Engineering a thermostable highly active glucose 6-phosphate dehydrogenase and its application to hydrogen production in vitro.

Glucose 6-phosphate dehydrogenase (G6PDH) is one of the most important dehydrogenases responsible for generating reduced NADPH for anabolism and is also the rate-limiting enzyme in the Entner-Doudoroff pathway. For in vitro biocatalysis, G6PDH must possess both high activity and good thermostability due to requirements of efficient use and low expense of biocatalyst. Here, we used directed evolution to improve thermostability of the highly active G6PDH from Zymomonas mobilis. Four generations of random mutagenesis and Petri-dish-based double-layer screening evolved the thermolabile wild-type enzyme to the thermostable mutant Mut 4-1, which showed a more than 124-fold increase in half-life time (t1/2) at 60 °C, a 3.4 °C increase in melting temperature (T m ), and a 5 °C increase in optimal temperature (Topt), without compromising the specific activity. In addition, the thermostable mutant was conducted to generate hydrogen from maltodextrin via in vitro synthetic biosystems (ivSB), gaining a more than 8-fold improvement of productivity rate with 76% of theoretical yield at 60 °C. Thus, the engineered G6PDH has been shown to effectively regenerate NADPH at high temperatures and will be applicable for NAD(P)H regeneration in numerous in vitro biocatalysis applications.

An activity transition from NADH dehydrogenase to NADH oxidase during protein denaturation.

A decrease in the specific activity of an enzyme is commonly observed when the enzyme is inappropriately handled or is stored over an extended period. Here, we reported a functional transition of an FMN-bound diaphorase (FMN-DI) that happened during the long-term storage process. It was found that FMN-DI did not simply lose its ß-nicotinamide adenine diphosphate (NADH) dehydrogenase activity after a long-time storage, but obtained a new enzyme activity of NADH oxidase. Further mechanistic studies suggested that the alteration of the binding strength of an FMN cofactor with a DI protein could be responsible for this functional switch of the enzyme.

Advanced water splitting for green hydrogen gas production through complete oxidation of starch by in vitro metabolic engineering.

Starch is a natural energy storage compound and is hypothesized to be a high-energy density chemical compound or solar fuel. In contrast to industrial hydrolysis of starch to glucose, an alternative ATP-free phosphorylation of starch was designed to generate cost-effective glucose 6-phosphate by using five thermophilic enzymes (i.e., isoamylase, alpha-glucan phosphorylase, 4-α-glucanotransferase, phosphoglucomutase, and polyphosphate glucokinase). This enzymatic phosphorolysis is energetically advantageous because the energy of α-1,4-glycosidic bonds among anhydroglucose units is conserved in the form of phosphorylated glucose. Furthermore, we demonstrated an in vitro 17-thermophilic enzyme pathway that can convert all glucose units of starch, regardless of branched and linear contents, with water to hydrogen at a theoretic yield (i.e., 12 H2 per glucose), three times of the theoretical yield from dark microbial fermentation. The use of a biomimetic electron transport chain enabled to achieve a maximum volumetric productivity of 90.2mmol of H2/L/h at 20g/L starch. The complete oxidation of starch to hydrogen by this in vitro synthetic (enzymatic) biosystem suggests that starch as a natural solar fuel becomes a high-density hydrogen storage compound with a gravimetric density of more than 14% H2-based mass and an electricity density of more than 3000Wh/kg of starch.

Systematic comparison of co-expression of multiple recombinant thermophilic enzymes in Escherichia coli BL21(DE3).

The precise control of multiple heterologous enzyme expression levels in one Escherichia coli strain is important for cascade biocatalysis, metabolic engineering, synthetic biology, natural product synthesis, and studies of complexed proteins. We systematically investigated the co-expression of up to four thermophilic enzymes (i.e., α-glucan phosphorylase (αGP), phosphoglucomutase (PGM), glucose 6-phosphate dehydrogenase (G6PDH), and 6-phosphogluconate dehydrogenase (6PGDH)) in E. coli BL21(DE3) by adding T7 promoter or T7 terminator of each gene for multiple genes in tandem, changing gene alignment, and comparing one or two plasmid systems. It was found that the addition of T7 terminator after each gene was useful to decrease the influence of the upstream gene. The co-expression of the four enzymes in E. coli BL21(DE3) was demonstrated to generate two NADPH molecules from one glucose unit of maltodextrin, where NADPH was oxidized to convert xylose to xylitol. The best four-gene co-expression system was based on two plasmids (pET and pACYC) which harbored two genes. As a result, apparent enzymatic activities of the four enzymes were regulated to be at similar levels and the overall four-enzyme activity was the highest based on the formation of xylitol. This study provides useful information for the precise control of multi-enzyme-coordinated expression in E. coli BL21(DE3).

In vitro metabolic engineering of bioelectricity generation by the complete oxidation of glucose.

The direct generation of electricity from the most abundant renewable sugar, glucose, is an appealing alternative to the production of liquid biofuels and biohydrogen. However, enzyme-catalyzed bioelectricity generation from glucose suffers from low yields due to the incomplete oxidation of the six-carbon compound glucose via one or few enzymes. Here, we demonstrate a synthetic ATP- and CoA-free 12-enzyme pathway to implement the complete oxidation of glucose in vitro. This pathway is comprised of glucose phosphorylation via polyphosphate glucokinase, NADH generation catalyzed by glucose 6-phosphate dehydrogenase (G6PDH) and 6-phosphogluconate dehydrogenase (6PGDH), electron transfer from NADH to the anode, and glucose 6-phosphate regeneration via the non-oxidative pentose phosphate pathway and gluconeogenesis. The faraday efficiency from glucose to electrons via this pathway was as high as 98.8%, suggesting the generation of nearly 24 electrons per molecule of glucose. The generated current density was greatly increased from 2.8 to 6.9mAcm-2 by replacing a low-activity G6PDH with a high-activity G6PDH and introducing a new enzyme, 6-phosphogluconolactonase, between G6PDH and 6PGDH. These results suggest the great potential of high-yield bioelectricity generation through in vitro metabolic engineering.

Exceptionally High Rates of Biological Hydrogen Production by Biomimetic In Vitro Synthetic Enzymatic Pathways.

Hydrogen production by water splitting energized by biomass sugars is one of the most promising technologies for distributed green H2 production. Direct H2 generation from NADPH, catalysed by an NADPH-dependent, soluble [NiFe]-hydrogenase (SH1) is thermodynamically unfavourable, resulting in slow volumetric productivity. We designed the biomimetic electron transport chain from NADPH to H2 by the introduction of an oxygen-insensitive electron mediator benzyl viologen (BV) and an enzyme (NADPH rubredoxin oxidoreductase, NROR), catalysing electron transport between NADPH and BV. The H2 generation rates using this biomimetic chain increased by approximately five-fold compared to those catalysed only by SH1. The peak volumetric H2 productivity via the in vitro enzymatic pathway comprised of hyperthermophilic glucose 6-phosphate dehydrogenase, 6-phosphogluconolactonase, and 6-phosphogluconate dehydrogenase, NROR, and SH1 was 310 mmol H2 /L h-1 , the highest rate yet reported. The concept of biomimetic electron transport chains could be applied to both in vitro and in vivo H2 production biosystems and artificial photosynthesis.

Facile Construction of Random Gene Mutagenesis Library for Directed Evolution Without the Use of Restriction Enzyme in Escherichia coli.

A foolproof protocol was developed for the construction of mutant DNA library for directed protein evolution. First, a library of linear mutant gene was generated by error-prone PCR or molecular shuffling, and a linear vector backbone was prepared by high-fidelity PCR. Second, the amplified insert and vector fragments were assembled by overlap-extension PCR with a pair of 5'-phosphorylated primers. Third, full-length linear plasmids with phosphorylated 5'-ends were self-ligated with T4 ligase, yielding circular plasmids encoding mutant variants suitable for high-efficiency transformation. Self-made competent Escherichia coli BL21(DE3) showed a transformation efficiency of 2.4 × 10(5) cfu/µg of the self-ligated circular plasmid. Using this method, three mutants of mCherry fluorescent protein were found to alter their colors and fluorescent intensities under visible and UV lights, respectively. Also, one mutant of 6-phosphorogluconate dehydrogenase from a thermophilic bacterium Moorella thermoacetica was found to show the 3.5-fold improved catalytic efficiency (kcat /Km ) on NAD(+) as compared to the wild-type. This protocol is DNA-sequence independent, and does not require restriction enzymes, special E. coli host, or labor-intensive optimization. In addition, this protocol can be used for subcloning the relatively long DNA sequences into any position of plasmids.

A Hidden Transhydrogen Activity of a FMN-Bound Diaphorase under Anaerobic Conditions.

BACKGROUND: Redox cofactors of NADH/NADPH participate in many cellular metabolic pathways for facilitating the electron transfer from one molecule to another in redox reactions. Transhydrogenase plays an important role in linking catabolism and anabolism, regulating the ratio of NADH/NADPH in cells. The cytoplasmic transhydrogenases could be useful to engineer synthetic biochemical pathways for the production of high-value chemicals and biofuels. METHODOLOGY/PRINCIPAL FINDINGS: A transhydrogenase activity was discovered for a FMN-bound diaphorase (DI) from Geobacillus stearothermophilus under anaerobic conditions. The DI-catalyzed hydride exchange were monitored and characterized between a NAD(P)H and a thio-modified NAD+ analogue. This new function of DI was demonstrated to transfer a hydride from NADPH to NAD+ that was consumed by NAD-specific lactate dehydrogenase and malic dehydrogenase. CONCLUSIONS/SIGNIFICANCE: We discover a novel transhydrogenase activity of a FMN-DI by stabilizing the reduced state of FMNH2 under anaerobic conditions. FMN-DI was demonstrated to catalyze the hydride transfer between NADPH and NAD+. In the future, it may be possible to incorporate this FMN-DI into synthetic enzymatic pathways for balancing NADH generation and NADPH consumption for anaerobic production of biofuels and biochemicals.

New artificial fluoro-cofactor of hydride transfer with novel fluorescence assay for redox biocatalysis.

A new artificial fluoro-cofactor was developed for the replacement of natural cofactors NAD(P), exhibiting a high hydride transfer ability. More importantly, we established a new and fast screening method for the evaluation of the properties of artificial cofactors based on the fluorescence assay and visible color change.

Production of Succinate from Acetate by Metabolically Engineered Escherichia coli.

Acetate, a major component of industrial biological wastewater and of lignocellulosic biomass hydrolysate, could potentially be a less costly alternative carbon source. Here we engineered Escherichia coli MG1655 strain for succinate production from acetate as the sole carbon source. Strategies of metabolic engineering included the blockage of the TCA cycle, redirection of the gluconeogenesis pathway, and enhancement of the glyoxylate shunt. The engineered strain MG03 featuring the deletion of genes: succinate dehydrogenase (sdhAB), isocitrate lyase regulator (iclR), and malic enzymes (maeB) accumulated 6.86 mM of succinate in 72 h. MG03(pTrc99a-gltA) overexpressing citrate synthase (gltA) accumulated 16.45 mM of succinate and the yield reached 0.46 mol/mol, about 92% of the maximum theoretical yield. Resting-cell was adopted for the conversion of acetate to succinate, and the highest concentration of succinate achieved 61.7 mM.

A simple assay for determining activities of phosphopentomutase from a hyperthermophilic bacterium Thermotoga maritima.

Phosphopentomutase (PPM) catalyzes the interconversion of α-D-(deoxy)-ribose 1-phosphate and α-D-(deoxy)-ribose 5-phosphate. We developed a coupled or uncoupled enzymatic assay with an enzyme nucleoside phosphorylase for determining PPM activities on D-ribose 5-phosphate at a broad temperature range from 30 to 90 °C. This assay not only is simple and highly sensitive but also does not require any costly special instrument. Via this technology, an open reading frame TM0167 from a thermophilic bacterium Thermotoga maritima putatively encoding PPM was cloned. The recombinant PPM was overexpressed in Escherichia coli Rosetta. This enzyme has the highest activity at 90 °C. MnCl2 (0.1 mM) and 50 µM α-D-glucose 1,6-bisphosphate are cofactors. The kinetic parameters of Km and kcat are 1.2 mM and 185 s(-1) at 90 °C, respectively. The enzyme has a half-life time of up to 156 min at 90 °C. This enzyme is the most active and thermostable PPM reported to date.

One-Pot Biosynthesis of High-Concentration α-Glucose 1-Phosphate from Starch by Sequential Addition of Three Hyperthermophilic Enzymes.

α-Glucose 1-phosphate (G1P) is synthesized from 5% (w/v) corn starch and 1 M phosphate mediated by α-glucan phosphorylase (αGP) from the thermophilic bacterium Thermotoga maritima at pH 7.2 and 70 °C. To increase G1P yield from corn starch containing branched amylopectin, a hyper-thermostable isoamylase from Sulfolobus tokodaii was added for simultaneous starch gelatinization and starch-debranching hydrolysis at 85 °C and pH 5.5 before αGP use. The pretreatment of isoamylase increased G1P titer from 120 mM to 170 mM. To increase maltose and maltotriose utilization, the third thermostable enzyme, 4-glucanotransferase (4GT) from Thermococcus litoralis, was added during the late stage of G1P biotransformation, further increasing G1P titer to 200 mM. This titer is the highest G1P level obtained on starch or its derived products (maltodextrin and soluble starch). This study suggests that in vitro multienzyme biotransformation has an advantage of great engineering flexibility in terms of space and time compared with microbial fermentation.

Novel Hydrogen Bioreactor and Detection Apparatus.

In vitro hydrogen generation represents a clear opportunity for novel bioreactor and system design. Hydrogen, already a globally important commodity chemical, has the potential to become the dominant transportation fuel of the future. Technologies such as in vitro synthetic pathway biotransformation (SyPaB)-the use of more than 10 purified enzymes to catalyze unnatural catabolic pathways-enable the storage of hydrogen in the form of carbohydrates. Biohydrogen production from local carbohydrate resources offers a solution to the most pressing challenges to vehicular and bioenergy uses: small-size distributed production, minimization of CO2 emissions, and potential low cost, driven by high yield and volumetric productivity. In this study, we introduce a novel bioreactor that provides the oxygen-free gas phase necessary for enzymatic hydrogen generation while regulating temperature and reactor volume. A variety of techniques are currently used for laboratory detection of biohydrogen, but the most information is provided by a continuous low-cost hydrogen sensor. Most such systems currently use electrolysis for calibration; here an alternative method, flow calibration, is introduced. This system is further demonstrated here with the conversion of glucose to hydrogen at a high rate, and the production of hydrogen from glucose 6-phosphate at a greatly increased reaction rate, 157 mmol/L/h at 60 °C.

Biosynthesis of D-xylulose 5-phosphate from D-xylose and polyphosphate through a minimized two-enzyme cascade.

Sugar phosphates cannot be produced easily by microbial fermentation because negatively-charged compounds cannot be secreted across intact cell membrane. D-xylulose 5-phosphate (Xu5P), a very expensive sugar phosphate, was synthesized from D-xylose and polyphosphate catalyzed by enzyme cascades in one pot. The synthetic enzymatic pathway comprised of xylose isomerase and xylulokinase was designed to produce Xu5P, along with a third enzyme, polyphosphate kinase, responsible for in site ATP regeneration. Due to the promiscuous activity of the ATP-based xylulokinase from a hyperthermophilic bacterium Thermotoga maritima on polyphosphate, the number of enzymes in the pathway was minimized to two without polyphosphate kinase. The reactions catalyzed by the two-enzyme and three-enzyme pathways were compared for Xu5P production, and the reaction conditions were optimized by examining effects of reaction temperature, enzyme ratio and substrate concentration. The optimized two-enzyme system produced 32 mM Xu5P from 50 mM xylose and polyphosphate after 36 h at 45°C. Biosynthesis of less costly Xu5P from D-xylose and polyphosphate could be highly feasible via this minimized two-enzyme pathway.

High-yield hydrogen production from biomass by in vitro metabolic engineering: Mixed sugars coutilization and kinetic modeling.

The use of hydrogen (H2) as a fuel offers enhanced energy conversion efficiency and tremendous potential to decrease greenhouse gas emissions, but producing it in a distributed, carbon-neutral, low-cost manner requires new technologies. Herein we demonstrate the complete conversion of glucose and xylose from plant biomass to H2 and CO2 based on an in vitro synthetic enzymatic pathway. Glucose and xylose were simultaneously converted to H2 with a yield of two H2 per carbon, the maximum possible yield. Parameters of a nonlinear kinetic model were fitted with experimental data using a genetic algorithm, and a global sensitivity analysis was used to identify the enzymes that have the greatest impact on reaction rate and yield. After optimizing enzyme loadings using this model, volumetric H2 productivity was increased 3-fold to 32 mmol H2⋅L(-1)⋅h(-1). The productivity was further enhanced to 54 mmol H2⋅L(-1)⋅h(-1) by increasing reaction temperature, substrate, and enzyme concentrations--an increase of 67-fold compared with the initial studies using this method. The production of hydrogen from locally produced biomass is a promising means to achieve global green energy production.

In vitro metabolic engineering of hydrogen production at theoretical yield from sucrose.

Hydrogen is one of the most important industrial chemicals and will be arguably the best fuel in the future. Hydrogen production from less costly renewable sugars can provide affordable hydrogen, decrease reliance on fossil fuels, and achieve nearly zero net greenhouse gas emissions, but current chemical and biological means suffer from low hydrogen yields and/or severe reaction conditions. An in vitro synthetic enzymatic pathway comprised of 15 enzymes was designed to split water powered by sucrose to hydrogen. Hydrogen and carbon dioxide were spontaneously generated from sucrose or glucose and water mediated by enzyme cocktails containing up to 15 enzymes under mild reaction conditions (i.e. 37°C and atm). In a batch reaction, the hydrogen yield was 23.2mol of dihydrogen per mole of sucrose, i.e., 96.7% of the theoretical yield (i.e., 12 dihydrogen per hexose). In a fed-batch reaction, increasing substrate concentration led to 3.3-fold enhancement in reaction rate to 9.74mmol of H2/L/h. These proof-of-concept results suggest that catabolic water splitting powered by sugars catalyzed by enzyme cocktails could be an appealing green hydrogen production approach.

Simple cloning and DNA assembly in Escherichia coli by prolonged overlap extension PCR.

We developed a simple method (Simple Cloning) for subcloning one, two, or three DNA fragments into any location of a targeted vector without the need for restriction enzyme, ligase, exonuclease, or recombinase. This cloning technology can be applied to a few common Escherichia coli hosts (e.g., BL21(DE3), DH5α, JM109, TOP10). The protocol includes three steps: (a) linear DNA fragments (i.e., the insert DNA and the vector backbone) with two overlap ends were generated by regular high-fidelity PCR, (b) the DNA multimers were generated based on these equimolar DNA templates by using prolonged overlap extension PCR (POE-PCR) without primers added, and (c) the POE-PCR product was transformed to E. coli strains directly. Because positive colony efficiencies are very high, it is not necessary to identify desired clones by using colony PCR. Simple Cloning provides a new cloning and DNA assembly method with great simplicity and flexibility.

Annexation of a high-activity enzyme in a synthetic three-enzyme complex greatly decreases the degree of substrate channeling.

The self-assembled three-enzyme complex containing triosephosphate isomerase (TIM), aldolase (ALD), and fructose 1,6-biphosphatase (FBP) was constructed via a mini-scaffoldin containing three different cohesins and the three dockerin-containing enzymes. This enzyme complex exhibited 1 order of magnitude higher initial reaction rates than the mixture of noncomplexed three enzymes. In this enzyme cascade reactions, the reaction mediated by ALD was the rate-limiting step. To understand the in-depth role of the rate-limiting enzyme ALD in influencing the substrate channeling effect of synthetic enzyme complexes, low-activity ALD from Thermotoga maritima was replaced with a similar-size ALD isolated from Thermus thermophilus, where the latter had more than 5 times specific activity of the former. The synthetic three-enzyme complexes annexed with either low-activity or high-activity ALDs exhibited higher initial reaction rates than the mixtures of the two-enzyme complex (TIM-FBP) and the nonbound low-activity or high activity ALD at the same enzyme concentration. It was also found that the annexation of more high-activity ALD in the synthetic enzyme complexes drastically decreased the degree of substrate channeling from 7.5 to 1.5. These results suggested that the degree of substrate channeling in synthetic enzyme complexes depended on the enzyme choice. This study implied that the construction of synthetic enzyme enzymes in synthetic cascade pathways could be a very important tool to accrelerate rate-limiting steps controlled by low-activity enzymes.

Overcoming biomass recalcitrance by combining genetically modified switchgrass and cellulose solvent-based lignocellulose pretreatment.

Decreasing lignin content of plant biomass by genetic engineering is believed to mitigate biomass recalcitrance and improve saccharification efficiency of plant biomass. In this study, we compared two different pretreatment methods (i.e., dilute acid and cellulose solvent) on transgenic plant biomass samples having different lignin contents and investigated biomass saccharification efficiency. Without pretreatment, no correlation was observed between lignin contents of plant biomass and saccharification efficiency. After dilute acid pretreatment, a strong negative correlation between lignin content of plant samples and overall glucose release was observed, wherein the highest overall enzymatic glucan digestibility was 70% for the low-lignin sample. After cellulose solvent- and organic solvent-based lignocellulose fractionation pretreatment, there was no strong correlation between lignin contents and high saccharification efficiencies obtained (i.e., 80-90%). These results suggest that the importance of decreasing lignin content in plant biomass to saccharification was largely dependent on pretreatment choice and conditions.

Cell-free Biosystems in the Production of Electricity and Bioenergy.

: Increasing needs of green energy and concerns of climate change are motivating intensive R&D efforts toward the low-cost production of electricity and bioenergy, such as hydrogen, alcohols, and jet fuel, from renewable sugars. Cell-free biosystems for biomanufacturing (CFB2) have been suggested as an emerging platform to replace mainstream microbial fermentation for the cost-effective production of some biocommodities. As compared to whole-cell factories, cell-free biosystems comprised of synthetic enzymatic pathways have numerous advantages, such as high product yield, fast reaction rate, broad reaction condition, easy process control and regulation, tolerance of toxic compound/product, and an unmatched capability of performing unnatural reactions. However, issues pertaining to high costs and low stabilities of enzymes and cofactors as well as compromised optimal conditions for different source enzymes need to be solved before cell-free biosystems are scaled up for biomanufacturing. Here, we review the current status of cell-free technology, update recent advances, and focus on its applications in the production of electricity and bioenergy.