Bioprocess Biosyst Eng (2004) 26: 325–330 DOI 10.1007/s00449-004-0373-1
O R I GI N A L P A P E R
Hikmet Geckil Æ Ze’ev Barak Æ David M. Chipman Sebnem O. Erenler Æ Dale A. Webster Benjamin C. Stark
Enhanced production of acetoin and butanediol in recombinant Enterobacter aerogenes carrying Vitreoscilla hemoglobin gene Received: 10 February 2004 / Accepted: 30 June 2004 / Published online: 7 August 2004 Springer-Verlag 2004
Abstract Microbial production of butanediol and acetoin has received increasing interest because of their diverse potential practical uses. Although both products are fermentative in nature, their optimal production requires a low level of oxygen. In this study, the use of a recombinant oxygen uptake system on production of these metabolites was investigated. Enterobacter aerogenes was transformed with a pUC8-based plasmid carrying the gene (vgb) encoding Vitreoscilla (bacterial) hemoglobin (VHb). The presence of vgb and production of VHb by this strain resulted in an increase in viability from 72 to 96 h in culture, but no overall increase in cell mass. Accumulation of the fermentation products acetoin and butanediol were enhanced (up to 83%) by the presence of vgb/VHb. This vgb/VHb related effect appears to be due to an increase of flux through the acetoin/butanediol pathway, but not at the expense of acid production. Keywords Acetoin Æ Bacterial hemoglobin Æ Butanediol Æ Enterobacter aerogenes Æ Vitreoscilla hemoglobin
H. Geckil (&) Æ S. O. Erenler Department of Biology, Inonu University, Malatya, 44069, Turkey E-mail:
[email protected] Tel.: +90-422-3410010 Fax: +90-422-3410037 Z. Barak Æ D. M. Chipman Department of Life Sciences, Ben-Gurion University of the Negev, Beer-Sheva, Israel D. A. Webster Æ B. C. Stark Biology Division, Department of Biological, Chemical, and Physical Sciences, Illinois Institute of Technology, Chicago, IL, USA
Introduction The hemoglobin (VHb) of the bacterium Vitreoscilla is the first discovered and probably best characterized of the microbial hemoglobins [1, 2]. Its primary function, for which experimental support but not proof exists, is most likely to bind oxygen at low extracellular concentrations and deliver it to the terminal respiratory oxidase, thus enhancing respiration under these conditions [3–6]. This presumably allows the native host, which is obligately aerobic, to survive in its normal hypoxic habitats [7]. Similar phenomena occur in heterologous microbial hosts into which the VHb gene (vgb) has been inserted. Especially under conditions of limited oxygen, these include increases in growth and production of products such as proteins [8–10] and antibiotics [11, 12]. Increases in ethanol production [13] and other changes in fermentation pathways [14–16] have also occurred in the presence of vgb/VHb, as has the enhancement of bioremediation of aromatic compounds [17–21]. The aim of the present work was to investigate the possibility that engineering of Enterobacter aerogenes to contain vgb and produce VHb can enhance the production of 2,3-butanediol (BD) and acetoin, both of which have commercial importance. Besides its use as a fuel, BD is important in making pharmaceuticals, cosmetics, and plastics. Acetoin, which is the immediate precursor to BD, can be oxidized to diacetyl for use as a natural flavor in dairy products [22, 23]. Here we show that expression of VHb in E. aerogenes can lead to increases in cell viability as well as production of these two fermentation products.
Materials and methods Bacterial strains, plasmids, and culture medium The host strain was E. aerogenes strain NRRL B-427 (USDA culture collection, Peoria, IL, USA). Plasmid pUC8:15 [24] has vgb (0.6 kb) inserted into vector pUC8
326
[25] on a Vitreoscilla DNA fragment of 2.3 kb. The untransformed strain is designated ‘‘Ea’’ and the transformed strains Ea[pUC8] and Ea[pUC8:15], respectively [26]. The basic medium was LB (pH 7.0) [27] containing peptone in place of tryptone; when noted it was supplemented with 2% glucose (autoclaved separately) or 100 lg/ml ampicillin. Cell growth and determination of fermentation products For growth experiments single colonies were inoculated into 10 ml of LB (Ea) or LB-containing ampicillin (the two plasmid-bearing strains) in 125 ml flasks and grown at 37C and 150 rpm overnight. Cells (20–50 ll) were then inoculated into 5 ml of LB plus 2% glucose, but without ampicillin (all three strains) in 250·10 mm tubes and grown at 37C, 200 rpm for 3–4 h. An equal A660nm of each culture (150–200 ll) was then used to inoculate 30 ml of LB plus 2% glucose (again without ampicillin for all three strains) in 125 ml flasks. These, which served as the experimental cultures, were grown at 37C and 150 rpm. Samples were taken at intervals for determination of viable cells by plating appropriate dilutions on LB without ampicillin (a separate set of experiments plating on LB both with and without ampicillin showed that plasmid stability was 100% for both transformants for at least 96 h in culture); A660nm (measured values were kept under 0.5 by dilution, if necessary, with fresh medium); and pH. Additional samples were assayed for BD and acetoin according to [28, 29]. In this method, acetoin is measured directly and BD oxidized to acetoin before measurement; the value measured directly (acetoin) is subtracted from the oxidized value to yield the BD level. VHb levels were determined from whole cell CO-difference spectra [24]. Acetolactate synthase assay The specific activity of acetolactate synthase (ALS) of strains was determined using a modified method of Stromer [30] in cultures grown as above for the production of acetoin and BD. Sampling of cultures was also at 24-h intervals. At each time point, 10 ml of culture was centrifuged at 10,000 rpm for 5 min and washed once with 50 mM potassium phosphate buffer, pH 7.6. Pellet was resuspended in 1 ml sonication buffer (50 mM sodium acetate pH 5.8, 0.5 mM MnCl2 and 0.1 mM TPP) with 10% glycerol and sonicated at half power for 1 min (three cycles), with 20 s sonication and 40-s pause (cooling) on ice. The cell debris was removed by centrifuging for 1 min on a bench-top centrifuge at maximum speed. To the 0.8 ml reaction mixture (50 mM sodium acetate pH 5.8, 0.5 mM MnCl2, 0.1 mM TPP and 50 mM sodium pyruvate (fresh)) we added 0.2 ml of cell-free extract (duplicates). The reaction mixture was incubated at 37C for 20 min and the
reaction was stopped by the addition of 0.1 ml 50% H2SO4. Incubation was continued for an additional 30 min at 37C, in order to decarboxylate all the formed acetolactate to acetoin. Of this mixture, 0.5 ml was taken into tubes containing 1 ml ddH2O, 1 ml 0.5% creatine solution and 1 ml 5% freshly made a-naphthol solution (prepared in 2.5 N NaOH), vortexed and incubated at 37C for 15 min. Tubes were briefly vortexed and A540 was read against the blanks to which H2SO4 was added before extract addition. Specific activity was determined as the micromole acetoin formed per minute (U) per milligram protein under the conditions stated, through utilizing an acetoin standard curve. Total protein was determined colorimetrically [31], using bovine serum albumin as the standard.
Results and discussion Both plasmids pUC8 and pUC8:15 were maintained stably in E. aerogenes. The presence of vgb in Ea(pUC8:15) resulted in stationary phase expression of VHb of 64 nmol/g wet weight. This is about twice the normal induced level in Vitreoscilla [24]. CO-difference spectra of Ea showed a small peak in the vicinity of that for VHb but with peak height only about 2% that of the vgb-bearing strain (possibly cytochrome bo or a native flavohemoglobin). All strains reached similar viable cell counts at 24 and 48 h of growth; the vgb-bearing strain, however, had increased viability at 72 and 96 h into culture. As measured by A660nm growth was fairly similar for all three strains, with only Ea, for reasons that are not yet clear, failing to gain after 72 h (Table 1). Because of the growth similarities among strains, relative levels of both acetoin and BD were generally quite similar whether normalized to millilitre of culture or A660nm units of cells (Figs. 1, 2, and 3). Accumulation of acetoin and BD in the medium increased for all strains from 24 to 48 h in culture, but thereafter leveled off or slowed; on a per A660nm of cells basis the levels of both chemicals peaked at 48 h and either leveled off (acetoin) or declined (BD) thereafter. Acetoin and BD levels for Ea and Ea(pUC8) were generally similar at all times. Levels of both acetoin and BD, however, were consistently higher for strain Ea(pUC8:15) than for the vgb strains; for example, the acetoin and BD totals for Ea(pUC8:15) averaged 38–83% and 48–83% greater than those for Ea(pUC8) on per millilitre of culture and A660nm bases, respectively. The maximum levels of acetoin and BD achieved by Ea(pUC8:15) (nmol/ml basis) were 435 and 357 (equivalent to 0.0383 and 0.0321 g/l), respectively. These figures are lower than levels produced by Serratia marcescens and Bacillus polymyxa grown aerobically at 2–5% glucose [22]; the reasons for this discrepancy are unknown, but presumably reflect strain and growth condition differences. Medium pH was measured for all strains every 2 h for 8 h and was identical (to within 0.09 units) for all strains at all times (Table 1). From an
327 Table 1 Growth and acid production of E. aerogenes and its recombinants in LB plus 2% glucose. The level of acid production was measured as the change in medium pH with time for the first 8 h of growth. All values are averages of three independent measureStrain
24 h
Viable cell number/ml culture Ea 2.10·109 (5.18·108) Ea(pUC8) 2.81·109 (9.43·108) Ea(pUC8:15) 1.79·109 (4.27·108) Total cell mass (A660nm) of cultures Ea 2.381 (0.22) Ea(pUC8) 2.721 (0.12) Ea(pUC8:15) 2.545 (0.28) a Culture pH Ea 6.35 (0.03) Ea(pUC8) 6.33 (0.03) Ea(pUC8:15) 6.36 (0.02) a
ments; standard deviations (rn 1) are in parentheses. Values for total cell mass are from the experiments, data from which appear in Figs. 1, 2, and 3.
48 h
72 h
96 h
1.85·109 (4.46·108) 2.20·109 (2.72·108) 1.61·109 (2.27·108)
7.57·104 (1.16·104) 6.20·104 (4.80·103) 4.42·107 (1.11·107)
0 0 3,863 (2,154)
2.390 (0.37) 2.738 (0.36) 2.555 (0.41)
2.564 (0.12) 2.801 (0.19) 2.731 (0.26)
2.521 (0.21) 3.132 (0.28) 3.152 (0.17)
4.74 (0.02) 4.74 (0.02) 4.77 (0.07)
4.59 (0.01) 4.62 (0.01) 4.64 (0.02)
4.56 (0.01) 4.59 (0.02) 4.65 (0.02)
The pH values for cultures at 24–96 h incubation periods are not given but they varied within the range of 4.4–4.6 for all three strains
initial medium pH 7.0, the average culture pH for all three strains was 6.35±0.015 and 4. 75±0.017 at 2 and 4 h incubation, respectively, and it leveled off at about 4.6 from 6 to 8 h; assuming that the acid produced is lactic and formic [32] (pKas of 3.86 and 3.75, respectively, at 25C) the 8 h data correspond to dissociated plus undissociated levels of about 26–32 nmol/ml. The pH values for cultures at 24–96 h incubation periods varied only within the range of 4.4–4.6 for all three strains. Furthermore, in a second set of experiments using LB plus 2% glucose initially at pH 7.5 the medium pH of the three strains ranged from 4.5 to 4.6 at 24 h
and then decreased slowly to 4.3–4.5 by 96 h; the 96 h data correspond to dissociated plus undissociated concentrations of 63, 58, and 43 nmol/ml for Ea, Ea(pUC8), and Ea(pUC8:15), respectively. Although engineering of various microorganisms with vgb to enhance protein and antibiotic production, and bioremediation has been broadly successful [8–12, 17–21], application of this strategy to systematically increase valuable fermentation products has had mixed results. Ethanol production, for example, was increased by up to about 50% in yeast-expressing VHb [13]. Attempts to increase the production of the final two
Fig. 1a, b Acetoin accumulation by the three strains as a function of time in culture. a Per millilitre of culture, b per A660nm of cells. Error bars represent ±STDEVs (rn 1) of n=3
Fig. 2a, b BD accumulation by the three strains as a function of time in culture. a Per millilitre of culture, b per A660nm of cells. Error bars represent ±STDEVs (rn 1) of n=3
328
Fig. 3a, b Acetoin plus BD accumulation by the three strains as a function of time in culture. a Per millilitre of culture, b per A660nm of cells. Error bars represent ±STDEVs (rn 1) of n=3
metabolites in the BD fermentation pathway (acetoin and BD) in S. marcescens using vgb/VHb, however, were medium-dependent, resulting, for LB plus 2% glucose for example, in decreased acetoin/BD production and increased accumulation of acid [14]. Like S. marcescens, E. aerogenes has BD fermentation, but has a nearly opposite response to the presence of vgb/VHb when grown in LB+2% glucose. The production of these two compounds is of interest because of the value of BD as a fuel and chemical feedstock, the ability of acetoin to be reduced to BD, and the uses of acetoin in flavoring in the food industry [22, 23, 33]. The precise reasons for the effects of VHb on metabolic pathways, including those of fermentation, are not yet known in S. marcescens and E. aerogenes, but they could be complex, resulting indirectly from VHb-induced changes in NAD+ /NADH and ADP/ATP ratios [14]. It has been shown that VHb makes the interior of the cell more oxidized and thereby alters the carbon flux within central carbon-metabolic pathways. In E. coli, VHb causes a decrease in NADH and an increase in ATP levels when grown under microaerobic conditions and a shift in the rate of NAD+ /NADH couple could well have effects on rates of key reactions, which determine total carbon flux [5]. The enhancement in acetoin and BD production by VHb-expressing E. aerogenes seen in this study may also be related to such an effect of VHb. An effective intracellular dissolved oxygen level,
achieved with VHb, may change respiratory metabolism of glucose (in favor of butanediol metabolism which requires a critical level of oxygen) through changed concentrations of NAD+ and NADH. The increase in cell viability afforded by vgb/VHb might also contribute to higher BD and acetoin production; presumably cells, which remain alive longer, remain metabolically active longer. A third possibility is that the VHb effect we see may be directly related to increased oxygen supply to the cells. Acetoin and BD are the main fermentation products of E. aerogenes grown under appropriate conditions [34]. Starting with two molecules of pyruvate, BD is produced in three enzymatic steps. The products a-acetolactate, acetoin, and BD are produced by the enzymes catabolic a-acetolactate synthase (also called pH 6.0 acetolactate forming enzyme or ALS), a-acetolactate decarboxylase and acetoin (diacetyl) reductase (also called BD dehydrogenase), respectively [35]. The activity of ALS, the first enzyme diverting pyruvate to the BD pathway, was higher in Ea(pUC8:15) than in the non-hemoglobin counterparts (Fig. 4). The difference was even more evident in cells at 72 h incubation period where the viable cells for the formers were highly reduced (Table 1). This difference between Ea(pUC8:15) and that of non-vgb strains might reflect the function of VHb in maintaining the cells in a healthy state for a longer period of time and thus contributing to higher ALS activity. With the exception of almost 50% higher ALS activity of the host strain (E. aerogenes) than the vgbstrain (Ea(pUC8)) at 24 h, both strains showed similar enzyme levels and also similar cell densities at further incubation periods. One explanation for the ALS activity difference at 24 h is plasmid burden in Ea(pUC8). This kind of plasmid burden, however, might not be apparent for cultures further deprived of nutrients in the medium. At 96 h incubation however, where no viable cells for non- vgb strains and substantially reduced number of viable cells for Ea(pUC8:15) were observed, no ALS activity was determined in any strain.
Fig. 4 Acetolactate synthase in E. aerogenes (open triangle), Ea(pUC8) (open circle), and Ea(pUC8:15) (filled circle). Cells were cultivated in LBG medium at 37C in a gyratory (150 rpm) airshaker. Error bars represent ±STDEVs (rn 1) of three independent trials made in duplicates
329
Aeration rate has been found to be of great importance in the optimization of BD and acetoin production during fermentations [22, 33–38]. In general, it has been found that an aeration/oxygen supply rate that is neither too high nor too low leads to this optimization. There appear to be two reasons for this, the first being the stimulation of cell growth countered by the general decrease in fermentation as a result of increasing aeration [35–38]. The second is aeration-mediated changes in metabolic fluxes, for example among the various fermentation pathways present in BD fermenters [35]. The aforementioned possibility that VHb affects NADH/ NAD+ ratios may be important in this context. Regarding the results reported here, the similarity in the stimulation due to vgb/VHb whether expressed on per millilitre of culture or per A660nm of cells bases, suggests that vgb/VHb is stimulating the flux through the acetoin–BD pathway, although apparently not at the expense of flux through the pathways of acid production. This contrasts with what we had seen with S. marcescens, in which vgb/VHb apparently caused a shift in flux from the acetoin-BD pathway to acid production [14]. In our case, then, VHb may be serving as a biological analogue of physical methods of oxygen control, which have been used to increase BD and acetoin production in other studies [37–40]. Although the VHb-associated increases in the levels of acetoin and butanediol that we measured are fairly substantial on a relative scale, their levels are below those produced by species such as Klebsiella oxytoca [34, 36] or B. polymyxa [35]. If VHb can be as effective in these latter species as in E. aerogenes, however, its use may prove to be a useful strategy for large-scale production of these solvents. Acknowledgements This work was supported in part by NSF grant number BES-9309759 and a postdoctoral fellowship from the Council for Higher Education of Israel to H.G. at Ben Gurion University.
References 1. Stark BC, Webster DA, Dikshit KL (1999) Vitreoscilla hemoglobin: molecular biology, biochemistry, and practical applications. Recent Res Dev Biotechnol Bioeng 2:155–174 2. Wakabayashi S, Matsubara H, Webster DA (1986) Primary sequence of a dimeric bacterial haemoglobin from Vitreoscilla. Nature 322:481–483 3. Park KW, Kim KJ, Howard AJ, Stark BC, Webster DA (2002) Vitreoscilla hemoglobin binds to subunit I of cytochrome bo ubiquinol oxidases. J Biol Chem 277:3334–3337 4. Geckil H, Gencer S, Kahraman H, Erenler SO (2003) Genetic engineering of Enterobacter aerogenes with Vitreoscilla hemoglobin gene: cell growth, survival, and antioxidant enzyme status under oxidative stress. Res Microbiol 154:425–431 5. Tsai PS, Rao G, Bailey JE (1995) Improvement of Escherichia coli microaerobic oxygen metabolism by Vitreoscilla hemoglobin: new insights from NAD(P)H fluorescence and culture redox potential. Biotechnol Bioeng 47:347–354 6. Geckil H, Stark BC, Webster DA (2001) Cell growth and oxygen uptake of Escherichia coli and Pseudomonas aeruginosa are differently effected by the genetically engineered Vitreoscilla hemoglobin gene. J Biotechnol 85:57–66
7. Pringsheim EG (1951) The Vitreoscillaceae: a family of colourless, gliding, filamentous organisms. J Gen Microbiol 5:124–149 8. Buddenhagen RE, Webster DA, Stark BC (1996) Enhancement by bacterial hemoglobin of amylase production in recombinant E. coli occurs under conditions of low O2. Biotechnol Lett 102:695–700 9. Kallio PT, Bailey JE (1996) Intracellular expression of Vitreoscilla hemoglobin (VHb) enhances total protein secretion and improves the production of a-amylase and neutral protease in Bacillus subtilis. Biotechnol Prog 12:31–39 10. Khosravi M, Webster DA, Stark BC (1990) Presence of the bacterial hemoglobin gene improves a-amylase production of a recombinant Escherichia coli strain. Plasmid 24:190–194 11. DeModena JA, Gutierrez S, Velasco J, Fernandez FJ, Fachini RA, Galazzo JL, Hughes DE, Martin JF (1993) The production of cephalosporin C by Acremonium chrysogenum is improved by the intracellular expression of a bacterial hemoglobin. Bio/Technol 11:926–929 12. Magnolo SK, Leenutaphong DL, DeModena JA, Curtis JE, Bailey JE, Galazzo JL, Hughes DE (1991) Bio/Technol 9:473– 476 13. Chen W, Hughes DE, Bailey JE (1994) Intracellular expression of Vitreoscilla hemoglobin alters the aerobic metabolism of Saccharomyces cerevisiae. Biotechnol Prog 10:308–313 14. Wei M-L, Webster DA, Stark BC (1998) Metabolic engineering of Serratia marcescens with the bacterial hemoglobin gene: alterations in fermentation pathways. Biotechnol Bioeng 59:640–646 15. Geckil H, Gencer S (2004) Production of L-asparaginase in Enterobacter aerogenes expressing Vitreoscilla hemoglobin for efficient oxygen uptake. Appl Microbiol Biotechnol 63:691–697 16. Geckil H, Ates B, Gencer S, Uckun M, Yilmaz I (2004) Membrane permeabilization of Gram-negative bacteria with a potassium phosphate/hexane aqueous phase system for the release of L-asparaginase: an enzyme used in cancer therapy. Process Biochem (in press) 17. Chung JW, Webster DA, Pagilla KR, Stark BC (2001) Chromosomal integration of the Vitreoscilla hemoglobin gene in Burkholderia and Pseudomonas for the purpose of producing stable engineered strains with enhanced bioremediating ability. J Ind Microbiol Biotechnol 27:27–33 18. Fish PA, Webster DA, Stark BC (2000) Vitreoscilla hemoglobin enhances the first step in 2,4-dinitrotoluene degradation in vitro and at low aeration in vivo. J Mol Catal B Enzymatic 9:75–82 19. Liu SC, Webster DA, Wei M-L, Stark BC (1996) Genetic engineering to contain the Vitreoscilla hemoglobin gene enhances degradation of benzoic acid by Xanthomonas maltophilia. Biotechnol Bioeng 49:101–105 20. Nasr MA, Hwang KW, Akbas M, Webster DA, Stark BC (2001) Effects of culture conditions on enhancement of 2,4dinitrotoluene degradation by Burkholderia engineered with the Vitreoscilla hemoglobin gene. Biotechnol Prog 17:359–361 21. Patel SM, Stark BC, Hwang KW, Dikshit KL, Webster DA (2000) Cloning and expression of the Vitreoscilla hemoglobin gene in Burkholderia sp. strain DNT for enhancement of 2,4dinitrotoluene degradation. Biotechnol Prog 16:26–30 22. Magee RJ, Kosaric N (1987) The microbial production of 2,3butanediol. Adv Appl Microbiol 32:89–161 23. Montville TJ, Hsu AH, Meyer ME (1987) High-efficiency conversion of pyruvate to acetoin by Lactobacillus plantarum during pH-controlled and fed-batch fermentations. Appl Environ Microbiol 53:1798–1802 24. Dikshit KL, Webster DA (1988) Cloning, characterization, and expression of the bacterial globin gene from Vitreoscilla in Escherichia coli. Gene 70:377–386 25. Messing J (1983) New M13 vectors for cloning. Methods Enzymol 101:20–78 26. Erenler SO, Gencer S, Geckil H, Stark BC, Webster DA (2004) Cloning and expression of the Vitreoscilla hemoglobin gene in Enterobacter aerogenes: effect on cell growth an oxygen uptake. Appl Biochem Microbiol 40(3):241–248
330 27. Miller JH (1972) Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, p 433 28. Westerfeld WW (1945) Colorimetric determination of blood acetoin. J Biol Chem 161:495–502 29. Keen AR, Walker NJ (1973) Separation of diacetyl, acetoin, and 2,3-butylene glycol by ion-exchange chromatography. Anal Biochem 52:475–481 30. Stormer FC (1972) 2,3-Butanediol biosynthetic system in Aerobacter aerogenes. Methods Enzymol 41:518–533 31. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ (1951) Protein measurement with the Folin phenol reagent. J Biol Chem 193:265–275 32. White D (1995) The physiology and biochemistry of prokaryotes. Oxford University Press, Oxford, pp 285–286 33. Bassit N, Boquien C, Picque D, Corrieu G (1993) Effect of initial oxygen concentration on diacetyl and acetoin production by Lactococcus lactis subsp. lactis biovar diacetylactis. Appl Environ Microbiol 59:1893–1897 34. Beronio PB, Tsao GT (1993) Optimization of 2,3-butanediol production by Klebsiella oxytoca through oxygen transfer rate control. Biotechnol Bioeng 42:1263–1269
35. de Mas C, Jansen NB, Tsao GT (1988) Production of optically active 2,3-butanediol by Bacillus polymyxa. Biotechnol Bioeng 31:366–377 36. Jansen NB, Flickenger MC, Tsao GT (1984) Production of 2,3butanediol from D-xylose by Klebsiella oxytoca ATCC 8724. Biotechnol Bioeng 26:362–369 37. Sablayrolles JM, Goma G (1984) Butanediol production by Aerobacter aerogenes NRRL B199: effects of initial substrate concentration and aeration agitation. Biotechnol Bioeng 26:148–155 38. Byun TG, Zeng AP, Deckwer WD (1994) Reactor comparison and scale-up for the microaerobic production of 2,3-butanediol by Enterobacter aerogenes at constant oxygen transfer rate. Bioproc Biosyst Eng 11:167–175 39. Zeng AP, Biebl H, Deckwer WD (1990) Effect of pH and acetic acid on growth and 2,3-butanediol production of Enterobacter aerogenes in continuous culture. Appl Microbiol Biotechnol 33:485–489 40. Mallonee DH, Speckman RA (1988) Development of a mutant strain of Bacillus polymyxa showing enhanced production of 2,3-butanediol. Appl Environ Microbiol 54:168–171
