SEARCH

SEARCH BY CITATION

Keywords:

  • 2,3-butanediol dehydrogenase;
  • 2,3-butanediol dehydrogenase gene;
  • l-2,3-butanediol;
  • l-acetoin;
  • chiral;
  • diacetyl;
  • diacetyl reductase;
  • stereospecificity

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

Aims:  A metabolic pathway for l-2,3-butanediol (BD) as the main product has not yet been found. To rectify this situation, we attempted to produce l-BD from diacetyl (DA) by producing simultaneous expression of diacetyl reductase (DAR) and l-2,3-butanediol dehydrogenase (BDH) using transgenic bacteria, Escherichia coli JM109/pBUD-comb.

Methods and Results:  The meso-BDH of Klebsiella pneumoniae was used for its DAR activity to convert DA to l-acetoin (AC) and the l-BDH of Brevibacterium saccharolyticum was used to reduce l-AC to l-BD. The respective gene coding each enzyme was connected in tandem to the MCS of pFLAG-CTC (pBUD-comb). The divided addition of DA as a source, addition of 2% glucose, and the combination of static and shaking culture was effective for the production.

Conclusions: l-BD (2200 mg l−1) was generated from 3000 mg l−1 added of DA, which corresponded to a 73% conversion rate. Meso-BD as a by-product was mixed by 2% at most.

Significance and Impact of the Study:  An enzyme system for converting DA to l-BD was constructed with a view to using DA-producing bacteria in the future.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

Chiral compounds are especially important to provide, for example, chiral groups in drugs or for liquid crystals. Micro-organisms produce various chiral compounds. Typical acetoinic compounds are acetoin (acetyl-methyl carbinol, AC) or 2,3-butanediol (BD). In particular, there are direct pathways from pyruvate to d-AC, d-BD and meso-BD (Taylor and Juni 1960; Hohn-Bentz and Radler 1978; Maddox 1988). As a result, efficient preparation of these isomers using genetic engineering has been achieved (Ui et al. 1996, 1998a). However, using such a pathway to produce l-AC or l-BD has not yet been carried out. We previously reported the conversion of diacetyl (butandione, DA) to l-BD in transgenic Escherichia coli containing the diacetyl reductase (DAR) gene as a basis for the construction of a l-BD production system in bacteria that produce large amounts of DA, for example, Streptococcus diacetilactis (Speckman and Collins 1968; Kaneko et al. 1986). For this system, meso-2,3-butanediol dehydrogenase (meso-BDH) from Klebsiella pneumoniae IAM 1063, which interconverts meso-BD to d-AC, also has a strong DAR (l-AC forming) activity (Ui et al. 1999). Moreover, the meso-BDH produced from a plasmid with an isopropyl β-d-thiogalactoside (IPTG)-dependent promoter, lacks stereospecificity to d-AC and also catalyses the conversion of l-AC to l-BD (Ui et al. 1996). Applying this knowledge, the conversion of DA to l-BD was achieved (Ui et al. 1999). However, this method requires careful timing and amounts of IPTG addition, which makes it an expensive production process. Therefore, it is not suitable as a method for extensive manufacturing. In order to dispense with IPTG, introduction of separate l-BDH activity was considered for the conversion of l-AC into l-BD, as shown in Fig. 1. By coupling this enzyme with the DAR activity described above, the production of l-BD from DA via simultaneous expression of DAR and l-BDH genes in transgenic E. coli was attempted.

image

Figure 1. The metabolic strategy proposed and the plasmid constructed, pBUD-comb, for the conversion of DA to l-BD

Download figure to PowerPoint

Bacterial strains used

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

Escherichia coli JM109 [hsdR 17, recA1, del(lac-proAB)] was used as the host strain for the plasmids used in this study. Escherichia coli JM109/pBUD119 was previously prepared in this laboratory (Ui et al. 1996). A 2·0-kbp PstI fragment containing the meso-BDH (DAR) gene from Klebsiella pneumoniae IAM1063 was cloned into pUC119 (Takara Shuzo Co., Kyoto, Japan) to give plasmid pBUD119. Escherichia coli JM109/pLBD2-CTC was also prepared previously in this laboratory (Ui et al. 1998b; Takusagawa et al. 2001). A 1-kbp HindIII–KpnI fragment containing the l-BDH gene from Brevibacterium saccharolyticum C-1012 was cloned into pFLAG-CTC to give plasmid pLBD2-CTC.

Media and culture conditions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

The bacteria were cultured in Bouillon medium containing (l−1 deionized water, pH 7·2): extract beef render (Mikuni Chemicals, Tokyo, Japan), 10 g; Polypepton (Nippon Pharmaceutical, Tokyo, Japan), 10 g and NaCl, 5 g as a basal medium. The selective antibiotic medium contained ampicillin (100 μg ml−1). A preculture (5 ml) was mixed into the medium containing the specified appropriate additions (100 ml medium in a 500-ml shaking flask) and cultured on a reciprocal shaker (160 strokes min−1) or left static at 37°C.

Plasmid construction and transformation

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

The pBUD-comb plasmid was constructed to connect the l-BDH gene and meso-BDH gene in tandem. First, the structural gene of l-BDH and its promoter were amplified by PCR using pLBD 2-CTC as a template and primers TD-L-F (5′-GTATAATGTGTGGAATTGTGAGCGG-3′) and TD-L-R (5′-CGGGAATTCAACCCTTAGTTGTAG-3′, EcoRI site added). The PCR product was cut by restriction enzymes, HindIII–EcoRI. Next, the structural gene of meso-BDH and its terminator were amplified using pBUD119 as a template and primers TD-M-F (5′-GCCGAATTCTTCGGGGCCAAAG-3′, EcoRI site added) and TD-M-R (5′-CCGGGAAGCTTACACTTACACTC-3′, HindIII-site added). The PCR product was cut by restriction enzymes, HindIII–EcoRI. Finally, the pFLAG-CTC vector was cut by restriction enzyme HindIII. The three gene fragments were ligated together. The resulting plasmid was transformed into E. coli JM109 by electroporation after checking the sequence using an ABI 377 DNA sequencer.

Analytical procedure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

Cell growth was measured from the attenuance at 660 nm. The protein concentration was determined according to Bradford (1976). Chiral AC and BD isomer concentrations were determined using a combination of enzyme (bacterial BDH) and gas chromatographs (Ui et al. 1984a,b). DA was analysed by a headspace technique using gas chromatography as reported previously (Postel and Meier 1981). All values were obtained from an average of three identical trials.

Preparation of cell-free extracts

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

Preparation of cell-free extracts was conducted as described previously (Taylor and Juni 1960). Assay of the BDH, DAR or ACR activity with NADH oxidation was also performed as described previously (Ui et al. 1983). The activity is given in units (1 U = 1 μmol NADH or NAD+ formed per minute).

Chemicals

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

DA and AC were purchased from Tokyo Kasei Kogyo Co., Ltd (Tokyo, Japan). BD isomers (d, l and meso) were purchased from Sigma-Aldrich (St Louis, MO, USA). Other chemicals, and restriction enzymes for genetic engineering were obtained from commercial sources.

Production and expression of a construct for l-BD formation

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

In order to convert DA into l-BD vial-AC, two steps are required, and an enzyme which can catalyse both steps efficiently has not been found. The l-BDH used not only reduces l-AC into l-BD, but also DA into l-AC, although with weak activity. However, DA may act as a competitive inhibitor to the reduction of l-AC. Therefore, if DA is added as a substrate for l-BDH, the main product formed would remain in AC and hardly be converted to l-BD (data not shown). Therefore, it is necessary to divide the process into two steps, with separate enzymes for the conversion of DA to l-AC as the first step, and l-AC to l-BD as the second step.

Thus, meso-BDH (d-AC forming) from K. pneumoniae IAM 1063, which also has a high DAR activity and a tolerance to high DA concentration (data not shown), was used as the enzyme for the first conversion, that is, the S-configuration to one carbonyl group of DA was added. The l-BDH from B. saccharolyticum C-1012 was used for the second conversion of l-AC to l-BD. The plasmid (pBUD-comb) containing the genes encoding both enzymes as shown in Fig. 1 was introduced to achieve their simultaneous expression in E. coli JM109. The time course of the expression is shown in Fig. 2. The pBUD-comb expressed the two enzymes, and the original plasmid (pBUD 119 or pLBD2-CTC) containing either one of the two enzymes showed only activity of DAR or BDH, respectively, in response to the gene introduced (Fig. 3).

image

Figure 2. Cell growth and expression of meso-BDH or l-BDH in the course of cultivation by Escherichia coli JM109/pBUD-comb. The specific activities were determined according to Materials and methods. Specific activity: (○) meso-BDH; (•) l-BDH; (bsl00066) growth

Download figure to PowerPoint

image

Figure 3. Activity disk-PAGE staining of the enzyme expressed in transformed Escherichia coli. The direction of migration was towards the bottom of the gel (anode). The bands were stained by the enzymatic reaction using a 2,3-butanediol (BD)-isomer, acetoin (AC) or diacetyl (DA) as the substrate as described in Materials and methods. Protein in the gel was detected by staining with 1% amino black-10B in 7% (v/v) acetic acid. In the gel for protein, a sample of the partial purified protein (30 μg) was applied. In the other gels for activity staining, ca 10 μg of the protein was added. [N, negative staining (reductive reaction, pH 6·2); P, positive staining (oxidative reaction, pH 9·8); DA, diacetyl; M-BD, meso-BD; Pr, protein staining]

Download figure to PowerPoint

Addition time and concentration of DA.  The growth of E. coli JM109/pBUD-comb was inhibited at more than 500 mg l−1 of DA added at the beginning of culture shaking. Therefore, DA (500, 1000, 1500, 2000, 2500 or 3000 mg l−1) was added to cultures after 12 h, corresponding to a late logarithmic phase in which the optimum specific activity with each enzyme had been obtained (Fig. 2), conversion rates were then measured. The most efficient conversion to l-BD was obtained with 1500 mg l−1 DA, and the amount produced reached ca 1000 mg l−1 after culturing for more 14 h. When below 1500 or above 2000 mg l−1 of DA added, the amount produced or the conversion rate decreased respectively.

Oxygen supply.  As the conversion of DA to BD is basically a reductive reaction, the conversion in a static culture was examined. When the culture was kept static for the whole postinoculation period and DA was added at 1000 mg l−1 in the late logarithmic phase, a 15% of conversion rate was obtained. However, the rate was low compared with that obtained (80%) by continuous shaking, probably because of the low growth of the strain. When the culture was shaken up to the addition point of DA (12 h), and thereafter maintained in a static condition, the amount converted from 1500 mg l−1 DA reached ca 1200 mg l−1, but still ca 1060 mg l−1 was retained by shaking at 48 h after the DA addition; therefore, the static condition was also effective for the conversion.

Sequential addition of DA.  As the conversion was prevented by a DA concentration of more than 2000 mg l−1, DA was added in increments over time, that is, the first addition (1500 mg l−1) was performed as described in the preceding section (after 12 h). At 14 h after the first addition, the residual DA concentration fell to almost zero and there was no further increase in l-BD concentration. The second addition (500, 1000 or 1500 mg l−1) was examined at this time. At 1000 mg l−1 DA for the second addition, l-BD formation efficiently occurred and reached a total of ca 1500 mg l−1 after 10 h. At 500 or 1500 mg l−1 DA addition, the l-BD formed totalled ca 1300 or 1550 mg l−1 respectively. However, the third addition of DA, 500 or 1000 mg l−1 at 10 h after the second addition, gave the same effect on the formation of l-BD at the point of a total of 60 h shaking. Although both enzyme activities, DAR and l-BDH, fell a little, they were present throughout the conversion course. However, at the second addition of DA, the addition time was a very important factor for the conversion, because if the addition was conducted in the middle of a conversion of l-AC to l-BD, the conversion to l-BD was inhibited, causing a low conversion rate ranging from 20 to 35%. The reason for this phenomenon was considered to be that DA demonstrated competitive inhibition for a reductive reaction of l-AC to l-BD by l-BDH as mentioned in preceding section. From this point of view, the interval between the DA addition points was important.

Effect of glucose.  As both genes are connected by a pFLAG-CTC vector, which is under the control of a tac promoter, the expression is inhibited in the presence of glucose. However, if glucose is not added to the culture, the cell growth is lowered and the total utilizable NADH for the conversion decreases. As a result, 2% glucose added at the beginning of the culture was effective for the conversion of l-BD from DA in the first addition. The up-conversion rate of 74% from 34% was estimated as a result of an applicable NADH increase arising from the glucose added. However, further additions of glucose (0·5%) over course of the conversion, which were conducted at 12, 24 or 36 h after the beginning of the culture, had no meaningful effect on the promotion of the amount of l-BD formed under the condition mentioned in the c-section, that is, DA additions were 1500 mg l−1 at 12 h (first addition), 1000 mg l−1 at 26 h (second addition), and 500 mg l−1 at 36 h (third addition) after the beginning of the culture. Conversely, the reduction of DA to l-BD was inhibited to a 20–30% degree by the interval addition of glucose. At 72 h after the first addition of DA, 2200 mg l−1 of l-BD was obtained from totally 3000 mg l−1 DA by these procedures (the addition of 2% glucose and the three DA additions as described above).

Preparation of l-BD

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References

The conversion of DA to l-BD was conducted based on the results obtained in the preceding section. In addition, the static condition was followed up to 84 h after 60 h shaking at the point of slowing the l-BD increase. The formation of l-BD is shown in Fig. 4. Consequently, l-AC (ca 580 mg l−1 with 98% optical purity), meso-BD (ca 45 mg l−1) and l-BD (ca 2200 mg l−1) were produced from the DA (3000 mg l−1 added in total at 12, 26 and 36 h in the culture course as described in the preceding section) in cultivation for 84 h, and the conversion rates of l-AC and l-BD from the DA were 19 and 73% respectively.

image

Figure 4. Formation progress of l-BD from DA by Escherichia coli JM109/pBUD-comb. Escherichia coli JM109/pBUD-comb was cultured by shaking (160 strokes min−1) at 37°C in Bouillon medium. DA was added to the culture after 12, 26 and 36 h respectively. (bsl00066) DA; (•) AC and (bsl00001) l-BD were measured as described previously (Postel and Meier 1981; Ui et al. 1984a,b)

Download figure to PowerPoint

These results not only excelled in terms of economic efficiency, because of being IPTG free and having a high conversion rate of 73%, but also the l-BD produced slightly increased (90 mg l−1) when compared with the previous method added of 0·25 g l−1 IPTG. Moreover, the success obtained here in one strain cultured in the absence of IPTG will make it possible to apply the host as a DA-producing bacteria in the future.

In this preparation, the production of ca 2% of meso-BD other than l-BD was confirmed. The meso-BD produced is thought to originate from d-AC; however, the rate of production was the same as that obtained by the IPTG addition method reported before. The reason for the appearance of d-AC could be the presence of an enzyme such as glycerol dehydrogenase having DAR activity for d-AC formed from DA in E. coli (Tang et al. 1979).

As the l-BDH used has little tolerance for DA, higher tolerance is required. That is, stability for DA is necessary to allow a higher DA concentration. If such a tolerance is achieved, one addition with a high concentration of DA will be possible and enable efficient production of l-BD.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Bacterial strains used
  6. Media and culture conditions
  7. Plasmid construction and transformation
  8. Analytical procedure
  9. Preparation of cell-free extracts
  10. Polyacrylamide gel electrophoresis
  11. Chemicals
  12. Results and discussion
  13. Production and expression of a construct for l-BD formation
  14. Examination of culture conditions
  15. Preparation of l-BD
  16. References
  • Bradford, M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248254.
  • Davis, B.J. (1964) Disk electrophoresis. II. Method and application to human serum proteins. Annals of the New York Academy of Sciences 121, 404427.
  • Hohn-Bentz, H. and Radler, F. (1978) Bacterial 2,3-butanediol dehydrogenases. Archives of Microbiology 116, 197203.
  • Kaneko, T., Suzuki, H. and Takahashi, T. (1986) Diacetyl formation and degradation by Streptococcus lactis subsp. diacetilactis 3022. Agricultural Biological Chemistry 50, 26392641.
  • Maddox, I.S. (1988) Microbial production of 2,3-butanediol. Biotechnology 6, 3250.
  • Ornstein, L. (1964) Disk electrophoresis. I. Background and theory. Annals of the New York Academy of Sciences 121, 321349.
  • Postel, W. and Meier, Z. (1981) Gaschromatographische bestimmung von 2-acetolactat, 2-acetohydroxybutyrat, diacetyl, 2,3-pentandion und acetoin in traubenmost und wein. Zeitschrift for Lebensmittel-Untersuchung und Forschung 173, 85–89.
  • Speckman, R.A. and Collins, E.B. (1968) Diacetyl biosynthesis in Streptococcus diacetilactis and Leuconostoc citrovorum. Journal of Bacteriology 95, 174180.
  • Takusagawa, Y., Otagiri, M., Ui, S., Ohtsuki, T., Mimura, A., Ohkuma, M. and Kudo, T. (2001) Purification and characterization of l-2,3-butanediol dehydrogenase of Brevibacterium saccharolyticum C-1012 expressed in Escherichia coli. Bioscience Biotechnology and Biochemistry 65, 18761878.
  • Tang, C.-T., Ruch, F.E. and Lin, E.C.C. (1979) Purification and properties of a nicotinamide adenine dinucleotide-linked dehydrogenase that serves an Escherichia coli mutant for glycerol catabolism. Journal of Bacteriology 140, 182187.
  • Taylor, M.B. and Juni, E. (1960) Stereoisomeric specificities of 2,3-butanediol dehydrogenases. Biochimica et Biophysica Acta 39, 448457.
  • Ui, S., Masuda, H. and Muraki, H. (1983) Stereospecific and electrophoretic natures of bacterial 2,3-butanediol dehydrogenases. Journal of Fermentation Technonogy 61, 467471.
  • Ui, S., Masuda, H. and Muraki, H. (1984a) Separation and quantitation of 2,3-butanediol isomers ((−), (+), and meso) by a combined use of enzyme and gas chromatography. Agricultural and Biological Chemistry 48, 28372838.
  • Ui, S., Masuda, H. and Muraki, H. (1984b) Separation and quantitation of acetoin isomers D(−) and L(+) by a combined use of enzyme and gas chromatography. Agricultural and Biological Chemistry 48, 28352836.
  • Ui, S., Otagiri, M., Mimura, A. Kanai, H., Kobayashi, T. and Kudo, T. (1996) Preparation of a chiral compound using transgenic Escherichia coli expressing the 2,3-butanediol. Journal of Fermentation Bioengineering 81, 386389.
  • Ui, S., Mimura, A., Okuma, M. and Kudo, T. (1998a) The production of d-acetoin by a transgenic Escherichia coli. Letters in Applied Microbiology 26, 275278.
  • Ui, S, Otagiri, M., Mimura, A. Dohmae, N., Takio, K., Ohkuma, M. and Kudo, T. (1998b) Cloning, expression and nucleotide sequence of the l-2,3-butanediol dehydrogenase gene from Brevibacterium saccharolyticum C-1012. Journal of Fermentation Bioengineering 86, 290295.
  • Ui, S., Mimura, A., Ohkuma, M. and Kudo, T. (1999) Formation of a chiral acetoinic compound from diacetyl by Escherichia coli expressing meso-2,3-butanediol dehydrogenase. Letters in Applied Microbiology 28, 457460.