Isolation and identification of an acetoin high production bacterium ...

World J Microbiol Biotechnol (2011) 27:2785–2790 DOI 10.1007/s11274-011-0754-y

ORIGINAL PAPER

Isolation and identification of an acetoin high production bacterium that can reverse transform 2,3-butanediol to acetoin at the decline phase of fermentation Xian Zhang • Tao-wei Yang • Qing Lin • Mei-juan Xu • Hai-feng Xia • Zheng-hong Xu Hua-zhong Li • Zhi-ming Rao

•

Received: 27 December 2010 / Accepted: 16 April 2011 / Published online: 28 April 2011 Ó Springer Science+Business Media B.V. 2011

Abstract Acetoin (3-hydroxy-2-butanone), a very popular food spice is now used in many industries (pharmaceuticals, chemicals, paint, etc.). In this study, an acetoin high producing strain, numbered as JNA-310, was newly isolated and identified as Bacillus subtilis which is safe on food industry, based on its physiological, biological tests and 16S rDNA sequence analysis. When glucose was used as carbon source in fermentation, the fermentation characterizations of this strain were analyzed, and a new phenomenon of reverse transforming 2,3-butanediol which was synthesized from glucose in the fermentation broth to acetoin was detected. Before 96 h, glucose which was mainly transformed to 2,3-butanediol and acetoin was totally consumed, and the yield of the two products were 41.7 and 21.0 g/l respectively. Acetoin was only a by product in the fermentation broth at prophase of fermentation. At the end of fermentation, the yield of acetoin was greatly improved and the yield of 2,3-butanediol was declined and the yield of them were about 42.2 and 15.8 g/ l, respectively. The results indicated that 2,3-butanediol was reversely transformed to acetoin. Keywords Bacillus subtilis 3-hydroxy-2-butanone Acetoin Fermentation

X. Zhang T. Yang Q. Lin M. Xu H. Xia Z. Xu H. Li Z. Rao (&) The Key Laboratory of Industrial Biotechnology of Ministry of Education, Lab of Applied Microbiology and Metabolic Engineering, School of Biotechnology, Jiangnan University, Wuxi 214122, Jiangsu Province, China e-mail: [email protected]

Introduction As a valuable flavor naturally, acetoin not only has multiple usages in foods, flavor, cosmetic, and chemical synthesis, but also works as one of the important primary metabolites like ethanol, acetate, and lactate, in various microorganisms (Xiao et al. 2009). It has a wide range of applications extremely served as a spice which is mainly used for the production of butter, milk, yogurt, etc. Furthermore, it is also widely used as reagent in a number of chemical syntheses. Acetoin can be diagnosed by the Voges Proskauer test, so it serves as a microbial classification marker. It has vital physiological meanings to these microbes mainly including avoiding acidification, participating in the regulation of NAD/NADH ratio, and storing carbon (Xiao and Xu 2007). The present methods for acetoin production are divided into chemical synthesis and microbial fermentation. There are many chemical synthetic methods for acetoin preparation. For example, it is prepared from diacetyl by partial deoxidization with zinc or other catalyst (Martin et al. 1998; Fumio Toda et al. 1989; Slipszenko et al. 1998). Another common approach of acetoin production is selective oxidation of 2, 3-butanediol (Blom 1945; Hilmi et al. 1997). Chemical synthetic methods mentioned above are aggressive to the environment, and the raw materials are also bulk chemical products. As a result, the production and environmental costs are greatly increased. Microbial fermentation has become an increasingly popular choice because of less environmental pollution and cheap raw materials. Thus, natural production using fermentative has been got attentions by researchers. Many species able to produce acetoin such as Klebsiella pneumoniae (Qin et al. 2006; Yu and Saddler 1983), Bacillus subtilis (Dettwiler et al. 1993). In relevant reports

123

2786

of these species, acetoin is only a minor by-product of 2,3butanediol production. It is a continual constraint to the mass production of acetoin and the consideration of food safety is another limiting factor. As a result, there is an urgent need for safe and high-yielding strains. Bacillus subtilis is one of the mostly studied Grampositive bacterial species. Xiao attained 36.9 g/l acetoin by flask fermentation after about 60 h with the strain Bacillus subtilis CICC 10025 (Xiao et al. 2007). To our knowledge, it is one of the highest levels of acetoin production by fermentation. Therefore, on the basis of the general methods for screening spore-forming Bacillus, this experiment was performed using high concentration of glucose as substrate to obtain high substrate tolerating Bacillus from which we could re-screen acetoin producing strains. A new strain JNA 3-10 was screened and identified as B. subtilis, which was capable to accumulate acetoin as a major product. In prophase fermentation of strain JNA,2,3butanediol was the major product, and acetoin was transferred to 2,3-butanediol. In the anaphase of fermentation, by the course of cell dissolution, 2,3-butanedio was transferred to acetoin. At the end of fermentation, the yield of acetoin was about 42.2 g/l.

Materials and methods Screening procedure and media The soil samples (Yan Cheng, China) were made into suspension. Enrichment of spore-forming bacteria from the samples was performed by transferring 1 g of sample to a 250 ml flask containing 50 ml enrichment medium (per liter: typtone 5 g, glucose 10 g, yeast extract 3 g, MgSO47H2O 0.5 g, K2HPO4 0.5 g). The mixture was incubated at 37°C on a rotary shaker at 160 rpm for 24 h, then heated for 30 min in 80°C water. The enrichment culture was serially diluted and spared on spore growth medium (per liter: typtone 2.5 g, glucose 150 g, yeast extract 1.5 g,MgSO47H2O 0.5 g, K2HPO4 0.5 g) agar plates (prepared by adding 2% of agar), and the plates were incubated at 37°C for 12 h. Subsequently, single isolated colonies were transferred to test tubes containing 10 ml of the spore growth medium. These procedures were repeated several times to ensure purity of the culture. The isolated pure colonies were grown aerobically at 37°C for 96 h in fermentation medium (per liter: glucose 150 g, beef extract 5 g, corn steep liquor 6 g, urea 2 g(natural pH)). The inoculum age was 10 h and the inoculum volume was 4% (v/v)). The cell free supernatants which were sterilized through 0.2 lm filter paper were harvested every 12 h by centrifugation 8,000 rpm for 10 min and then stored at -20°C for further use.

123

World J Microbiol Biotechnol (2011) 27:2785–2790

Identification and characterization The strain was identified by physiological, biochemical, and 16S rDNA sequence analysis. Physiological and biochemical identification were performed according to Bergey’s Manual of Systematic Bacteriology (Claus and Berkeley 1986). Gram reactions were determined according to standard microbiological procedures (Gerhardt et al. 1994). Spore formation was observed by a transmission electron microscope(HITACHI-7650 Japan). The 16S rDNA sequence was determined as described by (Li et al. 2004). The sequences obtained were compiled and compared with sequences in the GenBank databases using the BLAST program. The 16S rDNA sequences determined and reference sequences obtained from GenBank databases were aligned using multiple sequence alignment software CLUSTALX. A phylogenetic tree was constructed with MEGA4.0 software based on the homologous 16S rDNA sequences. Fermentation study For acetoin fermentation, the cells from LB plate (per liter: typtone 10 g, NaCl 10 g, yeast extract 10 g, and the medium plates were prepared by adding 2% of agar) were inoculated into 10 ml LB medium to be cultivated for 7 h on a rotary shaker at 160 rpm, and then 2 ml culture was transferred into 50 ml seed medium for preparation of seed cultures. After 10 h, the seed with an OD600 of approximately 8 was inoculated into 50 ml fermentation medium for 144 h. The inoculation volume was 4%. The cell free supernatants were harvested every 12 h by centrifugation 8,000 rpm for 10 min and then stored at -20°C for further use. Analytical methods Acetoin, 2,3-butanediol, acetic acid, lactic acid were monitored periodically over the experiment period. Bacteria were harvested by centrifugation at 8000 rpm for 10 min, and the supernatant was used for further analysis. Acetic acid and lactic acid were monitored by high performance liquid chromatograph (DIONEX) on an ODS of 250 mm 4.6 mm column at 30°C. The eluting solvent was composed of 50 mM KH2PO4 containing 5% (v/v) methanol. The flow rate was 1.0 ml/min; and the elution was monitored at 210 nm. Concentrations of acetoin and 2,3butanediol were monitored by gas chromatography (Dong Xi GC2000A, FID detector, 2 m 9 u5 mm packed column and operated with N2 as carrier gas at flow rate of 50 ml/ min, detector temperature 240°C and column temperature 220°C). Residual glucose was detected by a biological sensing analyzer (SBA China). Growth was measured at


2787

OD600 (UNICO UV-2000) of the culture broth after appropriate dilution with water. All assays were performed by duplicate or triplicate cultures.

Results

Table 2 Taxonomical characteristics of the strain JNA 3-10 Property

Characterization Property

Characterization

Nitrate reduction

?

Arabinose

?

Anaerobic growth ?

Sucrose

?

Starch hydrolysis ?

Fructose

?

Casein hydrolysis ?

Xylose

?

Isolation and identification

Gelation hydrolysis

?

Lactose

?

Various microbial strains were isolated from the samples through enrichment culture. After heating at 80°C, the diluted cultures were spared on the spores growth medium agar plates. With this method, 35 isolates which can grow under high concentration of glucose were selected. The 35 strains were then cultured in fermentation medium for 96 h. The cell free supernatants were monitored by GC. Table 1 shows the result of 6 strains that able to produce high concentration of acetoin. Strain 3–1 and 3–10 has 24.3 and 19.0 g/l of acetoin respectively. Though has higher yield of acetoin, strain 3–1 has low yield of 2,3-butanediol (3.8 g/l) and inefficient utilization of glucose compared to strain 3–10 (36.3 g/l of 2,3-butanediol). With high yield of acetoin, strain 3–1 and 3–10 was selected as the best strain for further fermentation. Figure 1 shows the fermentation results of strain 3–1 and 3–10. At about 132 h, the concentrations of acetoin are about 25.1 and 40.8 g/l, respectively. Strain 3–1 has a higher concentration of acetoin at

VP test

?

Maltose

?

Gas from glucose ? Catalase test ?

Mannitol Glucose

? ?

Protease tet

?

Motility

–

Gram stain

?

Spore forming

?

Table 1 Rescreening results of acetoin concentration Strain number

Acetoin 2,3-Butanediol concentration(g/l) concentration(g/l)

Residual glucose(g/l)

3-1

24.3 ± 0.3

3.8 ± 0.3

68 ± 2

3–10

19.0 ± 0.2

36.3 ± 0.1

4–2

9.5 ± 0.2

1.8 ± 0.2

70 ± 2

0

5–8

9.1 ± 0.3

1.6 ± 0.1

78 ± 1

8.6 ± 0.1

2.3 ± 0.3

89 ± 2

6.2 ± 0.3

4.8 ± 0.3

85 ± 3

Acetoin concentration(g/l)

6–4 6–7

Time(h)

Fig. 1 Compare of acetoin yields by strain 3–1 and 3–10. Strains were cultured in fermentation medium for 144 h. 3–1 (filled square) 3–10 (filled triangle)

?: positive; - : negative

the early time of fermentation, but does not increase later in contrast with strain 3–10. As a result of above, we chose strain 3–10 as the aim strain. Morphological and biochemical characteristics The detailed physiological characteristics analysis and conventional biochemical tests are summarized in Table 2. Cells were cultured on the LB agar plate for later experiments. Determination of gram reaction shows that strain JNA was a Gram positive bacterium. Figure 2 is the TEM picture of strain 3–10 and it indicates that strain 3–10 is a sporiparous bacterium. 16S rDNA sequence determination and phylogenetic analysis Partial of the 16S rDNA sequence of JNA was obtained and sequenced. The analysis of the 16S rDNA gene (GenBank accession no. HM357237) showed the closest sequence identity (98.0%) to Bacillus subtilis SCUT09 (accession no. FN869038.1). Phylogenetic relationships could be inferred through the alignment and cladistic analysis of homologous nucleotide sequences of known bacteria. Figure 2 shows the phylogenetic relationships that were established on the basis of the 16S rDNA gene sequences of strain JNA and its closely organisms. By examining physiological and biochemical characteristics and comparing its 16S rDNA gene sequence, strain JNA was identified as a strain of B. subtilis, and named B. subtilis JNA 3-10. Effect of glucose concentration on acetoin production The effect of glucose concentration on acetoin production is shown in Table 3. Because of the ability of rapid

123

2788


initial pH 4.0, the yield of acetoin was maximium. We choose pH 4.0 for subsequent fermentations by this strain. Fermentation characterization of Bacillus subtilis JNA 3-10

Fig. 2 TEM picture of strain 3–10. Cell size is about (0.6–0.9) lm wide and (2.2–3.2) lm long. (bar = 1 lm)

consumption of glucose, this experiment chose high concentration of glucose as substrate. At glucose concentrations of 10 and 15% (W/V), acetoin yield (g acetoin/g glucose) were higher than that of 20 and 25% (W/V). At concentration of glucose 20 and 25% (W/V), higher yield of acetoin were gained 48.2 and 47.8 g/l respectively, but longer time were required and the yield (g acetoin/g glucose) of acetoin were lower. Consequently, a glucose concentration of 15% (W/V) was chosen for subsequent fermentations. Effect of temperature and initial pH on acetoin production The effect of temperature on acetoin fermentation was examined in rotary flasks at 160 rpm. The results are summarized in Table 4. Acetoin production was optimal at 37°C. The utilization rate of glucose was slow at low temperature, and residual glucose was about 20 g/l at the end of fermentation. The culture pH increases with a concomitant increase of acetoin production. The effect of the initial pH (3.0–7.0) of the culture medium was investigated at 37°C and 160 rpm on 150 g/l glucose. The growth of this strain was poor at pH 3.0, and the utilization rate of substrate was slow. The final pH increase corresponded to the increase of initial pH. At

Table 3 Effect of glucose concentration on acetoin production by Bacillus subtilis JNA 3-10. The growth experiments were performed at natural pH, 37°C, and 160 rpm

123

Glucose (g/l)

Time of maximum acetoin yield (h)

The major metabolites of Bacillus subtilis JNA 3-10 are given in Fig. 3. Glucose was completely consumed in 96 h or so, accompanied by the rapidly accumulation of 2,3butanediol. Glucose was almost transformed to 2,3butanediol (41.7 g/l) and acetoin (21.0 g/l) at about 96 h, but what was acetoin transformed from later? A new phenomenon was first observed at the decline phase of fermentation by strain Bacillus subtilis JNA 3-10, thus, 2,3butanediol was transformed into acetoin, and at a time of 132 h the yield of acetoin was improved to a high level (about 42.2 g/l) (Fig. 4). It is a mixed acid fermentation. The dominating acids are lactic and acetic acid, and they affect the pH of fermentation. Figure 3 gives an apparently understanding of the interrelation of acids and pH. Lactic acid accumulated quickly in a few hours at the beginning of cell growth, and then declined slowly. The result was just in contrast with the pH movement. The affect of lactic acid on fermentation will be tried in the further studies. Another major byproduct acetic acid maintained steady after it first accumulated at 36 h. The biomass achieved to a max value at 72 h or so when cells get into the stationary phase. Then it declined quickly accompanied by the reverse transformation of 2,3-butanediol to acetoin.

Discussion Acetoin production by microorganism has not achieved to a high level that enough to edge out chemosynthetic product. This study isolated 35 spore-forming strains from soil samples. Strain JNA 3-10 was finally selected as an acetoin producer after fermented for 132 h. The concentration of acetoin reached about 42.2 g/l via GC determination of the cell free supernatants. It was identified by physiological, biochemical, and 16S rDNA sequence analysis, and was named Bacillus subtilis JNA 3-10.

Maximum acetoin (g/l)

Acetoin yield (g acetoin/g glucose)

Residual glucose at maximum acetoin yield (g/l)

100

132

28.1 ± 2.1

0.28 ± 0.02

0

150

132

41.9 ± 1.5

0.28 ± 0.01

0

200

168

48.2 ± 2.2

0.24 ± 0.01

250

168

47.8 ± 1.1

0.19 ± 0.004

0 48 ± 2


2789

Table 4 Effect of temperature and initial pH of on acetoin production by Bacillus subtilis JNA 3-10 Condition

Time of maximum acetoin yield (h)

Maximum acetoin (g/l)

Acetoin yield (g/g glucose)

Residual glucose at maximum acetoin yield (g/l)

pH at maximum acetoin yield

Temperature (°C) 25

168

32.2 ± 3.2

0.22 ± 0.023

20 ± 3

7.10 ± 0

30

144

38.3 ± 2.3

0.26 ± 0.015

0

7.22 ± 0.1

35

132

40.1 ± 1.2

0.27 ± 0.008

0

7.33 ± 0.1

37

132

42.0 ± 1.3

0.28 ± 0.009

0

7.35 ± 0

40

132

39.1 ± 2.1

0.26 ± 0.014

0

7.31 ± 0.1

Initial pH 3.0 4.0

132

32.9 ± 3.2

0.22 ± 0.021

12 ? 3

7.40 ± 0.1

132

42.2 ± 2.1

0.28 ± 0.014

0

7.39 ± 0.1

5.0

132

42.1 ± 1.2

0.28 ± 0.008

0

7.49 ± 0

6.0

144

41.4 ± 1.1

0.28 ± 0.007

0

7.67 ± 0.1

7.0

144

40.5 ± 2.1

0.27 ± 0.014

0

7.68 ± 0.1

Fig. 3 The phylogenetic dendrogram for JNA and related strains based on the 16S rDNA sequence. Numbers in parentheses are accession numbers of published sequences. Bootstrap values were based on 1,000 replicates. The tree was constructed by the neighbor-joining method

Bacillus subtilis is a well studied and stable bacterium, and has been used in industrial production. It is a good producer for acetoin not only because of its high yield of acetoin production but also because acetoin is mainly used as a flavor. The safe problem is especially notable and fortunately Bacillus subtilis is accepted as a food safe bacterium. Preliminary research on fermentation conditions about B. subtilis JNA 3-10 were studied, the initial pH, and fermentation temperature were included. Glucose, which is usually an excellent carbon source for bacterial growth was chose as the substrate for acetoin fermentation. 150 g/l glucose, pH 4.0, and 37°C were finally chose for fermentation conditions of this strain, and the fermentation process curve were studied on the conditions above. At the prophase, the pH of culture declined mainly because of the accumulation of lactic acid, and then it gradually increased to about 7.0 or more. Haavik reported that bacitracin production by Bacillus subtilis is pH dependent as a result of the accumulation of organic acids (Haavik 1974). Affect of lactic acid on acetoin fermentation and pH were studied, and the addition of pyruvate induced large increase in acetoin (Cardenas et al. 1989; Hegazi and Abo-Elnaga

B. cereu s NC_003909.8 100 98

B. thuringiensis NC_014171.1 B. weihenstephanensis NC_010184.1 B. megaterium NC_014103.1 B. pumilus NC_009848.1 B. licheniformis NC_006270.3 Strain JNA 3-10

94 91

B. subtilis FN869038.1

1980) by reason of a mechanism for disposing of toxic level of intracellular pyruvate (Collins 1972; Dessart and Steenson 1995). Researches have reported that the formation of acetoin is accompanied by alkalinization of the medium. It counteracted the acidification of the environment (McFall and Montville 1989), and/or is involved in energy generation (Starrenburg and Hugenholtz 1991; Lolkema et al. 1995). Fermentation conditions need to be further optimized and the production scale need to be expanded on fermentors. Dissolved oxygen and pH control are needed to shorten the fermentation period. In order to reduce production costs, considerations about raw materials had been in view. Xiao use low price substrate molasses and soybean which can supersede glucose obtained 37.9 g/l acetoin (Xiao et al. 2007). In this study, acetoin fermentation was improved to about 42.2 g/l by a newly isolated bacterium B. subtilis JNA 3-10, and a new phenomenon of reverse transformation of 2,3-butanediol to acetoin was observed accompanied with cell senescence and apoptosis. The mechanism of this phenomenon should be further studied, and related works had been carried out.

123

2790

a


150

50

125

40 35

100

30 25 20

75

15

50

10

Product g/l

Residual glucose g/l

45

5

25

0 0

-5

Time (h) 15

10

12

8

9

6

6

4

3

2

pH

OD600

b

0 0

12

24

36

48

60

72

84

96

0 108 120 132 144

Time (h)

Fig. 4 Time-course date of batch fermentation by Bacillus subtilis JNA. a Metabolite profiles; b Cell growth and pH residual glucose (square); acetoin (filled square); 2,3-butanediol (square); acetic acid (filled triangle); lactic acid (square); pH (circle); OD600 (filled circle) Acknowledgments This work was supported by the High-tech Research and Development Programs of China (2007AA02Z207), the National Basic Research Program of China (2007CB707800), the National Natural Science Foundation of China (30970056), the Fundamental Research Funds for the Central Universities (JUSRP31001), the Program for New Century Excellent Talents in University (NCET07-0380), A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions and the Programme of Introducing Talents of Discipline to Universities (111-206).

References Blom RH (1945) Configuration of acetylmethylcarbinol. Am Chem Soc 67 :494 Cardenas ILB, Ledesma OV, Oliver G (1989) Effect of lactic acid on diacetyl and acetoin production by Lactobacillus casei subsp. Rhamnosus ATCC 769. Current Microbiology 18:351–354 Claus D, Berkeley RWC (1986) Bergey’s manual of systematic bacteriology. In: Sneath PHA (ed) Genus bacillus cohn. Williams & Wilkins, Baltimore, pp 1105–1138 Collins EB (1972) Biosynthesis of flavor compounds by microorganisms. J Dairy Sci 55:1022–1028

123

Dessart SR, Steenson LR (1995) Biotechnology of dairy Leuconostoc. In: Hui YH, Khachatourians GG (eds) Food Biotechnology. Microorganisms, New York, pp 665–702 Dettwiler B, Dunn IJ, Heinzle E, Prenosil JE (1993) A simulation model for the continuous production of acetoin and butanediol using Bacillus subtilis with integrated pervaporation separation. Biotechnol Bioeng 41:791–800 Fumio T, Koichi T, Hiroshi T (1989) New reduction method of adiketones, oxo amides, and quinoned with Zn-EtOH in the presence of a salt. J Chem Soc, Perkin Trans 1:1555–1556 Gerhardt P, Murray RGE, Wood Willis A, Krieg Noel R (1994) Methods for general and molecular bacteriology. American Society for Microbiology, Washington DC ISBN: 1555810489 Haavik HI (1974) Studies on the formation of bacitracin by Bacillus licheniformis: Role of catabolite repression and organic acids. J Gen Microbiol 84:321–326 Hegazi FZ, Abo-Elnaga IG (1980) Production of acetoin and diacetyl by lactic acid bacteria in skimmed milk with added citrate and pyrauvate. Z Lebensm Unters Forsch 171:367–370 Hilmi A, Belgsir EM, Leger JM et al (1997) Electrocatalytic oxidation of aliphatic diols Part V. Electro-oxidation of butanediols on platinum based electrodes. J Electroanal Chem 435:69–75 Li X, Zhang D, Chen F, Ma J, Dong Y, Zhang L (2004) Klebsiella singaporensis sp. nov., a novel isomaltulose-producing bacterium. Int J Syst Evol Microbiol 54:2131–2136. doi:10.1099/ ijs.0.02690-0 Lolkema JS, Poolman B, Konings WN (1995) Role of scalar protons in metabolic energy generation in lactic acid bacteria. J Bioenerg Biomembr 27:467–473 McFall SM, Montville TJ (1989) pH-Mediated regulation of pyruvate catabolism in Lactobacillus plantarum chemostat cultures. J Ind Microbiol 4:335–340 Qin J, Xiao Z, Ma C, Xie N, Liu P, Xu P (2006) Production of 2, 3-butanediol by Klebsiella Pneumoniae using glucose and ammonium phosphate. Chin J Chem Eng 14:132–136 Slipszenko JA, Griffiths SP, Simons KE et al (1998) Enantioselective hydrogenation. J Catalyst 179(7):267–276 Starrenburg MJC, Hugenholtz J (1991) Citrate fermentation by Lactococcus and Leuconostoc spp. Appl Environ Microbiol 57:3535–3540 Studer M, Okafor V, Blaser HU (1998) Hydrogenation of butane-2, 3-dione with heterogeneous cinchona modified platinum catalysts: a combination of an enantioselective reaction and kinetic resolution. Chem Commun 13:1053–1054 Xiao Z, Xu P (2007) Acetoin metabolism in bacteria. Crit Rev Microbiol 33:127–140. doi:10.1080/10408410701364604 Xiao Z, Liu P, Qin J, Xu P (2007) Statistical optimization of medium components for enhanced acetoin production from molasses and soybean meal hydrolysate. Appl Microbiol Biotechnol 74:61–68. doi:10.1007/s00253-006-0646-5 Xiao Z, Ma C, Xu P, Lu JR (2009) Acetoin catabolism and acetylbutanediol formation by Bacillus pumilus in a chemically defined medium. PLoS ONE 4(5):e5627. doi:10.1371/journal. pone.0005627 Yu EK, Saddler JN (1983) Fed-batch approach to production of 2, 3-butanediol by Klebsiella pneumoniae grown on high substrate concentrations. Appl Environ Microbiol 46:630–635