Efficient production of acetoin in Saccharomyces cerevisiae by disruption of 2,3-butanediol dehydrogenase and expression of NADH oxidase

Acetoin is widely used in food and cosmetic industry as taste and fragrance enhancer. For acetoin production in this study, Saccharomyces cerevisiae JHY605 was used as a host strain, where the production of ethanol and glycerol was largely eliminated by deleting five alcohol dehydrogenase genes (ADH1, ADH2, ADH3, ADH4, and ADH5) and two glycerol 3-phosphate dehydrogenase genes (GPD1 and GPD2). To improve acetoin production, acetoin biosynthetic genes from Bacillus subtilis encoding α-acetolactate synthase (AlsS) and α-acetolactate decarboxylase (AlsD) were overexpressed, and BDH1 encoding butanediol dehydrogenase, which converts acetoin to 2,3-butanediol, was deleted. Furthermore, by NAD+ regeneration through overexpression of water-forming NADH oxidase (NoxE) from Lactococcus lactis, the cofactor imbalance generated during the acetoin production from glucose was successfully relieved. As a result, in fed-batch fermentation, the engineered strain JHY617-SDN produced 100.1 g/L acetoin with a yield of 0.44 g/g glucose.


Results and Discussion
Introduction of acetoin biosynthetic pathway in adh1-5Δgpd1Δgpd2Δ strain (JHY605). To produce acetoin in S. cerevisiae, α -acetolactate synthase (alsS) and α -acetolactate decarboxylase (alsD) genes from B. subtilis were introduced into JHY605 strain (adh1-5Δ gpd1Δ gpd2Δ ), which lacks five alcohol dehydrogenases (Adh1 to Adh5) and two glycerol 3-phosphate dehydrogenases (Gpd1 and Gpd2) (Fig. 1). The fermentation profiles of the control strain JHY605-C harboring empty vector p413GPD and strain JHY605-SD expressing alsS and alsD from p413-SD plasmid are shown in Fig. 2. JHY605-C produced only a trace amount of acetoin (0.1 g/L) and 1.4 g/L of 2,3-butanediol from 31.6 g/L of glucose after 96 h fermentation (Fig. 2a). Although five alcohol dehydrogenase genes were deleted, JHY605-C produced 3.8 g/L of ethanol as a major end product. This might be because pyruvate generated by glycolysis is mainly metabolized to ethanol production via pyruvate decarboxylases and remaining ADH isozymes including Sfa1, Adh6, and Adh7. In agreement with previous study, glycerol pathway was completely blocked in JHY605-C by the deletion of GPD1 and GPD2 14,18 .
Whereas, strain JHY605 harboring p413-SD plasmid (JHY605-SD) showed an increase in glucose consumption rate and produced up to 5.9 g/L acetoin (Fig. 2b). Furthermore, 2,3-butanediol production level increased to 9.3 g/L, even though 2,3-butanediol dehydrogenase was not overexpressed, reflecting the endogenous 2,3-butanediol dehydrogenase activity in S. cerevisiae. By introducing this competing pathway, ethanol production yield was reduced to 0.02 g/g glucose in JHY605-SD compared with that in JHY605-C (0.12 g/g glucose).
In S. cerevisiae, NAD + regeneration for glycolysis is mainly achieved by producing ethanol and glycerol (Fig. 1). Therefore, the growth defects of strain JHY605 might be due to the accumulation of NADH. In addition, the residual ADH activity might not be enough to prevent the accumulation of toxic acetaldehyde (Fig. 1). Introduction of acetoin biosynthetic pathway into JHY605 might relieve the growth defects by reducing acetaldehyde formation through efficient conversion of pyruvate to acetoin, and also by partly restoring NAD + regeneration through 2,3-butanediol production (Fig. 1).
Scientific RepoRts | 6:27667 | DOI: 10.1038/srep27667 Disruption of 2,3-butanediol dehydrogenase BDH1 to improve acetoin production. By introducing acetoin biosynthetic pathway into JHY605, pyruvate flux was successfully redirected toward acetoin pathway. However, acetoin was further converted to 2,3-butanediol, resulting in about 1.5-fold higher titer of 2,3-butanediol than that of acetoin. In S. cerevisiae, Bdh1 is a major enzyme catalyzing the reduction of acetoin to 2,3-butanediol ( Fig. 1). Therefore, we further deleted BDH1 gene in JHY605, resulting in strain JHY617. When acetoin biosynthetic pathway was introduced into strain JHY617 (JHY617-SD), 2,3-butanediol production from acetoin was significantly reduced to 0.2 g/L, which then contributed to the increase in acetoin production accordingly. As a result, up to 15.4 g/L acetoin was produced after 72 h fermentation in SC-His medium containing 50 g/L glucose, with a yield of 0.30 g/g glucose (Fig. 3a). The trace amount of 2,3-butanediol production in JHY617-SD might be mediated by other minor putative enzymes such as D-arabinose dehydrogenase (Ara1) having 2,3-butanediol dehydrogenase activity 19 . Recovering redox imbalance by expressing water-forming NADH oxidase noxE. Cofactor balance, especially NADH/NAD + ratio plays an important role in a large number of biochemical reactions 20,21 . Thus, maintaining the cofactor balance is an essential requirement for sustaining cellular metabolism and cell growth 22 . In acetoin production pathway, NADH produced from glycolysis could not be converted to NAD + , leading to a redox cofactor imbalance. Furthermore, since NADH-dependent metabolic pathways, related to the production of ethanol, glycerol, and 2,3-butanediol, were disrupted in strain JHY617-SD, the redox imbalance might be more severe. Therefore, as an effort to resolve the redox imbalance in JHY617-SD, we introduced noxE from L. lactis, encoding water-forming NADH oxidase. To this end, FBA1 promoter controlled-noxE was inserted to the acetoin biosynthetic plasmid p413-SD, resulting in p413-SDN. Strain JHY617 harboring p413-SDN (JHY617-SDN) showed a significant improvement in glucose consumption rate, thereby taking less time (~48 h) to completely ferment 50 g/L glucose than it was taken for JHY617-SD (~72 h) (Fig. 3). Moreover, acetoin production was improved up to 20.1 g/L with a yield of 0.39 g/g glucose, reaching 80% of maximum theoretical yield. Accordingly, strain JHY617-SDN exhibited about two-fold increase in acetoin productivity (0.42 g/(L·h)) compared with JHY617-SD (0.21 g/(L·h)), suggesting that redox imbalance caused by acetoin production was successfully alleviated by expressing NADH oxidase (Table 1). To confirm the effect of noxE expression on redox state, we analyzed intracellular NADH/NAD + ratios in JHY617-SD and JHY617-SDN. As expected, the NADH/NAD + ratios in JHY617-SDN were lower than those in JHY617-SD throughout the growth phase, demonstrating the efficient conversion of NADH to NAD + by NoxE (Fig. 3c). Fed-batch fermentation for acetoin production. To evaluate the potential of JHY617-SDN as a host strain for acetoin production, fed-batch fermentation was performed with intermittent feeding of glucose and pH control. JHY617-SDN was grown in YPD medium containing 100 g/L glucose with initial OD 600 of 9.5. In fed-batch fermentation, up to 100.1 g/L acetoin was produced with a yield of 0.44 g/g glucose after 55 h cultivation, reaching 90% of maximum theoretical yield (Fig. 4). Moreover, acetoin productivity was dramatically improved to 1.82 g/(L·h). Taken together, JHY617-SDN showed superior performance of acetoin production compared with the host strains reported in previous studies (Table 2). Notably, both acetoin titer and yield in this study are the highest among these studies. Although acetoin productivity reported in S. marcescens and E. cloacae were higher than that of our study 10,12 , these strains have potential pathogenicity 23,24 .
In this study, we developed S. cerevisiae strain for efficient production of acetoin by introducing heterologous acetoin pathway from B. subtilis and eliminating 2,3-butanediol dehydrogenase using JHY605 as a host strain, where the production of ethanol and glycerol was largely eliminated. In addition, cofactor imbalance generated during acetoin production was successfully alleviated by expressing NADH oxidase from L. lactis, leading to significantly enhanced acetoin production. As a result, to the best of our knowledge, the highest titer and yield in microbial production of acetoin were achieved. These results suggest that S. cerevisiae might be a promising host for the production of acetoin.

Strains and media.
All strains used in this study are described in Table 3. JHY617 strain, a BDH1 deletion mutant derived from JHY605 14 , was generated by PCR-mediated homologous recombination. The bdh1Δ ::KanMX6 cassette flanked by 300 bp upstream and 282 bp downstream of the BDH1 open reading frame was obtained by PCR amplification from genomic DNA of bdh1Δ strain (BY4741 bdh1Δ ::KanMX6, EUROSCARF) as a template, using the primer pair of d_BDH1 F (5′ -GATTTGCTCACGCTACTTTG-3′ ) and d_BDH1 R (5′ -GCCATGCTTTGTTTTAGACG-3′ ). The resulting PCR product was transformed into JHY605 strain and transformants were selected on YPD plate (10 g/L yeast extract, 20 g/L bacto-peptone, and 20 g/L glucose) supplemented with 200 μ g/mL G418 sulfate (AG Scientific, Inc.) Yeast cells were cultured in YPD medium or in synthetic complete medium lacking histidine (SC-His) (20 or 50 g/L glucose, 6.7 g/L yeast nitrogen base without amino acids, and 1.92 g/L amino acids mixture lacking histidine).
Plasmid construction. Plasmids used in this study are described in Table 4. The recombinant plasmids for acetoin pathway were constructed by using the multiple cloning system as previously described with minor modifications 14 . The alsS-expression cassette (P TDH3 -alsS-T CYC1 ) flanked by MauBI and and NotI sites was obtained by PCR from p413_P TDH3 -alsS-T CYC1 using the primers, Univ F2 (5′-GACTCGCGCGCGGGAACAAAAGCTGGAGCTC-3′ ) and Univ R (5′-GACTACGCGTGCGGCCGCTAATGGCGCGCCATAGGGCGAATTGGGTACC-3′ ), and  Strain JHY617-SDN was cultivated in YPD medium containing 100 g/L glucose with initial OD 600 of 9.5. Glucose was intermittently added into culture medium using the feeding solution (800 g/L glucose) before glucose was completely consumed.
Fermentation conditions. For flask fermentation, yeast cells harboring appropriate plasmids were pre-cultured in 5 mL of SC-His medium containing 20 g/L glucose in a 50 mL flask, inoculated to OD 600 of 0.3 in 10 mL of SC-His medium containing 50 g/L glucose in a 100 mL flask, and then cultivated at 30 °C with shaking at 170 rpm. Fed-batch fermentation was performed in 500 mL YPD medium containing 100 g/L glucose using a 1 L bench-top fermenter FMT-DS (Fermentec, Korea) at 30 °C with agitation speed of 500 rpm and air flow rate of 1.0 vvm. The pH of the culture medium was maintained at 5.5 by using 4 N NaOH. Strain JHY617-SDN was pre-cultured in SC-His medium containing 20 g/L glucose and inoculated into the fermenter with initial OD 600    of 9.5. The feeding solution (800 g/L glucose) was intermittently added to the culture medium when the glucose concentration was lower than 20 g/L.
Analytical methods. Cell growth was determined by measuring the optical density at 600 nm (OD 600 ).
To analyze profile of metabolites, 1 mL of culture supernatants were collected and filtered through a 0.22 μ m syringe filter. The concentrations of glucose, glycerol, acetoin, 2,3-butanediol, and ethanol were determined by high performance liquid chromatography (HPLC) using UltiMate 3000 HPLC system (Thermo fishers scientific) equipped with a BioRad Aminex HPX-87H column (300 mm× 7.8 mm, 5 μ m, Bio-rad) at 60 °C with 5 mM H 2 SO 4 at a flow rate of 0.6 mL/min and refractive index (RI) detector at 35 °C. The intracellular concentrations of NADH and NAD + were measured using EnzyChrom ™ NAD/NADH Assay Kit (E2ND-100, BioAssay Systems). Strains JHY617-SD and JHY617-SDN were harvested at different time points of fermentation and washed with cold phosphate-buffered saline (PBS; 8 g/L NaCl, 0.2 g/L KCl, 1.42 g/L Na 2 HPO 4 , 0.24 g/L KH 2 PO 4 [pH 7.4]) solution. Cells of OD 600 of 1.0 were pelleted and analyzed according to the manufacturer's instructions.