A balanced ATP driving force module for enhancing photosynthetic biosynthesis of 3-hydroxybutyrate from CO2

Author’s Accepted Manuscript A balanced ATP driving force module for enhancing photosynthetic biosynthesis of 3hydroxybutyrate from CO2 Jason T. Ku, Ethan I. Lan www.elsevier.com/locate/ymben

PII: DOI: Reference:

S1096-7176(17)30351-8 https://doi.org/10.1016/j.ymben.2018.02.004 YMBEN1343

To appear in: Metabolic Engineering Received date: 11 September 2017 Revised date: 4 January 2018 Accepted date: 13 February 2018 Cite this article as: Jason T. Ku and Ethan I. Lan, A balanced ATP driving force module for enhancing photosynthetic biosynthesis of 3-hydroxybutyrate from CO2, Metabolic Engineering, https://doi.org/10.1016/j.ymben.2018.02.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

A balanced ATP driving force module for enhancing photosynthetic biosynthesis of 3-hydroxybutyrate from CO2 Jason T. Ku b, Ethan I. Lan a* a

b

Department of Biological Science and Technology Institute of Molecular Medicine and Bioengineering, National Chiao Tung University, Hsinchu 30010, Taiwan

*Corresponding Author. Phone: +886-3-571-2121 ext.59719; Fax: +886-3-5729288. E-mail: [email protected]. Address: 1001 Daxue Road, Hsinchu 30010, Taiwan Abstract Using engineered photoautotrophic microorganisms for the direct chemical synthesis from CO2 is an attractive direction for both sustainability and CO2 mitigation. However, the behaviors of non-native metabolic pathways may be difficult to control due to the different intracellular contexts between natural and heterologous hosts. While most metabolic engineering efforts focus on strengthening driving forces in pathway design to favor biochemical production in these organisms, excessive driving force may be detrimental to product biosynthesis due to imbalanced cellular intermediate distribution. In this study, an ATP-hydrolysis based driving force module was engineered into cyanobacterium Synechococcus elongatus PCC 7942 to produce 3-hydroxybutyrate (3HB), a valuable chemical feedstock for the synthesis of biodegradable plastics and antibiotics. However, while the ATP driving force module is effective for increasing product formation, uncontrolled accumulation of intermediate metabolites likely led to metabolic imbalance and thus to cell growth inhibition. Therefore, the ATP driving force module was reengineered by providing a reversible outlet for excessive carbon flux. Upon expression of this balanced ATP driving force module with 3HB biosynthesis, engineered strain produced 3HB with a cumulative titer of 1.2 g/L, a significant increase over the initial strain. This result highlighted the importance of pathway reversibility as an effective design strategy for balancing driving force and intermediate accumulation, thereby achieving a self-regulated control for increased net flux towards product biosynthesis.

1

Keywords: metabolic engineering, cyanobacteria, 3-hydroxybutyric acid, hydroxyalkanoate, driving force, acetoacetyl-CoA synthase

2

1.

Introduction Overdependence on fossil resources and the rapid rise of CO2 emission are two significant

challenges in the 21st century. Replacement of fossil resources with renewable bio-based fuels and chemicals produced by engineered microbes is a promising solution. Among the bio-based chemical production strategies, direct utilization of CO2 via autotrophic organisms for the synthesis of industrially relevant compounds is an attractive method to address both challenges. Metabolically engineered autotrophs can directly secrete products in which they are programed for. Recent advances include the photosynthetic conversion of CO2 to various industrially important chemicals (Angermayr et al., 2015; Oliver et al., 2016), syngas and hydrogen based chemical production (Bengelsdorf et al., 2016; Nybo et al., 2015), and direct electricity based microbial CO2 capture and utilization (Li et al., 2012). However, a common challenge for the metabolic engineering of these microbes is the unpredictability of heterologous pathways. The metabolic difference between natural and heterologous hosts is especially apparent in photoautotrophic microorganisms such as cyanobacteria. Cyanobacteria are photosynthetic prokaryotes that have relatively higher photosynthetic efficiencies and growth rates compared to terrestrial plants, and simpler genetic manipulation (Ducat et al., 2011; Pisciotta et al., 2010) compared to microalgae. With these characteristics, cyanobacteria are highly attractive as the next generation cell factories (Lai and Lan, 2015; Lau et al., 2015). Non-native metabolic pathways frequently fail to perform in heterologous hosts. In addition to overexpressing the required enzymes, other engineering approaches including, but not limited to, substrate replenishment (Lan and Wei, 2016), enzyme fusions (Formighieri and Melis, 2015), cofactor balance and preference (Li and Liao, 2013), oxygen tolerance (Lan et al., 2013), and driving force (Shen et al., 2011) are often necessary to achieve improved titers. Among these strategies, engineering synthetic driving force is a powerful tool for developing synthetic pathways and increasing productivity. An ATP hydrolysis step is an effective driving force for a pathway because it irreversibly pushes carbon flux into a particular direction. We previously constructed an ATP driving force module by coupling acetyl-CoA carboxylase with acetoacetyl-CoA synthase (Lan and Liao, 2012). Together, these two enzymes catalyze the irreversible formation of the otherwise unfavorable two acetyl-CoA condensation at the expense of an ATP. This ATP driving force is applicable to pathways downstream of acetoacetyl-CoA, including higher alcohols, aldehydes, acids, terpenoids, and many others. Using this modified pathway, butanol biosynthesis made possible in cyanobacteria (Lan and Liao, 2012). Recently, a reversed glyoxylate pathway was designed to allow conversion of succinate into two acetyl-CoA (Mainguet et al., 2013). The glyoxylate pathway is typically irreversible due to the release of large free energy from malate synthase catalyzed reaction. To reverse this pathway. An ATP driving force was incorporated by using malyl-CoA synthetase with malyl-CoA lyase. Together, these efforts largely rely on recruiting ATP utilizing enzymes into the designed pathways. However, it is rarely mentioned what happens should driving force becomes 3

excessive. 3-Hydroxybutyrate (3HB) is a monomer for poly-hydroxyalkanoates (PHA), which are natural biodegradable polyesters used as renewable plastics (Sudesh, 2012; Sudesh and Iwata, 2008). In addition, 3HB may be used for treating seizures (Nichels and d'Oultremont, 1992; Niesen, 2000) and as a precursor to carbapenem antibiotics and pheromones (Chiba and Nakai, 1985; Chiba and Nakai, 1987; Schnurrenberger et al., 1987). Currently, 3HB may be produced via degradation of the natural poly-3-hydroxybutyrate (PHB) through acid (Seebach et al., 1992) or enzymatic hydrolysis (Lee et al., 1999). Bacteria such as Ralstonia eutropha and Synechocystis PCC 6803 produce considerable amount of PHB in response to nitrogen and phosphate shortage (Gostomski and Bungay, 1996; Grousseau et al., 2013; Panda et al., 2006; Wu et al., 2001). However, the depolymerization of PHB requires a multi-phase process, and therefore lowering the overall efficiency of 3HB production (de Roo et al., 2002; Lee and Lee, 2003). As an alternative, metabolic engineering efforts focus on installing de novo pathways for 3HB biosynthesis (Lee et al., 2008; Lee and Lee, 2003). 3HB biosynthesis starts with the condensation of two acetyl-CoA into acetoacetyl-CoA (Figure 1) catalyzed by thiolase PhaA. Acetoacetyl-CoA is subsequently reduced to 3-hydroxybutyryl-CoA by acetoacetyl-CoA reductase PhaB. The CoA moiety is then removed by either thioesterase (Liu et al., 2007) or a two-step reaction catalyzed by phosphotransbutyrylase and butyrate kinase (Gao et al., 2002) . CoA removal via thioesterase was found more favorable and enabled a 3HB production titer of 4 g/L from Escherichia coli using glucose (Liu et al., 2007; Tseng et al., 2009). Further fed-batch process achieved 3HB production titer up to 12 g/L (Gao et al., 2002; Liu et al., 2007). This 3HB producing pathway was recently constructed in a model cyanobacterium Synechocystis PCC 6803 (Wang et al., 2013). Utilizing the natural ability of Synechocystis PCC 6803 for PHB production, the engineered strain produced 533 mg/L of 3HB in 21 days. However, unlike in E. coli, the 3HB production using Synechocystis PCC 6803 occurs after a week-long lag period during which the culture accumulates biomass. This lag phase in production is likely due to insufficient acetyl-CoA pool during normal growth conditions. 3HB production begins when nitrogen in the culture medium is depleted. Upon nitrogen limitation, acetyl-CoA pool increases (Anfelt et al., 2015) and counteracts the thermodynamic barrier of the thiolase catalyzed reaction which has a ΔG´° of +26 kJ/mol (Flamholz et al., 2011; Noor et al., 2013), enabling the downstream 3HB biosynthesis. However, this pathway may not be directly transferable to other cyanobacteria such as Synechococcus elongatus PCC 7942 or UTEX 2973 that have superior growth rates (Yu et al., 2015) because these strains do not naturally accumulate PHB. Furthermore, it has been shown that acetyl-CoA concentration in S. elongatus PCC 7942 is relatively low (Gao et al., 2016), potentially limiting the biosynthesis of 3HB. Therefore, additional engineering strategies are needed. To engineer a favorable 3HB production pathway in S. elongatus PCC 7942, we applied the ATP driving force module that irreversibly pushes carbon flux towards acetoacetyl-CoA. Contrary to 4

expectation, while the resulting strains significantly improved its 3HB producing ability, they suffered from severe growth inhibition which limited the overall productivity. We subsequently solved this growth inhibition through simultaneous expression of both thiolase and acetoacetyl-CoA synthase, thereby achieving a self-regulated and balanced driving force that provides an outlet for potential intermediate accumulation. The resulting strain achieved 1.2 g/L of cumulative 3HB titer, representing a significant improvement over the base strain and previous works. Contrary to common expectation that higher driving force leads to higher production, these results indicate that proper balance between driving force and intermediate accumulation is more essential for a successful production. Furthermore, here we show that pathway reversibility could be an effective design strategy in cases of excessive driving force.

2. Material and method 2.1. Chemicals and reagents All chemicals were purchased from Sigma-Aldrich (Saint Louis, MO), Amresco (Solon, OH), or J.T. Baker (Center Valley, PA) unless otherwise specified. Bacto agar was purchased from BD Bioscience. T4 DNA polymerase, KOD DNA polymerase, and KOD Xtreme DNA polymerase were purchased from New England Biolabs (Ipswich, MA) and EMD Millipore, respectively. 2.2. Cultured mediums and conditions E. coli XL1-Blue was used for molecular cloning throughout this study. E. coli was cultured in LB medium for plasmid construction and propagation. All S. elongatus PCC 7942 and its derivatives were grown on modified BG-11 (1.5 g/L NaNO3, 0.0272 g/L CaCl2·2H2O, 0.012 g/L ferric ammonium citrate, 0.001 g/L Na2EDTA, 0.040 g/L K2HPO4, 0.0361 g/L MgSO4·7H2O, 0.020 g/L Na2CO3, 1,000X trace mineral (1.43 g/L H3BO3, 0.905 g/L MnCl2·4H2O, 0.111 g/L ZnSO4·7H2O, 0.195 g/L Na2MoO4·2H2O, 0.0395 g/L CuSO4·5H2O, 0.0245 g/L Co(NO3)2·6H2O), 0.00882 g/L sodium citrate dihydrate) agar (1.5% w/v) plates for maintenance. For liquid cultivation, 40 mL of modified BG-11 containing 50 mM NaHCO3 was used in 250 mL screw cap flasks. Cultures were grown under 40 μE s-1m-2 light at 30°C on a rotary incubator shaker (Hi-point 600SR) set at 110 rpm. Kanamycin and spectinomycin were added to the culture medium or agar when necessary with a final concentration of 10 μg/mL and 20 μg/mL, respectively. 2.3. Plasmids and strains construction The plasmids and strains used in this study are listed in Table 1. For plasmid construction, ligation independent cloning method (Aslanidis and de Jong, 1990) using T4 polymerase was used throughout. DNA fragments typically have homologous regions to neighboring fragments so that they can be assembled together. linear DNA fragments were amplified by PCR using the primers listed in Table S1. The NSI vector (based on pAM2991 (Bustos and Golden, 1991)), containing 5

spectinomycin resistance, was amplified from pEL258 (Lan and Wei, 2016). nphT7, codon optimized phaB, tesB, and NSII vector, which containing kanamycin resistance, were cloned from pSR36 (Lan et al., 2013). The other three nphT7 genes from Kitasatospora, S. cinnamonensis, and Actinoplanes, were codon optimized and cloned from pEL112, pEL113, and pEL111, respectively. The sequences of these codon optimized gene are listed in supplementary information. Other genes used in this study were amplified from the respective genomic DNA of source organism. The coding regions of all plasmids were sequenced before they transformed into S. elongatus PCC 7942. All S. elongatus strains were constructed through homologous recombination of plasmid DNA into S. elongatus PCC 7942 genome at neutral site I (NSI) (Bustos and Golden, 1992) and neutral site II (NSII) (Andersson et al., 2000). PCR was used to confirm full segregation of chromosomal DNA (Figure S1). To transform S. elongatus PCC 7942, 50 ml of culture grown to OD around 0.6 was centrifuged and concentrated into 2 mL. Then, 300 μL of this concentrated culture was used to incubate with 2 μg of plasmid overnight at 30°C in the dark. The overnight cultures were then selected on BG-11 agar plates containing appropriate antibiotics. 2.4. 3-Hydroxybutyrate (3HB) production An 1 μL inoculation loop was used to inoculate a loopful of S. elongatus cells from BG-11 plates into fresh 40 mL of modified BG-11 containing 50 mM NaHCO3. The cultures were grown under the conditions described in above section. IPTG with a final concentration of 1 mM was used to induce the cultures when cell optical density OD730nm reached between 0.4 and 0.6. After induction, 1 mL of culture was removed and analyzed for cell growth and 3HB concentration. Immediately after sampling, 1 mL of BG-11 containing 500 mM NaHCO3, appropriate antibiotics, and 1 mM IPTG was added back into the cultures, ensuring proper carbon supply. All experiments were done in triplicates. 2.5. Cell growth and product quantification Cell growth was monitored through measuring optical density of cultures at 730 nm wavelength (OD730nm) using a Biotek epoch 2 microplate spectrophotometer. For determining 3HB concentration, culture samples were centrifuged at 10,000 x g for 10 minutes. The supernatants were then collected for product analysis using Agilent 1260 HPLC equipped with a photodiode array detector. The injection volume was 20 μL. The mobile phase consisted of 5 mM H2SO4 with a flow rate of 0.6 mL/min. Separation of metabolites was achieved using Agilent HiPlex-H (300 x 7.7 mm, 8μm) organic acid analysis column at 65°C. A Bio-Rad Micro-Guard Cation H guard column (30×4.6 mm) was connected in front of the analysis column. 3HB was monitored by photodiode array detector at 210 nm. Concentration of 3HB in the collected samples was determined by standard curve constructed from HPLC analysis of standard 3HB solutions with concentrations of 0.1 to 10 mM.

6

2.6. Enzyme assay 40 mL of modified BG-11 was inoculated with a loopful of S. elongatus cell as described above. The Cultures were induced with 0.1 mM IPTG at OD730nm 0.4 - 0.6 and subsequently harvested one day after by centrifuge at 4,500 x g for 20 minutes. The pellets were then resuspended in 1 mL of 100 mM Tris-HCl pH 8.0. Next, these concentrated cultures were lysed using mini-beadbeater-16 (Biospec) with 0.1 mm beads. Three repeats cycles of 45 seconds homogenization and 2 minutes on ice were used for cell lysis. The homogenates were then centrifuged at 10,000 x g for 15 minutes to remove glass beads and cell debris. Supernatant was collected as the soluble crude extract. Protein concentration was determined by Bradford protein assay with BSA standard (BioRad). Microplate reader Epoch2 (Biotek) was used for spectrophotometric measurements. The reaction mixture with a total volume of 220 μL containing 100 mM Tris-HCl (pH 8.0), 100 μM acetyl-CoA, 100 μM malonyl-CoA, 100 μM NADPH, and 20 ng crude extract. The reaction was initiated by the addition of acetyl-CoA. After incubation for 5 min, the contents of reaction mixture were measured by HPLC with a 3 hours interval to determine the concentration of the different CoA species and NADPH involved in 3HB biosynthesis. The different CoA species and NADPH were separated using Agilent 1260 HPLC equipped with an ZORBAX Eclipse Plus (4.6 x 150 mm, 3.5 μm) column. An Agilent ZORBAX Eclipse (4.6 x 12.5 mm, 5 µm) guard column was connected before the analysis column. The solvent gradient program is summarized in Table S2. Two mobile phases were: buffer A (100 mM NaH2PO4, 75 mM Sodium Acetate, pH 4.6) and buffer B (40% buffer A and 60% methanol). Flow rate was 1 mL/min. Injection volume was 10 µL. Product elution was monitored at 254 nm using photo diode array detector. Column temperature was kept at 30 °C. Concentration of the CoA species involved were calculated using a standard curve constructed with 1 to 100 μM of each CoA species. Total NADPH consumed equals to that of 3HB-CoA and 3HB concentration. Therefore, concentration of 3HB was calculated based on consumption of NADPH subtract 3HB-CoA concentration observed. 2.7. Intracellular ATP and NADPH quantification For measuring intracellular ATP and NADPH, cyanobacteria cultures were cultivated according to production condition as described above. Cells were harvested two days after induction. Cell amounts were normalized by having 40 ml of 1 OD730 culture for ATP measurement and 8 ml of 1 OD730 for NADPH measurement. The cells were then lysed using EZLys™ Bacterial Protein Extraction Reagent (BioVision, USA) by directly resuspending the cell pellet in 2 ml of the protein extraction reagent. Proteins in the solution, which would potentially interfere the assay, were subsequently removed using Deproteinizing Sample Preparation Kit (BioVision, USA). The resulting mixtures were assayed using the ATP Colorimetric/Fluorometric Assay Kit (BioVision, USA) and NADP/NADPH Quantitation Colorimetric Kit (BioVision, USA) for ATP and NADPH quantification, 7

respectively, accordingly to the manufacturer’s instructions. 3. Results and discussion 3.1. Expression of phaA, phaB and tesB enables 3HB production in S. elongatus PCC 7942 First, we introduced the 3HB producing pathway previously constructed in E. coli and Synechocystis PCC 6803 into S. elongatus PCC 7942. This pathway consists of thiolase (PhaA), acetoacetyl-CoA reductase (PhaB), and thioesterase (TesB). Both PhaA and PhaB were selected from R. eutropha, while TesB was from E. coli. A synthetic operon expressing the genes coding for these three enzymes, phaA, phaB, and tesB, was constructed and placed under the control of an IPTG inducible promoter PTrc. This operon was then transformed into S. elongatus PCC 7942 via homologous recombination at NSI. The resulting and fully segregated strain KU4 produced 120 mg/L of 3HB in 5 days (Figure 2A), representing the first demonstration of 3HB production by engineered S. elongatus PCC 7942. Similar to previous work done in Synechocystis PCC 6803 (Wang et al., 2013), a lag phase in production was observed. However, in this case, the production lag phase was significantly shorter as strain KU4 showed detectable 3HB concentration of around 38 mg/L two days after induction. Cell growth of strain KU4 was comparable to the wildtype PCC 7942, indicating no metabolic stress. This result was to our expectation that productivity is low due to the lack of driving force. 3.2. ATP driving force module increases 3-hydroxybutyrate production Next, we introduced an ATP driving force module in the 3HB production pathway to increase the favorability of the pathway. It has been previously shown that thiolase reaction is highly unfavorable with a ΔGʹ° of +26 kJ/mol (Stern et al., 1953). This large thermodynamic barrier may be overcome by constructing an ATP driving force module. Previously, we showed that incorporation of this ATP driving force module enabled butanol production from CO2 (Lan and Liao, 2012). Since the 3HB pathway overlaps with the butanol pathway (Figure 1), we expected that this ATP driving force module should also be beneficial for 3HB production. To incorporate the ATP driving force module into our pathway, we substituted phaA with nphT7, a gene encoding for acetoacetyl-CoA synthase. NphT7 together with native acetyl-CoA carboxylase drives the irreversible synthesis of acetoacetyl-CoA. In addition to the NphT7 from Streptomyces sp. CL-190 previously identified (Okamura et al., 2010), we also tested three additional putative NphT7 homologues from Kitasatospora setae, Streptomyces cinnamonensis, and Actinoplanes sp. A40644. These nphT7 genes were individually cloned into an operon with phaB and tesB, and then introduced into S. elongatus PCC 7942. As shown in Figure 2A, except for the strain expressing nphT7 from K. Setae, all other nphT7 expressing strains significantly improved their 3HB production compared to the strain expressing phaA. Furthermore, 65 mg/L of 3HB was observed from these strains on the first day while 3HB was undetected from strain expressing phaA. This result represents the first report that NphT7 homologues from S. cinnamonensis and Actinoplanes are functional acetoacetyl-CoA 8

synthases. The best performing strain KU5, expressing nphT7 from Streptomyces sp. CL190, produced around 250 mg/L of 3HB in 5 days, representing a two-fold improvement over the strain expressing phaA. Together, results showed that NphT7 was effective for pushing carbon flux into the 3HB pathway. Although the ATP driving force module was effective for pushing flux into 3HB pathway and increasing production, a significant and unexpected growth inhibition was observed (Figure 2B). While the OD730nm of wild type S. elongatus PCC 7942 reached 2.5 in 5 days, the strains expressing nphT7 reached only about 1. These nphT7 expressing strains stopped growing 2 days after induction, and eventually died as OD730nm decreases. The culture color turned from green to yellow and eventually completely bleached. nphT7 from K. Setae most likely was not expressed in S. elongatus PCC 7942 as strain KU10 expressing it exhibited no growth inhibition and did not produce 3HB. Growth inhibition upon introduction of the ATP driving force module was surprising. This module was previously used to drive butanol biosynthesis (Lan and Liao, 2012), but without growth retardation. The same NphT7 and PhaB were used between the previous and this study. By comparing the butanol and 3HB producing strains, we hypothesized that the growth limitation may be due to inefficient recycle of CoA. As NphT7 irreversibly synthesizes acetoacetyl-CoA, it is likely that a flux trap is formed if TesB cannot efficiently hydrolyze 3-hydroxybutyryl-CoA (3HB-CoA) and liberate CoA. As a result, CoA becomes limiting, halting the biosynthesis of acetyl-CoA from pyruvate and resulting in cell death. Similar behavior was observed for hydroxymethylglutaryl-CoA (HMG-CoA) toxicity in the engineering of terpenoid biosynthesis via mevalonate pathway (Martin et al., 2003). HMG-CoA reductase is less efficient than enzymes before it, thus leading to toxic accumulation of HMG-CoA. This difficulty was solved through overexpression of HMG-CoA reductase (Pitera et al., 2007) and coordinated scaffolding of HMG-CoA reductase with HMG-CoA synthase (Zheng et al., 2013). Here, we approached this difficulty via increasing thioesterase expression and constructing a reversible outlet to alleviate intermediate accumulation. 3.3. Overexpression of additional thioesterase improved 3-hydroxybutyrate titer To increase the overall activity of thioesterase, we introduced additional thioesterase genes into the engineered strains with ATP driving force module. Here we individually cloned pte1 from Saccharomyces cerevisiae, tesB (hereafter referred to as pptesB to avoid confusion with tesB from E. coli) from pseudomonas putida, and yciA from E. coli, into a synthetic operon driven by IPTG-inducible PLlacO1 promoter and inserted into NSII. PTE1 was selected for its demonstrated in vitro activity on 3HB-CoA (Seto et al., 2010). ppTesB was selected because of its activity on both butyryl-CoA and 3-hydroxyvaleryl-CoA (McMahon and Prather, 2014), suggesting the potential activity on 3HB-CoA. Similarly, YciA was chosen for its activity on butyryl-CoA and isobutyryl-CoA (Zhuang et al., 2008). The 3HB production and growth results are shown in Figure 3. 9

Regardless of which NphT7 is used, strains expressing pptesB in addition to nphT7, phaB, and tesB consistently improved their 3HB production. On the other hand, expression of the other thioesterase achieved only marginal improvement for 3HB production. The best strain KU12 achieved around 370 mg/L of 3HB in 5 days, representing a 75% improvement over its parent strain without additional thioesterase expression. This result represents the first report for ppTesB having activity towards the hydrolysis of 3HB-CoA. Corresponding to the improved 3HB production, the growth of strain KU12 was partially restored by about 70% compared to its parent strain. Despite the improved 3HB productivity and growth, growth inhibition remained severe compared to wildtype (Figure S2), indicating insufficient reduction of toxic intermediate accumulation. 3.4.Construction of a Balanced driving force module Next, we modified the ATP driving force module by incorporating a reversible outlet back to acetyl-CoA to form a self-regulated and balanced driving force. While improving overall thioesterase activity through enhancing tesB gene expression may be effective, manipulation of promoter and RBS strengths for tesB or pptesB expression is time-consuming and difficult, particularly for cyanobacteria. Therefore, as an alternative approach to reduce 3HB-CoA accumulation, we simultaneously expressed both NphT7 and PhaA in the same strain. We expected that by providing a reversible outlet for acetoacetyl-CoA back to acetyl-CoA, toxic intermediate accumulation would be alleviated. We placed phaA gene in the fourth and last position in the synthetic operon in NSI to avoid excessive acetoacetyl-CoA flux back to acetyl-CoA by PhaA. To test this strategy, we designed an in vitro enzyme assay to look at the kinetic distribution of the CoA species in the 3HB production pathway. Crude extracts from strains KU4, KU5 and KU7, representing the strain without, with, and with balanced ATP driving force module, respectively. We incubated these crude extracts individually in a solution containing 100 μM of acetyl-CoA, malonyl-CoA, and NADPH. We observed the concentration changes of these CoA species as a function of time. As shown in Figure 4, 3HB-CoA concentration was lower in the KU7 reaction than that of the KU5 at all time points, indicating the functional construction of a balanced ATP driving force module. Corresponding to the lower increase of 3HB-CoA concentration, acetyl-CoA drain was significantly lowered in the KU7 reaction. Interestingly, while acetyl-CoA and NADPH concentration decreased and free CoA concentration increased with time in the KU4 reaction, 3HB-CoA was not detected. This result indicated that NADPH drives the synthesis of 3HB. However, end concentration of NADPH in KU4 reaction is significantly higher than that of KU5 and KU7, indicating the lack of ATP driving force led to faster and unfavorable equilibrium for 3HB biosynthesis. Together these results showed the importance of driving force for pushing carbon flux down to 3HB-CoA. With the reversible outlet in the balanced ATP driving force module, less 3HB-CoA would accumulate, and less acetyl-CoA would be drained. Both of which may help to reduce the growth retardation of strain KU7 by lowering toxic effects. 10

3.5. Sustained growth and improved production via balanced driving force module After validation of the balanced ATP driving force module, we next evaluated its effect on 3HB production and cell growth. Using this modified driving force module, strain KU7 produced 204 mg/L of 3HB in 5 days (Figure 5A) and is not inhibited in growth (Figure 5B). While strain KU7 produced slightly less 3HB compared to strain KU5 expressing NphT7 (ATP driving force module), it’s 3HB titer is significantly higher than the 120 mg/L produced by strain KU4 expressing PhaA (without ATP driving force). These results show that co-expression of NphT7 and PhaA forms a self-controlled metabolic loop that pushes carbon flux towards 3HB-CoA while reliving any accumulation of 3HB-CoA by providing it a reversible outlet back to acetyl-CoA, resulting in sustained growth. Strain KU7 was observed with a continuous 3HB production up to around 650 mg/L (Figure 5A). Since ppTesB is effective in 3HB-CoA hydrolysis and improves 3HB production, we combined the balanced ATP driving force module with ppTesB expression to see if 3HB production would further increase. Strain KU21 was constructed to express nphT7, phaB, tesB, and phaA under PTrc promoter in NSI and pptesB under PLlacO1 promoter in NSII. As shown in Figure 5A, strain KU21 produced over 840 mg/L of 3HB, representing a 30% improvement over the strain without ppTesB expression. Accompanied with higher 3HB productivity, a slowed but sustained growth rate (Figure 5B) was observed for strain KU21. Byproduct analysis revealed that strain KU21 secreted small amounts of acetate, reaching over 50 mg/L (Figure 5C). Prior to day 10 post-induction, the level of acetate was below our detection limit. This may be the result of less specific thioesterase activity of ppTesB and may have caused extra burden to cell as it directly affected acetyl-CoA availability for growth. The exact cause to the slower growth of KU21 is unclear at this time. More detailed analysis, especially in the forms of targeted metabolic profiling or metabolomics may be useful to identify the exact reason. Nonetheless, the sustainable growth promoted continuous production of 3HB. Due to sampling and feeding, dilutions were made to the culture, leading to inevitable halt in observed titer increase. Accounting the dilutions made, the cumulative titer of 3HB produced by KU21 over the period tested reached 1.22 g/L (Figure 5D). Average productivity was about 44 mg/L/d with peak productivity occurring at day 2 post-induction with 91 mg/L/d (Figure 5D). Together, these results show that the balanced ATP driving force module is effective for maintaining sustained growth and aids in overall improvement in 3HB production. This strategy may also be applicable to other metabolic engineering cases where toxic intermediate accumulates due to unbalanced fluxes. 3.6. Analysis of intracellular ATP and NADPH concentration Although the balanced ATP driving force module presented in this work aided in increasing 3HB production, it formed a futile cycle. To evaluate the effect of this driving force module on 11

intracellular ATP balance, we measured ATP concentration between strains KU4 expressing no ATP driving force, KU5 expressing ATP driving force, KU7 expressing the balanced version of driving force module, and KU21 expressing the balanced driving force module with additional thioesterase and also the best 3HB producing strain in this work. Interestingly, those strains expressing either ATP driving force module or its balanced version all contained a slightly higher intracellular ATP concentration compared to the strain without driving force (Figure 6). Intracellular ATP concentration increased with the amount of 3HB produced. To gain a better understanding of potential reasons why ATP concentration slightly increased, we measured intracellular NADPH. It has been previously documented that extra consumption of NADPH increases the photosynthetic efficiency and the concentration of ATP in cyanobacteria (Zhou et al., 2016). We expected the production of 3HB would lead to a lower level of intracellular NADPH because it is used in the biosynthesis of 3HB. As indicated in figure 6, with the exception of KU21, the strains containing driving force modules exhibited small, but visible, lowered amounts of intracellular NADPH levels. Strain KU21, expressing both the balanced ATP driving force module and extra thioesterase, did not exhibit a significant drop in NADPH level but has the highest increase in ATP concentration. The exact cause of this behavior is unclear at this point. However, it is likely that due to better recycling of CoA through the expression of additional thioesterase, pyruvate dehydrogenase activity may be increased, leading to formation of more reducing equivalents. The detailed mechanism requires further metabolic profiling of the CoA species and relevant upstream precursors to elucidate. Here, these results indicated that while the balanced ATP driving force module represented a futile cycle, it has minimal negative impact on the intracellular ATP concentration for cyanobacteria. Therefore, it is a viable strategy for alleviating bottleneck accumulation of toxic intermediate 3HB-CoA. 4.

Conclusion While majority of the pathway engineering efforts focuses on enhancing driving force, excessive driving force could to accumulation of intermediate, causing unbalanced fluxes and resulting in growth inhibition. The most obvious solution to intermediate accumulation would be to increase the flux of downstream steps. However, this task is especially difficult for organisms without ample genetic tools such as cyanobacteria. Therefore, in this study, we modified the ATP driving force module such that, by incorporating a reversible outlet in the driving force module, the module becomes self-regulated. It retains the ability to push downstream towards product formation, meanwhile can return excessive flux back upstream when intermediate accumulates. Using this balanced ATP driving force module, we achieved the highest 3HB production of 840 mg/L observed and 1.2 g/L cumulative titer reported in literature thus far.

Acknowledgement This work was supported by the Ministry of Science and Technology, R.O.C., Taiwan (MOST) 12

through grant MOST-106-3113-E-007-002 and MOST-106-2221-E-009-128-MY2.

Reference Andersson, C. R., Tsinoremas, N. F., Shelton, J., Lebedeva, N. V., Yarrow, J., Min, H., Golden, S. S., 2000. Application of bioluminescence to the study of circadian rhythms in cyanobacteria. Methods in enzymology. 305, 527-542. Anfelt, J., Kaczmarzyk, D., Shabestary, K., Renberg, B., Rockberg, J., Nielsen, J., Uhlen, M., Hudson, E. P., 2015. Genetic and nutrient modulation of acetyl-CoA levels in Synechocystis for n-butanol production. Microb Cell Fact. 14, 167. Angermayr, S. A., Gorchs Rovira, A., Hellingwerf, K. J., 2015. Metabolic engineering of cyanobacteria for the synthesis of commodity products. Trends in biotechnology. 33, 352-61. Aslanidis, C., de Jong, P. J., 1990. Ligation-independent cloning of PCR products (LIC-PCR). Nucleic Acids Res. 18, 6069-6074. Bengelsdorf, F. R., Poehlein, A., Linder, S., Erz, C., Hummel, T., Hoffmeister, S., Daniel, R., Durre, P., 2016. Industrial Acetogenic Biocatalysts: A Comparative Metabolic and Genomic Analysis. Front Microbiol. 7. Bustos, S. A., Golden, S. S., 1991. Expression of the Psbdii Gene in Synechococcus Sp Strain-Pcc-7942 Requires Sequences Downstream of the Transcription Start Site. Journal of bacteriology. 173, 7525-7533. Bustos, S. A., Golden, S. S., 1992. Light-regulated expression of the psbD gene family in Synechococcus sp. strain PCC 7942: evidence for the role of duplicated psbD genes in cyanobacteria. Molecular and General Genetics MGG. 232, 221-230. Chiba, T., Nakai, T., 1985. A synthetic approach to (+)-thienamycin from methyl (R)-3-hydroxybutanoate. A new entry to (3R, 4R)-3-((R)-1-hydroxyethyl)-4-acetoxy-2-azetidinone. Chemistry Letters. 651-654. Chiba, T., Nakai, T., 1987. A new synthetic approach to the carbapenem antibiotic PS-5 from ethyl (S)-3-hydroxybutanoate. Chemistry Letters. 2187-2188. de Roo, G., Kellerhals, M. B., Ren, Q., Witholt, B., Kessler, B., 2002. Production of chiral R‐3‐ hydroxyalkanoic acids and R‐3‐hydroxyalkanoic acid methylesters via hydrolytic degradation of polyhydroxyalkanoate synthesized by pseudomonads. Biotechnol Bioeng. 77, 717-722. Ducat, D. C., Way, J. C., Silver, P. A., 2011. Engineering cyanobacteria to generate high-value products. Trends in biotechnology. 29, 95-103. Flamholz, A., Noor, E., Bar-Even, A., Milo, R., 2011. eQuilibrator—the biochemical thermodynamics calculator. Nucleic Acids Res. 40, D770-D775. Formighieri, C., Melis, A., 2015. A phycocyanin center dot phellandrene synthase fusion enhances recombinant protein expression and beta-phellandrene (monoterpene) hydrocarbons production in Synechocystis (cyanobacteria). Metabolic engineering. 32, 116-124. 13

Gao, H.-J., Wu, Q., Chen, G.-Q., 2002. Enhanced production of D-(−)-3-hydroxybutyric acid by recombinant Escherichia coli. Fems Microbiol Lett. 213, 59-65. Gao, X., Gao, F., Liu, D., Zhang, H., Nie, X., Yang, C., 2016. Engineering the methylerythritol phosphate pathway in cyanobacteria for photosynthetic isoprene production from CO2. Energy & Environmental Science. 9, 1400-1411. Gostomski, P. A., Bungay, H. R., 1996. Effect of Glucose and NH4+ Levels on Poly (β‐hydroxybutyrate) Production and Growth in a Continuous Culture of Alcaligenes eutrophus. Biotechnol Progr. 12, 234-239. Grousseau, E., Blanchet, E., Déléris, S., Albuquerque, M. G., Paul, E., Uribelarrea, J.-L., 2013. Impact of sustaining a controlled residual growth on polyhydroxybutyrate yield and production kinetics in Cupriavidus necator. Bioresource technology. 148, 30-38. Lai, M. C., Lan, E. I., 2015. Advances in Metabolic Engineering of Cyanobacteria for Photosynthetic Biochemical Production. Metabolites. 5, 636-58. Lan, E. I., Liao, J. C., 2012. ATP drives direct photosynthetic production of 1-butanol in cyanobacteria. P Natl Acad Sci USA. 109, 6018-6023. Lan, E. I., Ro, S. Y., Liao, J. C., 2013. Oxygen-tolerant coenzyme A-acylating aldehyde dehydrogenase facilitates efficient photosynthetic n-butanol biosynthesis in cyanobacteria. Energy & environmental science. Lan, E. I., Wei, C. T., 2016. Metabolic engineering of cyanobacteria for the photosynthetic production of succinate. Metabolic engineering. 38, 483-493. Lau, N.-S., Matsui, M., Abdullah, A. A.-A., 2015. Cyanobacteria: photoautotrophic microbial factories for the sustainable synthesis of industrial products. BioMed research international. 2015. Lee, S.-H., Park, S. J., Lee, S. Y., Hong, S. H., 2008. Biosynthesis of enantiopure (S)-3-hydroxybutyric acid in metabolically engineered Escherichia coli. Applied microbiology and biotechnology. 79, 633-641. Lee, S. Y., Lee, Y., 2003. Metabolic engineering of Escherichia coli for production of enantiomerically pure (R)-(−)-hydroxycarboxylic acids. Applied and environmental microbiology. 69, 3421-3426. Lee, S. Y., Lee, Y., Wang, F., 1999. Chiral compounds from bacterial polyesters: sugars to plastics to fine chemicals. Biotechnol Bioeng. 65, 363-368. Li, H., Liao, J. C., 2013. Engineering a cyanobacterium as the catalyst for the photosynthetic conversion of CO2 to 1,2-propanediol. Microb Cell Fact. 12. Li, H., Opgenorth, P. H., Wernick, D. G., Rogers, S., Wu, T. Y., Higashide, W., Malati, P., Huo, Y. X., Cho, K. M., Liao, J. C., 2012. Integrated Electromicrobial Conversion of CO2 to Higher Alcohols. Science. 335, 1596-1596. Liu, Q., Ouyang, S.-P., Chung, A., Wu, Q., Chen, G.-Q., 2007. Microbial production of R-3-hydroxybutyric acid by recombinant E. coli harboring genes of phbA, phbB, and tesB. Applied microbiology and biotechnology. 76, 811-818. 14

Mainguet, S. E., Gronenberg, L. S., Wong, S. S., Liao, J. C., 2013. A reverse glyoxylate shunt to build a non-native route from C4 to C2 in Escherichia coli. Metabolic engineering. 19, 116-27. Martin, V. J., Pitera, D. J., Withers, S. T., Newman, J. D., Keasling, J. D., 2003. Engineering a mevalonate pathway in Escherichia coli for production of terpenoids. Nature biotechnology. 21, 796-802. McMahon, M. D., Prather, K. L., 2014. Functional screening and in vitro analysis reveal thioesterases with enhanced substrate specificity profiles that improve short-chain fatty acid production in Escherichia coli. Applied and environmental microbiology. 80, 1042-1050. Nichels, W., d'Oultremont, P., Pharmaceutical compositions containing a derivative of 3-hydroxybutanoic acid chosen from oligomers of this acid and esters of this acid or of these oligomers with 1, 3-butanediol. Google Patents, 1992. Niesen, C., 2000. Use of beta-hydroxybutyrate for treating a chronic seizure disorder such as generalized or partial epilepsy. PCT patent, WO200028985-A1. Noor, E., Haraldsdóttir, H. S., Milo, R., Fleming, R. M., 2013. Consistent estimation of Gibbs energy using component contributions. Plos Comput Biol. 9, e1003098. Nybo, S. E., Khan, N. E., Woolston, B. M., Curtis, W. R., 2015. Metabolic engineering in chemolithoautotrophic hosts for the production of fuels and chemicals. Metabolic engineering. 30, 105-120. Okamura, E., Tomita, T., Sawa, R., Nishiyama, M., Kuzuyama, T., 2010. Unprecedented acetoacetyl-coenzyme A synthesizing enzyme of the thiolase superfamily involved in the mevalonate pathway. Proceedings of the National Academy of Sciences of the United States of America. 107, 11265-70. Oliver, N. J., Rabinovitch-Deere, C. A., Carroll, A. L., Nozzi, N. E., Case, A. E., Atsumi, S., 2016. Cyanobacterial metabolic engineering for biofuel and chemical production. Curr Opin Chem Biol. 35, 43-50. Panda, B., Jain, P., Sharma, L., Mallick, N., 2006. Optimization of cultural and nutritional conditions for accumulation of poly-β-hydroxybutyrate in Synechocystis sp. PCC 6803. Bioresource technology. 97, 1296-1301. Pisciotta, J. M., Zou, Y., Baskakov, I. V., 2010. Light-dependent electrogenic activity of cyanobacteria. PloS one. 5, e10821. Pitera, D. J., Paddon, C. J., Newman, J. D., Keasling, J. D., 2007. Balancing a heterologous mevalonate pathway for improved isoprenoid production in Escherichia coli. Metabolic engineering. 9, 193-207. Schnurrenberger, P., Hungerbühler, E., Seebach, D., 1987. Total Synthesis of (+)‐Colletodiol from (S, S)‐Tartrate and (R)‐3‐Hydroxybutanoate. Liebigs Annalen der Chemie. 1987, 733-744. Seebach, D., Beck, A. K., Breitschuh, R., Job, K., 1992. Direct Degradation of the Biopolymer Poly [(R)‐3‐Hydroxybutyric Acid] to (R)‐3‐Hydroxybutanoic Acid and its Methyl Ester. Organic Syntheses. 39-39. 15

Seto, Y., Kang, J., Ming, L., Habu, N., Nihei, K.-i., Ueda, S., Maeda, I., 2010. Genetic replacement of tesB with PTE1 affects chain-length proportions of 3-hydroxyalkanoic acids produced through β-oxidation of oleic acid in Escherichia coli. J Biosci Bioeng. 110, 392-396. Shen, C. R., Lan, E. I., Dekishima, Y., Baez, A., Cho, K. M., Liao, J. C., 2011. Driving forces enable high-titer anaerobic 1-butanol synthesis in Escherichia coli. Applied and Environmental Microbiology. 77, 2905-15. Stern, J. R., Coon, M. J., Delcampillo, A., 1953. Acetoacetyl Coenzyme-a as Intermediate in the Enzymatic Breakdown and Synthesis of Acetoacetate. J Am Chem Soc. 75, 1517-1518. Sudesh, K., 2012. Polyhydroxyalkanoates from palm oil: biodegradable plastics. Springer Science & Business Media. Sudesh, K., Iwata, T., 2008. Sustainability of biobased and biodegradable plastics. CLEAN–Soil, Air, Water. 36, 433-442. Tseng, H.-C., Martin, C. H., Nielsen, D. R., Prather, K. L. J., 2009. Metabolic engineering of Escherichia coli for enhanced production of (R)-and (S)-3-hydroxybutyrate. Applied and environmental microbiology. 75, 3137-3145. Wang, B., Pugh, S., Nielsen, D. R., Zhang, W., Meldrum, D. R., 2013. Engineering cyanobacteria for photosynthetic production of 3-hydroxybutyrate directly from CO2. Metabolic engineering. 16C, 68-77. Wu, G., Wu, Q., Shen, Z., 2001. Accumulation of poly-β-hydroxybutyrate in cyanobacterium Synechocystis sp. PCC6803. Bioresource technology. 76, 85-90. Yu, J., Liberton, M., Cliften, P. F., Head, R. D., Jacobs, J. M., Smith, R. D., Koppenaal, D. W., Brand, J. J., Pakrasi, H. B., 2015. Synechococcus elongatus UTEX 2973, a fast growing cyanobacterial chassis for biosynthesis using light and CO2. Sci Rep-Uk. 5. Zheng, Y., Liu, Q., Li, L., Qin, W., Yang, J., Zhang, H., Jiang, X., Cheng, T., Liu, W., Xu, X., 2013. Metabolic engineering of Escherichia coli for high-specificity production of isoprenol and prenol as next generation of biofuels. Biotechnology for biofuels. 6, 57. Zhou, J., Zhang, F. L., Meng, H. K., Zhang, Y. P., Li, Y., 2016. Introducing extra NADPH consumption ability significantly increases the photosynthetic efficiency and biomass production of cyanobacteria. Metab Eng. 38, 217-227. Zhuang, Z., Song, F., Zhao, H., Li, L., Cao, J., Eisenstein, E., Herzberg, O., Dunaway-Mariano, D., 2008. Divergence of function in the hot dog fold enzyme superfamily: the bacterial thioesterase YciA†. Biochemistry. 47, 2789-2796.

16

Table 1 Strains and plasmids used in this study Strain PCC 7942

Relevant Genotype

Reference

Wild-type Synechococcus elongatus PCC 7942

PCC

KU4

Ptrc::phaA, tesB, phaB integrated at NSI in PCC 7942 genome

This work

KU5

Ptrc::nphT7, tesB, phaB integrated at NSI in PCC 7942 genome

This work

KU7

Ptrc::nphT7, tesB, phaB, phaA integrated at NSI in PCC 7942 genome

This work

KU9

Ptrc::npht7 (Kitasatospora), tesB, phaB integrated at NSI in PCC 7942 genome

This work

KU10

Ptrc::npht7 (S. cinnamonensis), tesB, phaB integrated at NSI in PCC 7942 genome

This work

KU11

Ptrc::npht7 (Actinoplanes), tesB, phaB integrated at NSI in PCC 7942 genome Ptrc::nphT7, tesB, phaB integrated at NSI and PLlacO1::pptesB integrated at NSII in PCC 7942 genome Ptrc::nphT7, tesB, phaB integrated at NSI and PLlacO1:: yciA integrated at NSII in PCC 7942 genome Ptrc::nphT7, tesB, phaB integrated at NSI and PLlacO1:: pte1 integrated at NSII in PCC 7942 genome Ptrc::npht7 (S. cinnamonensis), tesB, phaB integrated at NSI and PLlacO1::pptesB integrated at NSII in PCC 7942 genome Ptrc::npht7 (S. cinnamonensis), tesB, phaB integrated at NSI and PLlacO1:: yciA integrated at NSII in PCC 7942 genome Ptrc::npht7 (S. cinnamonensis), tesB, phaB integrated at NSI and PLlacO1:: pte1 integrated at NSII in PCC 7942 genome Ptrc::npht7 (Actinoplanes), tesB, phaB integrated at NSI and PLlacO1::pptesB integrated at NSII in PCC 7942 genome Ptrc::npht7 (Actinoplanes), tesB, phaB integrated at NSI and PLlacO1:: yciA integrated at NSII in PCC 7942 genome Ptrc::npht7 (Actinoplanes), tesB, phaB integrated at NSI and PLlacO1:: pte1 integrated at NSII in PCC 7942 genome Ptrc::nphT7, tesB, phaB, phaA integrated at NSI and PLlacO1::pptesB integrated at NSII in PCC 7942 genome Ptrc::phaA, tesB, phaB integrated at NSI and PLlacO1::pptesB integrated at NSII in PCC 7942 genome Ptrc::phaA, tesB, phaB integrated at NSI and PLlacO1:: yciA integrated at NSII in PCC 7942 genome Ptrc::phaA, tesB, phaB integrated at NSI and PLlacO1:: pte1 integrated at NSII in PCC 7942 genome Relevant Genotype

This work

Reference

pKU4

Ptrc::phaA, tesB, phaB

This work

pKU5

Ptrc::nphT7, tesB, phaB

This work

pKU7

Ptrc::nphT7, tesB, phaB, phaA

This work

pKU9

Ptrc::nphT7(Kitasatospora), tesB, phaB

This work

pKU10

Ptrc::nphT7(S. cinnamonensis), tesB, phaB

This work

pKU11

Ptrc::nphT7(Actinoplanes), tesB, phaB

This work

pKU15

PLlacO1::pptesB

This work

pKU16

PLlacO1::yciA

This work

pKU17

PLlacO1::pte1

This work

KU12 KU13 KU14 KU15 KU16 KU17 KU18 KU19 KU20 KU21 KU33 KU34 KU35 Plasmid

17

This work This work This work This work This work This work This work This work This work This work This work This work This work

ATP

acetyl-CoA

CoA

Calvin Cycle

acetyl-CoA

HCO3-

CO2

G3P

O

phaA

Photosynthesis O2

ATP

hν NADPH

accABCD

O -O

H2O O

O CoA

malonyl-CoA

nphT7

acetyl-CoA

NADPH

O CoA

acetoacetyl-CoA

OH O

OH

O O-

CoA

phaB

CO2

Balanced ATP driving force module

3-Hydroxybutyryl-CoA

tesB 3-Hydroxybutyrate

Butanol

Figure 1. Schematic for the construction of a balance driving force module to aid 3-hydroxybutyrate biosynthesis. Expression of phaA alone lacks driving force. Expression of nphT7 is an effective driving force, but causes toxic effects. Co-expression of both nphT7 and phaA creates a balanced driving force module which uses ATP to driving biosynthesis of acetoacetyl-CoA, but with a reversible outlet back to acetyl-CoA should acetoacetyl-CoA accumulates. Dashed lines represent multiple enzymatic stesp. Butanol pathway is shown as a reference.

18

B

300

3.5 3.0

250

2.5

200

OD730

3HB concentration (mg/L)

A

150 100

2.0 1.5 1.0

50

0.5 0.0

0 0

1 2 3 4 Time since induction (days)

PCC7942 KU9 (NphT7-K.set)

0

5 KU4 (phaA) KU10 (NphT7-S.cin)

1 2 3 4 Time since induction (days)

5

KU5 (NphT7-CL190) KU11 (NphT7-Act)

Figure 2. Time course for (A) 3-hydroxybutyrate production and (B) growth of strains expressing different NphT7 enzymes and PhaA. All strains express PhaB and TesB. Production was done by cultivation of engineered cyanobacteria under continuous 40 µE light with bicarbonate as carbon source. See material and methods for details.

19

2.5

OD730

2.0 1.5 1.0 0.5

nphT7 (CL190)

nphT7 (S. cin)

nphT7 (Act)

ppTesB YciA PTE1

o

o o

o o

o

phaA

o o

o

KU4 KU33 KU34 KU35

KU11 KU18 KU19 KU20

Acetoacetyl-CoA formation gene Additional thioesterase

Strain name

KU10 KU15 KU16 KU17

400 350 300 250 200 150 100 50 0 KU5 KU12 KU13 KU14

3HB concentration (mg/L)

0.0

o o

o

Figure 3. Growth and 3HB production by strains expressing PhaA or different NphT7, serving as ATP driving force module, and additional thioesterases. Data shown is from day 5 post induction. Detailed time course of this experiment is shown in Figure S2.

20

Malonyl-CoA

100 80 O

40

O

-O

20

CoA

O

Malonyl-CoA

CoA

0 0

8

Acetyl-CoA

12 16 20 24 28 Time (Hr)

NphT7

Acetoacetyl-CoA

1.0

concentration (uM)

4

PhaA

60 40 20

0

0.6

O

4

8 12 16 20 24 28 Time (Hr)

O

0.4

CoA

0.2 0

4

8 12 16 20 24 28 Time (Hr)

NADPH

PhaB

3HB-CoA

20

CoA

200

Acetoacetyl-CoA + CoA-SH

0.0

concentration (μM)

80

0

0.8

15

150 100

50 0

OH O

10

0

4

8

CoA

12 16 20 24 28 Time (Hr)

3-Hydroxybutyryl-CoA

5 0

NADPH

120 4

8

12 16 20 24 28 Time (Hr)

TesB

concentration (uM)

0

3-Hydroxybutyrate

70 concentration (uM)

Acetyl-CoA

100 concentration (uM)

60

concentration (uM)

concentration (uM)

120

60 50

OH

O

40

O-

30

3-Hydroxybutyrate

20 10

100 80 60 40

20 0 0

4

8

12 16 20 24 28 Time (Hr)

0 0

4

8

12 16 20 24 28 Time (Hr)

Figure 4. In vitro reaction using crude extracts of KU4, KU5, and KU7, representing strain without, with, and with balanced ATP driving force module, respectively, to compare the effect of driving force module on the distribution of CoA species, 3HB, and NADPH involved in the 3HB production pathway.

21

3.0 2.5 2.0

OD730

600

1.5

400

1.0

200

0.5

0

0.0

4

KU7 KU21

60 50

40 30 20 10 0

0

8 12 16 20 24 28 Time since induction (days)

70

C

KU7 KU21

4 8 12 16 20 24 Time since induction (days)

28

0

4 8 12 16 20 24 Time since induction (days)

Cumulative 3HB (mg/L)

1,400

100 90 80 70 60 50 40 30 20 10 0

1,200 1,000 800

600 400 200 0 0

1

2

3

4

5

6

7

8

28

Productivity (mg/L/day)

800

0

D

B

KU7 KU21

Acetate (mg/L)

1,000

Observed 3HB titer (mg/L)

A

9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Time since induction (days)

Figure 5. Long term production by engineered strains expressing the balanced ATP driving force module. Strain KU7 expresses nphT7, phaB, tesB, and phaA. Strain KU21 expresses the same genes as KU7 but with additional pptesB expression. A) Observed titer, B) growth profile, and C) byproduct acetate secretion of strain KU7 and KU21. D) Cumulative titer and productivity of the best producer in this study KU21. Cumulative titer accounts for dilutions made to the culture during feeding and represents the sum of the productivities.

22

Strains NADPH ATP

No driving force KU4 ATP driving force KU5 Balanced driving force KU7 Balanced driving force KU21 With additional thioesterase 0.0

0.5 1.0 1.5 Relative NADPH & ATP concentration

2.0

Figure 6. Intracellular concentration of ATP and NADPH. All strains express phaB and tesB. KU4 additionally expresses phaA. KU5 additionally expresses nphT7. KU7 additionally expresses nphT7 and phaA. KU21 additionally expresses nphT7, phaA, and pptesB.

Highlights 

Synthetic 3-hydroxybutyrate producing pathway was engineered in Synechococcus elongatus PCC 7942



ATP driving force composed of acetyl-CoA carboxylase and acetoacetyl-CoA synthase inhibited cellular growth



ATP driving force was redesigned with reversibility between acetoacetyl-CoA and acetyl-CoA



Resulting strain produced cumulative titer of 1.3 g/L 3HB with the balanced driving force module

23