Fed-batch with biomass recycle and batch production of 2,3-butanediol from glucose by Bacillus polymyxa

The production of 2,3-butanediol from glucose by Bacillus polymyxa in batch was sensitive to both protein concentration and aeration rate. Two fed-batch experiments which were resupplied from a reservoir containing urea as the sole source of nitrogen, and incorporated total biomass recycle resulted in yields of 65.84 mM and 69.66 mM of 2,3-butanediol per 100 mM of glucose utilized. No massive sporulation events were detected. In addition, fed-batch with recycle yielded more butanediol and less acetate than any batch run.     

Enhanced 2,3-butanediol production by Klebsiella oxytoca using a two-stage agitation speed control strategy

Batch fermentative production of 2,3-butanediol by Klebsiella oxytoca was investigated using various oxygen supply methods though varying agitation speed. Based on the analysis of three kinetic parameters including specific cell growth rate (μ), specific glucose consumption rate (q_s) and specific 2,3-butanediol formation rate (q_p), a two-stage agitation speed control strategy, aimed at achieving high concentration, high yield and high productivity of 2,3-butanediol, was proposed. At the first 15 h, agitation speed was controlled at 300 rpm to obtain high μ for cell growth, subsequently agitation speed was controlled at 200 rpm to maintain high q_p for high 2,3-butanediol accumulation. Finally, the maximum concentration of 2,3-butanediol reached 95.5 g l⁻¹ with the yield of 0.478 g g⁻¹ and the productivity of 1.71 g l⁻¹ h⁻¹, which were 6.23%, 6.22% and 22.14% over the best results controlled by constant agitation speeds.     

Effect of oxygen supply and dilution rate on the production of 2,3-butanediol in continuous bioreactor by Klebsiella pneumoniae

Kinetics of 2,3-butanediol production by Klebsiella pneumoniae (NRRL B199) from glucose have been studied in a continuous bioreactor. The effect of oxygen supply rate and dilution rate on the product output rate and yield of 2,3-butanediol were investigated. For a feed glucose concentration of 100 g l⁻¹, the optimum oxygen transfer rate is between 25.0–35.0 mmol l⁻¹ h⁻¹. Under these conditions, maximum product concentration obtained was 35 g l⁻¹ at a dilution rate of 0.1 h⁻¹ and the maximum product output rate obtained was 4.25 g l⁻¹ h⁻¹. The product yield based on the substrate utilized approached the theoretical value (50%) at low values of oxygen transfer rate but decreased with increasing oxygen transfer rate.     

Engineered <Emphasis Type="Italic">E. coli</Emphasis> W enables efficient 2,3-butanediol production from glucose and sugar beet molasses using defined minimal medium as economic basis

Background

Efficient microbial production of chemicals is often hindered by the cytotoxicity of the products or by the pathogenicity of the host strains. Hence 2,3-butanediol, an important drop-in chemical, is an interesting alternative target molecule for microbial synthesis since it is non-cytotoxic. Metabolic engineering of non-pathogenic and industrially relevant microorganisms, such as Escherichia coli, have already yielded in promising 2,3-butanediol titers showing the potential of microbial synthesis of 2,3-butanediol. However, current microbial 2,3-butanediol production processes often rely on yeast extract as expensive additive, rendering these processes infeasible for industrial production.

Results

The aim of this study was to develop an efficient 2,3-butanediol production process with E. coli operating on the premise of using cost-effective medium without complex supplements, considering second generation feedstocks. Different gene donors and promoter fine-tuning allowed for construction of a potent E. coli strain for the production of 2,3-butanediol as important drop-in chemical. Pulsed fed-batch cultivations of E. coli W using microaerobic conditions showed high diol productivity of 4.5 g l^?1 h^?1. Optimizing oxygen supply and elimination of acetoin and by-product formation improved the 2,3-butanediol titer to 68 g l^?1, 76% of the theoretical maximum yield, however, at the expense of productivity. Sugar beet molasses was tested as a potential substrate for industrial production of chemicals. Pulsed fed-batch cultivations produced 56 g l^?1 2,3-butanediol, underlining the great potential of E. coli W as production organism for high value-added chemicals.

Conclusion

A potent 2,3-butanediol producing E. coli strain was generated by considering promoter fine-tuning to balance cell fitness and production capacity. For the first time, 2,3-butanediol production was achieved with promising titer, rate and yield and no acetoin formation from glucose in pulsed fed-batch cultivations using chemically defined medium without complex hydrolysates. Furthermore, versatility of E. coli W as production host was demonstrated by efficiently converting sucrose from sugar beet molasses into 2,3-butanediol.

     

Production of 2,3-butanediol in a membrane bioreactor with cell recycle

Summary The production of 2,3-butanediol by Enterobacter aerogenes DSM 30053 was studied in a cell recycle system with a microfiltration module. Emphasis was put on the influence of oxygen supply, cell residence time, dilution rate, and pH. Under optimal conditions a productivity as high as 14.6 g butanediol + acetoin/l per hour was achieved with a product concentration of 54 g/l and a product yield of 88%. This productivity is three times higher than that of an ordinary continuous culture. The achievable final product concentration of a cell recycle system was limited by the accumulation of the inhibiting by-product acetic acid, which increased very rapidly at low dilution rate. To maximize product concentration a fed-batch fermentation was carried out with stepwise pH adaption at high cell density. A final product concentration of 110 g/l was obtained with a productivity of 5.4 g/l per hour and a yield of 97%.     

Effects of oxygen supply on yeast growth and metabolism in continuous fermentation

The yield changes in cell mass and metabolites with changes in the oxygen supply rate were investigated in continuous ethanol fermentation. With increases in oxygen concentration in the purging gas from 5.3 to 39.3 %, the specific oxygen uptake rate (qO₂) increased from 0.158 to 1.24 mmol/g/h. With this change, cell mass increased from 13.2 to 14.9 g/l and glycerol decreased from 4.8 to 0.99 g/l, although little change in ethanol yield was observed. At a higher oxygen concentration and/or at a lower respiratory quotient (RQ), glycerol disappeared, acetaldehyde, acetoin and 2,3-butanediol increased, and ethanol started to decrease. The yields of iso-butylalcohol and iso-amylalcohol also increased with increases in the oxygen supply rate when RQ was lower than approximately 10. Reduction in the redox balance (NADH/NAD) in the cells by qO₂, appeared to reduce initially the rate of glycerol-3-phosphate formation and next the rate of ethanol formation, resulting in the accumulation of acetaldehyde and formation of 2,3-butanediol through acetoin. Fatty acid composition changed with changes in the oxygen supply rate. The value for unsaturation, Δ mol⁻¹, increased from 0.745 to 0.836 with the increase in qO₂ from 0.158 to 1.79 mmol/g/h. Increases in oleic acid (C18:1) and decreases in palmitic acid (C16:0) were the major changes with the increases in Δ mol⁻¹.     

Effect of deletion of 2,3-butanediol dehydrogenase gene (bdhA) on acetoin production of Bacillus subtilis

The present work aims to block 2,3-butanediol synthesis in acetoin fermentation of Bacillus subtilis. First, we constructed a recombinant strain BS168D by deleting the 2,3-butanediol dehydrogenase gene bdhA of the B. subtilis168, and there was almost no 2,3-butanediol production in 20?g/L of glucose media. The acetoin yield of BS168D reached 6.61?g/L, which was about 1.5 times higher than that of the control B. subtilis168 (4.47?g/L). Then, when the glucose concentration was increased to 100?g/L, the acetoin yield reached 24.6?g/L, but 2.4?g/L of 2,3-butanediol was detected at the end of fermentation. The analysis of 2,3-butanediol chiral structure indicated that the main 2,3-butanediol production of BS168D was meso-2,3-butanediol, and the bdhA gene was only responsible for (2R,3R)-2,3-butanediol synthesis. Therefore, we speculated that there may exit another pathway relating to the meso-2,3-butanediol synthesis in the B. subtilis. In addition, the results of low oxygen condition fermentation showed that deletion of bdhA gene successfully blocked the reversible transformation between acetoin and 2,3-butanediol and eliminated the effect of dissolved oxygen on the transformation.     

Improvement of acetoin reductase activity enhances bacitracin production by Bacillus licheniformis

Bacitracin fermentation by Bacillus licheniformis in this work showed three characteristics: (1) the extracellular propionate, butyrate, acetoin and 2,3-butanediol accumulates under conditions of low dissolved oxygen (zero after 4 h cultivation), reaching a total content of approximately 11.1 g/L; (2) cell growth occurs quickly subsequent to cell autolysis and the second growth; and (3) there is a low content of 2,3-butanediol, a reduced product of acetoin catalyzed by acetoin reductase, in the culture process. In this study, addition of MnCl₂ (0.3 mg/L) to the production medium increased the acetoin reductase activity, redirected the NADH oxidation partly from the propionate- and butyrate-production pathways to the 2,3-butanediol synthesis pathway, reduced the intracellular NADH/NAD⁺ ratio, and facilitated cell growth, ultimately achieving a 11.6% increase in bacitracin production (1076 U/mL) versus the control. The results provide useful information regarding large-scale bacitracin production by B. licheniformis.     

Reconstruction of an Acetogenic 2,3-Butanediol Pathway Involving a Novel NADPH-Dependent Primary-Secondary Alcohol Dehydrogenase

Acetogenic bacteria use CO and/or CO₂ plus H₂ as their sole carbon and energy sources. Fermentation processes with these organisms hold promise for producing chemicals and biofuels from abundant waste gas feedstocks while simultaneously reducing industrial greenhouse gas emissions. The acetogen Clostridium autoethanogenum is known to synthesize the pyruvate-derived metabolites lactate and 2,3-butanediol during gas fermentation. Industrially, 2,3-butanediol is valuable for chemical production. Here we identify and characterize the C. autoethanogenum enzymes for lactate and 2,3-butanediol biosynthesis. The putative C. autoethanogenum lactate dehydrogenase was active when expressed in Escherichia coli. The 2,3-butanediol pathway was reconstituted in E. coli by cloning and expressing the candidate genes for acetolactate synthase, acetolactate decarboxylase, and 2,3-butanediol dehydrogenase. Under anaerobic conditions, the resulting E. coli strain produced 1.1 ± 0.2 mM 2R,3R-butanediol (23 μM h⁻¹ optical density unit⁻¹), which is comparable to the level produced by C. autoethanogenum during growth on CO-containing waste gases. In addition to the 2,3-butanediol dehydrogenase, we identified a strictly NADPH-dependent primary-secondary alcohol dehydrogenase (CaADH) that could reduce acetoin to 2,3-butanediol. Detailed kinetic analysis revealed that CaADH accepts a range of 2-, 3-, and 4-carbon substrates, including the nonphysiological ketones acetone and butanone. The high activity of CaADH toward acetone led us to predict, and confirm experimentally, that C. autoethanogenum can act as a whole-cell biocatalyst for converting exogenous acetone to isopropanol. Together, our results functionally validate the 2,3-butanediol pathway from C. autoethanogenum, identify CaADH as a target for further engineering, and demonstrate the potential of C. autoethanogenum as a platform for sustainable chemical production.     

Cloning,expression and characterization of glycerol dehydrogenase involved in 2,3-butanediol formation in Serratia marcescens H30

The meso-2,3-butanediol dehydrogenase (meso-BDH) from S. marcescens H30 is responsible for converting acetoin into 2,3-butanediol during sugar fermentation. Inactivation of the meso-BDH encoded by budC gene does not completely abolish 2,3-butanediol production, which suggests that another similar enzyme involved in 2,3-butanediol formation exists in S. marcescens H30. In the present study, a glycerol dehydrogenase (GDH) encoded by gldA gene from S. marcescens H30 was expressed in Escherichia coli BL21(DE3), purified and characterized for its properties. In vitro conversion indicated that the purified GDH could catalyze the interconversion of (3S)-acetoin/meso-2,3-butanediol and (3R)-acetoin/(2R,3R)-2,3-butanediol. (2S,3S)-2,3-Butanediol was not a substrate for the GDH at all. Kinetic parameters of the GDH enzyme showed lower K _m value and higher catalytic efficiency for (3S/3R)-acetoin in comparison to those for (2R,3R)-2,3-butanediol and meso-2,3-butanediol, implying its physiological role in favor of 2,3-butanediol formation. Maximum activity for reduction of (3S/3R)-acetoin and oxidations of meso-2,3-butanediol and glycerol was observed at pH 8.0, while it was pH 7.0 for diacetyl reduction. The enzyme exhibited relative high thermotolerance with optimum temperature of 60 °C in the oxidation–reduction reactions. Over 60 % of maximum activity was retained at 70 °C. Additionally, the GDH activity was significantly enhanced for meso-2,3-BD oxidation in the presence of Fe²⁺ and for (3S/3R)-acetoin reduction in the presence of Mn²⁺, while several cations inhibited its activity, particularly Fe²⁺ and Fe³⁺ for (3S/3R)-acetoin reduction. The properties provided potential application for single configuration production of acetoin and 2,3-butanediol .     

Improved 2,3-butanediol production from corncob acid hydrolysate by fed-batch fermentation using Klebsiella oxytoca

Corncob acid hydrolysate, detoxed by sequently boiling, overliming and activated charcoal adsorption, was used for 2,3-butanediol production by Klebsiella oxytoca ACCC 10370. The effects of acetate in hydrolysate and pH on 2,3-butanediol production were investigated. It was found that acetic acid in hydrolysate inhibited the growth of K. oxytoca while benefited the 2,3-butanediol yield. With the increase in acetic acid concentration in medium from 0 to 4 g/l, the lag phase was prolonged and the specific growth rate decreased. The acetic acid inhibition on cell growth can be alleviated by adjusting pH to 6.3 prior to fermentation and a substrate fed-batch strategy with a low initial acetic acid concentration. Under the optimum condition, a maximal 2,3-butanediol concentration of 35.7 g/l was obtained after 60 h of fed-batch fermentation, giving a yield of 0.5 g/g reducing sugar and a productivity of 0.59 g/h l.     

2,3-Butanediol production from Jerusalem artichoke,Helianthus tuberosus,by Bacillus polymyxa ATCC 12 321. Optimization of k L a profile

Summary Optimization of d-(-)-2,3-butanediol production from the Jerusalem artichoke, Helianthus tuberosus, by Bacillus polymyxa ATCC 12 321 is described. The effects of initial sugar concentration and oxygen transfer rate were examined. The latter appears to be the most important parameter affecting the kinetics of the process. The best results (44 g·l^-1 2,3-butanediol, productivity of 0.79 g·l^-1·h^-1) were obtained by setting an optimal k _L a profile during batch culture.     

2,3-Butanediol Metabolism in the Acetogen Acetobacterium woodii

The acetogenic bacterium Acetobacterium woodii is able to reduce CO₂ to acetate via the Wood-Ljungdahl pathway. Only recently we demonstrated that degradation of 1,2-propanediol by A. woodii was not dependent on acetogenesis, but that it is disproportionated to propanol and propionate. Here, we analyzed the metabolism of A. woodii on another diol, 2,3-butanediol. Experiments with growing and resting cells, metabolite analysis and enzymatic measurements revealed that 2,3-butanediol is oxidized in an NAD⁺-dependent manner to acetate via the intermediates acetoin, acetaldehyde, and acetyl coenzyme A. Ethanol was not detected as an end product, either in growing cultures or in cell suspensions. Apparently, all reducing equivalents originating from the oxidation of 2,3-butanediol were funneled into the Wood-Ljungdahl pathway to reduce CO₂ to another acetate. Thus, the metabolism of 2,3-butanediol requires the Wood-Ljungdahl pathway.     

The Bacillus subtilis ydjL (bdhA) Gene Encodes Acetoin Reductase/2,3-Butanediol Dehydrogenase

Bacillus subtilis is capable of producing 2,3-butanediol from acetoin by fermentation, but to date, the gene encoding the enzyme responsible, acetoin reductase/2,3-butanediol dehydrogenase (AR/BDH), has remained unknown. A search of the B. subtilis genome database with the amino acid sequences of functional AR/BDHs from Saccharomyces cerevisiae and Bacillus cereus resulted in the identification of a highly similar protein encoded by the B. subtilis ydjL gene. A knockout strain carrying a ydjL::cat insertion mutation was constructed, which (i) abolished 2,3-butanediol production in early stationary phase, (ii) produced no detectable AR or BDH activity in vitro, and (iii) accumulated the precursor acetoin in early stationary phase. The ydjL::cat mutation also affected the kinetics of lactate but not acetate production during stationary-phase cultivation with glucose under oxygen limitation. A very small amount of 2,3-butanediol was detected in very-late-stationary-phase (96-hour) cultures of the ydjL::cat mutant, suggesting the existence of a second gene encoding a minor AR activity. From the data, it is proposed that the major AR/BDH-encoding gene ydjL be renamed bdhA.     

Engineering of the 2,3-butanediol pathway of Paenibacillus polymyxa DSM 365

Paenibacillus polymyxa is a Gram-positive, non-pathogenic soil bacterium that has been extensively investigated for the production of R-,R-2,3-butanediol in exceptionally high enantiomeric purity. Rational metabolic engineering efforts to increase productivity and product titers were restricted due to limited genetic accessibility of the organism up to now. By use of CRISPR-Cas9 mediated genome editing, six metabolic mutant variants were generated and compared in batch fermentations for the first time. Downstream processing was facilitated by completely eliminating exopolysaccharide formation through the combined knockout of the sacB gene and the clu1 region, encoding for the underlying enzymatic machinery of levan and paenan synthesis. Spore formation was inhibited by deletion of spoIIE, thereby disrupting the sporulation cascade of P. polymyxa. Optimization of the carbon flux towards 2,3-butanediol was achieved by deletion of the lactate dehydrogenase ldh1 and decoupling of the butanediol dehydrogenase from its natural regulation via constitutive episomal expression. The improved strain showed 45 % increased productivity, reaching a final concentration of 43.8 g L⁻¹ butanediol. A yield of 0.43 g g⁻¹ glucose was achieved, accounting for 86 % of the theoretical maximum.     

Effect of pH on the metabolic flux of Klebsiella oxytoca producing 2,3-butanediol in continuous cultures at different dilution rates

The efficiency of the bioconversion process and the achievable end-product concentration decides the economic feasibility of microbial 2,3-butanediol (2,3-BDO) production. In 2,3-BDO production, optimization of culture condition is required for cell growth and metabolism. Also, the pH is an important factor that influences microbial performance. For different microorganisms and substrates, it has been shown that the distribution of the metabolites in 2,3-BDO fermentation is greatly affected by pH, and the optimum pH for 2,3-BDO production seems dependently linked to the particular strain and the substrate employed. Quantification analysis of intracellular metabolites and metabolic flux analysis (MFA) were used to investigate the effect of pH on the Klebsiella oxytoca producing 2,3-BDO and other organic acids. The main objectives of MFA are the estimation of intracellular metabolic fluxes and the identification of rate-limiting step and the key enzymes. This study was conducted under continuous aerobic conditions at different dilution rates (0.1, 0.2, and 0.3 h^?1) and different pH values (pH 5.5 and 7.0) for the steady-state experimental data. In order to obtain the flux distribution, the extracellular specific rates were calculated from the experimental data using the metabolic network model of K. oxytoca. Intracellular metabolite concentration profiles were generated using ultra-performance liquid chromatography/quadrupole time-of-flight mass spectrometry.     

Enhanced production of 2,3-butanediol by engineered <Emphasis Type="Italic">Bacillus subtilis</Emphasis>

Production of 2,3-butanediol by Bacillus subtilis takes place in late-log or stationary phase, depending on the expression of bdhA gene encoding acetoin reductase, which converts acetoin to 2,3-butanediol. The present work focuses on the development of a strain of B. subtilis for enhanced production of 2,3-butanediol in early log phase of growth cycle. For this, the bdhA gene was expressed under the control of P _alsSD promoter of AlsSD operon for acetoin fermentation which served the substrate for 2,3-butanediol production. Addition of acetic acid in the medium induced the production of 2,3-butanediol by 2-fold. Two-step aerobic–anaerobic fermentation further enhanced 2,3-butanediol production by 4-fold in comparison to the control parental strain. Thus, addition of acetic acid and low dissolved oxygen in the medium are involved in activation of bdhA gene expression from P _alsSD promoter in early log phase. Under the conditions tested in this work, the maximum production of 2,3-butanediol, 2.1 g/l from 10 g/l glucose, was obtained at 24 h. Furthermore, under the optimized microaerophilic condition, the production of 2,3-butanediol improved up to 6.1 g/l and overall productivity increased by 6.7-fold to 0.4 g/l h in the engineered strain compared to that in the parental control.     

Glycerol Overproduction by Engineered Saccharomyces cerevisiae Wine Yeast Strains Leads to Substantial Changes in By-Product Formation and to a Stimulation of Fermentation Rate in Stationary Phase

Six commercial wine yeast strains and three nonindustrial strains (two laboratory strains and one haploid strain derived from a wine yeast strain) were engineered to produce large amounts of glycerol with a lower ethanol yield. Overexpression of the GPD1 gene, encoding a glycerol-3-phosphate dehydrogenase, resulted in a 1.5- to 2.5-fold increase in glycerol production and a slight decrease in ethanol formation under conditions simulating wine fermentation. All the strains overexpressing GPD1 produced a larger amount of succinate and acetate, with marked differences in the level of these compounds between industrial and nonindustrial engineered strains. Acetoin and 2,3-butanediol formation was enhanced with significant variation between strains and in relation to the level of glycerol produced. Wine strains overproducing glycerol at moderate levels (12 to 18 g/liter) reduced acetoin almost completely to 2,3-butanediol. A lower biomass concentration was attained by GPD1-overexpressing strains, probably due to high acetaldehyde production during the growth phase. Despite the reduction in cell numbers, complete sugar exhaustion was achieved during fermentation in a sugar-rich medium. Surprisingly, the engineered wine yeast strains exhibited a significant increase in the fermentation rate in the stationary phase, which reduced the time of fermentation.     

A Novel Whole-Cell Biocatalyst with NAD+ Regeneration for Production of Chiral Chemicals

Background

The high costs of pyridine nucleotide cofactors have limited the applications of NAD(P)-dependent oxidoreductases on an industrial scale. Although NAD(P)H regeneration systems have been widely studied, NAD(P)⁺ regeneration, which is required in reactions where the oxidized form of the cofactor is used, has been less well explored, particularly in whole-cell biocatalytic processes.

Methodology/Principal Findings

Simultaneous overexpression of an NAD⁺ dependent enzyme and an NAD⁺ regenerating enzyme (H₂O producing NADH oxidase from Lactobacillus brevis) in a whole-cell biocatalyst was studied for application in the NAD⁺-dependent oxidation system. The whole-cell biocatalyst with (2R,3R)-2,3-butanediol dehydrogenase as the catalyzing enzyme was used to produce (3R)-acetoin, (3S)-acetoin and (2S,3S)-2,3-butanediol.

Conclusions/Significance

A recombinant strain, in which an NAD⁺ regeneration enzyme was coexpressed, displayed significantly higher biocatalytic efficiency in terms of the production of chiral acetoin and (2S,3S)-2,3-butanediol. The application of this coexpression system to the production of other chiral chemicals could be extended by using different NAD(P)-dependent dehydrogenases that require NAD(P)⁺ for catalysis.     

Occurrence of multiple forms of alcohol dehydrogenase in Penicillium supplemented with 2,3-butanediol

The NAD-dependent oxidation of ethanol, 2,3-butanediol, and other primary and secondary alcohols, catalyzed by alcohol dehydrogenases derived from Penicillium charlesii, was investigated. Alcohol dehydrogenase, ADH-I, was purified to homogeneity in a yield of 54%. The enzyme utilizes several primary alcohols as substrates, with K_m values of the order of 10^?4m. A K_m value of 60 mm was obtained for R,R,-2,3-butanediol. The stereospecificity of the oxidation of 2-butanol was investigated, and S-(+)-2-butanol was found to be oxidized 2.4 times faster than was R-(?)-2-butanol. The reduction of 2-butanone was shown to produce S-(+)-2-butanol and R-(?)-butanol in a ratio of 7:3. ADH-I is the primary isozyme of alcohol dehydrogenase present in cultures utilizing glucose as the sole carbon source. The level of alcohol dehydrogenase activity increased 7.6-fold in mycelia from cultures grown with glucose and 2,3-butanediol (0.5%) as carbon sources compared with the activity in cultures grown on only glucose. Two additional forms of alcohol dehydrogenase, ADH-II and ADH-III, were present in the cultures supplemented with 2,3-butanediol. These forms of alcohol dehydrogenase catalyze the oxidation of ethanol and 2,3-butanediol. These data suggest that P. charlesii carries out an oxidation of 2,3-butanediol which may constitute the first reaction in the degradation of 2,3-butanediol as well as the last reaction in the mixed-acid fermentation. Alcohol dehydrogenase activities in P. charlesii may be encoded by multiple genes, one which is expressed constitutively and others whose expression is inducible by 2,3-butanediol.