Enhanced production of 2,3-butanediol by engineered Saccharomyces cerevisiae through fine-tuning of pyruvate decarboxylase and NADH oxidase activities
© The Author(s) 2016
Received: 19 September 2016
Accepted: 1 December 2016
Published: 9 December 2016
2,3-Butanediol (2,3-BD) is a promising compound for various applications in chemical, cosmetic, and agricultural industries. Pyruvate decarboxylase (Pdc)-deficient Saccharomyces cerevisiae is an attractive host strain for producing 2,3-BD because a large amount of pyruvate could be shunted to 2,3-BD production instead of ethanol synthesis. However, 2,3-BD yield, productivity, and titer by engineered yeast were inferior to native bacterial producers because of the following metabolic limitations. First, the Pdc-deficient yeast showed growth defect due to a shortage of C2-compounds. Second, redox imbalance during the 2,3-BD production led to glycerol formation that lowered the yield.
To overcome these problems, the expression levels of Pdc from a Crabtree-negative yeast were optimized in S. cerevisiae. Specifically, Candida tropicalis PDC1 (CtPDC1) was used to minimize the production of ethanol but maximize cell growth and 2,3-BD productivity. As a result, productivity of the BD5_G1CtPDC1 strain expressing an optimal level of Pdc was 2.3 folds higher than that of the control strain in flask cultivation. Through a fed-batch fermentation, 121.8 g/L 2,3-BD was produced in 80 h. NADH oxidase from Lactococcus lactis (noxE) was additionally expressed in the engineered yeast with an optimal activity of Pdc. The fed-batch fermentation with the optimized 2-stage aeration control led to production of 154.3 g/L 2,3-BD in 78 h. The overall yield of 2,3-BD was 0.404 g 2,3-BD/g glucose which corresponds to 80.7% of theoretical yield.
A massive metabolic shift in the engineered S. cerevisiae (BD5_G1CtPDC1_nox) expressing NADH oxidase was observed, suggesting that redox imbalance was a major bottleneck for efficient production of 2,3-BD by engineered yeast. Maximum 2,3-BD titer in this study was close to the highest among the reported microbial production studies. The results demonstrate that resolving both C2-compound limitation and redox imbalance is critical to increase 2,3-BD production in the Pdc-deficient S. cerevisiae. Our strategy to express fine-tuned PDC and noxE could be applicable not only to 2,3-BD production, but also other chemical production systems using Pdc-deficient S. cerevisiae.
2,3-Butanediol (2,3-BD) is a potential feedstock for use by the chemical, cosmetic, and agricultural industries. It is used as a precursor for synthesizing 1,3-butadiene which is a monomer for producing synthetic rubber. In addition, 2,3-BD could be used as anti-freeze agent (R,R-BD), solvents, and its derivatives could be used as flavoring agent, moisturizer, liquid fuel additive, humectant in cosmetics, and plant growth promoter [1, 2].
Many research groups have attempted producing 2,3-BD via microbial fermentations with native bacterial producers such as Klebsiella pneumoniae, K. oxytoca, Paenibacillus polymyxa, and Enterobacter species [3, 4]. These bacterial strains are able to produce 2,3-BD with high productivity, but formation of biofilm consisting of exopolysaccharides , optical impurity of 2,3-BD , and production of various by-products such as succinate, lactate, acetate, and ethanol  hampered the use of the strains for industrial fermentations. Commercialization is also constrained by most of 2,3-BD-producing bacteria belonging to class II (pathogenic) microorganisms, which requires tight safety regulations for industrial-scale fermentations . In contrast, Saccharomyces cerevisiae is a GRAS (generally recognized as safe) microorganism and has been widely employed in industrial-scale fermentation processes for producing various chemicals and fuels. Thus, S. cerevisiae would be an appropriate microorganism for industrial production of 2,3-BD. Nonetheless, it is necessary to delete the genes coding for pyruvate decarboxylase (Pdc) for 2,3-BD production because S. cerevisiae produces ethanol as a major product.
Pdc-deficient S. cerevisiae is a promising metabolic background for producing non-ethanol products such as 2,3-BD, 3-hydroxypropionic acid, and lactic acid. It accumulates pyruvate which is a precursor of numerous chemical molecules instead of producing ethanol from glucose . However, impaired growth of Pdc-deficient S. cerevisiae on glucose has been a major obstacle to exploit Pdc-deficient S. cerevisiae for 2,3-BD production. The reasons for the growth defect are (1) lack of acetyl-CoA in the cytosol [9, 10] and (2) a redox imbalance due to accumulation of cytosolic NADH [11, 12]. Cytosolic acetyl-CoA is indispensable for growth of S. cerevisiae because it is used for synthesizing lysine and fatty acids in the cytosol [8–10]. Pdc-deficient S. cerevisiae cannot synthesize cytosolic acetyl-CoA from glucose because the deletion of PDC leads to elimination of cytosolic C2-compounds (e.g., acetaldehyde, acetate, ethanol). Within mitochondria, the pyruvate dehydrogenase (Pdh) complex converts pyruvate into acetyl-CoA, but mitochondrial acetyl-CoA cannot pass through the inner membrane of mitochondria . Although the YBR219C and YBR220C are known as putative genes coding for acetyl-CoA transporter, activities of these enzymes were not sufficient for supplying enough acetyl-CoA to cytosol . Therefore, cell growth and carbon utilization of Pdc-deficient S. cerevisiae strains were greatly inhibited by insufficient supply of cytosolic acetyl-CoA, which is often termed as ‘C2-auxotrophy’. Redox imbalance is another reason for growth defect of Pdc-deficient S. cerevisiae on glucose. Excess NADH is generated in Pdc-deficient S. cerevisiae because oxidation of cytosolic NADH via the ethanol production pathway is blocked. NADH generated by converting glucose to pyruvate should be re-oxidized to NAD+ to maintain cellular redox metabolism. However, insufficient activity of the respiratory pathway because of the glucose-induced Crabtree effect  and absence of transhydrogenase activity [15, 16] aggravate the redox imbalance of Pdc-deficient S. cerevisiae.
There have been several reports about the production of 2,3-BD with engineered S. cerevisiae harboring the bacterial 2,3-BD biosynthetic enzymes [12, 17, 18]. By introduction of B. subtilis α-acetolactate synthase (alsS) and α-acetolactate decarboxylase (alsD), and overexpression of endogenous 2,3-butanediol dehydrogenase (BDH1), pyruvate was converted into optically pure (2R,3R)-BD by the Pdc-deficient S. cerevisiae with MTH1 mutation .
The mutation (G241C) in MTH1 has been reported to suppress the growth defect of the Pdc-deficient S. cerevisiae strain . In the presence of extracellular glucose, signal transduction via the glucose sensors (Rgt2/Snf3) and casein kinases (Yck1/2) induces phosphorylation of Mth1 to be degraded . The degradation of Mth1 led to the down-regulation of hexose transporter genes (HXTs) , resulting in decreased glucose influx rate in spite of extracellular glucose . A slow glucose uptake rate caused by the MTH1 mutation might be responsible for restoration of growth defect by the Pdc-deficient strains on glucose . Additionally, although the exact mechanism remains unknown, the mutant MTH1 could partially relieve the C2-auxotrophy of Pdc-deficient S. cerevisiae [12, 21]. The mutation in MTH1 might be regarded as an indispensable strategy for Pdc-deficient S. cerevisiae to grow on glucose, but the slow glucose consumption rates caused by the MTH1 mutation resulted in much lower 2,3-BD productivity  than that by bacterial 2,3-BD production systems [22, 23]. The lower glucose consumption rates and 2,3-BD productivity by the Pdc-deficient S. cerevisiae are putatively caused by reduced expression levels of HXTs by MTH1 mutation [19, 20, 24]. As such, metabolic engineering strategies to alleviate the growth defect of Pdc-deficient S. cerevisiae without mutations in MTH1 needs to be devised for efficient and rapid production of 2,3-BD by engineered yeast.
In this study, we present metabolic engineering strategies for rapid production of 2,3-BD by the Pdc-deficient S. cerevisiae strain without mutating MTH1. To this end, we attempted to address both C2-auxotrophy and redox imbalance. For the construction of C2-independent strains, various PDC genes from the Crabtree-negative yeast were expressed at a minimal level to eliminate the C2-auxotrophy of Pdc-deficient S. cerevisiae strains. As excess expression of PDC produces a large amount of ethanol instead of producing 2,3-BD, we hypothesized that expression levels of PDC need to be fine-tuned to balance metabolic fluxes between acetyl-CoA synthesis and 2,3-BD production. Specifically, PDC expression levels were optimized through the variations of PDC gene sources, promoters, and copy numbers. Additionally, Lactococcus lactis NADH oxidase (noxE) was co-expressed to resolve the issue of redox imbalance . The resulting engineered S. cerevisiae strain expressing both PDC and noxE at optimal levels produced 2,3-BD with much higher titer, productivity, and yield than those of the other engineered yeast capable of producing 2,3-BD. The results presented in this study suggest that utilization of the engineered S. cerevisiae strain for 2,3-BD production could be an attractive alternative to bacterial 2,3-BD production systems .
A mutation in MTH1 decreased glucose uptake rate in Pdc-deficient S. cerevisiae
The mutations in MTH1 were identified from evolved strains exhibiting improved growth phenotypes of Pdc-deficient S. cerevisiae. In particular, a mutation (G241C) was employed for improving 2,3-BD production by the S. cerevisiae strain because the MTH1 mutation could overcome the C2-auxotrophy of Pdc-deficient S. cerevisiae [12, 21]. However, the MTH1 mutation resulted in reduced expression levels of hexose transporters in S. cerevisiae [11, 20], causing slow glucose uptake rates. Batch fermentations were performed for comparing 2,3-BD production by 2,3-BD-producing strains with the MTH1 mutation (the BD4 strain) and without the MTH1 mutation (the BD5 strain) (Additional file 1: Fig. S1). When glucose was used as a sole carbon source (Additional file 1: Fig. S1A, B), the maximum dry cell weight (DCW) of the BD4 strain was 5.5-fold higher (2.2 g/L) than that of the BD5 strain (0.4 g/L). The BD4 strain (with the MTH1 mutation) showed higher volumetric glucose uptake rate and 2,3-BD productivity (0.289 g glucose/L·h and 0.087 g 2,3-BD/L·h) than those (0.155 g glucose/L·h and 0.051 g 2,3-BD/L·h) of the BD5 strain (without the MTH1 mutation). However, when a small amount of ethanol (0.5 g/L) was supplemented as a C2-compound (Additional file 1: Fig. S1C, D), the maximum DCW of the both strains were similar (3.9 and 4.2 g/L), but the BD5 strain (without the MTH1 mutation) showed a much higher volumetric glucose uptake rate and 2,3-BD productivity (1.23 g glucose/L·h and 0.305 g 2,3-BD/L·h) than the BD4 strain (0.665 g glucose/L·h and 0.205 g 2,3-BD/L·h). Thus, we concluded that the BD5 strain without the MTH1 mutation can grow and consume glucose with comparable to the wild type and produce 2,3-BD efficiently as long as C2-compounds are properly supplied.
Optimization of PDC expression levels led to enhanced 2,3-BD production through relieving C2-auxotrophy in Pdc-deficient S. cerevisiae
Strains and plasmids used in this study
Strains and plasmids
Candida tropicalis ATCC20336
Source for CtPDC1
In this study
Kluyveromyces marxianus KCTC17555
Source for KmPDC1
In this study
Saccharomyces cerevisiae D452-2
Source for ScPDC1, ScPDC5, ScPDC6 MATα leu2 his3 ura3
D452-2, pdc1Δ, pdc5Δ, evolved 2,3-BD-producing strain
D452-2, pdc1Δ, pdc5Δ, pdc6Δ
In this study
In this study
In this study
In this study
In this study
SOS5, p423_alsSalsD, p425_BDH1
In this study
In this study
In this study
In this study
BD5, p406GPD2_CtPDC1, pAUR_Llnox
In this study
URA3 2 μm origin
pRS406, GPD2 prom CYC1 term
pRS426, CYC1 prom CYC1 term
URA3 2 μm origin, TDH3 prom CYC1 term
HIS3 2 μm origin, TDH3 prom-alsS-CYC1 term TDH3 prom-alsD-CYC1 term
LEU2 2 μm origin, TDH3 prom-BDH1-CYC1 term
pRS426, TDH3 prom-Llnox-CYC1 term
pRS426, TDH3 prom-CtPDC1-CYC1 term
In this study
pRS426, TDH3 prom-KmPDC1-CYC1 term
In this study
pRS426, TDH3 prom-ScPDC1-CYC1 term
In this study
pRS426, TDH3 prom-ScPDC5-CYC1 term
In this study
pRS426, TDH3 prom-ScPDC6-CYC1 term
In this study
pRS406, CYC1 prom-CtPDC1-CYC1 term
In this study
pRS406, GPD2 prom-CtPDC1-CYC1 term
In this study
pRS426, CYC1 prom-CtPDC1-CYC1 term
In this study
AUR1-C CEN6 ARS4, TDH3 prom-Llnox-CYC1 term
In this study
Expression of NADH oxidase increased 2,3-BD yield and productivity
In vivo activity of NADH oxidase influence on intracellular metabolites
Our results from NADH oxidase expression in the Pdc-deficient strain suggested that cofactor imbalance induced by the deletion of PDC genes in S. cerevisiae could be relieved by the expression of NADH oxidase and control of aeration conditions. To investigate the effects of NADH oxidase at metabolite levels, intracellular metabolites were analyzed from batch fermentations with the BD5_G1CtPDC1 and BD5_G1CtPDC1_nox strains. The batch fermentations were carried out under different conditions including aerobic cultivation of the BD5_G1CtPDC1 strain (Ct100), and microaerobic (N50) and aerobic (N100) cultivation of the BD5_G1CtPDC1_nox strain (Additional file 1: Fig. S6). Additional expression of NADH oxidase led to reduced glycerol yields and increased acetoin yields regardless of aeration conditions. Moreover, glucose consumption rate (2.68 g glucose/L·h) of the Ct100 condition increased to 4.35 g glucose/L·h for the N50 condition and 3.98 g glucose/L·h for the N100 condition. Maximum DCWs (2.5 g cell/L) under the Ct100 and N50 conditions were similar, but it increased to 3.9 g/L under the N100 condition.
From the BD5_G1CtPDC1 and BD5_ G1CtPDC1_nox cells harvested at 22 h, 129 intracellular metabolites were identified and quantified (Additional file 1: Table S2). These metabolites belonged to different chemical classes including amino acids, organic acids, sugars and sugar alcohols, fatty acids, phosphates, amines, nucleotides. Multivariate data were analyzed by principal component analysis (PCA) (Additional file 1: Fig. S7) and hierarchical clustering analysis (HCA) (Additional file 1: Fig. S8) to explore the variations of intracellular metabolites in response to NADH oxidase expression. As shown in Additional file 1: Fig. S7 and Additional file 1: Fig. S8, the samples from each condition (Ct100, N50, and N100) were clearly distinguished. It indicated that in vivo activity of NADH oxidase changed intracellular metabolism and displayed distinctive metabolic characteristics. The differences in intracellular metabolites of glycolysis, TCA cycle, and amino acid synthesis between Ct100 and N50 conditions were depicted in Additional file 1: Fig. S9. Intermediate metabolites in the glycolytic pathway including glucose, glucose-6-phosphate, fructose-6-phosphate, and pyruvate, and in the TCA cycle including isocitrate and fumarate significantly increased in Ct100 condition compared to N50 condition.
Fed-batch fermentation with the engineered strain co-expressing CtPDC1 and noxE
It has been known that Pdc-deficient S. cerevisiae cannot grow on glucose as a sole carbon source because of C2-auxotrophy and redox imbalance in the cytosol . However, as shown in Additional file 1: Fig. S1B, the engineered strain with the 2,3-BD pathway grew slowly in synthetic medium with glucose as a sole carbon source. The growth recovery of Pdc-deficient S. cerevisiae by introduction of the 2,3-BD pathway was also observed in the other studies . In the engineered strains harboring the 2,3-BD biosynthetic genes, C2-compounds might be supplied by α-acetolactate synthase (AlsS) which has a decarboxylase activity on pyruvate . In addition, redox imbalance could be relieved by the 2,3-BD biosynthetic pathway because NADH could be additionally oxidized by producing 2,3-BD from pyruvate. However, the growth recovery by introduction of the 2,3-BD pathway is still insufficient for efficient production of 2,3-BD, and hence additional metabolic engineering is needed for enhancing growth rate, 2,3-BD yield, and productivity.
In other studies, the use of the MTH1 variant has been regarded as an efficient way to overcome the growth defect of Pdc-deficient S. cerevisiae on glucose [12, 21]. The specific growth rate was substantially increased by the mutation on the MTH1 gene (ref, Additional file 1: Fig. S1). However, suppressed expression of hexose transporters by mutation on the MTH1 gene [34, 35] became another problem for the efficient production of 2,3-BD. The specific glucose uptake rate of the BD4 strain harboring the MTH1 mutation was only 20% that of the BD5 strain with the wild-type MTH1 gene (Additional file 1: Fig. S10). As a result, the volumetric productivity of the BD5 strain was much higher than that of the BD4 strain when the growth rate of the BD5 strain was enhanced by supplementation of ethanol as a C2-compound (Additional file 1: Fig. S1C, D). Therefore, we constructed the engineered Pdc-deficient S. cerevisiae strain to overcome C2-auxotrophy and redox balance without MTH1 mutation.
As shown in Fig. 1, the product yields and 2,3-BD productivities were changed according to the levels of Pdc activity. The BD5_Con strain produced 2,3-BD dominantly, but ethanol production was positively correlated with Pdc activity. The BD5_G1CtPDC1 strain was selected as the best 2,3-BD production strain because ethanol production was minimized along with the sufficient supplementation of a C2-compound for cell growth. The production of ethanol by additional PDC expression led to reduced glycerol production because cofactor imbalance might be relieved by ethanol biosynthesis which is a redox-neutral process . As a result, through fine-tuning expression levels of Pdc, C2-compounds were efficiently supplied for synthesizing acetyl-CoA and 2,3-BD or ethanol production could be easily modulated.
Expression of NADH oxidase in 2,3-BD-producing Pdc-deficient S. cerevisiae reduced glycerol production as a by-product by relieving cofactor balance in the cytosol (Fig. 3). L. lactis NADH oxidase is an H2O-forming NADH oxidase involved in the oxygen defense mechanism in facultative anaerobic bacteria . Several studies about the effect of water-forming NADH oxidase in S. cerevisiae strains have been reported [18, 38–40]. Especially, L. lactis NADH oxidase has a lower Km value (4.1 µM) for NADH  than other metabolic enzymes such as Gpd (23 µM) and Bdh (55 µM) [42, 43], the expression of NADH oxidase caused a large change in metabolic fluxes. The expression of NADH oxidase in S. cerevisiae reduced overflow metabolism  and decreased production of reduced metabolites such as ethanol, succinate, 2-hydroxyglutarate, glycerol, and xylitol [39, 40]. In some cases, the introduction of NADH oxidase led to reduced biomass formation because of decreasing the NADH pool and accumulating acetaldehyde at toxic levels [38, 40]. However, introduction of NADH oxidase efficiently increased 2,3-BD yield and productivity by relieving cofactor imbalance. The possible reasons for this difference are as follows. First, Pdc-deficient S. cerevisiae with the 2,3-BD production pathway produces excess NADH in the cytosol. Two moles of NADH were generated by converting one mole of glucose into two moles of pyruvate, but each mole of NADH could be oxidized by producing one mole of 2,3-BD from two moles of pyruvate. The depletion of cytosolic NADH by NADH oxidase might be prevented because the excess cytosolic NADH was generated in the engineered strain. Second, as shown in Fig. 3 and Additional file 1: Fig. S5, the in vivo activity of NADH oxidase could be controlled by aeration condition. As the cells expressing L. lactis NADH oxidase had a high demand for dissolved oxygen (DO)  and rapidly reduced DO levels in fed-batch fermentation , the oxygen-limited cultivation could be carried out regardless of aerobic condition. Thus, the optimized aeration condition was able to efficiently enhance 2,3-BD yield and productivity and to reduce by-product formation (Fig. 4). Third, in fully aerated condition, acetoin which is the oxidized form of 2,3-BD was over-produced by the 2,3-BD-producing Pdc-deficient S. cerevisiae strain. Acetoin exerts no deleterious effect in contrast to acetaldehyde, which can inhibit growth and carbon utilization at concentrations above 1.0 g/L . In addition, acetoin could be converted into 2,3-BD under oxygen-limited condition (Fig. 4). Thus, acetoin could be acted as a redox sink in the engineered strain for maintaining redox balance that could be induced by excessive NADH oxidase activity.
Interestingly, the expression of NADH oxidase increased glucose uptake rate as well as altered product yields. Similarly, the presence of NADH oxidase or alternative oxidase improved specific glucose uptake rate and subsequent metabolism in the wild-type background . The metabolomic analysis led to a possible hypothesis for this. The concentrations of glucose, and glucose-6-phosphate, and pyruvate (intermediates of glycolysis) increased in the BD5_G1CtPDC1 strain (Ct100) compared with the BD5_G1CtPDC1_nox strain (N50). Pyruvate formation by glycolysis is NADH-generating process. Since cofactor imbalance by reduced oxidation of cytosolic NADH inhibited glycolytic flux and cell growth rate in Pdc-deficient S. cerevisiae [12, 21], the inhibition of glycolytic flux in the BD5_G1CtPDC1 strain might reduce the accumulation of glycolytic intermediates. Metabolic reactions with isocitrate and fumarate as substrate are also NADH-generating steps in the TCA cycle. Similar to glycolytic intermediates, these compounds might accumulate because of intracellular cofactor imbalance of NADH in the BD5_G1CtPDC1 strain (Ct100). From the comparison of intracellular metabolites in the BD5_G1CtPDC1 and BD5_G1CtPDC1_nox strains, the relieved cofactor imbalance by NADH oxidase changed glycolytic and TCA cycle intermediates. These data could explain the improved glucose consumption rate by the expression of NADH oxidase in Pdc-deficient S. cerevisiae.
In this study, a supplementation of a C2-compound by optimized expression of Pdc and relieved cofactor imbalance by controlled in vivo activity of NADH oxidase enhanced 2,3-BD yield, productivity, and titer in 2,3-BD-producing Pdc-deficient S. cerevisiae. The 2,3-BD productivity was increased fivefold compared to the previous report . In addition, final titer of 2,3-BD (154.3 g/L) is one of the highest concentration reported for production of 2,3-BD by engineered yeasts and bacterial native producers. To improve 2,3-BD production performance by the engineered yeast strains, efficient utilization of various sugars such as galactose, xylose, and cellobiose or elimination of glycerol formation are required. In addition, optimizing aeration condition is important in large-scale fermentations. The fine-tuned expression of PDC and co-expression of NADH oxidase in Pdc-deficient S. cerevisiae could be applied to improving other chemical production systems using pyruvate as a precursor compound.
Construction of plasmids
Primers used in this study
Cloning of PDC genes
Cloning of AUR1-C gene
Yeast transformation and construction of recombinant S. cerevisiae strains
Transformation of plasmids for constructing engineered strains was performed using a spheroplast transformation kit (BIO 101, Vista, CA). To select transformants, S. cerevisiae strains were routinely cultivated aerobically at 30 °C in YNB medium (6.7 g/L yeast nitrogen base and appropriate nucleotides and amino acid). 20 g/L of ethanol was used as a carbon source for the cultivation of the SOS5 strain and 20 g/L of glucose and 1 g/L of ethanol was used for cultivation of the other strains. The C. tropicalis PDC gene (CtPDC1), K. marxianus PDC gene (KmPDC1), and S. cerevisiae PDC genes (ScPDC1, ScPDC5, ScPDC6) were expressed under TDH3 promoter and multi-copy plasmid into the SOS5 strain. The C. tropicalis PDC gene was expressed under different native promoters (TDH3, CYC1, and GPD2) and copy number (single and multi-copy). The PDC expression plasmids in Table 1 were introduced into the SOS5 and BD5 strains and the engineered S. cerevisiae strains expressing different levels of PDC were constructed. Integrative PDC expression plasmids (p406CYC1_CtPDC1, p406GPD2_CtPDC1, and p406TDH3_CtPDC1) were digested with StuI before use and integrated into the URA3 locus. For additional expression of L. lactis NADH oxidase, pAUR_Llnox plasmid was transformed to the BD5_G1CtPDC1 strain. The resulting recombinant S. cerevisiae strains are listed in Table 1.
All cultures were carried out at 30 °C. Pre-cultures of yeast cells were conducted aerobically in 250-mL baffled flasks. Main flask batch cultures were conducted under microaerobic conditions in 250-mL flasks at 80 rpm. In order to prepare inoculums, engineered S. cerevisiae cells were cultivated for 48–72 h in 5 mL YNB medium containing 20 g/L glucose and 1 g/L ethanol. The grown cells were transferred to 100 mL YNB medium containing 20 g/L glucose and 0.5 g/L ethanol. After 24 h cultivation, the mid-exponential growing cells (OD600 < 3) were harvested and washed twice with double-distilled water (DDW). Cells were inoculated into the main culture at the initial concentration of 0.20 g/L. The main culture YP medium (10 g/L yeast extract and 20 g/L peptone) contains 100 g/L glucose, and YNB medium contains 90 g/L glucose and 50 mM potassium phthalate at pH 5.5 adjusted by NaOH. Ethanol 0.5 g/L was added if necessary. Batch fermentations with bioreactor were performed in 500 mL YP medium containing 100 g/L glucose using 1 L bench-top fermentor (Fermentec, Korea). For controlling aeration conditions, different ratios of mixed inlet gas (pure air, 1:1 or 1:3 of air and nitrogen gas mixture) were used at constant agitation speed of 500 rpm and air flow rate of 2 vvm. Fed-batch fermentations with bioreactor were carried out in 500 mL YP medium containing 300 g/L glucose using 1 L bench-top fermentor (Fermentec, Korea) at 30 °C. The medium pH was maintained at 5.5 with 5 N NaOH solution, and dissolved oxygen (DO) levels were monitored with O2 sensor (Mettler Toledo, Switzerland). The culture medium was agitated at 300–500 rpm and aerated with air flow rate of 1–2 vvm according to DO levels in medium. During the fed-batch fermentation, DO levels were kept under 2.0. The grown cells prepared from flask culture were inoculated at the initial concentration of 2.0 g/L.
Analysis of dry cell weight and metabolites
Cell growth was monitored by optical density at 600 nm (OD600) using a spectrophotometer (UV-1601, Shimadzu, Japan). Dry cell weight (DCW) was calculated using a pre-determined factor of 0.20 g DCW/L/OD600 . Glucose, glycerol, acetoin, 2,3-BD, and ethanol were analyzed by a high-performance liquid chromatography (1100 series, Agilent, CO) equipped with a Rezex ROA-organic acid column (Phenomenex, CA). Metabolites were detected by a refractive index (RI) detector. For measurement of intracellular NADH and NAD+, about 4 × 107 of exponentially growing cells in flask batch cultivation were used. NADH and NAD+ concentration assays were conducted following manufacturer-provided methods (BioAssay Systems, CA).
In vitro Pdc activity analysis
To prepare crude extracts, about 1 × 109 mid-exponential phase cells grown on the YNB medium with 80 g/L glucose and 0.5 g/L ethanol in a flask culture were harvested and washed twice with DDW. Protease inhibitor (Roche, Switzerland) was added, and the harvested cells were lysed with Yeast Protein Extraction Reagent (Y-PER, Thermo Scientific, MA). After centrifugation for 20 min at 12,000 rpm and 4 °C, the supernatants were used to determine the Pdc activity within 3 h and diluted with DDW if necessary. The NADH oxidase activity assays were performed at 30 °C with the reaction mixture containing 40 mM imidazole hydrochloride buffer (pH 6.5), 5 mM MgCl2, 0.2 mM TPP, 10 U alcohol dehydrogenase from S. cerevisiae, 0.4 mM NADH, and 50 mM pyruvate . The reactions were initiated by adding pyruvate, and a decrease of absorbance at 340 nm was measured. One unit of activity was defined as the amount of enzyme oxidizing 1 μmol NADH per minute at the corresponding reaction conditions. The protein concentration of crude extracts was determined by the Bradford method . For determination of kinetic values (K m and V max) of the Pdc enzymes, Pdc activities were measured at various concentrations of pyruvate (1.56, 3.13, 6.25, 12.5, 25.0, 50.0, 100 mM). The K m and V max values for Pdc were determined from Lineweaver–Burk plots of the data.
Sample preparation for metabolite analysis
For the extraction of metabolome sampling, a fast filtration method was performed with a slight modification of the method used in a previous study . Briefly, cell pellets on each conditions were collected by vacuum-filtering of 1-mL culture broth using a nylon membrane filter (0.45 μm pore size, 30 mm diameter; Whatman, Piscataway, NJ) and washed using 5 mL of distilled water at room temperature. The membrane filter and loaded cells were mixed using 10 mL of acetonitrile/water (1:1, v/v) at −20 °C. After that, the extraction mixture was immersed in liquid nitrogen. These steps were completed in less than 30 s. The extraction mixture was thawed on ice, vortexed for 3 min, and centrifuged at 16,100×g for 5 min at 4 °C. The supernatant was collected and vacuum-dried.
GC/TOF MS analysis
For the GC/TOF MS analysis, the derivative metabolite samples were prepared using 5 μL of 40 mg/mL methoxyamine hydrochloride in pyridine (Sigma-Aldrich, St. Louis, MO) at 30 °C for 90 min and 45 μL of N-methyl-N-trimethylsilyl trifluoroacetamide (Fluka, Buchs, Switzerland) at 37 °C for 30 min. As retention index markers, a mixture of fatty acid methyl esters was added to the derivative samples. An Agilent 7890A FX (Agilent Technologies, Wilmington, DE) coupled with a Pegasus HT TOF MS (LECO, St. Joseph, MI) was used for metabolite analysis. A 0.5 μL of derivative samples was injected into the GC in splitless mode and separated on an RTX-5Sil MS column (30 m × 0.25 mm i.d., 0.25 μm film thickness, Restek, Bellefonte, PA) with an additional 10-m integrated guard column. The initial oven temperature was set at 50 °C for 1 min and then increased to 330 °C at a rate of 20 °C/min and held at 330 °C for 5 min. Mass spectra were recorded in the mass range of 85–500 m/z at an acquisition rate of ten spectra/s. The temperatures of the ion source and transfer line were set at 250 and 280 °C, respectively, and the sample ionization was performed electron impact at 70 eV.
Data processing and statistical analysis for metabolomics analysis
For the processing of GC/TOF MS data, the software LECO Chroma TOF was used to detect peaks and to deconvolute the mass spectra. The processed data were further processed using an inhouse library, BinBase . The raw metabolite data were normalized by the median of the sum of the peak intensities of all identified metabolites in each sample. The normalized data were used for multivariate and univariate statistical analysis , and MultiExperiment Viewer was used for the hierarchical clustering analysis (HCA).
JWK conceived and designed the study, analyzed and interpreted data, conducted experiments, and wrote the manuscript. JK and KHK operated GC/TOF MS and did the analysis of metabolomics study part. SOS contributed to design and construct recombinant strains. YSJ and JHS analyzed data and revised the manuscript. All authors read and approved the final manuscript.
This work was supported by the Advanced Biomass R&D Center (ABC) of Global Frontier Project (2011-0031359) and the National Research Foundation of Korea Grant (2014M1A2A2069904) funded by the Ministry of Science, ICT, and Future Planning.
The authors declare that they have no competing interests.
Availability of data and materials
Data made available to all interested researchers upon request.
Consent for publication
All authors approved the manuscript.
This study is funded by National Research Foundation of Korea.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Syu M-J. Biological production of 2, 3-butanediol. Appl Microbiol Biotechnol. 2001;55(1):10–8.View ArticleGoogle Scholar
- Ryu C-M, Farag MA, Hu C-H, Reddy MS, Wei H-X, Paré PW, Kloepper JW. Bacterial volatiles promote growth in Arabidopsis. Proc Natl Acad Sci USA. 2003;100(8):4927–32.View ArticleGoogle Scholar
- Ji X-J, Huang H, Ouyang P-K. Microbial 2, 3-butanediol production: a state-of-the-art review. Biotechnol Adv. 2011;29(3):351–64.View ArticleGoogle Scholar
- Cho S, Kim T, Woo HM, Kim Y, Lee J, Um Y. High production of 2, 3-butanediol from biodiesel-derived crude glycerol by metabolically engineered Klebsiella oxytoca M1. Biotechnol Biofuels. 2015;8(1):1.View ArticleGoogle Scholar
- Häßler T, Schieder D, Pfaller R, Faulstich M, Sieber V. Enhanced fed-batch fermentation of 2, 3-butanediol by Paenibacillus polymyxa DSM 365. Bioresour Technol. 2012;124:237–44.View ArticleGoogle Scholar
- Jantama K, Polyiam P, Khunnonkwao P, Chan S, Sangproo M, Khor K, Jantama SS, Kanchanatawee S. Efficient reduction of the formation of by-products and improvement of production yield of 2, 3-butanediol by a combined deletion of alcohol dehydrogenase, acetate kinase-phosphotransacetylase, and lactate dehydrogenase genes in metabolically engineered Klebsiella oxytoca in mineral salts medium. Metab Eng. 2015;30:16–26.View ArticleGoogle Scholar
- Celińska E, Grajek W. Biotechnological production of 2, 3-butanediol—current state and prospects. Biotechnol Adv. 2009;27(6):715–25.View ArticleGoogle Scholar
- Pronk JT, Steensma HY, Van Dijken JP. Pyruvate metabolism in Saccharomyces cerevisiae. Yeast. 1996;12(16):1607–33.View ArticleGoogle Scholar
- Flikweert MT, van der Zanden L, Janssen WMTM, Yde Steensma H, van Dijken JP, Pronk JT. Pyruvate decarboxylase: an indispensable enzyme for growth of Saccharomyces cerevisiae on glucose. Yeast. 1996;12(3):247–57.View ArticleGoogle Scholar
- Flikweert MT, Swaaf M, Dijken JP, Pronk JT. Growth requirements of pyruvate-decarboxylase-negative Saccharomyces cerevisiae. FEMS Microbiol Lett. 1999;174(1):73–9.View ArticleGoogle Scholar
- van Maris AJ, Geertman JMA, Vermeulen A, Groothuizen MK, Winkler AA, Piper MD, van Dijken JP, Pronk JT. Directed evolution of pyruvate decarboxylase-negative Saccharomyces cerevisiae, yielding a C2-independent, glucose-tolerant, and pyruvate-hyperproducing yeast. Appl Environ Microbiol. 2004;70(1):159–66.View ArticleGoogle Scholar
- Kim S-J, Seo S-O, Jin Y-S, Seo J-H. Production of 2, 3-butanediol by engineered Saccharomyces cerevisiae. Bioresour Technol. 2013;146:274–81.View ArticleGoogle Scholar
- Van Roermund C, Elgersma Y, Singh N, Wanders R, Tabak H. The membrane of peroxisomes in Saccharomyces cerevisiae is impermeable to NAD (H) and acetyl-CoA under in vivo conditions. EMBO J. 1995;14(14):3480.Google Scholar
- Giaever G, Chu AM, Ni L, Connelly C, Riles L, Veronneau S, Dow S, Lucau-Danila A, Anderson K, Andre B. Functional profiling of the Saccharomyces cerevisiae genome. Nature. 2002;418(6896):387–91.View ArticleGoogle Scholar
- Bruinenberg PM, Jonker R, van Dijken JP, Scheffers WA. Utilization of formate as an additional energy source by glucose-limited chemostat cultures of Candida utilis CBS 621 and Saccharomyces cerevisiae CBS 8066. Arch Microbiol. 1985;142(3):302–6.View ArticleGoogle Scholar
- Lagunas R, Gancedo JM. Reduced pyridine-nucleotides balance in glucose-growing Saccharomyces cerevisiae. Eur J Biochem. 1973;37(1):90–4.View ArticleGoogle Scholar
- Ng CY, Jung M-y, Lee J, Oh M-K. Production of 2, 3-butanediol in Saccharomyces cerevisiae by in silico aided metabolic engineering. Microb Cell Fact. 2012;11:68.View ArticleGoogle Scholar
- Kim J-W, Seo S-O, Zhang G-C, Jin Y-S, Seo J-H. Expression of Lactococcus lactis NADH oxidase increases 2, 3-butanediol production in Pdc-deficient Saccharomyces cerevisiae. Bioresour Technol. 2015;191:512–9.View ArticleGoogle Scholar
- Lafuente MJ, Gancedo C, Jauniaux JC, Gancedo JM. Mth1 receives the signal given by the glucose sensors Snf3 and Rgt2 in Saccharomyces cerevisiae. Mol Microbiol. 2000;35(1):161–72.View ArticleGoogle Scholar
- Moriya H, Johnston M. Glucose sensing and signaling in Saccharomyces cerevisiae through the Rgt2 glucose sensor and casein kinase I. Proc Natl Acad Sci USA. 2004;101(6):1572–7.View ArticleGoogle Scholar
- Oud B, Flores C-L, Gancedo C, Zhang X, Trueheart J, Daran J-M, Pronk JT, van Maris A. An internal deletion in MTH1 enables growth on glucose of pyruvate-decarboxylase negative, non-fermentative Saccharomyces cerevisiae. Microb Cell Fact. 2012;11:131.View ArticleGoogle Scholar
- Wang A, Xu Y, Ma C, Gao C, Li L, Wang Y, Tao F, Xu P. Efficient 2, 3-butanediol production from cassava powder by a crop-biomass-utilizer, Enterobacter cloacae subsp. dissolvens SDM. PLoS ONE. 2012;7(7):e40442.View ArticleGoogle Scholar
- Petrov K, Petrova P. High production of 2, 3-butanediol from glycerol by Klebsiella pneumoniae G31. Appl Microb Biotech. 2009;84(4):659–65.View ArticleGoogle Scholar
- Kim J-H. DNA-binding properties of the yeast Rgt1 repressor. Biochimie. 2009;91(2):300–3.View ArticleGoogle Scholar
- Fletcher E, Krivoruchko A, Nielsen J. Industrial systems biology and its impact on synthetic biology of yeast cell factories. Biotechnol Bioeng. 2016;113(6):1164–70.View ArticleGoogle Scholar
- van Urk H, Voll WL, Scheffers WA, Van Dijken JP. Transient-state analysis of metabolic fluxes in Crabtree-positive and Crabtree-negative yeasts. Appl Environ Microbiol. 1990;56(1):281–7.Google Scholar
- Postma E, Verduyn C, Scheffers WA, Van Dijken JP. Enzymic analysis of the crabtree effect in glucose-limited chemostat cultures of Saccharomyces cerevisiae. Appl Environ Microbiol. 1989;55(2):468–77.Google Scholar
- De Deken R. The Crabtree effect: a regulatory system in yeast. J Gen Microbiol. 1966;44(2):149–56.View ArticleGoogle Scholar
- Arjunan P, Umland T, Dyda F, Swaminathan S, Furey W, Sax M, Farrenkopf B, Gao Y, Zhang D, Jordan F. Crystal structure of the thiamin diphosphate-dependent enzyme pyruvate decarboxylase from the yeast Saccharomyces cerevisiae at 2.3Å resolution. J Mol Biol. 1996;256(3):590–600.View ArticleGoogle Scholar
- Pohl M. Protein design on pyruvate decarboxylase (PDC) by site-directed mutagenesis. In: Scheper T, editor. New enzymes for organic synthesis. Heidelberg: Springer; 1997.Google Scholar
- Baburina I, Gao Y, Hu Z, Jordan F, Hohmann S, Furey W. Substrate activation of brewers’ yeast pyruvate decarboxylase is abolished by mutation of cysteine 221 to serine. Biochemistry. 1994;33(18):5630–5.View ArticleGoogle Scholar
- Nan H, Seo S-O, Oh EJ, Seo J-H, Cate JH, Jin Y-S. 2, 3-Butanediol production from cellobiose by engineered Saccharomyces cerevisiae. Appl Microbiol Biotechnol. 2014;98:5757–64.View ArticleGoogle Scholar
- Atsumi S, Li Z, Liao JC. Acetolactate synthase from Bacillus subtilis serves as a 2-ketoisovalerate decarboxylase for isobutanol biosynthesis in Escherichia coli. Appl Environ Microbiol. 2009;75(19):6306–11.View ArticleGoogle Scholar
- Flick KM, Spielewoy N, Kalashnikova TI, Guaderrama M, Zhu Q, Chang H-C, Wittenberg C. Grr1-dependent inactivation of Mth1 mediates glucose-induced dissociation of Rgt1 from HXT gene promoters. Mol Biol Cell. 2003;14(8):3230–41.View ArticleGoogle Scholar
- Kim J-H, Brachet V, Moriya H, Johnston M. Integration of transcriptional and posttranslational regulation in a glucose signal transduction pathway in Saccharomyces cerevisiae. Eukaryot Cell. 2006;5(1):167–73.View ArticleGoogle Scholar
- Kim J-W, Chin Y-W, Park Y-C, Seo J-H. Effects of deletion of glycerol-3-phosphate dehydrogenase and glutamate dehydrogenase genes on glycerol and ethanol metabolism in recombinant Saccharomyces cerevisiae. Bioproc Biosyst Eng. 2012;35(1):49–54.View ArticleGoogle Scholar
- Jiang R, Riebel BR, Bommarius AS. Comparison of alkyl hydroperoxide reductase (AhpR) and water-forming NADH oxidase from Lactococcus lactis ATCC 19435. Adv Synth Catal. 2005;347(7–8):1139–46.View ArticleGoogle Scholar
- Vemuri G, Eiteman M, McEwen J, Olsson L, Nielsen J. Increasing NADH oxidation reduces overflow metabolism in Saccharomyces cerevisiae. Proc Natl Acad Sci USA. 2007;104(7):2402–7.View ArticleGoogle Scholar
- Zhang G-C, Liu J-J, Ding W-T. Decreased xylitol formation during xylose fermentation in Saccharomyces cerevisiae due to overexpression of water-forming NADH oxidase. Appl Environ Microbiol. 2012;78(4):1081–6.View ArticleGoogle Scholar
- Heux S, Cachon R, Dequin S. Cofactor engineering in Saccharomyces cerevisiae: expression of a H2O-forming NADH oxidase and impact on redox metabolism. Metab Eng. 2006;8(4):303–14.View ArticleGoogle Scholar
- LopezdeFelipe F, Hugenholtz J. Purification and characterisation of the water forming NADH-oxidase from Lactococcus lactis. Int Dairy J. 2001;11(1):37–44.View ArticleGoogle Scholar
- González E, Fernández MR, Larroy C, Solà LS, Pericàs MA, Parés X, Biosca JA. Characterization of a (2R, 3R)-2, 3-butanediol dehydrogenase as the Saccharomyces cerevisiae YAL060W gene product. J Biol Chem. 2000;275(46):35876–85.View ArticleGoogle Scholar
- Albertyn J, van Tonder A, Prior BA. Purification and characterization of glycerol-3-phosphate dehydrogenase of Saccharomyces cerevisiae. FEBS Lett. 1992;308(2):130–2.View ArticleGoogle Scholar
- Roustan JL, Sablayrolles J-M. Modification of the acetaldehyde concentration during alcoholic fermentation and effects on fermentation kinetics. J Biosci Bioeng. 2002;93(4):367–75.View ArticleGoogle Scholar
- Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem. 1976;72(1):248–54.View ArticleGoogle Scholar
- Kim S, Lee DY, Wohlgemuth G, Park HS, Fiehn O, Kim KH. Evaluation and optimization of metabolome sample preparation methods for Saccharomyces cerevisiae. Anal Chem. 2013;85(4):2169–76.View ArticleGoogle Scholar
- Lee DY, Fiehn O. High quality metabolomic data for Chlamydomonas reinhardtii. Plant Methods. 2008;4(1):7.View ArticleGoogle Scholar
- Denkert C, Budczies J, Weichert W, Wohlgemuth G, Scholz M, Kind T, Niesporek S, Noske A, Buckendahl A, Dietel M. Metabolite profiling of human colon carcinoma–deregulation of TCA cycle and amino acid turnover. Mol Cancer. 2008;7(1):72.View ArticleGoogle Scholar
- Nikawa J-I, Tsukagoshi Y, Yamashita S. Isolation and characterization of two distinct myo-inositol transporter genes of Saccharomyces cerevisiae. J Biol Chem. 1991;266(17):11184–91.Google Scholar
- Mumberg D, Müller R, Funk M. Yeast vectors for the controlled expression of heterologous proteins in different genetic backgrounds. Gene. 1995;156(1):119–22.View ArticleGoogle Scholar
- Kim S-J, Seo S-O, Park Y-C, Jin Y-S, Seo J-H. Production of 2, 3-butanediol from xylose by engineered Saccharomyces cerevisiae. J Biotechnol. 2014;192:374–82.Google Scholar