
pFL-AN BioBrick “T” vector with AvrII and NheI on each side (42)
pWD42 AmprKmr, containing the selection cassette (43)
pWD71 pFL-AN derivative, Ampr, containing mdh upstream and downstream homologous regions this study
pWD72 pFL-AN derivative, AmprKmr, containing the selection cassette flanked by mdh upstream and downstream homologous regions this study
pWD73 pFL-AN derivative, Ampr, containing me upstream and downstream homologous regions this study
pWD74 pFL-AN derivative, AmprKmr, containing the selection cassette flanked by me upstream and downstream homologous regions this study
pHKH001 AmprKmr, integration vector at slr0168 genomic locus (3)
pBB1 pHKH001 derivative, AmprKmr, fumC (from Escherichia coli) expressed under the cpcBA promoter this study
Synechocystis sp. PCC6803 Synechocystis sp. PCC6803 wild type D. Bhaya
WD163 Synechocystis sp. PCC6803 mdh gene knockout mutant this study
WD169 Synechocystis sp. PCC6803 me gene knockout mutant this study
WD170 Synechocystis sp. PCC6803 mdh and me gene knockout mutant this study
WD198 Synechocystis sp. PCC6803 fumC overexpressing mutant this study
WD199 Synechocystis sp. PCC6803 mdh and me gene knockout, fumC overexpressing mutant this study
SAA023 Synechocystis sp. PCC6803 expressing L-ldh gene fromLactococcus lactis (36)
Open in a new tab

Table 2. Growth Rate and Product Yields for Malate and Fumarate Relative to Biomass Either Predicted by FBA or Experimentally Observed during the Exponential Phase in Wild-Type Synechocystis and Derivative Strains Impaired in Malate Consumption Reactionsa2.

  growth rate (μ, h–1)
 malate yield (Yp/x, mM gDW–1)
 fumarate yield (Yp/x, mM gDW–1)
strains model prediction measured model prediction measured model prediction measured
wild type 0.052 0.055 ± 0.001 0 0 0 0
Δme 0.052 0.052 ± 0.000 0 0.067 ± 0.019 0 0.274 ± 0.014
Δmdh 0.052 0.055 ± 0.000 0 0 0 0
ΔmeΔmdh 0.050 0.050 ± 0.001 0–0.860 0.607 ± 0.091 0–0.860 0.557 ± 0.002
Open in a new tab
a

FVA was used for evaluating the robustness of the network and predicting the expected range for malate and/or fumarate production. The biomass equation was always used as the primary objective function in all FBA and FVA simulations. The experimental data are referred to the batch cultivation reported in Figure 2.

To experimentally validate the feasibility of our approach by introducing the necessary genetic modifications, three markerless deletion mutants of Synechocystis as Δme, Δmdh, and ΔmeΔmdh were constructed (Figure 1c) and further characterized.

Malate Production of Δme, Δmdh, and ΔmeΔmdh in a Batch Photobioreactor

Growth and production capabilities of engineered and WT strains were monitored by photoautotrophic cultivation under continuous light illumination in the Multi-Cultivator. Δmdh grew similarly to the WT until the end of the experiment, while Δme and ΔmeΔmdh showed a measurably slower growth rate (0.052 and 0.050 h–1, respectively) in the exponential phase compared to the WT (0.055 h–1) (Figure 2a and Table 2). A similar negative effect on the growth rate due to the disruption of the gene encoding for the malic enzyme has been observed before.32 Comparison of fumarate and malate extracellular production between the WT and engineered strains allowed assessing the impact of deleting me and mdh genes on the metabolism of Synechocystis. As expected from previous work investigating the secretion of organic metabolites by Synechocystis grown under photoautotrophic conditions,33 neither fumarate nor malate was detected in significant amount in the supernatant of WT (Figure 2b,c). In contrast to the in silico analysis carried out, which predicted a null malate yield for the Δme strain (Table 2), this mutant was able to accumulate modest amounts of malate, suggesting that the conversion of malate into oxaloacetate by Mdh is less effective than the conversion of malate into pyruvate mediated by Me. This result is in partial agreement with 13C flux analysis showing that during continuous illumination, similar to that of our setup, the high levels of ATP generated inhibit the pyruvate kinase, which converts phosphoenolpyruvate (PEP) into pyruvate (Figure 1a),34,35 inducing cells to convert malate for supplying the necessary pyruvate. In this context, indeed, the higher activity of Me compared to Mdh could be functional to provide pyruvate, thus compensating the bottleneck through the pyruvate kinase. In accordance with model predictions (Table 2), both compounds were released in the extracellular broth by ΔmeΔmdh strain, which also showed higher yields compared to Δme strain (Figure 2b,c). For the ΔmeΔmdh strain, the maximum titers of malate and fumarate observed were to 1.2 and 1.0 mM, respectively, after 14 days of batch cultivation (Figures 2b,c and S1) and the maximum malate to fumarate molar ratio was about 1 (Figure 2d).

Figure 2.

Figure 2

Open in a new tab

Characterization of fumC overexpressing strains of Synechocystis compared to the WT and to the double knockout ΔmeΔmdh cultivated under continuous light in a photobioreactor. (a) Growth curves and extracellular concentration of (b) malate and (c) fumarate of different strains. (d) Profiles of the malate/fumarate molar ratio in Δme and ΔmeΔmdh strains. Values are the mean of at least two biological replicates. Error bars are standard error in panels (a)–(c) and combined standard uncertainty in panel (d). Concentration values below the detection limit of the analytical techniques used (i.e., 2 μM for malate and 20 μM for fumarate) are reported as zero in panels (b) and (c). Malate/fumarate molar ratios are not reported whenever either malate or fumarate concentration falls below the detection limit.

Overexpression of FumC for Improving Malate Accumulation

When considering the ΔmeΔmdh strain, which was the mutant with the highest malate production (Figures 2b and S1), the presence of a considerable amount of fumarate in the culturing broth (Figure 2c) suggests that the intracellular conversion of fumarate into malate catalyzed by FumC might be limited. The possible enzymatic bottleneck related to the low availability of the enzyme and/or restricted access to the substrate has been explored by overexpressing fumC. Two strains based on the WT strain and on the background of ΔmeΔmdh mutant, i.e., ΔNSI::fumC and ΔmeΔmdhΔNSI::fumC, have been engineered by inserting the fumC gene from E. coli in the chromosomal neutral site slr0168 (Figure 1d). The E. colifumC gene overexpressed is only 59.4% identical to the native one of Synechocystis, which avoids the occurrence of undesired recombination events.

The fumC overexpressing strains were cultured photoautotrophically in the Multi-Cultivator under the same conditions as the deletion mutants. The growth curve of ΔNSI::fumC was similar to that of the WT, whereas ΔmeΔmdhΔNSI::fumC showed a slower growth during the exponential phase when compared to the ΔNSI::fumC (Figure 3a). Conversely, ΔmeΔmdhΔNSI::fumC grew similarly to the ΔmeΔmdh in the first stages of growth, reaching the steady state earlier than the double mutant (Figure 3a). Concerning malate and fumarate extracellular production, the overexpression of fumC alone in Synechocystis was not sufficient to enhance the secretion of detectable quantities of neither of the two compounds (Figure 3b). On the other hand, ΔmeΔmdhΔNSI::fumC was capable of releasing in the extracellular environment higher titers of malate and lower titers of fumarate compared to the ΔmeΔmdh strain (Figure 3b; Figure S1 normalized by biomass content), determining a higher molar ratio of malate over fumarate in the extracellular broth throughout the entire cultivation (Figure 3c). When analyzing the malate productivity (calculated as the ratio between the variation of the malate production and the difference in biomass observed between two consecutive sampling points), the highest productivity was observed during the exponential phase for both the ΔmeΔmdhΔNSI::fumC and ΔmeΔmdh strains (Figure 3d). In this growth phase, the malate productivity of the overexpressing triple mutant was significantly higher than that of the double deletion mutant, once more indicating an improved ability of conversion of fumarate into malate in the ΔmeΔmdhΔNSI::fumC mutant.

Figure 3.

Figure 3

Open in a new tab

Characterization of fumC overexpressing strains of Synechocystis compared to the double knockout ΔmeΔmdh cultivated under continuous light in a photobioreactor. (a) Growth curves and (b) extracellular production of malate and fumarate of different strains. Variation of (c) the malate/fumarate molar ratio and (d) the malate productivity of ΔmeΔmdh and ΔmeΔmdhΔNSI::fumC strains. Dry weight concentration was calculated from OD730 measurements using a conversion factor of 148 mg L–1 OD–1 obtained in a similar setup.44 Values are the mean of at least two biological replicates. Error bars are standard errors in panels (a) and (b) and combined standard uncertainty in panels (c) and (d). Concentration values below the detection limit of the analytical techniques used (i.e., 2 μM for malate and 20 μM for fumarate) are reported as zero in panel (b). Malate/fumarate molar ratios are not reported whenever either malate or fumarate concentration falls below the detection limit.

Testing the Stability of ΔmeΔmdh and ΔmeΔmdhΔNSI::fumC

The engineering approach adopted to build the ΔmeΔmdh strain is based on the removal of the two consumption pathways of malate. Cyanobacterial strains engineered adopting an analogue deletion method have been shown to be stable and productive over dozens of generations.24,25 Consequently, also the ΔmeΔmdh strain is expected to be phenotypically stable. Most importantly, malate production should be stably maintained due to the grounds that its precursor, fumarate, is a growth-coupled metabolite. In the ΔmeΔmdhΔNSI::fumC strain, more cellular resources are involved in the synthesis of an extra amount of heterologous FumC, thus potentially introducing an additional protein burden to the cell. Therefore, assessing the phenotypic stability of this strain is of utmost pertinence.

Phenotypic stability of the most productive strains, ΔmeΔmdh and ΔmeΔmdhΔNSI::fumC, was assessed through a serial propagation experiment in which cultures were kept in exponential growth (OD730 between ∼0.6 and 1) over a period of ∼2 months to ensure a constant selective pressure.10 As a control strain, we used SAA023, a lactate-producing mutant of Synechocystis, which carries L-LDH that converts pyruvate into l-lactate.36 SAA023 has been shown to be unstable during prolonged culturing since spontaneous revertants with mutations in the ldh cassette become dominants after a few generations.15 The production instability of this strain is associated with the metabolic engineering strategy used that imposes a high trade-off between biomass growth and product formation because an essential metabolite of the central metabolism, here pyruvate, is depleted.

Throughout our laboratory evolution experiment, each culture was monitored between 42 and 50 generations, depending on the specific growth rate of the strain. ΔmeΔmdh and ΔmeΔmdhΔNSI::fumC maintained a similar ability to accumulate malate in the extracellular environment until the end of their serial propagations (Figure 4a), albeit with some fluctuations along time. In contrast, a clear drop in lactate extracellular concentration was recorded for the SAA023 strain (Figure 4b). To check whether this decrease in lactate was due to the presence of mutations in the ldh cassette, the gene encoding for the l-LDH and the upstream promoter sequence were amplified from 15 single colonies (five from each of the three biological replicates) and subjected to Sanger sequencing. Mutations in the ldh cassette were found in all of the analyzed sequences (Table S2 and Figure S2). Interestingly, the mutations were found only in the coding sequence, directly hampering the correct translation of the l-LDH protein. On the other hand, no mutations were found in the fumC cassette tested with an analogous approach.

Figure 4.

Figure 4

Open in a new tab

Malate and lactate extracellular production of different strains during the serial propagation experiment. (a) Malate production normalized by the OD730 of ΔmeΔmdh and ΔmeΔmdhΔNSI::fumC strains. (b) Lactate production normalized by the OD730 of SAA023 strain. The values are the mean of three biological replicates, the y error bars are combined uncertainties, and the x error bars are standard errors.

Taken together, these results show the promise of strategies aiming at producing compounds (here, malate) derived from growth-coupled metabolites (here, fumarate), rather than from central C-compounds necessary for growth.24,25 The nonappearance of mutations in the fumC gene indicates that (i) malate and fumarate, at these concentrations, do not pose a significant challenge to be transported out of the cell; (ii) nor is the accumulation of neither of them at these levels inhibitory for the cell; and (iii) at least in the absence of nutrient limitations, the protein overexpression per se does not seem to pose a sufficiently large fitness cost for the cell. Although this finding can be somewhat unexpected a priori, it is supported by the fact that no decrease in the growth capability (Figure 3a) and no loss in malate production (Figure 4a) were recorded for the ΔmeΔmdhΔNSI::fumC strain.

Conclusions

The feasibility of a metabolic engineering strategy aiming at exploiting a growth-coupled compound, fumarate, to drive the stable production of a downstream metabolite, malate, has been proved in this study. By implementing this strategy, a malate-producing strain of Synechocystis, ΔmeΔmdh, has been engineered. The conversion of fumarate into malate has been further improved by overexpressing heterologous FumC in the ΔmeΔmdh strain. The synthesis of an extra amount of this protein did not appear to compromise the fitness of the microorganism and did not result in a progressive loss in malate productivity due to phenotypic instability. These results indicate that the strategy proposed here in principle could be applied to other compounds deriving from a growth-coupled metabolite that is a side product of anabolism.

Methods

In Silico Simulation and Tools

The genome-scale metabolic model (GSMM) iJN678 was used to simulate light-limited photoautotrophic growth of Synechocystis. All of the details concerning the constraints imposed to perform metabolic flux analysis through flux variability analysis (FVA) and flux balance analysis (FBA)37 are reported in ref (24). FBA and FVA were performed using PySCeS-CBMPy38 (http://cbmpy.sourceforge.net) in combination with ILOG CPLEX Optimization Studio (IBM) under an academic license. Additional visualization of modeling simulations was carried out using a resource specifically developed for Synechocystis(39) available at the FAME online modeling environment.40

Strains and Growth Conditions

Molecular cloning was carried out in E. coli XL-1 blue grown at 37 °C in LB medium, either liquid in a flask with a shaking speed of 200 rpm or solidified in plates (by adding agar 1.5% w/v). To propagate a specific plasmid, the appropriate antibiotics were added to the culture medium, with a final concentration of 100 μg mL–1 for ampicillin and 50 μg mL–1 for kanamycin.

Glucose-tolerant wild-type (WT) strain of Synechocystis sp. PCC6803, obtained from D. Bhaya, Stanford University, Stanford, CA, was used in this study. All of the strains of Synechocystis (i.e., wild type and constructed mutants) have been grown in BG-11 medium41 supplemented with 10 mM TES–NaOH (pH 8.0) as a buffer, at 30 °C in a shaking incubator (Innova 43, New Brunswick Scientific) at 120 rpm, under constant moderate white-light illumination (∼40 μmol photons m–2 s–1, measured with a LI-250A light meter, LI-COR).

For Synechocystis mutants’ construction, the medium was supplemented with kanamycin (final concentration of 50 μg mL–1) or nickel sulfate (final concentration of 20 μM). Cell growth was monitored by measuring the optical density at 730 nm wavelength (OD730) in the spectrophotometer (Biochrom WPA, Lightwave II).

Batch Cultivation

Batch cultivation experiments were carried out in a Multi-Cultivator (MC1000-OD, PSI, Czech Republic), with emitted light intensity controlled through an LED panel, equipped with “cool-white” LEDs (PSI, CZ). BG-11 supplemented with 10 mM TES–NaOH (pH 8.0) was used as a medium to grow Synechocystis in the reactor. The cultures grew at 30 °C bubbled with a mixture (v/v) of 99% air and 1% CO2 supplied at a flow rate of 150 mL min–1. Precultures from shake flasks (OD730 ∼1–1.5) were used for inoculation in the Multi-Cultivator, working with a volume of ∼60 mL (area exposed to the light = 0.0028 m2, height of the water column = 10.32 cm) and initial OD730 of 0.05. After inoculation, a continuous irradiation of 30 μmol photons m–2 s–1 was applied until the cultures reached an OD730 of 0.6, then increased to 120 μmol photons m–2 s–1 and kept constant to the end of the experiment. Samples were collected daily and used for OD730 determination and product quantification in the extracellular broth.

Serial Propagation Experiments

Serial propagation experiments were performed in 100 mL Erlenmeyer flasks at 30 °C in a shaking incubator (Innova 43, New Brunswick Scientific) at 120 rpm, under constant moderate white-light illumination (∼40 μmol photons m–2 s–1, measured with a LI-250A light meter, LI-COR). Cultures have been grown in BG-11 medium41 supplemented with 10 mM TES–NaOH (pH 8.0) as a buffer and with kanamycin (50 μg mL–1) when necessary.

Precultures derived from the original stock of each strain (OD730 ∼1–1.5) were used for inoculation of the experimental batches. Every strain has been tested in triplicate. Every day, each culture at OD730 ∼1 has been diluted to OD730 ∼0.6 by keeping a constant volume of 20 mL. Samples were collected at periodic intervals before the dilution procedure and used for product quantification in the extracellular broth.

The Supporting Information provides details of plasmids and generation of Synechocystis mutants; quantification of extracellular fumarate, malate, and lactate; malate yield on biomass; sequencing of the ldh and fumC cassettes; and sequences of the main plasmid constructs.

Acknowledgments

F.B.d.S. and W.D. were supported by the European Union’s EFRO grant “Kansen voor West II.” The Netherlands Organization for Scientific Research (NWO) supported F.B.d.S. and J.A.J. through Solar-2-product grant 733 000 005. The research leading to this publication has received funding from the European Union’s Horizon 2020 Research, an innovation programme under Grant Agreement No. 760994 (ENGICOIN project).

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acssynbio.1c00440.

  • (i) Additional methods on the generation of plasmids and Synechocystis mutants; quantification of extracellular fumarate, malate, and lactate; and sequencing of the ldh and fumC cassettes; (ii) tables of the primers used in this study and the mutational analysis of the ldh expression cassette; (iii) figures of the malate yield on biomass and the identity and position of mutations found in the ldh cassete; and (iv) detailed sequence information of the cassetes used in different plasmids to obtain the Synechocystis mutants used in this study (PDF)

Author Contributions

∥ B.B. and W.D. contributed equally to this work. W.D. and F.B.d.S. first conceived this study. B.B. and W.D. carried out all experiments with the assistance of J.A.J. F.B.d.S., C.P., A.R., and G.S. provided additional supervision. B.B., W.D., and F.B.d.S. wrote the first draft. All authors read and approved the final manuscript.

The authors declare no competing financial interest.

Supplementary Material

sb1c00440_si_001.pdf (482.6KB, pdf)

References

  1. Gao Z.; Zhao H.; Li Z.; Tan X.; Lu X. Photosynthetic Production of Ethanol from Carbon Dioxide in Genetically Engineered Cyanobacteria. Energy Environ. Sci. 2012, 5, 9857–9865. 10.1039/C2EE22675H. [DOI] [Google Scholar]
  2. Liu X.; Miao R.; Lindberg P.; Lindblad P. Modular Engineering for Efficient Photosynthetic Biosynthesis of 1-Butanol from CO2 in Cyanobacteria. Energy Environ. Sci. 2019, 12, 2765–2777. 10.1039/c9ee01214a. [DOI] [Google Scholar]
  3. Angermayr S. A.; Paszota M.; Hellingwerf K. J. Engineering a Cyanobacterial Cell Factory for Production of Lactic Acid. Appl. Environ. Microbiol. 2012, 78, 7098–7106. 10.1128/AEM.01587-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Drosg B.; Fritz I.; Gattermayr F.; Silvestrini L. Photo-Autotrophic Production of Poly(Hydroxyalkanoates) in Cyanobacteria. Chem. Biochem. Eng. Q. 2015, 29, 145. 10.15255/CABEQ.2014.2254. [DOI] [Google Scholar]
  5. Lindberg P.; Park S.; Melis A. Engineering a Platform for Photosynthetic Isoprene Production in Cyanobacteria, Using Synechocystis as the Model Organism. Metab. Eng. 2010, 12, 70–79. 10.1016/j.ymben.2009.10.001. [DOI] [PubMed] [Google Scholar]
  6. Chen G.; Qu S.; Wang Q.; Bian F.; Peng Z.; Zhang Y.; Ge H.; Yu J.; Xuan N.; Bi Y.; et al. Transgenic Expression of Delta-6 and Delta-15 Fatty Acid Desaturases Enhances Omega-3 Polyunsaturated Fatty Acid Accumulation in Synechocystissp. PCC6803. Biotechnol. Biofuels 2014, 7, 32 10.1186/1754-6834-7-32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Menin B.; Santabarbara S.; Lami A.; Musazzi S.; Villafiorita Monteleone F.; Casazza A. P. Non-Endogenous Ketocarotenoid Accumulation in Engineered Synechocystis Sp. PCC 6803. Physiol. Plant 2019, 166, 403–412. 10.1111/ppl.12900. [DOI] [PubMed] [Google Scholar]
  8. Davy A. M.; Kildegaard H. F.; Andersen M. R. Cell Factory Engineering. Cell Syst. 2017, 4, 262–275. 10.1016/j.cels.2017.02.010. [DOI] [PubMed] [Google Scholar]
  9. Lee S. Y.; Kim H. U. Systems Strategies for Developing Industrial Microbial Strains. Nat. Biotechnol. 2015, 33, 1061–1072. 10.1038/nbt.3365. [DOI] [PubMed] [Google Scholar]
  10. Bachmann H.; Molenaar D.; Branco Dos Santos F.; Teusink B. Experimental Evolution and the Adjustment of Metabolic Strategies in Lactic Acid Bacteria. FEMS Microbiol. Rev. 2017, 41, S201–S219. 10.1093/femsre/fux024. [DOI] [PubMed] [Google Scholar]
  11. Burnap R. L. Systems and Photosystems: Cellular Limits of Autotrophic Productivity in Cyanobacteria. Front. Bioeng. Biotechnol. 2015, 3, 1 10.3389/fbioe.2015.00001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cano M.; Holland S. C.; Artier J.; Burnap R. L.; Ghirardi M.; Morgan J. A.; Yu J. Glycogen Synthesis and Metabolite Overflow Contribute to Energy Balancing in Cyanobacteria. Cell Rep. 2018, 23, 667–672. 10.1016/j.celrep.2018.03.083. [DOI] [PubMed] [Google Scholar]
  13. Janasch M.; Asplund-Samuelsson J.; Steuer R.; Hudson E. P. Kinetic Modeling of the Calvin Cycle Identifies Flux Control and Stable Metabolomes in Synechocystis Carbon Fixation. J. Exp. Bot. 2019, 70, 973–983. 10.1093/jxb/ery382. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Zheng X.-y.; O’Shea E. K. Cyanobacteria Maintain Constant Protein Concentration despite Genome Copy-Number Variation. Cell Rep. 2017, 19, 497–504. 10.1016/j.celrep.2017.03.067. [DOI] [PubMed] [Google Scholar]
  15. Du W.; Angermayr S. A.; Jongbloets J. A.; Molenaar D.; Bachmann H.; Hellingwerf K. J.; Branco dos Santos F. Nonhierarchical Flux Regulation Exposes the Fitness Burden Associated with Lactate Production in Synechocystis Sp. PCC6803. ACS Synth. Biol. 2017, 6, 395–401. 10.1021/acssynbio.6b00235. [DOI] [PubMed] [Google Scholar]
  16. Molenaar D.; van Berlo R.; de Ridder D.; Teusink B. Shifts in Growth Strategies Reflect Tradeoffs in Cellular Economics. Mol. Syst. Biol. 2009, 5, 323 10.1038/msb.2009.82. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Nielsen J.; Keasling J. D. Engineering Cellular Metabolism. Cell 2016, 164, 1185–1197. 10.1016/j.cell.2016.02.004. [DOI] [PubMed] [Google Scholar]
  18. Scott M.; Gunderson C. W.; Mateescu E. M.; Zhang Z.; Hwa T. Interdependence of Cell Growth and Gene Expression: Origins and Consequences. Science 2010, 330, 1099–1102. 10.1126/science.1192588. [DOI] [PubMed] [Google Scholar]
  19. Jones P. R. Genetic Instability in Cyanobacteria - An Elephant in the Room?. Front. Bioeng. Biotechnol. 2014, 2, 12 10.3389/fbioe.2014.00012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Du W.; Burbano P. C.; Hellingwerf K. J.; Branco dos Santos F.. Challenges in the Application of Synthetic Biology Toward Synthesis of Commodity Products by Cyanobacteria via “Direct Conversion” BT – Synthetic Biology of Cyanobacteria. In Advances in Experimental Medicine and Biology; Zhang W.; Song X., Eds.; Springer Singapore: Singapore, 2018; pp 3–26. [DOI] [PubMed] [Google Scholar]
  21. Kusakabe T.; Tatsuke T.; Tsuruno K.; Hirokawa Y.; Atsumi S.; Liao J. C.; Hanai T. Engineering a Synthetic Pathway in Cyanobacteria for Isopropanol Production Directly from Carbon Dioxide and Light. Metab. Eng. 2013, 20, 101–108. 10.1016/j.ymben.2013.09.007. [DOI] [PubMed] [Google Scholar]
  22. Jacobsen J. H.; Frigaard N. U. Engineering of Photosynthetic Mannitol Biosynthesis from CO2 in a Cyanobacterium. Metab. Eng. 2014, 21, 60–70. 10.1016/j.ymben.2013.11.004. [DOI] [PubMed] [Google Scholar]
  23. Yunus I. S.; Jones P. R. Photosynthesis-Dependent Biosynthesis of Medium Chain-Length Fatty Acids and Alcohols. Metab. Eng. 2018, 49, 59–68. 10.1016/j.ymben.2018.07.015. [DOI] [PubMed] [Google Scholar]
  24. Du W.; Jongbloets J. A.; van Boxtel C.; Pineda Hernández H.; Lips D.; Oliver B. G.; Hellingwerf K. J.; Branco dos Santos F. Alignment of Microbial Fitness with Engineered Product Formation: Obligatory Coupling between Acetate Production and Photoautotrophic Growth. Biotechnol. Biofuels 2018, 11, 38 10.1186/s13068-018-1037-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Du W.; Jongbloets J. A.; Guillaume M.; van de Putte B.; Battaglino B.; Hellingwerf K. J.; Branco dos Santos F. Exploiting Day- and Night-Time Metabolism of Synechocystis Sp. PCC 6803 for Fitness-Coupled Fumarate Production around the Clock. ACS Synth. Biol. 2019, 8, 2263–2269. 10.1021/acssynbio.9b00289. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Nogales J.; Gudmundsson S.; Knight E. M.; Palsson B. O.; Thiele I. Detailing the Optimality of Photosynthesis in Cyanobacteria through Systems Biology Analysis. Proc. Natl. Acad. Sci. U.S.A. 2012, 109, 2678–2683. 10.1073/pnas.1117907109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Thakker C.; Martínez I.; Li W.; San K. Y.; Bennett G. N. Metabolic Engineering of Carbon and Redox Flow in the Production of Small Organic Acids. J. Ind. Microbiol. Biotechnol. 2015, 42, 403–422. 10.1007/s10295-014-1560-y. [DOI] [PubMed] [Google Scholar]
  28. Zou X.; Cheng C.; Feng J.; Song X.; Lin M.; Yang S. T. Biosynthesis of Polymalic Acid in Fermentation: Advances and Prospects for Industrial Application. Crit. Rev. Biotechnol. 2019, 39, 408–421. 10.1080/07388551.2019.1571008. [DOI] [PubMed] [Google Scholar]
  29. Hu G.; Zhou J.; Chen X.; Qian Y.; Gao C.; Guo L.; Xu P.; Chen W.; Chen J.; Li Y.; et al. Engineering Synergetic CO2-Fixing Pathways for Malate Production. Metab. Eng. 2018, 47, 496–504. 10.1016/j.ymben.2018.05.007. [DOI] [PubMed] [Google Scholar]
  30. Mahadevan R.; Schilling C. H. The Effects of Alternate Optimal Solutions in Constraint-Based Genome-Scale Metabolic Models. Metab. Eng. 2003, 5, 264–276. 10.1016/j.ymben.2003.09.002. [DOI] [PubMed] [Google Scholar]
  31. Flamholz A.; Noor E.; Bar-Even A.; Milo R. EQuilibrator - The Biochemical Thermodynamics Calculator. Nucleic Acids Res. 2012, 40, D770–D775. 10.1093/nar/gkr874. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Bricker T. M.; Zhang S. L.; Laborde S. M.; Mayer P. R.; Frankel L. K.; Moroney J. V. The Malic Enzyme Is Required for Optimal Photo Autotrophic Growth of Synechocystis Sp Strain PCC 6803 under Continuous Light but Not under a Diurnal Light Regimen. J. Bacteriol. 2004, 186, 8144–8148. 10.1128/Jb.186.23.8144-8148.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Young J. D.; Shastri A. A.; Stephanopoulos G.; Morgan J. A. Mapping Photoautotrophic Metabolism with Isotopically Nonstationary 13C Flux Analysis. Metab. Eng. 2011, 13, 656–665. 10.1016/j.ymben.2011.08.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Wan N.; DeLorenzo D. M.; He L.; You L.; Immethun C. M.; Wang G.; Baidoo E. E. K.; Hollinshead W.; Keasling J. D.; Moon T. S.; et al. Cyanobacterial Carbon Metabolism: Fluxome Plasticity and Oxygen Dependence. Biotechnol. Bioeng. 2017, 114, 1593–1602. 10.1002/bit.26287. [DOI] [PubMed] [Google Scholar]
  35. Jazmin L. J.; Xu Y.; Cheah Y. E.; Adebiyi A. O.; Johnson C. H.; Young J. D. Isotopically Nonstationary 13C Flux Analysis of Cyanobacterial Isobutyraldehyde Production. Metab. Eng. 2017, 42, 9–18. 10.1016/j.ymben.2017.05.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Angermayr S. A.; Hellingwerf K. J. On the Use of Metabolic Control Analysis in the Optimization of Cyanobacterial Biosolar Cell Factories. J. Phys. Chem. B 2013, 117, 11169–11175. 10.1021/jp4013152. [DOI] [PubMed] [Google Scholar]
  37. Orth J. D.; Thiele I.; Palsson B. O. What Is Flux Balance Analysis?. Nat. Biotechnol. 2010, 28, 245–248. 10.1038/nbt.1614. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Olivier B. G.; Rohwer J. M.; Hofmeyr J. H. S. Modelling Cellular Systems with PySCeS. Bioinformatics 2005, 21, 560–561. 10.1093/bioinformatics/bti046. [DOI] [PubMed] [Google Scholar]
  39. Maarleveld T. R.; Boele J.; Bruggeman F. J.; Teusink B. A Data Integration and Visualization Resource for the Metabolic Network of Synechocystis Sp. PCC 6803. Plant Physiol. 2014, 164, 1111–1121. 10.1104/pp.113.224394. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Boele J.; Olivier B. G.; Teusink B. FAME, the Flux Analysis and Modeling Environment. BMC Syst. Biol. 2012, 6, 8 10.1186/1752-0509-6-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. van Alphen P.; Abedini Najafabadi H.; Branco dos Santos F.; Hellingwerf K. J. Increasing the Photoautotrophic Growth Rate of Synechocystis Sp. PCC 6803 by Identifying the Limitations of Its Cultivation. Biotechnol. J. 2018, 13, 1700764 10.1002/biot.201700764. [DOI] [PubMed] [Google Scholar]
  42. Zhu T.; Xie X.; Li Z.; Tan X.; Lu X. Enhancing Photosynthetic Production of Ethylene in Genetically Engineered Synechocystis Sp. PCC 6803. Green Chem. 2015, 17, 421–434. 10.1039/C4GC01730G. [DOI] [Google Scholar]
  43. Carpine R.; Du W.; Olivieri G.; Pollio A.; Hellingwerf K. J.; Marzocchella A.; Branco dos Santos F. Genetic Engineering of Synechocystis Sp. PCC6803 for Poly-β-Hydroxybutyrate Overproduction. Algal Res. 2017, 25, 117–127. 10.1016/j.algal.2017.05.013. [DOI] [Google Scholar]
  44. Du W.; Jongbloets J. A.; Pineda Hernandez H.; Bruggeman F. J.; Hellingwerf K. J.; Branco dos Santos F. Photonfluxostat: A Method for Light-Limited Batch Cultivation of Cyanobacteria at Different, yet Constant, Growth Rates. Algal Res. 2016, 20, 118–125. 10.1016/j.algal.2016.10.004. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

sb1c00440_si_001.pdf (482.6KB, pdf)

Articles from ACS Synthetic Biology are provided here courtesy of American Chemical Society

RESOURCES